UNIVERSITY OF CALGARY Speaking Their Language:

UNIVERSITY OF CALGARY
Speaking Their Language:
A Randomized Pilot Study of Synchronous Online Chat Groups for Young Adults with Cancer
by
Michael Lang
A THESIS
SUBMITTED TO THE FACULTY OF GRADUATE STUDIES
IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE
DEGREE OF MASTERS OF SCIENCE
GRADUATE PROGRAM IN COMMUNITY HEALTH SCIENCES
CALGARY, ALBERTA
JULY, 2015
© Michael Lang 2015
Abstract
Background: In Canada, over 7,000 adolescents and young adults or “AYA” (ages 18 to 39
years) are diagnosed with cancer annually. Few evidence-based psychosocial interventions exist
for this population.
Methods: We conducted a randomized pilot study to compare a new AYA-specific
psychotherapy model to a standard psycho-educational model, using CancerChatCanada (CCC)
synchronous online chat groups.
Results: 34 AYAs in Alberta participated in the study and all feasibility targets were met.
Participants rated the new AYA model more suitable and it appeared to improve important group
processes and psychosocial outcomes. A power calculation providing sample size estimation for
an RCT indicated that more than 90 participants would be needed to demonstrate differential
efficacy between groups.
Discussion: This randomized study design is feasible with some important modifications.
Conclusion: Online synchronous chat groups can be used to provide psychosocial care for
AYAs. Future RCTs could provide evidence of efficacy for age-specific psychotherapy models.
Words: 150
ii
Acknowledgements
I would like to acknowledge Dr. Janine Giese-Davis for the incredible mentorship and teaching
concerning all aspects of this thesis. Dr. Joseph Dort for “giving me a shot” and his perspective
and support during my time at the U of C. Dr. Marilynne Hebert for her involvement and
guidance concerning the qualitative aspects of this thesis and Dr. Joanne Stephen for cultivating
such a wonderful online chat platform and helpful research team at CancerChatCanada. Finally, I
would like to acknowledge all the incredible study participants who filled out many surveys to
help me better understand their chat group experience and ultimately, help improve psychosocial
care for other young adults with cancer.
iii
Dedication
To my wife, Bonnie. Thanks for feeding me when I forget to eat, and loving me when I don’t
deserve it.
iv
Table of Contents
Abstract ........................................................................................................................................... ii Dedication ...................................................................................................................................... iv Table of Contents............................................................................................................................ v List of Tables ................................................................................................................................. ix List of Figures and Illustrations ...................................................................................................... x List of Symbols, Abbreviations and Nomenclature....................................................................... xi Epigraph........................................................................................................................................ xii CHAPTER ONE: THESIS OVERVIEW.................................................................................................................1 1.1 Outline of Candidature.......................................................................................................... 1 1.2 Background........................................................................................................................... 1 1.3 Rationale ............................................................................................................................... 2 1.4 Primary Research Question and Study Aims........................................................................ 3 1.5 Study Design......................................................................................................................... 4 CHAPTER TWO: LITERATURE REVIEW .........................................................................................................6 2.1 Introduction........................................................................................................................... 6 2.2 The Field of Psychosocial Oncology .................................................................................... 6 2.2.1 Importance of Psychosocial Interventions for People with Cancer.............................. 7 2.2.2 Types of Psychosocial Interventions and their Purpose ............................................... 7 2.2.3 Technology-based Psychosocial Interventions in Cancer Care.................................... 9 2.3 Adolescent and Young Adult Psychosocial Oncology ....................................................... 10 2.3.1 Emerging Definitions of AYA ................................................................................... 11 2.3.2 Unique Aspects of AYA Psychosocial Care .............................................................. 11 2.3.3 Rationale for Developing Online Interventions Specifically for AYAs .................... 12 2.3.4 Rationale for Group Psychotherapy with AYAs ........................................................ 13 2.4 Therapeutic Models of Online Psychosocial Interventions for AYAs ............................... 14 2.4.1 Group Psychotherapy ................................................................................................. 14 2.4.2 Psycho-Educational Group Psychotherapy (CCC Standard Model) .......................... 15 2.4.3 Narrative-informed Emotion-focused Group Psychotherapy (Modified Model)....... 17 2.4.3.1 Awareness, Contextualization and Symbolization of Emotions....................... 18 2.4.3.2 Narrative Construction, Transformation of Story Outcomes and Identity
Reconstruction ................................................................................................... 19 2.4.3.3 Potential Differential Psychosocial Outcomes of NIEFP in AYAs.................. 19 2.5 Theoretical Evaluation of Online Group Psychotherapy Interventions (Study Aim #1).... 20 2.5.1 Evaluating Content Sensitivity ................................................................................... 20 2.5.2 Evaluating Changes in Group Processes .................................................................... 20 2.6 Outcome Evaluation of Online Psychosocial Interventions (Study Aim #2) ..................... 21 2.6.1 Distress, Depression and Anxiety in AYAs ............................................................... 21 2.6.2 Loneliness in AYAs ................................................................................................... 22 2.6.3 Trauma in AYAs ........................................................................................................ 22 2.6.4 Posttraumatic Growth in AYAs.................................................................................. 23 2.6.5 Coping Self-Efficacy in AYA’s ................................................................................. 23 2.7 Objectives of External Pilot Studies in Online RCT Interventions (Study Aim #3) .......... 24 2.7.1 Recruitment and Retention ......................................................................................... 24 v
2.7.2 Attendance.................................................................................................................. 25 2.7.3 System Usability......................................................................................................... 25 2.7.4 Participant Satisfaction............................................................................................... 25 2.7.5 Statistical Power and Appropriateness of Psychosocial Outcomes............................ 26 2.8 Connection to Study Aims.................................................................................................. 26 CHAPTER THREE: METHODOLOGY .............................................................................................................. 27 3.1 Overview............................................................................................................................. 27 3.2 Participant Recruitment, Eligibility Confirmation, and Randomization ............................ 29 3.2.1 Participant Recruitment and Enrolment Procedures................................................... 29 3.2.2 Eligibility Confirmation ............................................................................................. 30 3.2.3 Randomization............................................................................................................ 31 3.3 Interventions ....................................................................................................................... 31 3.3.1 Standard CCC Psycho-Educational Model (Workbook)............................................ 32 3.3.2 Modified CCC Emotion-Focused Model (Video)...................................................... 33 3.4 Study Integrity and Participant Safety ................................................................................ 34 3.5 Data Collection and Management....................................................................................... 35 3.6 Measures ............................................................................................................................. 36 3.6.1 Demographic and Medical Variables ......................................................................... 36 3.6.2 Study Aim #1 – “Proof of Concept” Measures .......................................................... 37 3.6.2.1 Topic-Specific Suitability ................................................................................. 37 3.6.2.2 Overall Suitability............................................................................................. 37 3.6.2.3 Group Cohesiveness ......................................................................................... 38 3.6.2.4 Group Experiences Inventory ........................................................................... 38 3.6.3 Study Aim #2 – “Appropriateness of Psychosocial Outcomes” Measures ................ 39 3.6.3.1 Hospital Anxiety and Depression Scale (HADS) ............................................. 39 3.6.3.2 UCLA Loneliness Scale (UCLA-LS) ............................................................... 39 3.6.3.3 Posttraumatic Stress Disorder Checklist – Civilian (PCL-C)........................... 40 3.6.3.4 Posttraumatic Growth Inventory (PTGI) .......................................................... 41 3.6.3.5 Cancer Behaviour Inventory – Brief (CBI-B) .................................................. 41 3.6.4 Study Aim #3 – Feasibility Measures ........................................................................ 42 3.6.4.1 Recruitment and Retention ............................................................................... 42 3.6.4.2 Attendance ........................................................................................................ 42 3.6.4.3 System Usability ............................................................................................... 42 3.6.4.4 Participant Satisfaction ..................................................................................... 43 3.7 Data Analyses ..................................................................................................................... 43 3.7.1 Demographic and Medical Variable Analysis............................................................ 44 3.7.2 Study Aim #1 – “Proof of Concept” Analysis............................................................ 44 3.7.3 Study Aim #2 – “Appropriateness of Psychosocial Outcomes” Analysis ................. 44 3.7.4 Study Aim #3 – Feasibility Analysis.......................................................................... 45 3.7.5 Reporting of Qualitative Comments........................................................................... 46 CHAPTER FOUR: RESULTS ................................................................................................................................ 47 4.1 Overview............................................................................................................................. 47 4.2 Demographic, Medical and Psychosocial Characteristics of Sample................................. 47 4.3 Study Aim #1 – “Proof of Concept”................................................................................... 52 4.3.1 Content Sensitivity ..................................................................................................... 52 vi
4.3.1.1 Topic-Specific Suitability ................................................................................. 52 4.3.1.2 Overall Suitability............................................................................................. 56 4.3.2 Group Processes ......................................................................................................... 58 4.3.2.1 Group Cohesiveness ......................................................................................... 58 4.3.2.2 Group Experiences Inventory ........................................................................... 59 4.4 Study Aim #2 – “Appropriateness of Psychosocial Outcomes”......................................... 61 4.4.1 Pre/Post Analysis of Initial Treatment and Control Groups....................................... 61 4.4.2 Longitudinal Analysis of Initial Treatment Groups ................................................... 65 4.5 Study Aim #3 – Feasibility of Study Design and Procedure .............................................. 67 4.5.1 Recruitment and Retention ......................................................................................... 67 4.5.2 Attendance.................................................................................................................. 67 4.5.3 System Usability......................................................................................................... 71 4.5.4 Overall Participant Satisfaction.................................................................................. 72 4.5.5 Statistical Power Calculation...................................................................................... 75 4.5.6 Matched Qualitative Responses and Psychosocial Outcome Data............................. 75 CHAPTER FIVE: DISCUSSION ........................................................................................................................... 78 5.1 Overview............................................................................................................................. 78 5.2 Study Aim #1 – “Proof of Concept”................................................................................... 78 5.3 Study Aim #2 – Appropriateness of Psychosocial Outcomes ............................................ 82 5.4 Study Aim #3 – Feasibility of Study Design and Procedure .............................................. 84 5.5 Limitations .......................................................................................................................... 88 5.6 Future Directions ................................................................................................................ 89 APPENDIX A: RECRUITMENT DOCUMENTS.............................................................................................. 91 A.1. Letters of Initial Contact ................................................................................................... 91 A.2. Follow up Letter................................................................................................................ 93 A.3. Recruitment Poster............................................................................................................ 94 A.4. CancerChatCanada Online Registration Form.................................................................. 95 A.5. Intake Phone Interview Script........................................................................................... 96 A.6. Participant Information and Consent Form....................................................................... 99 A.7. CancerChat Canada Enrolment Questionnaire ............................................................... 104 APPENDIX B: LINKS TO MODELS .................................................................................................................107 B.1. Workbook Links.............................................................................................................. 107 B.2. Video Links and Discussion Questions........................................................................... 107 Appendix C: FOLLOW-­‐UP INTERVIEW SCRIPTS....................................................................................109 C.1. Low Attendance Interview Script ................................................................................... 109 C.2. Follow-Up Interview Script ............................................................................................ 109 APPENDIX D: RESEARCH STUDY LOGIC MODEL...................................................................................111 APPENDIX E: ONLINE SURVEYS....................................................................................................................112 E.1. Weekly Post Chat Surveys .............................................................................................. 112 E.1.1. Standard Survey - Week 2,3,5,6,8 .......................................................................... 112 E.1.2. Extended Survey - Week 1,4,7,9 (Additional to Standard Survey) ........................ 114 vii
E.2. Final Post Chat Survey – Week 10.................................................................................. 118 E.3. Psychosocial Outcomes Questionnaire – Baseline, 3, 6, 9 Months ................................ 126 REFERENCES .........................................................................................................................................................136 viii
List of Tables
Table 1: Baseline Demographic, Medical and Psychosocial Characteristics of AYA
Participants (n=34) and Non-Respondents (n=383) ............................................................. 50 Table 2: AYA Endorsement of Important Reasons for Participation in Trial at Intake (n=34) ... 51 Table 3: During the Intervention, AYA Cancer Participants’ (n=34) Ratings of Weekly
Topic-Specific Relevance by PEP/NIEFP Group (Winter/Spring Chat-groups
Combined, Ranked by Median) ............................................................................................ 55 Table 4: Valued Group Experiences Inventory by PEP/NIEFP Intervention Groups
(Winter/Spring Chat-groups Combined, n=20) .................................................................... 60 Table 5: For Initial Randomization of AYA Cancer Participants (n=34), Estimated Pre/Post
Mean Change Scores for Psychosocial Outcomes................................................................ 63 Table 6: For Initial Randomization of AYA Cancer Participants, Estimated Mean Slope of
Change for PEP/NIEFP Intervention Groups (n=16) over 9 Months for Psychosocial
Outcomes .............................................................................................................................. 65 ix
List of Figures and Illustrations
Figure 1: Pilot Study Conceptual Model ........................................................................................ 5 Figure 2: Pilot Study Consort Diagram with Recruitment and Retention Targets ....................... 28 Figure 3: Simplified Version of CONSORT Diagram Highlighting Overall Demographics of
Initial Randomization............................................................................................................ 48 Figure 4: Simplified CONSORT Diagram Demonstrating Combination of Initial
Randomization (Winter) and Re-randomized (Spring) Chat Groups to Explore Aim #1 .... 52 Figure 5: Overall Change in Group Cohesion for the PEP/NIEFP Groups Over 10 Week
Intervention (Winter/Spring Chat-groups Combined, n=28)................................................ 58 Figure 6: Radar Chart of Valued Group Experiences for PEP/NIEFP Intervention Groups
(Winter/Spring Chat-groups Combined, n=20) .................................................................... 59 Figure 7: Simplified CONSORT Diagram Demonstrating Samples Used for LMM Analysis
of Psychosocial Outcome Data ............................................................................................. 61 Figure 8: Linear Mixed Models of Pre/Post Change in Psychosocial Outcomes between
Initial Randomization of PEP/NIEFP Intervention Groups and TAU Control (n=34)......... 64 Figure 9: Linear Mixed Models of Longitudinal Change in Psychosocial Outcomes for Initial
Randomization PEP/NIEFP Intervention Groups over 9 Months (n=16) ............................ 66 Figure 10: Trajectory of Weekly Attendance Percentage Overall and by PEP/NIEFP
Intervention Groups (Winter/Spring Chat-groups Combined) ............................................. 68 Figure 11: Categorical Attendance Levels of Published CCC data (n=351) and AYA Pilot
Study (n=34) by Number of Sessions Attended ................................................................... 69 Figure 12: Categorical Attendance Levels between PEP/NIEFP Intervention Groups by
Number of Sessions Attended (Winter/Spring Chat-groups Combined).............................. 70 x
List of Symbols, Abbreviations and Nomenclature
Symbol
Definition
AYA
Adolescent and young adult cancer patient or
survivor
RCT
Randomized control trial
TAU
Treatment as usual (control group)
CCC
CancerChatCanada
PEP
Psycho-educational group psychotherapy
NIEFP
Narrative-informed emotion-focused group
psychotherapy
HADS
Hospital Anxiety and Depression Scale
UCLA-LS
University of California Los Angeles – Loneliness
Scale
PCL-C
Posttraumatic Stress Check List – Civilian Version
PTGI
Posttraumatic Growth Index
CBI-B
Cancer Behaviour Inventory – Brief Version
QoL
Quality of Life
LMM
Linear Mixed Model (analysis)
ITT
Intent to Treat (analysis)
SET
Supportive-expressive group therapy
WW2H
Wrong Way to Hope: An Inspiring Story of
Young Adults and Cancer (video)
VGE
Valued group experience
xi
Epigraph
“Facts bring us to knowledge, but stories lead to wisdom.”
- Rachel Naomi Remen, M.D.
xii
Chapter One: Thesis Overview
1.1 Outline of Candidature
This thesis fulfills in part the requirements of a Master’s of Science in the Faculty of
Medicine, Department of Community Health Services at the University of Calgary. It examines
the feasibility of comparing two models of online group therapy for Adolescents and Young
Adults (AYA) diagnosed with cancer, setting the stage for a larger randomized clinical trial
(RCT). This chapter will contextualize the research and give rationale for piloting different
models of online psychosocial interventions for AYAs with cancer. It will include a brief
background and give an overview of the primary research question, study aims and design.
Chapter 2 will provide an in-depth review of the topics relevant to the study aims,
beginning with an overview of the field of Psychosocial Oncology and the burgeoning
subspecialty of AYA Oncology, including the theoretical foundations, structure and evaluation
of online group-therapy models. The chapter ends with an overview of the general principles and
measures used in randomized pilot studies to test the feasibility of online interventions.
Chapter 3 will outline the methods used in this study including 1) participant accrual,
eligibility, and randomization, 2) intervention design, 3) study integrity and safety, 4) data
collection and management, and 5) assessment and data analysis. Chapter 4 will present the
results, categorized according to the three primary aims of the study. Chapter 5 will discuss the
results of the two online group therapy models in detail, present the lessons learned about the
study design and feasibility, highlight the limitations of this study, and make recommendations
for future online psychosocial interventions for AYAs with cancer.
Throughout Chapter 4 and 5 qualitative comments, from open-ended survey questions
and phone interviews, will add richness to the observed quantitative data and further insight
useful to the aims of this study.
1.2 Background
In Canada, over 7,000 adolescents and young adults or “AYAs” (ages 15 to 39 years) (1)
are diagnosed with cancer annually with deaths in this population accounting for approximately
49,400 of potential years of life lost due to cancer each year (2). Cancer incidence in AYAs has
been increasing but improvements in survival rates have been slow in contrast to the outcomes of
1
both children and older adults (3). Of the 20 types of cancers in 15-39 year-olds evaluable for
survival-rate trends, only eight had significant improvements in their age-adjusted 5-year
survival rate from 1985-2011 (4). The factors most cited for slow survival-rate improvements are
delayed diagnosis (5), age-specific psychological factors (6), low enrolment in clinical trials (7),
unique tumour biology (8) and minimal communication between adult and paediatric oncologists
leading to sub-optimal treatment protocols (9).
The high prevalence of clinical distress in AYAs, caused by age-specific psychological
and practical problems, has been associated with poorer treatment and psychosocial outcomes (6,
10, 11). For a number of reasons very few age-appropriate interventions addressing these unique
AYA psychosocial issues have been trialed and evaluated (12, 13). First of all, the AYA
population is relatively small and geographically dispersed (14). Secondly, the high prevalence
of practical barriers in this age group, such as time restraints (due to work/family commitments)
and prohibitive program costs (due to lower incomes), keep them from participating in research
studies (15). Finally, there is a poor understanding of the unique needs of AYAs which may lead
to inappropriate program design (16, 17). AYAs prefer peer support and technology-based
interventions (video, phone, internet, social media) (6) over traditional psychosocial
interventions, such as written psycho-educational materials and in-person group therapy (6, 18),
and these preferences may limit their participation in psycho-oncology studies.
1.3 Rationale
To effectively meet the psychosocial needs of Canadian AYAs with cancer, the ideal
intervention would be easily accessible, reach a wide geographic population, combine peer and
professional support and be tailored to AYA patients’ needs (19). The established
CancerChatCanada program (cancerchatcanada.ca) meets the first three of these four criteria.
However, an AYA-specific modification of the CancerChatCanada (CCC) program might be
necessary because the current therapy model is designed for the primary user group, adults over
50 years of age (20).
The standard CCC therapeutic model is a hybrid of group psychotherapy and psychoeducational psychotherapy, a combination commonly used in psycho-oncology interventions
with the adult oncology population (21). However, AYAs have different needs and desires than
the standard oncology population (50+) (22). Specifically, AYAs consistently indicate that
2
connecting with other AYA cancer peers is more important than gaining new knowledge or
information about cancer (6, 23). Therefore, a model of group psychotherapy focusing on
building “connections” (emotion-focused), rather than “teaching” (psycho-educational), could
potentially improve the efficacy of the CCC synchronous online chat groups for AYAs.
Alternatively, offering AYA-only groups, using the standard psycho-educational group
psychotherapy model designed for older adults, might be equally efficacious.
A pilot study to determine the feasibility of a randomized study design to test this
question is a necessary first step to further exploration of online, AYA-specific psychosocial
interventions. RCTs of online interventions have been highlighted as a priority for the future of
supportive cancer care in Canada (24) and a pilot study will lay the necessary groundwork for
this to take place.
1.4 Primary Research Question and Study Aims
The primary research question is whether an narrative-informed emotion-focused group
psychotherapy model (NIEFP) in online synchronous chat groups with AYAs (18-39 years old)
is more effective than the standard psycho-educational psychotherapy (PEP) model currently
used with older adults with cancer (40+ years old). However, to begin to answer this research
question, this small randomized pilot study investigated three primary aims:
1. “Proof of Concept” - To test the theory that different models of online group
psychotherapy can impact a) content sensitivity (topic-specific and overall suitability)
and b) group processes (group cohesion, valued group experiences).
2. “Appropriateness of Outcomes” - To assess the appropriateness of specific
psychometrically validated psychosocial measures relevant to AYAs (distress,
loneliness, posttraumatic stress, coping self-efficacy and post-traumatic growth).
3. “Study Design Feasibility ” - To determine the feasibility of the study design and
procedures using pilot study data (recruitment and retention, attendance, usability,
satisfaction, statistical power calculation).
The goals of this study are to determine if this new narrative-informed emotion-focused group
psychotherapy model 1) provides a different experience for participants, 2) positively influences
outcomes, and 3) if this study design is promising for a larger RCT.
3
1.5 Study Design
This prospective, three-arm pilot study compared two models of online group
psychotherapy: the standard 10-week CancerChatCanada synchronous online group therapy
model using a standard psycho-educational workbook (PEP), and a modified online group
therapy model using video clips from a patient-created video, Wrong Way to Hope: An Inspiring
Story of Young Adults and Cancer (NIEFP). We randomized participants to the two treatment
arms and a Treatment as Usual (TAU) waitlist control group. At 3 months, we re-randomized the
TAU control group to receive one of the two interventions. We collected psychosocial outcomes
at four common time points (Baseline, 3, 6, 9 months) and “proof of concept” measures during
intervention participation.
The first 3-month comparison of the PEP/NIEFP groups to each other and to the TAU
control allows for estimation of effect sizes, with the TAU group accounting for secular trends in
outcomes over time. The re-randomization of the TAU control collects more data about
participants chat-group experiences for comparisons of “proof of concept” measures, but does
not provide a TAU control for longitudinal comparisons across the full 9 months. As the goal of
a pilot study is to understand thoroughly the intervention in question, data collection included
open-ended survey questions and follow-up interviews to provide additional insight regarding
both study aims.
Overall, this pilot study design enabled us to highlight potential pitfalls and successes
before the initiation of a larger randomized clinical trial. The conceptual model used to organize
this pilot study appears in Figure 1. The clear list of objectives added methodological rigour to
this pilot study so that, despite its limitations, it could make a meaningful contribution to the
growing body of literature exploring innovative online interventions in clinical and supportive
care for AYAs with cancer.
4
Figure 1: Pilot Study Conceptual Model
5
Chapter Two: Literature Review
2.1 Introduction
As described in Chapter One, the three primary purposes of this study are to provide pilot
data testing two different models of online group psychotherapy for AYAs in a larger RCT.
Specifically the study aims are to examine 1) the impact on content sensitivity and group
processes of the two online group psychotherapy models, 2) the appropriateness of using specific
psychometrically validated psychosocial measures with AYAs and 3) to determine the feasibility
of the study design used to comparing these two models. This chapter builds the case for
conducting this pilot study by providing a brief overview of the field of psychosocial oncology
(2.2) and the sub-specialty of AYA oncology (2.3), exploring the theoretical foundations (2.4),
structure (2.5) and evaluation (2.6, 2.7) of online group therapy models and reviewing the
general principles and measures used in pilot studies testing the feasibility of online interventions
(2.8).
2.2 The Field of Psychosocial Oncology
A recent summary of the history and development of Psychosocial Oncology describes it
as a relatively new field (25). In the first half of 20th century there were no successful treatments
for cancer and therefore it was something to be feared and hidden away. Only in the second half
of the 20th century, when patient advocacy groups worked to de-stigmatize the disease and
successful treatments led to longer survivorship, was there room for a discussion about the
psychosocial sequelae of cancer diagnosis and treatment (25). The first studies on psychological
reactions to cancer emerged in the 1950s. Throughout the next 30 years, the organization of
professional societies, such as the International Psycho-Oncology Society (1984), and the
establishment of Psycho-oncology specific journals, helped define the sub-specialty and began a
shift towards evidence-based care models. In the 1990s many validated scales emerged to
measure the subjective symptoms of psychosocial distress, allowing for the creation of clinical
practice guidelines (26-28). The 2008 Institute of Medicine Report called Care for the Whole
Patient: Meeting Psychosocial Health Needs (29), found strong evidence for numerous types of
psychosocial interventions. This provided impetus for most developed countries to require their
cancer centers to offer their patients appropriate psychosocial care in order to meet accreditation
standards (25).
6
While the discipline of Psychosocial Oncology is relatively new, its ethos is ancient, and
in many ways it has been a powerful driver in the renewed emphasis on “whole-person care” or
“person-centered care”. To “capture the essence” of psychosocial oncology one researcher (25)
quoted an 87-year-old article from JAMA:
The good physician knows his patient through and through and his
knowledge is bought dearly. Time, sympathy and understanding
must be lavishly dispensed. One of the essential qualities of the
clinician is interest in humanity, for the secret of the care of the
patient is in caring for the patient (30).
Psychosocial Oncology has brought the biopsychosocial-spiritual model of care back into
medical care and significantly shifted cancer care away from the strictly biomedical focus of the
20th century. Psychosocial Distress is now recognized as the 6th vital sign in many medical
disciplines due to the pioneering work of psychosocial oncology (10). This significant
achievement helps patients to identify their needs and receive appropriate referrals. This section
provides an overview of why patients need psychosocial interventions in oncology (2.2.1), what
types of psychosocial interventions are commonly used (2.2.2) and the emerging role of
technology-based interventions in psycho-oncology (2.2.3).
2.2.1 Importance of Psychosocial Interventions for People with Cancer
A diagnosis of cancer causes long-term, mental, and spiritual challenges in addition to the
physical challenges of treatment (29). Clinical psychosocial interventions are central components
of patient-centered care within oncology and routine integration of psychosocial with medical
care--from diagnosis through survivorship--is considered a standard of care (31, 32).
Psychosocial interventions with cancer patients may reduce the risk of cancer reoccurrence (33),
decrease the incidence of debilitating depression and anxiety (34) and some may improve
survival (35-37). There is emerging evidence of biological pathways through which certain
psychosocial interventions may contribute to improved medical outcomes; in particular, they
may positively influence the HPA axis and immune function in cancer patients and survivors
(38-40). Overall, there is evidence that psychological interventions in cancer patients improve
psychosocial, behavioural, and physiological outcomes (41).
2.2.2 Types of Psychosocial Interventions and their Purpose
Psychosocial oncology interventions can take many forms and can combine multiple
techniques, curriculums, or models, but in general they are grouped into four broad categories:
7
1) education, 2) individual (or couple’s and family) psychotherapy, 3) behavioural training, and
4) group-therapy interventions (42). Almost all interventions include some aspect of psychoeducation. The goal of psycho-educational interventions is to reduce feelings of helplessness and
inadequacy due to lack of knowledge as well as build a sense of mastery, control, and selfefficacy that helps people adapt better to their cancer experience (42, 43). Seminars providing
information about the psychosocial sequele of cancer (i.e. anxiety, fear, depression) and different
ways manage them (i.e. healthy lifestyle, social support, ways of thinking) are examples of this
basic intervention. The purpose of individual or couple’s psychotherapy is to “facilitate growth
of self-knowledge, emotional acceptance and personal resources” (42). Psychotherapy can
involve interactions with a psychosocial oncology professional in a traditional face-to-face
setting or remotely, such as in an online chat group.
Behavioural training aims to reduce psychological stress and/or physical complications
due to treatments (42). Examples of this type of intervention include stress management and
relaxation training, such as Mindfulness-based Stress Reduction (44), cognitive-behavioural
therapy (45), communication-skills training (46) and guided imagery (47). Often, psychoeducation and behavioural training occur in a group setting. Other group-based interventions,
such as supportive-expressive group therapy, expressive writing, journaling and other arts-based
therapies, are designed to promote psychological adjustment by encouraging emotional
expression in a supportive setting (42). Use of behavioural training and group-based
interventions is higher in those with cancer (23.7%) than those with other chronic health
conditions (14.5%) as most cancer centers offer evidenced-based Supportive-Expressive Therapy
and/or Cognitive Behavioural Therapy group programs (48). These face-to-face group
interventions decrease distress, depression and anxiety (49) and increase overall well being,
leading to better health outcomes (50, 51).
The types of measures used to assess the outcomes of these psychosocial interventions
are as diverse as the interventions themselves. Common indicators include generalized distress,
mood and anxiety, anger, coping, self-efficacy, quality-of-life, post-traumatic growth or stress,
fatigue, insomnia, pain, social functioning, sexual functioning, loneliness, existential distress,
vitality, emotional wellbeing and cancer-related knowledge.
Historically in psycho-oncology research, there has been a greater focus on quantitative
strategies (52); however, qualitative methods have been recognized as useful to discover
8
important needs and constructs, and to generate hypotheses. In some cases qualitative data is
helpful to understand the full impact of psychosocial interventions as some quantitative measures
can be constrained to a single construct that may or may not be impacted by the intervention
(52). A mixed-method approach, with both quantitative and qualitative data, especially in the
early stages of intervention development, can provide a more holistic picture of the impact and
provide direction for future studies (53).
Not everyone benefits in the same way from psychosocial interventions. Standard
moderators, such as gender, socioeconomic status, age, marital status, psychological
characteristics and medical variables, such as type or duration of treatment, have been found to
have an impact on psychosocial intervention efficacy (54, 55). A systematic review of potential
moderators of psychosocial treatment outcome found that personality traits, quality-of-life, social
environment and self-efficacy are all potential moderators depending on the intervention (43).
Patients with poorer quality of life, interpersonal relationships, and sense of control benefited
more from psychosocial interventions. Additionally, patients with low levels of optimism and
neuroticism, high emotional expressiveness and interpersonal sensitivity showed greater benefits
from certain interventions (43). Others have found that initial distress level strongly predicts who
benefits from many standard psychosocial oncology interventions (56-58). The growing
knowledgebase about the differential impact of psychosocial interventions provides a greater
ability to personalize a patient’s psychosocial care by targeting interventions based on specific
characteristics, like age.
2.2.3 Technology-based Psychosocial Interventions in Cancer Care
Equitable access is an issue in every aspect of health care including psychosocial
oncology. However, with the invention of the Internet and its accompanying technologies like
Smartphone’s and Tablets, new ways to deliver care and education to underserved populations
are materializing in rapid succession (19, 59). Patients can receive many psychosocial
interventions remotely that were once limited to large tertiary centers. Some examples of this
would be online Mindfulness-Based Cancer Recovery (60), peer and professionally facilitated
online chat programs (61, 62), individual counselling via webcam or “telehealth” (63) and many
online psycho-educational programs using websites and social-media tools (24, 64, 65). One
common criticism of moving towards these online care-delivery platforms is that the primary
demographic of cancer patients is older adults who do not use these mediums. However, as
9
generations age into the Internet era, these methods are becoming more acceptable to older
cancer patients (66). In particular, the AYA cancer population has shown a preference for these
types of technology-based interventions (13).
2.3 Adolescent and Young Adult Psychosocial Oncology
Recognition of Adolescents and Young Adults (AYA) as a distinct sub-population in
oncology originated from a biological, not a psychosocial, foundation. The first milestone article
used the National Cancer Institutes (NCI) Surveillance, Epidemiology and End Result (SEER)
data (67). It demonstrated that between the ages 15 and 19 there was a significantly lower
survival-rate improvement from 1975-1995 as compared to younger children (67). Expanding on
these findings, in 2006 the National Cancer Institute released a groundbreaking analysis using
the NCI SEER database that clearly identified the “AYA Gap” - a significantly lower
improvement in survival rates from age 15-39 as compared with the younger and older age
ranges (68). This report has been the impetus behind much of the AYA research that exists today
and has led to important discoveries of unique tumour biology as well as increased clinical trial
accrual and AYA-specific treatment protocols.
In the psychosocial domain, AYA self-reported unmet needs have been identified and
compared with clinicians perceptions, highlighting distinct discrepancies between the practical
and psychosocial supports AYAs want and what clinician’s think they need (17). This
awareness, among other things, led to a strong patient/provider alliance, with AYA patients and
community-based organizations working very closely with cancer-care professionals in the
development and trialing of new psychosocial interventions.
Recently, a heightened awareness of AYA uniqueness has been instrumental in
motivating North American cancer centres and others all over the world to create specific
programs and services for AYAs. The National Comprehensive Cancer Network (NCCN)
published AYA clinical-care guidelines in 2012 (69, 70) and in Canada, the Canadian
Partnership Against Cancer (CPAC) has been a driving force in building awareness of the gap in
care for AYAs with the publication of clinical-care guidelines (71). One of the most contentious
problems still facing the young AYA oncology movement is defining an age range that is
biomedically, operationally, and psychosocially meaningful. This section provides an overview
of the emerging definitions of AYA (2.3.1), the unique psychosocial challenges of cancer in the
10
AYA age range (2.3.2), and the rational for developing AYA-specific online psychosocial
interventions (2.3.3).
2.3.1 Emerging Definitions of AYA
There has been much debate in the literature about the age range that ought to comprise
AYA, with the majority of studies determining their own cut-points without considering the
resulting fragmentation of this relatively small and specific literature (72). As highlighted in the
CPAC AYA Task Force Report in 2011 (71), paediatric clinicians/researchers commonly use 1529 as their age group, adult oncologists commonly use 18-29 or 18-35, and adult psychosocial
oncologists, social workers and community-based organizations most often use 18-39 as their
formal definition. It has been argued that the problems of defining this group are due in part due
to the arbitrary nature of using age as a proxy measure to determine the more specific underlying
construct of developmental life stage (72).
AYAs recruited to this study were 18-39 years old for both operational and theoretical
reasons. Specifically, 1) including participants 15-17 would involve many logistical challenges
with consent and ethics procedures as those participants are not above the legal age of majority,
2) using age 39 as the upper age limit allows for the widest potential participant group to reach
recruitment targets and 3) the National Cancer Institutes Special report on AYA Cancers in 2006
(1) recommended age 39 as the upper age limit to this demographic. Thus, this study conformed
to the usual definition used by adult psychosocial oncology professionals, rather than the 15-39
age range recommended by NCI (69).
2.3.2 Unique Aspects of AYA Psychosocial Care
AYAs with cancer face unique psychosocial challenges distinct from those of children or
older adults (73). Transitioning to adulthood consists of important developmental tasks including
establishing identity, developing a positive body image and sexual identity, separating from
parents, increasing involvement with peers and dating, and beginning to make decisions about
careers, higher education, and/or family (74). Consequently, cancer-related issues such as
premature confrontation with mortality, changes in physical appearance, increased dependence
on parents, treatment related disruptions in social life and school/employment, loss of fertility,
and health-related concerns about the future are particularly distressing for adolescents and
young adults (75, 76). These unique issues lead to high levels of psychosocial and practical
needs unmet by current supportive cancer care practices that are often biased toward older
11
populations (13, 77). Developing interventions to reduce distress in AYAs is a high priority
because of the association between high distress and poor treatment outcomes (41). In the AYA
context, where very long-term survivorship is possible (i.e. 50+ years), physical, emotional, and
social quality of life need to be protected so that AYAs can be productive members of society
over the long-term (7, 78). Appropriate psychosocial care for AYAs needs to support key
developmental life tasks to ensure these long-term goals are realized (22).
2.3.3 Rationale for Developing Online Interventions Specifically for AYAs
A national Canadian survey (unpublished data) revealed that few paediatric and adult
centers have psychosocial support staff experienced with the AYA population and most do not
have programs or services dedicated to AYAs (78). Virtual interventions can span the paediatric
and adult health-care systems, reach across a wide geographic area and be implemented by
professionals with expertise in AYA psychosocial care. The small numbers of AYA
psychosocial intervention studies using the Internet have generally had high levels of
participation and satisfaction and have been associated with increased self-efficacy,
improvement in symptom stress, coping and quality of life (79, 80). Although face-to-face and
online support groups with AYAs have different therapeutic strengths and weaknesses (81),
using internet-based interventions to involve AYAs in their own care through means that are
familiar to them is not only empowering, but could become a cost-effective and inclusive model
for psychosocial support (82-84).
There are many different methods to utilize the internet in psychosocial interventions.
Chat-groups are commonly used and can be generally categorized as synchronous (facilitator and
participants all chat together at the same time) or asynchronous (discussion boards where people
leave comments and respond to each other at any time). Synchronous chat groups may be able to
provide the benefit of face-to-face group interactions and online interactions more efficiently
than asynchronous groups (81, 85). One AYA psycho-oncology study suggested that face-to-face
groups may facilitate more storytelling, positive feedback and sense-making while asynchronous
online discussion boards may provide more space to express negative feelings, engage in
information seeking and discuss sensitive topics (81). The combination of active discussion and
anonymity in synchronous online chat interventions could be an optimal method to engage with
AYA cancer patients and survivors.
12
Researchers in other disease groups have piloted online interventions with AYAs
producing positive outcomes. Despite a limited sample size (n=22), one study of AYAs with
spina bifida and cerebral palsy found trends toward participants reporting more contact with
peers with disabilities (F(2,19) = 9.91, P < .001), decreased loneliness (F(2, 20) = 0.69, not sig.)
and increased social acceptance and confidence (F(2, 20) = 2.31, not sig.) (86). Another larger
study (n=244) of AYAs with depression found that an online intervention group showed
significantly greater improvement in depressive symptoms at 3 months than the control group (t
187 = 6.62, P < .001) with a large between-group effect size of d = 0.94 (95% CI 0.64–1.23)
(87). The online intervention group also showed greater improvement in anxiety (t(187) = 3.80,
P < .001, d = 0.49 95% CI 0.24–0.75) and mastery (t(187) = 3.36, P = .001, d = 0.44 95% CI
0.19–0.70) (87).
To add further weight to the importance of developing online support, one exploratory
study compared the depression scores of people enrolling in a traditional face-to-face support
group verses an online support group (88). They found that 92% of those enrolled in the online
group were clinically depressed as compared to 0% of those enrolled in the face-to-face group
(88). This demonstrates that depressed people might be more likely to withdraw socially, but
look for support online. If this is true more generally, online support should be highlighted as a
priority area for clinical psychosocial interventions with AYAs (16).
Overall, online modalities have great potential to begin to meet the current unmet needs
for AYA psychosocial care in oncology (89). There is a need for high-quality randomized
clinical trials of the efficacy of internet-based group interventions with AYA cancer patients to
inform consumers, practitioners, policy makers, and other users as well as to improve
psychosocial outcomes for this population (18, 90). A pilot study is a first step toward creating
such a trial.
2.3.4 Rationale for Group Psychotherapy with AYAs
Group psychotherapy with other AYAs may be an ideal psychosocial intervention model
as AYAs consistently endorse peer support from other young adults with cancer as one of their
highest unmet needs (6, 91). In one study, 100% of participants (n=26) endorsed opportunities to
communicate with other survivors as a top-5 need with only 50% endorsing support from friends
and family as a top-5 need (17). In a larger study (n=523) 43% of the participants reported a
desire to meet with peer survivors (91). A study of AYAs with advanced cancer (n=71) found
13
that providing opportunities to talk with their advanced-cancer peers had the most global impact
on their psychological health above material aid (92). Therefore, the ideal psychosocial
intervention for AYAs needs to involve direct peer-to-peer communication, preferably in the safe
environment that professional facilitation provides (22). Group Psychotherapy, as a therapeutic
model of psychosocial care, can provide the peer-to-peer interaction and psychosocial support
that AYAs desperately need.
2.4 Therapeutic Models of Online Psychosocial Interventions for AYAs
As noted earlier, AYAs are low users of psycho-oncology services, despite high unmet
needs, and therefore new, more suitable methods of supporting AYAs may be required (13, 77,
93). A systematic review of psychological interventions for AYAs with chronic illness found that
diabetes programs demonstrated a higher rate of success (78% with significant results) relative to
cancer interventions (71%) (21), and had more RCT designed studies. This demonstrates that
AYA psycho-oncology is behind in both our practical and theoretical conceptualization of
psychosocial interventions. A primary goal of this study is to expand this knowledge by
providing pilot results useful for the design of a larger RCT testing online group therapy models;
specifically exploring if a narrative-informed emotion-focused psychotherapy (NIEFP) model
(2.4.3) with age-appropriate content, will be more effective for AYAs than the standard CCC
psycho-educational psychotherapy (PEP) model (2.4.2). This section also provides an overview
of the overarching concepts of group psychotherapy research (2.4.1).
2.4.1 Group Psychotherapy
Group psychotherapy has a set of recognizable factors that make it distinct from other
“gatherings” of people, namely, 1) appropriately referred group members, 2) skilled leaders and
3) defined goals (94). The roots of psychotherapy go back to the beginning of the 1900s (94) but
Yalom first described the therapeutic “factors” of Group Psychotherapy in 1970 (95), with
further definition by Lieberman et al. in 1973 (96). They named the hypothesized therapeutic
mechanisms that may make group psychotherapy effective; these include the instillation of hope,
the universality of experiences, a feeling of belonging, interpersonal learning, self-understanding
and others (12 factors total, not all listed). All these factors are important aspects of supporting
the psychological health of cancer patients and survivors and therefore group psychotherapy is
commonly used in psycho-oncology (48). In cancer patients and survivors, group psychotherapy
14
has demonstrated improvements in distress (mood disturbance, depression, anxiety, and trauma
symptoms) (41, 97-100), pain (101, 102), social support (41), emotion regulation (103), and
immune function (41, 104).
Researchers have explored many theories and models of group psychotherapy over the
past 100 years with continual refinement of which group therapy model works best for specific
patients (94). In practice, many psychotherapy practitioners rarely ascribe to a single model or
theory, but draw from many different models and theories to deliver the best intervention for
their specific context and population (94). For example, clinicians use PEP models with adult
cancer patients to transmit knowledge and build a sense of mastery and control while also
improving their sense of belonging and self-understanding facilitated by the group setting (42,
43).
Specific combinations of group psychotherapy models may have shown specific
outcomes. A Cochrane Review identified emotion-focused supportive-expressive group
psychotherapy model as the standard-of-care for metastatic breast cancer patients (105).
Similarly, the cognitive-behavioural stress management model has been found to best improve
immune and cortisol function (106). Supportive-expressive group therapy, as one of the most
researched and disseminated group models for cancer patients, shares with other emotionfocused therapies (107) an emphasis on emotional expression and regulation (108) and
storytelling (109) within the context of existential processing (110). Although both PEP and
NIEFP models have been tested individually and in groups with different patients, in AYA
psycho-oncology there has yet to be a study comparing these different models in a three-arm
randomized trial (21). It is unclear which group psychotherapy model might help AYAs adapt
more effectively to their cancer experience.
2.4.2 Psycho-Educational Group Psychotherapy (CCC Standard Model)
A standard model used in many cancer psychotherapy groups combines the methods and
principles of group psychotherapy with psycho-education (PEP); teaching important concepts in
the context of group discussions (42). One meta-analysis found that a PEP model was more
effective in primary breast cancer patients than supportive-expressive group therapy, however,
the age range of the samples in this study did not include patients younger then 40 years old (55).
A systematic review of psychological interventions for AYAs living with chronic illness found
PEP to be one of the most common forms of group psychotherapy used with AYAs (21).
15
Psycho-educational groups can discuss a large variety of topics and are similar to the
combination of an academic class and a counselling/therapy group (111). The types of psychoeducational groups can generally be categorized as 1) personal development, 2) support and
therapy-related, and 3) life transitions. Support and therapy-related groups (of interest in this
study) can be further delineated by medical characteristics (cancer, diabetes, cardiac),
psychological/emotional characteristics (anxiety, hyper-activity disorder, posttraumatic stress
disorder) or recovery characteristics (addiction, eating disorders, work related issues) (111). The
content of a support and therapy-related psycho-education group is often similar regardless of the
patient characteristics, and includes information about many pertinent issues such as
communication skills, self-esteem, relationship issues, and managing difficult emotions.
Psycho-educational groups are generally “time-limited” psychotherapy interventions
lasting on average 6-10 sessions (111); essentially until all the content is covered. With AYAs in
particular, “successful” PEP interventions (i.e. improved significantly on one psychosocial
outcome measures) had more intervention sessions on average than did studies with no
significant improvement (number of sessions M=8.4, SD=6.3 compared with M=5.3, SD=2.9;
p=0.022) (111). The association between increased length of the intervention and efficacy could
either indicate that more information leads to better outcomes or that increased time for the
development of therapeutic group relationships (95) leads to better outcomes.
The primary goal of PEP groups is to improve psychosocial adaptation to cancer through
the acquisition of knowledge or skills, emphasizing learning over increased emotional
engagement. This emphasis on the educational content requires that health care professionals or
scientists with content specific expertise develop the materials. However, scientists or
professionals may focus primarily on the quality of the educational content and not on the
method used to communicate that content to their target audience. This could present challenges
when working with AYAs.
Developmentally, AYAs are making important shifts in their identities and expectations
(112) leaving many professionals and organizations working with this population struggling to
stay relevant. Traditional psychosocial cancer care initiatives do not attract this population to the
same extent as older demographics, often because AYAs believe that they share little in common
with the majority of cancer patients who are much older (93). Therefore, a PEP curriculum
16
developed by scientists and clinicians for older adults may not be an effective way to
communicate important educational information to AYAs.
The standard PEP model in this pilot study used a workbook developed with adult cancer
survivors in mind (NuCare) (113), however, no one has previously examined outcomes for
delivery of this content within an AYA peer psychotherapy group. This pilot study provides an
opportunity to evaluate the potential impact of using this standard CCC PEP methodology with
an AYA group.
2.4.3 Narrative-informed Emotion-focused Group Psychotherapy (Modified Model)
The underlying premise of emotion-focused psychotherapy is that emotion is
“foundational in the construction of the self and is a key determinant of self-organization” (114,
115). Emotion-focused research shows that becoming an observer of personal emotions allows
people to “have the emotion”, rather than it “having them” (96). This is an essential agentic
position for adaptation to an environment, promoting wellbeing and creating meaning (114, 115).
Emotion-focused psychotherapy’s primary therapeutic goals are “to increase emotional
intelligence and to transform maladaptive emotions into adaptive emotional responses…
resulting in new story outcomes” (115).
Emotion-focused psychotherapy, just like PEP, can take many forms. As mentioned
previously, one well-researched emotion-focused group psychotherapy model for cancer patients
is called supportive-expressive group therapy (SET) (116). Many of the theoretical
underpinnings of SET, and some of the research on therapeutic mechanisms (108, 109), align
well with emotion-focused therapy theory (117). As AYAs are not drawn to traditional face-toface SET groups (13, 77, 93), there may be different types of emotion-focused group
psychotherapy that would work better with this demographic.
Lynne Angus and Leslie Greenberg have researched how personal narratives and emotion
work together in psychotherapy (115). In their book, Working with Narrative and Emotion
Processes in Emotion-focused Therapy, they identify a specific type of emotion-focused
psychotherapy called narrative-informed emotion-focused therapy (NIEFP) (115). Though their
model may also benefit cancer patients, none of their trials specifically researches its use in this
population or in group therapy for AYA cancer survivors. In this pilot study aspects of Angus
and Greenberg’s work were used to develop an AYA-specific NIEFP model.
17
A narrative framework allows critical life events to be described, “re-experienced”
emotionally, and “re-storied” so that psychosocial trauma can heal. This process allows
participants to make sense of, or ascribe meaning to, their traumatic experience and ultimately
move forward in their lives (115). This model borrows from standard-of-care trauma models that
focus heavily on the emotional expression of vulnerable emotions as people re-live trauma (118).
According to Angus and Greenberg, there are five overlapping NIEFP processes that promote
change: 1) awareness and contextualization of emotions, 2) symbolizing emotions, 3) narrative
construction of emotional experience, 4) transformation of emotion and story outcomes, and 5)
identity reconstruction (115). Other researchers have posited that the emotional expression itself
is an important aspect of how emotion-focused psychotherapy improves outcomes (119, 120).
2.4.3.1 Awareness, Contextualization and Symbolization of Emotions
One of the essential activities of the AYA developmental life stage is strengthening
emotional intelligence and articulation (22, 112). For many AYAs differentiating and naming
their emotions can be a difficult task. In NIEFP, “helping participants to narrate and then to
subjectively enter and situate their most emotionally vulnerable and painful experience” is a
central focus (115). A video sharing AYA cancers stories may help facilitate the process of
participants “entering” their own story. It may also help participants “name” their emotions by
allowing them to reference directly the stories from the video in weekly chat-group discussions.
Video is a powerful and suitable medium for engaging with AYAs as Canadians 20-34
years old watched on average 450 online videos in 2011 (121). For this reasons, video may be a
more suitable modality for AYAs than the typical clinician-created written resources such as
brochures and workbooks. Prior to entering upon this pilot study, the master’s student created a
video called Wrong Way to Hope: An Inspiring Story of Young Adults and Cancer (WW2H). He
screened this video in most cancer centres across Canada (34 screenings), and at multiple
conferences across North America. Many AYA cancer patients and professionals viewed this
video and provided feedback. These viewer surveys indicated that all the themes addressed in the
video were highly relevant to AYAs and 95.6% indicated that they would recommend that other
AYAs watch the video (122). In addition to video being easy to watch and absorb, the agerelevant themes in WW2H may help AYAs identify and articulate the deep primary emotions
within their stories; a crucial process for the creation of new meaning and may be important to
improving their social support (109). 18
2.4.3.2 Narrative Construction, Transformation of Story Outcomes and Identity Reconstruction
Cancer is incredibly disruptive to the idealized narrative of young adulthood. From an
AYA’s perspective, cancer is something that happens to older people or young children, and for
this reason a cancer diagnosis at this stage of life can be extremely disorienting (22, 112). Reconstructing their narratives into a coherent story so that they can move forward in life after
cancer can be incredibly difficult.
NIEFP encourages the organization and incorporation of traumatic experiences into the
unfolding life story (115). Using the narratives of other AYAs can help facilitate the participants’
construction of their own stories, which can then be told to others and reflected on for further
understanding and personal-meaning construction (115). Using a video with the stories of other
AYAs could promote the assimilation of cancer into their ongoing self-narratives as well as act
as a guide for them to share their stories with their friends and family. Ultimately, it could help
them make sense of their experience and articulate it to others in a meaningful way.
Narrative construction and identity reconstruction are closely linked (115). Primary
processes of the AYA developmental life stage is separating from their parents, asserting their
independence and creating their own identities (22, 112). Therefore, cancer in this phase of life
can potentially have a large negative or positive impact on identity formation. Stories from this
AYA-created video may provide other AYAs with a better opportunity to “enter” their stories,
“name” their emotions, “assimilate” the cancer experience into their life narrative, and provide a
“guide” for them as they reconstruct their identity after cancer.
2.4.3.3 Potential Differential Psychosocial Outcomes of NIEFP in AYAs
Personal narratives are capable of eliciting stronger emotion while still communicating
important health information (123). In one study of African American women with breast cancer,
those who watched a narrative video were more likely to report feeling attentive, inspired, proud
and less likely to feel upset (all P < 0.05), while still absorbing important cancer related
information (123). A NIEFP model using a patient-created video could increase emotional
expression which in turn could lead to more group cohesion and engagement (109); helping
group members to feel more connected, understood, and supported by their peers. During a
developmental life stage where peers are of upmost importance, AYAs may benefit more from
sharing emotions with their peer group during group psychotherapy, than from following a
workbook to gain more knowledge about their cancer and techniques to manage psychosocial
19
challenges. It is apparent that NIEFP is very developmentally appropriate for AYAs with cancer
in both its therapeutic goals and methods. This pilot study explores further if this group
psychotherapy model is able to impact group processes and psychosocial outcomes.
2.5 Theoretical Evaluation of Online Group Psychotherapy Interventions (Study Aim #1)
Group psychotherapy research examines both outcomes and processes to help clarify
how, why, and when interventions work (94). Therefore, studies designed to explore new models
of group psychotherapy need to determine not only if these models produce different outcomes,
but also examine specific group processes, which may mediate outcomes. Despite the importance
of understanding therapeutic processes that contribute to the efficacy of psychosocial
interventions for cancer patients, few studies have explored processes and instead focus
primarily on outcomes (124). In this study, the exploration of group-process differences between
the group psychotherapy models compared (PEP/NIEFP) has been termed “Proof of Concept”.
This section reviews the methods that can be used to evaluate content sensitivity (2.6.1) and
changes within group processes (2.6.2).
2.5.1 Evaluating Content Sensitivity
The suitability and sensitivity of intervention content is an important consideration
because it can have a large impact on participant outcomes (125). In order for content to be
considered “sensitive” it must address topics that mirror the participant’s own experiences,
communicate in an engaging way, enhance personal reflection and, in the online chat group
setting, help stimulate group discussion. As previously discussed, AYAs face very different
challenges when diagnosed with cancer than both older and younger age groups, and therefore
content designed with these specific challenges in mind, may be rated as more suitable/relevant/
sensitive by participants. Unfortunately, there are no universally applicable measures of content
sensitivity easily identified in the literature and the development of measures specific to this pilot
study was necessary.
2.5.2 Evaluating Changes in Group Processes
Group processes are core elements of group psychotherapy that contribute to intervention
effectiveness. Broadly defined these are group purpose, group composition, group cohesiveness,
group development, and group communication (126). There are measures of each of these
20
processes, however Group Cohesion in particular has been highlighted as one of the most
important indicators of successful group psychotherapy (95). Group cohesion is defined as the
therapeutic relationship within group psychotherapy emerging from the aggregate of memberleader, member-member, and member-group relationships (127). Group cohesiveness has been
studied intensively since the 1950s (128, 129) and is important because a “feeling of belonging”
encourages greater openness and emotional expression within the group (128, 130). In many
studies, group cohesion and engagement are the only process measures that reliably predicted
psychosocial outcomes (127, 131).
Another way to characterize the processes involved in a psychotherapy group is through
identification of Valued Group Experiences (VGE). VGEs are a set of variables identified by
group therapy participants as important aspects of their support-group experience (132). In the
cancer context, these include the following: expressing true feelings, discussing sexual concerns,
developing a new attitude, establishing supportive contact, and accessing information and advice
(132, 133). Cancer patients may value different aspects of their support-group experiences
depending on variation in group model (132). Increasing emotional expression and encouraging
the open and honest discussion of personal cancer experiences ought to enhance the “feeling of
belonging” (group cohesion) and positively influence participants’ VGEs. These two measures
will allow for a comparison of group processes between PEP/NIEFP intervention groups.
2.6 Outcome Evaluation of Online Psychosocial Interventions (Study Aim #2)
There are standard psychosocial outcomes, such as distress and trauma, which are
consistently evaluated by psycho-oncology interventions. However, there are many
psychometrically validated tools used to measure these constructs, each with their own strengths
and weaknesses. This can make choosing the appropriate tool difficult. In addition, there are
psychosocial outcomes that are of interest specifically to AYAs because of the unique
psychosocial challenges they face. This section reviews the psychosocial outcomes of interest in
this study including distress (2.7.1), loneliness (2.7.2), trauma (2.7.3), post-traumatic growth
(2.7.4) and coping self-efficacy (2.7.5).
2.6.1 Distress, Depression and Anxiety in AYAs
The constructs of distress, depression and anxiety are closely connected. The National
Comprehensive Cancer Network has defined distress as a multi-factorial unpleasant emotional
21
experience that “extends along a continuum, ranging from common normal feelings of
vulnerability, sadness and fears to problems that can become disabling, such as depression,
anxiety, panic, social isolation and existential and spiritual crisis” (26). Therefore, depression
and anxiety are considered clinically relevant manifestations of generalized emotional distress
that occur at the more severe end of the distress continuum (134). High levels of distress are very
common among cancer patients, with estimates ranging from 20% (135) to 43% (136), and it has
been correlated with worsening treatment and psychosocial outcomes (137). Younger patients
and those who are unmarried often report greater distress (138), and “Younger age” is
consistently cited in the literature as a significant risk factor for distress, depression and anxiety
(72). This intervention has the potential to decrease distress, depression and anxiety levels, and
therefore these important outcomes will be part of the assessment.
2.6.2 Loneliness in AYAs
Developmentally, AYAs value friend relationships and social support (139); therefore,
isolation from friends, or loneliness, due to “illness-related situations”, can become a primary
disabling issue for young adults with cancer (140). Not surprisingly, feelings of loneliness and an
inability to connect with other young adults have been commonly reported as one of the most
distressing aspects of a cancer experience for AYAs (13). There is potential for this intervention
to decrease loneliness as participants connect with their peers and build relationships in the
groups. Additionally, the NIEFP modification is expected to increase group cohesion and
participant engagement, and therefore we expect that group differences in loneliness and feelings
of isolation could be observed between the styles of intervention groups.
2.6.3 Trauma in AYAs
Trauma is known to be an underlying cause of many psychological disorders (141). A
cancer diagnosis or other life-threatening illness can often cause a Criterion A2 Response
(intense fear, helplessness, or horror) and the evidence of a high rate of PTSD symptoms in
cancer survivors is growing (142). It is estimated that 10-15% of cancer patients might meet
criteria for a diagnosis of cancer-related PTSD at some time within five years of diagnosis (143).
There is a documented inverse relationship between adult survivors’ age and PTSD-like
symptoms demonstrating that young-adult cancer survivors have a higher risk of PTSD (144146). Many young adults express that seeking social support, and perceiving it as satisfying, is an
important factor in preventing the development of PTSD (147), while less social support is
22
associated with higher PSTD rates (143). The chat-group interventions may increase the social
support experienced by these young adults leading to reduction in PTSD symptoms and therefore
this measure should be included in any psychosocial oncology intervention study of AYAs.
2.6.4 Posttraumatic Growth in AYAs
Many studies have reported that younger patients and minorities experienced more
posttraumatic growth, as did patients with less education and higher income (148). It is important
to include this measure to understand further where and how post-traumatic growth occurs and
whether change over time is related to group psychotherapy model differences (149). The
emotional response and opportunity to re-experience the cancer journey vicariously through the
AYAs’ candid reflections in the video could lead to differences in post-traumatic growth
between the video-based and workbook-based participants. A recent study of the relationship
between posttraumatic stress and posttraumatic growth in young adults with cancer, found that
re-experiencing the trauma of cancer was an adaptive cognitive process necessary to achieve
personal growth for AYAs after cancer (150). It concluded that providing opportunities for
AYAs to engage in candid discussions about their fears, worries and concerns might contribute
to posttraumatic growth and a greater understanding of their cancer journey by allowing them to
re-experience the trauma in a safe and controlled environment.
2.6.5 Coping Self-Efficacy in AYA’s
Cancer patients who feel more able to cope are better adjusted (151) and experience less
anxiety (152) than those who do not feel like they are able to cope. Self-efficacy is defined as a
person’s perception of their capability to execute the actions necessary to achieve a desired goal
(153). Therefore, coping self-efficacy defined in the cancer context, is a person’s perception of
their ability to cope with the challenges of a cancer diagnosis and treatment. Many AYAs face
significant challenges when coping with cancer and any successful psychosocial interventions
will strengthen their internal and external coping strategies (74). Interactions with the
professional facilitator, video content and peer group may influence a participant’s perception of
their coping ability and therefore this construct needs to be assessed.
23
2.7 Objectives of External Pilot Studies in Online RCT Interventions (Study Aim #3)
An external pilot study is defined as a “stand-alone piece of work planned and carried out
independently to the main study”; contrasted with an internal pilot study that is incorporated into
the main study design of an RCT (125). In general, pilot studies are designed to explore the
integrity of a study protocol including: 1) data collection, randomization and consent procedures,
2) recruitment and retention (2.7.1), 3) attendance (2.7.2), 4) system usability and participant
satisfaction with the intervention (2.7.3, 2.7.4), 5) appropriateness of outcome measures and the
sample size necessary to detect a significant change in these outcomes (power calculation, 2.7.5)
(125). The lessons learned can prevent major unanticipated problems before collecting patient
data in a larger trial.
Specifically in online interventions, which are heavily reliant on technology and involve
coordination across a wide geographic area, concerns around patient-safety, privacy, logistics,
communication and recruitment procedures could occur (125). General learning about these
important processes, procedures and technical considerations are collected during the pilot study.
A well-executed pilot study, with a clear logic model and list of objectives, will ensure that the
work is scientifically valid and publishable and ultimately lead to higher quality randomized
trials (125). This section explores the measures of feasibility used in this study.
2.7.1 Recruitment and Retention
Recruitment and retention, as measures of feasibility, are concerned with the ease of
finding and enrolling eligible participants and the collection of primary psychosocial outcome
data. Specifically, retention is not concerned with adherence to the treatment protocol (i.e.
attendance at weekly chat session and reading/watching the workbook/video), but with the
proportion of participants who filled out the psychosocial outcome questionnaires over the 9month study period. It is important to determine if randomization has an impact on retention (i.e.
are TAU waitlist control group participants retained to the same degree as intervention groups)
(125). Recruitment and retention targets based on published interventions that are similar in form
and geographic location can be used to measure feasibility (125). Identifying the potential
challenges of recruitment and retention are extremely important to the success of a future
randomized trial.
24
2.7.2 Attendance
Attendance is a separate construct than recruitment and retention as it is concerned with
attendance at individual sessions and not whether participants completed study surveys and
questionnaires. Attendance is an important indicator of intervention feasibility because in
psychosocial interventions with AYAs, better attendance leads to more positive outcomes (21).
Poor attendance therefore, could have a strong influence on outcomes and impact the feasibility
of the study design. A pilot study can also explore reasons for poor attendance and take steps to
mitigate these challenges in a larger study.
2.7.3 System Usability
Incorporating a validated scale of system usability does not appear to be common in
online psychosocial intervention trials, despite being highlighted as “best practice” in the
evaluation of online interventions (89). Among the online interventions reviewed for this study
none used a validated scale of usability, instead relying on a single usability item or an overall
satisfaction measure, neither of which delineated between suitability and usability (60, 154, 155).
A platform or system that has ‘good’ usability is one where the user can easily understand, use,
and navigate without undue stress or difficulty (156). The technology industry has developed
many validated usability instruments, with the most widely used being the System Usability
Scale (SUS) developed by John Brooke in 1996 (157). The SUS has the advantage of being
technology agnostic, which enables it to be compared across many different products and
services (158). Routine use of a scale like the SUS within online psychosocial oncology
interventions, especially those that are technology dependant, will allow for a more objective
rating of usability within and across interventions.
2.7.4 Participant Satisfaction
Participant satisfaction does not always correlate highly with psychosocial outcomes
(124), however, it is important to collect this information in a pilot study from a feasibility
perspective. Participant satisfaction can help to determine the likelihood of potential participants
enrolling in the future, and once enrolled, the likelihood of completing the interventions and
follow-up questionnaires. If participants are not generally satisfied with the intervention,
recruitment, retention and attendance could be an issue for a larger study.
25
2.7.5 Statistical Power and Appropriateness of Psychosocial Outcomes
Outcome data collected during a pilot study has been used to 1) select the most
appropriate outcome measures that assess the most important constructs and 2) use effect sizes to
approximate the sample size needed for a larger randomized trial (statistical power calculation)
(60, 125). Many psychosocial outcome scales have been developed and validated; however, very
few of these have been validated specifically with an AYA population (159, 160). It is possible
that standard psychosocial outcome measures will not be sensitive to change over time in AYAs
or capture appropriately the unique issues and challenges of this developmental life stage.
Therefore, it is important to assess the appropriateness of outcome measures before the initiation
of a randomized trial with this population.
Although there is some disagreement around using pilot study data for sample size
calculations in psycho-oncology intervention design (161, 162), it may help to provide some
context for the discussion of study design feasibility. For example, a power calculation that
produces an extremely large sample size estimate (i.e. 400 participants), no matter the inherent
problems with the calculation or the potential variability of the estimate, may still have
feasibility implications. Therefore, with the potential problems in mind, a power calculation was
completed to provide a rough indication of study feasibility.
2.8 Connection to Study Aims
The results of this literature review indicate the need for further study of AYA online
psychosocial interventions and in particular the study of developmentally appropriate group
psychotherapy models. The aims of this pilot study highlighted in Chapter 1 (Figure 1) are
designed specifically to address this gap in the psycho-oncology literature. The three primary
purposes of this study are: 1) to test the theory that different therapeutic models of online group
psychotherapy (PEP/NIEFP) for AYAs impact content sensitivity and group processes, 2) to
determine the appropriateness of using specific psychometrically validated psychosocial
measures with AYAs and 3) to determine the feasibility of a larger RCT comparing these two
models. Chapter 3 discusses the methods used to fulfill these study aims.
26
Chapter Three: Methodology
3.1 Overview
This prospective, three-arm pilot study (with re-randomization of the TAU control) used
the 10-week CancerChatCanada synchronous online chat group program to explore two different
group psychotherapy models; 1) a PEP model using a standard workbook (called NuCare) (113)
and 2) a modified NIEFP model using a video (WW2H) in place of the workbook. The enrolment
target was set at 36 participants with 9 allocated to each of 4 chat groups. The standard
enrolment for a CancerChatCanada (CCC) group is between 5-9 participants (85) with 7 being
the ideal number according to many facilitators (more participants it can difficult to track the
rolling online text conversation of too many participants at once). The sample size for this pilot
study was calculated using an estimated 10% low attendance rate in the initial chat groups and
20% low attendance rate within the re-randomized groups which would give each chat group the
desired size of 7-9 participants. It has been suggested that a minimum group size of 10
participants is necessary to find any meaningful estimates of a parameter in a pilot study (163),
but unfortunately this would interfere with the chat group experience. We provide the
recruitment, retention and attendance totals in Figure 2 (CONSORT Diagram) and discuss them
in detail in the results and discussion sections.
27
Figure 2: Pilot Study Consort Diagram with Recruitment and Retention Targets
CCC = CancerChatCanada, PEP = Standard Psychoeducational Psychotherapy, NIEFP = Narrative-Informed
Emotion-Focused Psychotherapy, TAU Control = Treatment As Usual Control Group, ITT = Intent to Treat, LMM
= Linear Mixed Model
28
3.2 Participant Recruitment, Eligibility Confirmation, and Randomization
3.2.1 Participant Recruitment and Enrolment Procedures
We recruited participants over a six-week period (January 16 – February 28, 2014)
through letters sent from the Alberta Cancer Registry (ACR), posters in cancer-treatment centers
and direct referrals from cancer-care professionals (Figure 2). As the recruitment strategies used
in online interventions can have a profound impact on the type of sample recruited (164), this
combination of methods allowed for a more representative sample.
Participants were instructed to register on the CCC website confirming their intention to
participate. As potential participants registered, the CCC central-office staff transmitted
registration and contact information via phone to the study coordinator (ML) who followed up
with an intake interview within two days of initial participant registration. The intake interview
was 10 minutes on average and conducted using a predetermined script (Appendix A.5). Once a
participant met inclusion criteria and gave informed verbal consent during the intake interview,
we sent the participant a final enrolment form as well as a link to the online baseline assessment.
At the completion of the recruitment period, we randomized participants to their treatment
groups and sent an email confirming the start date of their chat group. This email also included
instructions for how to use the chat group and login information. The week before their group
began each participant received a welcome email from the chat-group facilitator and a reminder
of chat-group login instructions. Appendix A provides all recruitment materials.
Historical (unpublished) ACR data (2000-2009) shows that approximately 700 AYAs are
diagnosed and treated in Alberta each year. Since eligibility required patients to have been
treated within the past six months, we estimated 350 potential participants. We anticipated that
the ACR mailing recruitment method would yield a 10% response rate, based on previous
experience and recommendations from CancerControl Alberta Patient Education Specialists in
both Calgary and Edmonton (who had been using this same method to mail symposium
invitations to patients within 6 months of treatment end). The initial registry mailing consisted of
a study invitation letter and two pieces of information about the ACR (See Appendix A.1 &
A.2). We sent a follow-up letter after 3 weeks to all non-respondents (Appendix A.3). Assuming
a conservative 10% response rate to all recruitment strategies (posters, direct referrals and ACR
mailing), we expected to meet the enrolment target of 36 participants within 2 months.
29
The CONSORT diagram outlining the participant flow through initial contact, screening,
eligibility, consent and retention, with associated targets for this feasibility study, is outlined in
Figure 2. In total, we recruited 34 participants to the study. The initial ACR mailing was sent to
414 potential participants and resulted in 16 respondents. The follow-up letter was sent to 395
potential participants resulting in 12 respondents. An additional 12 participants were recruited
primarily through direct-referrals from health care professionals equalling a total of 40
respondents. Of these 40, 37 were eligible (2 poor English, 1 deteriorating health) and 34
consented to participate (2 did not complete enrolment procedure, 1 could not commit). The
feasibility of using these recruitment procedures in a larger RCT and the sample characteristics
are discussed further in Chapter 4 and 5.
3.2.2 Eligibility Confirmation
We confirmed eligibility in two stages with the study intake interview confirming
inclusion criteria and the CancerChatCanada online enrolment form assessing exclusion criteria.
Inclusion criteria were as follows: 1) between ages 18-39 at enrolment, 2) treated for any
type or stage of cancer within the past 6 months, 3) ability to read and write in English
sufficiently to complete questionnaires and participate in chat groups, 4) access to a computer
and reliable internet connection in a private environment and 5) willing to provide the name and
phone number of a support person as their emergency contact.
Exclusion criteria were as follows: 1) indication of cognitive or motor skill impairment
that would keep them from participating in an online chat group, 2) endorsement of suicidal
ideation or substance abuse and 3) undiagnosed clinical depression (as per CancerChatCanada
policies). To determine undiagnosed depression the Hospital Anxiety and Depression Scale
(HADS) was included in the online enrolment questionnaire in addition to a self-report question
of a previous anxiety or depressive disorder diagnosis. Participants with a HADS score above
clinical cut-offs that did not indicate a previous diagnosis of clinical anxiety or depression by a
health care professional, met this exclusion criteria; however, a score above clinical cut-offs
combined with a self-reported diagnosis of depressive or anxiety disorder did not prevent
enrolment. It was determined that those who were aware of their clinical anxiety or depression
would have the necessary resources available for an optimal level of safety as they participated
in the online chat group. Conversely, those who were unaware of their condition required a
30
consult with a psychosocial professional in their area to ensure the availability of resources for
optimal participant safety. All participants excluded from the study were directly referred to a
psychosocial professional in the appropriate area of the province.
3.2.3 Randomization
Upon enrolment, each participant was assigned a three-digit participant identification
number (PID). Once enrolment closed, participants were stratified by gender, to insure an
equitable distribution among groups, and randomly allocated to the standard model (PEP),
modified model (NIEFP) or TAU waitlist control condition by a third-party research assistant
drawing PIDs from a hat. This method of stratified randomization is considered the best method
of randomizing small samples (165). The nature of the intervention was not conducive to the
masking of participants or researchers; however, they remained blinded to group allocation until
after completion of the baseline assessment.
3.3 Interventions
CancerChatCanada (CCC) is a pan-Canadian program, based at the BC Cancer Agency,
offering professionally led, live synchronous chat support groups for cancer patients, survivors
and caregivers. These text-only groups occur in real time and have provided psychosocial
benefits to a diverse demographic of participants across a wide geographic area (20). CCC
groups are available to Canadians at no cost and work in conjunction with a number of provincial
health systems including Alberta Health Services. Each online chat group follows the same basic
format (166):
•
Professional psychosocial oncology counsellors facilitate each chat group session.
•
Groups have 5-9 members and meet weekly for 10 weeks.
•
Online chats are hosted on a password-protected website and only members of the group
have a password to log in. Participants use a chat name so true identities are anonymous.
•
Each session is 90 minutes. Group discussions focus on common experiences, concerns
and questions.
The online groups provide emotional support and a safe place to discuss personal topics. In a
qualitative study examining the experiences of the CCC professional facilitators, researchers
reported that clinically important issues, such as death and dying, pain, spirituality and coping,
31
were discussed in-depth during the chat groups (167). A primary component of this intervention
is professional facilitation, as the therapist guides, focuses and summarizes group discussions,
enhances participant engagement, models group norms and ensures participant safety just as in a
traditional face-to-face group (85). The professional facilitation and synchronous format of this
intervention are the primary and essential differences between it and the many other online
“support” platforms that use only one or neither of these components. In past CCC studies the
presence of a professional facilitator was seen to be very important with 91% of participants
indicating “high” or “moderate” satisfaction or benefit with the professional facilitator in their
group (20).
Additional materials have been developed by CCC to help the facilitator provide a
starting point for, or to stimulate, discussion in the chat groups. The NuCare workbook
(described below) has been used for this purpose by CCC facilitators in the past. In this study,
participants were sent a weekly reminder email two days before their chat session that contained
a link to the content for one of the two therapeutic models. They were encouraged to read the
weekly NuCare chapter (standard PEP model) or watch the weekly WW2H video clip (modified
NIEFP model) before the chat session, but were also able to review the content during the
session or participate without having reviewed it at all.
3.3.1 Standard CCC Psycho-Educational Model (Workbook)
The NuCare workbook was developed in 1992 as a psychoeducational tool for cancer
patients within one year of initial diagnosis (113). It was used in interventional studies at McGill
University and the Jewish General Hospital (168), and was adopted for use with CCC groups in
2010. NuCare is composed of 8 chapters addressing the following 8 topics: Good Coping and
Mindfulness, Relaxation and Imagery, Ways of Thinking, Communication, Problem Solving,
Social Support, Healthy Lifestyle, and Goal Setting. When delivered in individual or group
settings, NuCare has led to modest quality-of-life improvements (168). In the initial design of
this current study, consultations with various CCC facilitators determined that the NuCare
Manual is used to varying degrees, with some finding it a useful starting point for discussion and
others simply mentioning it but not actively promoting its use. In this study the primary goal of
the PEP group was to improve psychosocial adaptation to cancer through the acquisition of
32
knowledge or skills, emphasizing learning rather than increased emotional intelligence and
engagement.
In the PEP arm of this study, the weekly chat-group reminder emails (week 2-9)
contained a link to the subsequent chapter from the NuCare manual. Participants were
encouraged to read the chapter and complete the exercises because it would form the basis of the
chat-group discussion that week; however, the facilitator still urged them to attend the chat-group
whether or not they read the weekly workbook chapter. Links to the individual workbook
chapters are provided in Appendix B.1.
3.3.2 Modified CCC Emotion-Focused Model (Video)
Due to the unique developmental life stage of AYAs, standard cancer education
materials, such as the NuCare workbook, may not be attractive or suitable for young adults
despite containing helpful information. For this reason, the modified model (NIEFP) arm used an
AYA patient-created video with high face validity (discussed in Chapter 2). The primary goal of
the NIEFP group was to improve psychosocial adaptation to cancer through increased emotional
intelligence and engagement with peers, emphasizing emotional support and storytelling rather
than the acquisition of knowledge or skills.
Preliminary research using the video Wrong Way to Hope: An Inspiring Story of Young
Adults and Cancer, has demonstrated high clinical utility and relevance within the AYA
population (122). The video follows a group of young adult survivors on a wilderness kayaking
expedition and the beautiful and engaging scenes on the river are interspersed with participants’
reflections on their cancer experiences. The video is chronological, and with each new day the
young adult survivors address a new topic about their struggle with cancer. These 7 themes, in
order, form the substance of the video clips used in the NIEFP for this study: Transitions, Fear
and Uncertainty, Isolation, Relationships, Identity, Reflection and Reintrajectorization (a word to
describe Cancer Survivorship). The NIEFP chat-group followed the flow of the kayaking trip
with each new week using the theme addressed in the video to stimulate discussion.
Using the same method as with the NuCare workbook, the weekly chat group reminder
emails (week 2-9) contained a private link to the YouTube video clip of the video. Participants
were encouraged to watch each video clip but still urged to attend the chat-group whether or not
33
they were able to watch it. Links to the video clips and discussion questions are provided in
Appendix B.2.
3.4 Study Integrity and Participant Safety
CancerChatCanada incorporated this pilot study into their ongoing and successful
program that has a robust research arm and highly trained facilitators (167). This study therefore,
was subject to the same high standards of quality and confidentiality. It used standardized
measures, replicable facilitation/procedures and random allocation.
The unique nature of online chat groups allows for complete privacy of the individuals
involved. In order to access a chat group all participants must be approved by the program
administrator and use their username and password to sign on. All chat-group participants are
encouraged to use a “screen name” that is not their given name and no participant is required to
disclose any personally identifiable information to other chat-group participants. The CCC
website itself uses the same level of security and encryption as many websites in the banking
industry, keeping all IP addresses and other potentially private information safe from others
online. All email addresses collected during the enrolment process and used for communication,
were never available to other participants, unless the participant posted it themselves, and all
group email communications blind carbon copied all email addresses.
CCC facilitators, in addition to practicing a minimum of 4 years in Oncology and
maintaining their professional certifications, undergo chat-group-specific training. This includes
a minimum 10 weeks of training and peer-supervision sessions (167). This training includes the
identification of highly distressed individuals in a text-based chat-group setting. In this study, the
facilitator reviewed the chat-group transcripts and recorded weekly clinical notes that staff at the
central CCC office monitored. The facilitator was able to follow up by phone with participants
exhibiting drastic changes in distress levels or consistently high, sustained levels of distress and
refer them to an appropriate health-care professional in their geographic area. In addition, we
provided the facilitator’s office phone number and email to participants in case they wished to
address an issue in a personal conversation. If a participant missed two consecutive chat groups,
the facilitator contacted them to understand the reason for the absence and encourage them to
continue attending.
34
There is potential for bias in psychotherapy trials when the same therapist facilitates both
arms and when different therapists facilitate each arm, because each of these practices adds error
variance but in different ways. In a larger trial, using multiple therapists would be necessary. As
this was an unfunded pilot study, one therapist facilitated both arms of the trial. This clinical
social worker (MSW) facilitates CCC groups and also works with AYA patients in her role as a
staff member in the Psychosocial Resources Division of The Tom Baker Cancer Centre in
Calgary, AB.
All participants (including the waitlist controls) filled out primary psychosocial outcome
measures at the same time points. However, only active treatment groups filled out the weekly
post-chat group questionnaires after each session. As this study was a pilot RCT and involved a
relatively high survey burden, researchers offered an incentive to participants to encourage them
to complete all the Primary Outcome Assessments at baseline, 3 months, 6 months and 9 months.
The incentive was a $25 gift card to Tim Horton’s/Starbucks at the end of the study. Two study
participants chose not to receive the incentive despite having completed all the surveys.
Researchers obtained ethics approval from the conjoint health research ethics board
(REB13-0482) at the University of Calgary and were included under the CCC ethics approval
through the BCCA Research Ethics Board at the University of British Columbia.
3.5 Data Collection and Management
Online data collection is quickly becoming the preferred method of data collection for
most AYA participants (169). CancerChatCanada.ca website facilitated online enrolment and
registration data collection while participants completed the feasibility, suitability and usability
(weekly post-chat group surveys) and psychosocial outcome assessments (baseline, 3, 6, 9,
months) through FluidSurveys.ca. The weekly post-chat group surveys varied from 3 – 15
questions and required a median time of 1 minute (range 0:12 – 6:50 min) to complete. The
psychosocial outcome questionnaires had 86 items requiring a median time of 11.5 minutes
(range 4:55 – 22:10) to complete.
We stored participant registration and enrolment information on the secure CCC servers
at the BC Cancer Agency and stored online survey and assessment data on the Fluid Survey’s
secure, Canadian-located server until the study was complete. Upon study completion, we de35
identified registration and enrolment data using PIDs, placed it in a password protected Excel file
by a CCC research assistant, and emailed it to the primary investigator. We used the participant’s
chosen “screen name” as the primary identifier in all surveys and assessments completed on
Fluid Surveys and at study completion we downloaded the data and further de-identified using
PIDs. These procedures created an anonymous final dataset used in all analyses. We stored the
identification key on a secure Alberta Health Services computer and the primary investigator
stored and analyzed all de-identified data on a password-protected hard drive.
We collected all qualitative low attendance and follow-up interview data (standardized
phone interview scripts, Appendix C.1 & C.2) over the phone and used PIDs as the only
identifier. All low attendees and a randomly selected 35% of participants (n=12) who completed
the intervention were contacted for a follow-up interview. We also stored the qualitative data on
the password protected hard drive with the quantitative data. Chat-group transcript data is stored
on the secure and encrypted CCC computers at the BC Cancer Agency as per CCCs policies and
procedures, with the understanding that researchers may complete a qualitative study of the
transcripts in the future. No personally identifiable information was ever transmitted using
electronic sources and the primary investigator was the only person with access to all the study
data.
3.6 Measures
Appendix D gives a concise overview of the study aims and connects them directly to the
measurement tool used, data collection method and analysis plan.
3.6.1 Demographic and Medical Variables
CancerChatCanada.ca collected demographic and medical-status variables through online
pre-registration (Appendix A) and the Alberta Cancer Registry also provided some demographic
data. The data collected through CCC include age, gender, marital status, number of dependants,
education level, employment status, residential location, cancer diagnosis, metastatic diagnosis,
therapy received, current treatment status, self-reported functional status (Eastern Cooperative
Oncology Group Performance Status – ECOG-PS) (170) and time from initial diagnosis to
baseline. The demographic data collected through the ACR was more limited and included age,
gender, cancer type and therapy received. It included aggregate data of non-responders only and
36
was collected to explore a possible response bias inherent in the 9.6% response rate of the 414
potential participants.
3.6.2 Study Aim #1 – “Proof of Concept” Measures
3.6.2.1 Topic-Specific Suitability
We assessed “Proof of Concept” through both content sensitivity and group process
measures. The two measures developed specifically for this study to assess content sensitivity
included: 1) the weekly post-chat group Topic-Specific Suitability measure (Appendix E.1.1 &
E.1.2) and 2) the post-intervention Overall Suitability measure (Appendix E.2).
The weekly Topic-Specific Suitability measure (Appendix. E.1.1) was delivered 8 times
during the 10 weeks of the chat-group experience (week 2-9) and included 6 items (3 Likertscaled items, 2 multiple-choice items and one qualitative item). For the Likert-type questions
(“This topic was applicable to my cancer experience”, “I experienced challenges similar to what
was discussed”, “This week’s video/workbook chapter was helpful for me”, 4-point scale, Not at
all=0 to Extremely=4) we conducted an exploratory factor analysis using principle axis factoring
and Varimax rotation. This analysis revealed a single factor; “Relevance of Content” (α =0.77).
The first multiple-choice question (“I felt ________after watching/reading this weeks
video clip/workbook chapter”), included 5 potential emotional responses (bored, frustrated,
neutral, validated, excited to discuss). The second multiple-choice questions about weekly use of
the model content (“This week’s video clip/workbook chapter I watched/read…) had 3
endorsement levels (before chat, during chat, not at all). The one open-ended qualitative response
(“Any further comments you would like to make about this weeks chapter/video clip”) was able
to accept any length of text responses.
3.6.2.2 Overall Suitability
The Overall Suitability measure (Appendix E.2) was delivered once, immediately postintervention, and included 5 items using a 6-point Likert-type scale (Strongly Disagree=0 to
Strongly Agree=5, “Overall, I feel the video/workbook was useful for me”, “I am satisfied with
using the video/workbook content as a basis for discussion”). All items loaded on a single factor
in an exploratory factor analysis using principle axis factoring and Varimax rotation; “Overall
Suitability for Learning and Discussion” (5-items, α=0.95). We explored qualitative comments
37
from the weekly post-chat survey and follow-up interviews for common themes to provide
further information on model suitability.
3.6.2.3 Group Cohesiveness
We also assessed “Proof of Concept” using two group process measures. The first group
process measure was the Group Cohesiveness Scale (GCS) (130). As mentioned earlier, a
suitable psychotherapeutic model for AYAs should help build group cohesion and engagement,
more than transmit knowledge, and therefore increased within-group cohesion demonstrates that
participants experienced the NIEFP group in the way we intended.. The GCS is brief (7 items)
and has good internal reliability (Cronbach α = 0.87) and concurrent validity (positive correlation
with Cohesion to Therapist Scale r = 0.77, p < .001 and Group Benefit Questionnaire r = 0.71, p
< .001, and intraclass correlation = .82, p < .001) (130). All 7 items load on a single factor (factor
loadings = 0.52 - 0.83) indicating good internal consistency (130). The GCS was delivered 4
times across the 10-week intervention (week 1, 4, 7, 9) to measure changes in group
cohesiveness over time (Appendix E.1.2). It was not administered weekly due to concerns of
decreased scale sensitivity. In this sample the GCS score Cronbach’s α = 0.80.
3.6.2.4 Group Experiences Inventory
We used the Group Experiences Inventory (GEI) as a second group process measure to
explore the “Proof of Concept” study aim. It has been used in other group psychotherapy
interventions to determine what aspects of their group experience participants value the most
(132, 171). The measure includes 25 statements rated on a 6-point Likert-type scale (0 = not
applicable to 5 = One of the most important group experiences, “Developing a new attitude
towards life”, “Expressing my true feelings”, “Talking about fears of death”). Participants were
asked to rate the top 3 most important experiences as fives and then rate the rest of the statements
to the degree to which each statement was an important aspect of their group experience.
Exploratory factor analysis determined 5 subscales: Expressing-True-Feelings (factor loadings =
0.37 - 0.65); Developing-A-New-Attitude (0.42 - 0.78); Accessing-Resources-And-Advice (0.71
- 0.74); Establishing-Supportive-Contact (0.20 - 0.67); Discussing-Sexual-Concerns (single item)
(171). One-week test-retest reliability of each identified subscale was reasonable (r = 0.51- 0.64,
Cronbach’s α (range) = 0.66 - 0.84) (171). This measure was chosen for its ability to
comprehensively characterize the participant’s group psychotherapy experience and was
38
delivered once immediately post-intervention to measure the impact of the psychotherapy model
on valued group experiences (Appendix E.2). In this sample, Cronbach’s α for each sub-scale is
0.75, 0.88, 0.84, 0.68 respectively.
3.6.3 Study Aim #2 – “Appropriateness of Psychosocial Outcomes” Measures
3.6.3.1 Hospital Anxiety and Depression Scale (HADS)
We measured distress, depression and anxiety levels using the Hospital Anxiety and
Depression scale (HADS) (172). It is a 14-item multidimensional self-report measure and is the
most commonly used distress-screening tool as well as being the most used benchmark for the
validation of other distress measures (134, 173). The term 'hospital' in its title suggests that it is
only valid in such a setting but many studies conducted throughout the world have confirmed
that it is valid when used outside of an in-patient setting (174). A review of over 200 studies
using the HADS found it to have good internal consistency for both the anxiety (Cronbach’s α
(range) = 0.80 - 0.93) and depression subscales (Cronbach’s α (range) = 0.81 - 0.90) (175). Testretest reliability for each subscale was strong (r>0.80) and because the two subscales correlate
fairly highly (r=0.63 in 18 studies with n=8160) there is evidence that using the total HADS
score as a general measure of distress is also feasible (175). Cut-off scores for each subscale
have been developed and debated over time, however a recent meta-analysis recommended
against using the HADS as a case-finding instrument but instead as a continuous measure of
changes in depression and anxiety levels over time or as a generalized screening for distress
measure (176). Higher scores are associated with greater distress, depression or anxiety. In this
sample the HADS Total score Cronbach’s α = 0.88.
3.6.3.2 UCLA Loneliness Scale (UCLA-LS)
We measured participant’s subjective feelings of loneliness and social isolation using the
UCLA Loneliness scale. It is a 20-item unidimensional self-report measure that provides a valid
and reliable assessment of loneliness across a variety of populations and data-collection methods
(177). The reliability of this measure was assessed using groups of college students (n=489),
nurses (n=310), teachers (n=316) and the elderly (n=301) (177). Internal consistency for each
sample had a Cronbach α = 0.92, 0.94, 0.89 and 0.89 respectively and demonstrates that the
measure is highly reliable (177). Convergent validity for this measure was confirmed by
39
significant correlations with other measures of loneliness such as the NYU Loneliness Scale
(r=0.65), the Differential Loneliness Scale (r=0.72) and a negative association with a measure of
social support, the Social Provisions Scale (r=-0.68) (177). In addition, construct validity was
confirmed with measures of the adequacy of interpersonal relationships, as measured by the
Social Provisions Scale, and by correlations between loneliness and measures of health and
wellbeing such as the presence of chronic illnesses (r=0.18, p<0.01) and depression (r=0.45,
p<0.001)(177). Finally, a confirmatory factor analysis found all of the items loaded significantly
on a bipolar global loneliness factor, and confirmed this as a unidimensional measure. Higher
scores indicated greater feelings of loneliness. In this sample the UCLA-LS Cronbach’s α = 0.91.
3.6.3.3 Posttraumatic Stress Disorder Checklist – Civilian (PCL-C)
We measured the presence of trauma or PTSD symptomatology in study participants
using the PCL-C, keyed to the diagnosis of cancer. The PCL-C was developed by the Veterans
Affairs National Center for PTSD and is a 17-item self-report measure of the 17 DSM-IV
symptoms of PTSD divided into three clusters: re-experiencing the trauma, avoidance/numbing
in response to trauma reminders, and hyperarousal keyed to a specific event. The PCL-C is the
most commonly used measure of PTSD in the cancer population and its reliability and validity
have been long established (142, 143, 178). A systematic review of the psychometric properties
of the PCL-C found all studies indicated acceptable 1-week test-retest reliability (r > 0.75),
internal consistency (Cronbach α > 0.92), convergent validity (CAPS r = 0.79 , and Mississippi
PTSD Scale r = 0.90) and divergent validity (BDI-II r (range female – male) = 0.71- 0.76) (179).
All studies affirmed the originally postulated 4-factor structure (179).
Two sets of criteria are used to identify respondents likely to merit a formal diagnosis of
cancer related PTSD (143). The cut-off method uses a total score of 50 or more to identify
PTSD. The symptom method defines ratings of 3 (moderately), 4 (quite a bit), or 5 (extremely)
as endorsement of a particular symptom following DSM-IV criteria. Individuals are considered
likely candidates for a diagnosis of PTSD if they endorse one or more re-experiencing
symptoms, three or more avoidance and/or numbing symptoms, and two or more arousal
symptoms (143). In this sample the PCL-C total score Cronbach’s α = 0.91.
40
3.6.3.4 Posttraumatic Growth Inventory (PTGI)
We measured posttraumatic growth using the PTGI. This measure is a 21-item
multidimensional self-report measure of posttraumatic growth in trauma survivors (180). Posttraumatic growth is due in part to certain personality traits, however, the PTGI was only
moderately associated with two traits indicating it is an objective measurement of personal
growth (181). A confirmatory factor analysis indicates an oblique five-factor structure, including
the sub-scales of 1) relating to others, 2) new possibilities, 3) personal strength, 4) spiritual
change and 5) appreciation for life, demonstrating that all of these, as well as the total PTGI
score, should be considered in analyses (149).
The PTGI has demonstrated good reliability and internal consistency (Cronbach’s α =
0.90) in adolescent populations who have suffered traumatic experiences (182). Concurrent
validity was established using the Social Desirability Scale (r=0-.15, p<0.01), Life Orientation
Test (r=0.23, p<0.01), and the NEO Personality Inventory (r=0.16-0.29, p<0.01) and the index
of religious participation (r=0.25, p<0.01) (181). The PTGI is considered to be most appropriate
if a researcher thinks it is important to measure new opportunities that may open up after a
challenging life event (183). Higher scores indicate greater levels of posttraumatic growth. In
this sample the PTGI Total score Cronbach’s α = 0.95.
3.6.3.5 Cancer Behaviour Inventory – Brief (CBI-B)
We measured coping self-efficacy in study participants was measured using the CBI-B, a
12-item multidimensional self-report measure used specifically with cancer patients (184). The
CBI-B assesses areas important to coping with cancer including (a) the respondent’s confidence
in maintaining independence and a positive attitude, (b) confidence in their ability to participate
in medical care, (c) confidence in the skills important for coping and stress management, and (d)
their confidence to manage their emotions/affect in difficult situations (184).
The reliability of this measure was assessed using a university-based cancer centre cohort
(n=735), a community-case cancer program (n=199) and a nationally representative sample
(n=370) (184). Internal consistency for each sample had a Cronbach α = 0.84, 0.84 and 0.88
respectively (184). Validity of this measure was verified using an Exploratory Factor Analysis
with oblique rotation of data from the first sample (n=735). This produced a four-factor structure
that was then supported using confirmatory factor analysis with data from the second (n=199)
41
and third (n=370) samples (184). Also, there were positive relationships between the CBI-B and
measures of quality and satisfaction of life as well as optimism, demonstrating concurrent
validity (184). Cut-off scores and population norms have not been developed for this measure
however, a higher score is indicative of stronger endorsement of the respective subscales as well
as overall coping self-efficacy. In this sample the CBI score Cronbach’s α = 0.90.
3.6.4 Study Aim #3 – Feasibility Measures
3.6.4.1 Recruitment and Retention
We measured recruitment and retention using the following: a) the proportion of AYAs
interested in the study, as estimated by number of respondents to the ACR mailing and word-ofmouth recruitment over (divided by) the number of potential participants identified through the
registry b) proportion of AYA respondents eligible to participate, as estimated by the number of
interested participants who met study eligibility, c) proportion of AYA respondents consenting to
participate, estimated by the number of eligible participants who consented, and d) proportion of
consented AYAs who completed the 3, 6 and 9 month psychosocial outcome questionnaires (See
Fig. 1 - CONSORT Diagram).
3.6.4.2 Attendance
We assessed attendance both quantitatively, by reviewing the attendance records in the
post-chat facilitator notes to get overall attendance numbers, and qualitatively by contacting all
low attendees (those who attended less than two chat-groups) to ascertain their primary and
secondary reasons for discontinuing the study.
3.6.4.3 System Usability
We assessed the usability of the CCC online chat platform in conjunction with the
standard (PEP) and modified (NIEFP) models using the System Usability Scale (SUS) and a
single-item “Ease of Use” question. Both SUS and the “Ease of Use” responses were collected
once, immediately post-intervention with the single “Ease of Use” item (“It was easy for me to
watch/read the video/workbook content”) being measured using a 6-point Likert-type scale
(Strong Disagree=0 to Strongly Agree=5) (Appendix E.2).
As noted earlier in Chapter 2, the SUS is the most widely used usability scale and has the
advantage of being technology agnostic, which enables it to be compared across many different
42
products and services (158). It is easy to administer (10 question, 5-point Likert-type scale), has
good internal reliability (Cronbach α = 0.91) and construct validity (expert rated) and has
benchmarks to aid in interpretation (157, 158). It is a unidimensional scale with factor analysis
revealing only one significant factor for the 10 SUS statements (158). In this sample the SUS
score Cronbach’s α = 0.70.
The single number generated by System Usability Scale is easily used for relative
judgments of system usability, where comparison with other platforms or iterative versions is the
primary aim; however, as a single measure of usability a “standard” must be applied. As the
scale is from 0 – 100, a simple and useful standard adopted by most usability professionals is the
“university grade analog” where SUS scores are correlated with a letter grade to provide realworld relevance to the number (158): a score of 90-100 is an A, 80-89 is a B, etc. This means
that systems that are passable have SUS scores above 70, with better products scoring in the 80s
and superior products scoring better than 90 (158). Qualitative comments from follow-up
interviews provide context to usability scores.
3.6.4.4 Participant Satisfaction
We measured participant satisfaction using a single item (“Overall, I am satisfied with
my chat group experience”), collected immediately post intervention, using a 6-point Likert-type
scale (Strong Disagree=0 to Strongly Agree=5). Additionally, we used qualitative comments
from the final post-chat group survey (“What was the best thing about this group for you as a
cancer patient/survivor?”) and follow-up interviews (Satisfaction with “Online Communication,
Privacy/Confidentiality, Understanding of Group, Facilitator) to explore the what aspects may
have contributed to the participant’s satisfaction levels.
3.7 Data Analyses
The small sample size of the study proposed a serious threat to most tests of significance,
specifically the analysis of psychosocial changes over time, as the assumptions of these tests
were not met. Therefore, we conducted most statistical tests to examine trajectories in the data
and to assist in characterizing the data for discussion or future analyses. All data analyses were
completed using SPSS v.19 (2013) except for 1) the slope calculations for change in Group
43
Cohesion over time and 2) the LMM analysis of the psychosocial outcome measures which were
conducted using SAS Proc Mixed v. 9.2 (2014).
3.7.1 Demographic and Medical Variable Analysis
We analyzed these data using standard descriptive statistics, such as proportion, mean,
standard deviation, confidence intervals, median and range where appropriate, to characterize
intervention groups within the larger AYA cancer population in Alberta. No tests of significance
compared group differences due to small group sizes.
3.7.2 Study Aim #1 – “Proof of Concept” Analysis
We aggregated data from both the initial randomization chat groups and TAU waitlist rerandomization groups for all proof-of-concept analyses. We compared the two psychotherapy
models (PEP/NIEFP) on weekly Topic-Specific Suitability questions and used valid percents to
characterize differences in the “Use of Content” and “Feeling about Content” items across all 8
weeks. For the composite “Relevance of Content” factor we compared the two models using
median scores to rank what topics were most relevant within each resource. The “Overall
Suitability for Learning and Discussion” factor was not normally distributed so we used the
Mann-Whitney U Test to compare differences in overall suitability of the PEP/NIEFP models.
We analyzed the Group Cohesiveness Scale by comparing the mean slopes and pre/post
intervention change scores between groups using Cohen’s d as the effect size to characterize
changes in group cohesiveness over time. We used the Mann-Whitney U Test for nonparametric
data to test differences in the Group Experiences Inventory subscales between the PEP/NIEFP
groups.
3.7.3 Study Aim #2 – “Appropriateness of Psychosocial Outcomes” Analysis
All analyses of psychosocial outcomes used continuous scales despite the presence of
established clinical cut-offs for certain scales, because change-over-time was the phenomenon of
interest, not a clinical diagnosis. Exploratory analyses investigated effect sizes. Using the initial
study randomization (three groups – Standard n=8, Modified n=8, Control n=18), we conducted
a Linear Mixed Model (LMM) analysis to explore changes over time in psychosocial outcomes
from baseline to 3 months of all three groups, and from baseline to 9 months for only the
standard and modified intervention groups. As it is resilient to small samples, we used Cohen’s d
44
effect sizes to examine differences between PEP and NIEFP groups and controls. To assist in
determining appropriateness of the chosen outcomes, we explored the qualitative responses from
three follow-up interview questions for comments related to psychosocial constructs of interest
(“Can you identify anything that has changed for you?”, “Have you made any practical changes
as a result of the group?”, “Do you expect anything to be different for you now?”) .
3.7.4 Study Aim #3 – Feasibility Analysis
Recruitment and retention targets were set based on other similar interventions, in
geographic location or intervention type (20, 60), and we considered feasibility achieved if actual
percentages were within 5% of the target (60). We compared attendance numbers with
unpublished CCC operational data, as well as published CCC operation data where applicable
(20), to assess feasibility using: 1) the average weekly attendance across all groups, 2) the
average number of sessions attended by each participant across all groups. To further
characterize participant attendance, we compared categorical attendance levels (number of
sessions attended = 9-10, 7-8, 5-6, 3-4, and 2 or less) with published CCC data (20) and between
PEP/EFP groups. We analyzed low attendance quantitatively comparing the proportion of low
attendees with unpublished CCC operational data for the same period as the study (2014). We
analyzed low-attendee follow-up interviews by summarizing common primary and secondary
reasons for low attendance and reported the proportion of participants endorsing each reason. For
both the average weekly attendance and low attendance rate, we considered feasibility as
achieved if actual percentages were within 5% of the CCC operational percentages.
The goal of the SUS was to provide a point estimate of usability for the CCC platform
and explore if this varied significantly by psychotherapy model. If there was no significant
difference, the different psychotherapy models might not have an effect on system usability.
Unlike psychosocial outcome measures, system usability does not change over 9 months if there
is no change to the system (158). Therefore, we aggregated the SUS by psychotherapy model for
inter-group comparison, using Student’s t-test to measure significant differences, and compared
using the University Grade Analog described in Chapter 2. We analyzed the single item “Ease of
Use” question comparing PEP/NIEFP using the Mann-Whitney U Test for nonparametric data
and explored qualitative comments from the follow-up interviews for common themes that
provided context and depth to the quantitative results.
45
To determine how satisfied the AYA participants were with their chat group experience,
we compared overall participant satisfaction with unpublished CCC normative data on other chat
groups. We also compared PEP/NIEFP groups on participant satisfaction using the MannWhitney U Test and used qualitative responses to explore different aspects of participant
satisfaction. We performed a statistical power calculation using the PCL-C outcome measure to
explore sample size considerations.
3.7.5 Reporting of Qualitative Comments
In order to provide clarity, context and depth to the proof of concept, impact and
feasibility aspects of this pilot study, we used qualitative comments from the weekly TopicSpecific post-chat survey (week 2-9), the final post-chat survey (week 10) and follow-up
interviews. Although rigorous qualitative methodology was not used to explore these data, we
have included summaries of common themes and participant quotes where appropriate. A second
reader (committee member) reviewed all participant comments and corroborated the general
themes and qualitative comments selected for inclusion in this thesis. The NVIVO v. 10 (2014)
and Microsoft Word (2009) was used for the thematic coding of participant feedback.
46
Chapter Four: Results
4.1 Overview
Organization of the results follows the study aims of determining 1) proof of concept for
the difference in PEP/NIEFP models using content sensitivity and group process (4.3), 2) the
impact of PEP/NIEFP models on psychosocial outcomes (4.4) and 3) feasibility of study design
and procedures (4.5). Quantitative and qualitative results appear together under each study aim.
The goal of this section is to provide an in depth exploration of AYA participants’ online chatgroup experience.
4.2 Demographic, Medical and Psychosocial Characteristics of Sample
We recruited a total of 34 participants to the study. Table 1 contains the demographic and
medical characteristics of the study participants and Figure 3 provides a simplified version of the
CONSORT diagram highlighting overall group demographics. Due to small sample sizes
(standard model n=8, modified model n=8, and control n=18), no tests of significance compare
group differences. However, randomization was likely effective as there were no observable
differences in demographic characteristic proportions between the three arms. Additionally, there
were no observable differences in age and gender between study participants and nonresponders, indicating that the sample is likely representative of the larger AYA population in
Alberta.
The mean age of the sample was 30.7 (SD=5.5, Range=21-39). The majority of study
participants were female (n=23, 67.6%) with approximately half married (n=16, 47.1%). A
smaller number of participants had children (n=10, 33.3%), and a large number of participants
had either a university undergraduate or graduate degree (n=15, 44.1% and n=8, 23.6%,
respectively). The majority of participants were on disability (n=20, 58.8%) and lived in an
urban setting (n=22, 64.7%).
The proportion of breast tumour group representation was lower in the control group
(16.7%) as compared to the two treatment groups (standard = 37.5%, modified = 25.0%).
Overall, study participants had a higher proportion of Hematopoietic and Reticuloendothelial
cancer’s (38.2%) and Eye, Brain and Other CNS cancers (17.6%) then non-respondents (27.4%
47
and 5.5% respectively). Breast (23.5%) cancers were less prevalent in study participants then
non-respondents (31.1%).
Figure 3: Simplified Version of CONSORT Diagram Highlighting Overall Demographics
of Initial Randomization
* Not statistically analyzed for differences between groups due to small sample size.
The majority of participants indicated they were still in treatment (n=21, 61.8%), with a
relatively large of participants indicating a metastatic diagnosis (n=11, 32.4%), however, only
one participant indicated that they were not able to take care of themselves. There was a large
observable difference in the time from initial diagnosis to baseline between groups. Specifically
the median for the PEP group was 21.5 months from diagnosis as compared to 7.5 months and
8.0 months for the modified model and control groups. This difference is due to 4 participants (of
8) randomized to this group who had their original diagnosis more than 24 months prior to study
enrolment.
Due to the small group sizes, there was large variance in the psychosocial outcome scores
at baseline as demonstrated by the large standard deviations and 95% confidence intervals. Using
the standardized cut-off scores for each scale, only the PCL-C and the UCLA-LS indicated
clinically significant scores. At baseline this sample of AYAs was moderately traumatized (PCL-
48
C cut-off = 26-43) and moderately lonely (UCLA-LS cut-off = 41-60) but only mildly distressed
(HADS cut-off = 9-25) and had a relatively high level of self-efficacy and post-traumatic growth.
Finally, other sample characteristics of particular interest were explored using the
standard CCC enrolment questionnaire data. In general, this sample indicated it was a socially
connected group with 91% endorsing having a close friend or relative with whom they could
talk. Half of the sample (n=17) had received professional counselling before, although 70.6%
(n=12) of those had only attended between 1-4 sessions. Only 5.9% (n=2) of the entire sample
had ever attended a face-to-face support group. Almost half the participants (n=15, 44.1%) had
previously visited a cancer forum or blog and only 1 had participated in a chat group. Most
participants (n=29, 85.3%) identified convenience as a main reason for participation in the
online chat with a smaller number highlighting the added comfort of participating from home
(n=12, 35.3%), increased privacy (n=10, 29.4%) and no other local supports available (n=10,
29.4%) as reasons for choosing online support. Only 2 participants listed their physical condition
(5.9%) as a reason for participating in the online group over other face-to-face psychosocial
supports. Table 2 contains the participant reasons, at enrolment, for choosing the online support
modality, instead of other face-to-face of psychosocial supports.
49
Table 1: Baseline Demographic, Medical and Psychosocial Characteristics of AYA Participants (n=34) and Non-Respondents (n=383)
Demographic
Age at Baseline, mean SD (range)
Female Gender, No. (%)
Marital Status, No. (%)
Married/Common Law
Single
Separated/Divorced
Children at Home (Yes), No. (%)
Education, No. (%)
High School
Technical College
University Degree
Graduate Degree
Employment Status, No. (%)
Full-Time
Part-Time
On Disability
Geographic Location – Urban, No. (%)
Cancer Type, No. (%)
Hematopoietic & Reticuloendothelial
Breast
Melanoma
Eye, Brain & Other CNS
Male/Female Genital Organs
Digestive Organs
Bones & Connective Tissue
Head & Neck
Thyroid & Endocrine Glands
Other
Metastatic (Yes), No. (%)
Therapy Received
Surgery
Chemotherapy
Radiation
Other
Current Treatment Status (Yes), No. (%)
Initial Diag. to Baseline (Months),
Median (Q1-Q3)
Self-Reported Functional Status
(ECOG-PS)
Fully Active
Light Work
Take Care of Self
Not Able to Take Care of Self
HADS, Mean SD (95% CI)
PCL-C, Mean SD (95% CI)
UCLA-LS, Mean SD (95% CI)
CBI-B, Mean SD (95% CI)
PTGI, Mean SD (95% CI)
28.9, 4.3 (23-35)
5 (62.5)
TAU Control
(n=18)
29.8, 5.8 (21-39)
12 (66.7)
Total Sample
(n=34)
30.7, 5.5 (21-39)
23 (67.6)
5 (62.5)
3 (37.5)
0
3 (37.5)
3 (37.5)
5 (62.5)
0
3 (37.5)
8 (44.4)
9 (50.0)
1 (5.6)
4 (22.2)
16 (47.1)
17 (50.0)
1 (2.9)
10 (29.4)
1 (12.5)
1 (12.5)
3 (37.5)
3 (37.5)
0
2 (25.0)
4 (50.0)
2 (25.0)
3 (16.7)
4 (22.2)
8 (44.4)
3 (16.7)
4 (11.8)
7 (20.6)
15 (44.1)
8 (23.5)
2 (25.0)
0
6 (75.0)
6 (75.0)
1 (12.5)
2 (25.0)
5 (62.5)
5 (62.5)
Medical
7 (38.9)
2 (11.1)
9 (50.0)
11 (61.1)
10 (29.4)
4 (11.8)
20 (58.8)
22 (64.7)
2 (25.0)
3 (37.5)
0 (0.0)
2 (25.0)
0 (0.0)
1 (12.5)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
4 (50.0)
2 (25.0)
2 (25.0)
1 (12.5)
2 (25.0)
1 (12.5)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
2 (25.0)
9 (50.0)
3 (16.7)
1 (5.6)
2 (11.1)
2 (11.1)
1 (5.6)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
5 (32.4)
13 (38.2)
8 (23.5)
2 (5.9)
6 (17.6)
3 (8.8)
2 (5.9)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
11 (32.4)
4 (50)
7 (87.5)
3 (37.5)
1 (12.5)
6 (75.0)
5 (62.5)
6 (75.0)
5 (62.5)
2 (25.0)
4 (50.0)
8 (44.4)
16 (88.9)
9 (50.0)
3 (16.7)
11 (61.1)
17 (50.0)
29 (85.3)
17 (50.0)
6 (17.6)
21 (61.8)
21.5 (2.3 – 38.3)
7.5 (4.0 – 11.3)
8.0 (5.0 – 17.0)
8.0 (4.0 – 19.0)
PEP (n=8)
NIEFP (n=8)
34.5, 4.6 (27-39)
6 (75.0)
4 (50.0)
2 (25.0)
2 (25.0)
6 (75.0)
2 (25.0)
0 (0,0)
0 (0.0)
0 (0.0)
Primary Outcome at Baseline
10.8 (6.2),
5.6-15.9
38.0 (13.9),
26.4-49.7
53.9 (7.0),
48.0-59.8
87.9 (10.1),
79.4-96.4
75.9 (17.7),
61.1-90.7
12 (6.6),
6.5-17.5
36.0 (9.5),
28.1-43.9
55.0 (4.6),
51.2-58.8
78.8 (18.3),
63.5-94.1
73.7 (23.6),
53.9-93.3
4 (22.2)
8 (44.4)
5 (27.8)
1 (5.6)
10 (29.4)
16 (47.1)
7 (20.6)
1 (2.9)
12.1 (7.1),
8.6-15.6
39.1 (13.9),
32.2-46.0
56.6 (3.7),
54.8-58.5
81.9 (13.2),
73.4-88.0
60.3 (24.4),
48.1-72.4
11.8 (6.6),
9.5-14.1
38.1 (12.7),
33.7-42.6
55.6 (4.8),
53.9-57.3
81.9 (14.7)
76.8-87.1
67.1 (23.4),
58.9-75.2
Non-Responders
(n=383)
32.0, 6.0 (18-39)
237 (61.9)
105 (27.4)
119 (31.1)
13 (3.4)
21 (5.5)
45 (11.7)
34 (8.9)
23 (6.0)
8 (2.1)
5 (1.3)
10 (2.6)
212 (55.4)
271 (70.8)
164 (42.8)
143 (37.3)
* HADS = Hospital Anxiety and Depression Scale, UCLA-LS = University of California Los Angeles – Loneliness Scale, PCL-C = Posttraumatic Stress Checklist – Civilian
Version, PTGI = Posttraumatic Growth Inventory, CBI-B = Cancer Behaviour Inventory – Brief Version, PEP = Standard Psychoeducational Psychotherapy Group, NIEFP =
Narrative-Informed Emotion-Focused Psychotherapy Group, TAU Control = Treatment As Usual Control Group
50
Table 2: AYA Endorsement of Important Reasons for Participation in Trial at Intake (n=34) Level of Importance (percentage)
Reason for Participation
Very
Somewhat
Not at all
Be with people who understand me
70.6
23.5
5.9
A place to express feelings
61.8
35.3
2.9
Learn how others cope
61.8
38.2
0
Get relief from troubling feelings
55.9
41.2
2.9
Learn how I can better cope
55.9
32.4
11.7
Receive support from others
55.9
35.3
8.8
Make friends
55.9
32.4
11.8
Make important changes in my life
58.8
32.4
8.8
Feel less alone or isolated
52.9
29.4
17.6
Get information about cancer
47.1
41.2
11.8
Seek increased meaning in life
41.2
35.3
23.5
Get away from problems
29.4
38.2
32.4
51
4.3 Study Aim #1 – “Proof of Concept”
To explore the Proof of Concept study aim, data from the Winter and Spring session
groups were aggregated to provide a larger sample size for comparisons. This is outlined in
Figure 4.
Figure 4: Simplified CONSORT Diagram Demonstrating Combination of Initial
Randomization (Winter) and Re-randomized (Spring) Chat Groups to Explore Aim #1
*Collected week 2-8 **Collected week 10 ***Collected Week 1,4,7,9
4.3.1 Content Sensitivity
4.3.1.1 Topic-Specific Suitability
Participants rated the suitability of each weekly topic (workbook chapter or video clip)
immediately following each chat-group session. As not all participants attended each session, all
percentages reported below are valid percentages, with weekly topic-specific suitability data not
appearing for non-attendees although they may have still viewed the PEP/NIEFP content.
52
Overall participants highly endorsed weekly “use” of the PEP/NIEFP content. Of AYAs
attending the PEP chat group each week, the mean percentage of participants who read the
workbook before the chat group was 76.5% (Range=66.7%-91.7%) with 20.6% (Range=14.3%33.3%) reading the chapter during the chat group. Only in week 3 and 7 were there chat-group
participants who did not read the workbook chapter at all (7.7% and 8.3% respectively). For
AYAs attending the NIEFP chat group each week, the mean percentage of participants who
viewed the video before the chat group was 88.2% (Range=75.0%-100.0%) with 11.8%
(Range=7.7%-25.0%) viewing the video during the chat group and no chat-group attendees
indicating that they did not view the video that week.
The weekly “feeling” item offered 6 potential responses including bored, frustrated,
neutral, validated and excited to discuss with the group. In the PEP group, overall endorsement
of each “feeling” across all 8 chat groups that used a workbook chapter was bored=4.4%,
frustrated=1.5%, neutral=39.7%, validated=32.4% and excited to discuss with the group=22.1%.
In the NIEFP group, overall endorsement of each “feeling” across the 8 chat groups that used a
video clip was bored=0%, frustrated=7.8%, neutral=37.7%, validated=33.8%, excited to discuss
with the group=20.8%. Participants in the PEP group were most bored with the workbook
chapter related to Healthy Lifestyle (22.2%), frustrated with the workbook chapter related to
Social Support (8.3%), neutral about the Coping and Mindfulness workbook chapter (75.0%),
validated by the Problem Solving chapter (71.4%) and were most excited to discuss the
Communication chapter with the group (50.0%). Participants in the NIEFP group were not bored
with any of the video clips but were most frustrated by the Relationship video clip (16.7%),
neutral about the Reintrajectorization (survivorship) video clip (55.6%), validated by the
Transitions video clip (46.2%) and excited to discuss the Time to Reflect, Refocus and Rebuild
video clip with the group (33.3%).
The “Relevance of Content” factor (3 items) was not normally distributed and therefore
we used the median and range to rank the relevance of each weekly topic (Table 3). Overall,
participant responses to the PEP/NIEFP content ranged from “slightly” to “highly” relevant (2-4)
with the median of most topics falling within the moderately to highly relevant range (3-4) and
no topics being deemed as irrelevant by any participant (1). The top three most relevant topics
from the PEP group were Communication (n=6, M=3.83, Range=2.3-4.0), Ways of Thinking
53
(n=7, M=3.33, Range=2.3-4.0) and Relaxation and Imagery (n=13, M=3.33, Range=2.3-4.0).
The top three most relevant topics from the NIEFP group were Isolation (n=8, M=3.67, Range
2.3-4.0), Taking Time to Reflect, Refocus and Rebuild Your Life (n=6, M=3.67, Range=3.0-4.0)
and Relationships (n=6, M=3.50, Range= 2.3-4.0). The least relevant topics, both with medians
in the “slightly” relevant range, were Coping and Mindfulness (n=8, M=2.67, Range=2.3-4.0)
for the PEP group and Identity (n=10, M=2.83, Range=2.0-4.0) for the NIEFP group.
54
Table 3: During the Intervention, AYA Cancer Participants’ (n=34) Ratings of Weekly Topic-Specific Relevance by
PEP/NIEFP Group (Winter/Spring Chat-groups Combined, Ranked by Median)
Standard PEP Model (n=17)
“Relevance”
Topic
Median (Range)
Communication (n=6)
3.83 (2.3-4.0)
Ways of Thinking (n=7)
3.33 (2.3-4.0)
Relaxation and Imagery (n=13)
3.33 (2.3-4.0)
Social Support (n=12)
3.17 (2.3-4.0)
Healthy Lifestyle (n=9)
3.33 (1.3-4.0)
Goal Setting (n=6)
3.00 (2.3-4.0)
Problem Solving (n=7)
3.00 (2.3-3.7)
Coping and Mindfulness (n=8)
2.67 (2.33-4)
Modified NIEFP Model (n=17)
Topic
Time to Reflect, Refocus, Rebuild (n=6)
Isolation (n=8)
Relationships (n=6)
Transitions (n=13)
Fear and Uncertainty (n=13)
Reintrajectorization (n=9)
Video Trailer (n=12)
Identity (n=10)
55
“Relevance”
Median (Range)
3.67 (3.0-4.0)
3.67 (2.3-4.0)
3.50 (2.3-4.0)
3.33 (2.0-4.0)
3.33 (2.0-4.0)
3.00 (3.0-4.0)
3.00 (2.0-4.0)
2.83 (2.0-4.0)
4.3.1.2 Overall Suitability
The “suitability for learning and discussion” factor significantly differed between
PEP/NIEFP groups (n=20, U=78.5, p=0.029) with the video content deemed more suitable (PEP
median=2.9, IQR = 2.35; NIEFP median=4.1, IQR = 2.40). Participants wrote responses to
open-ended questions in the weekly Topic-Specific Suitability surveys. Analyses of these
responses related to the PEP group (n=30 responses over 8 weeks) identified 3 general themes:
too general, good for future reference and did not guide discussions.
Within the “too general” theme, 7 responses indicated that the weekly workbook chapter
was too general and lacked specifics on how to apply the concepts in daily life.
The workbook selection (mindfulness) is applicable to me, but the
use of a workbook to promote mindfulness wasn't all that helpful.
There were just too many platitudes and simple answers in the
reading without focus on how to actually apply these concepts to
my life. i.e. - "Find the silver lining" - what does that even mean?
How am I supposed to do that?
The comments in the workbook were so general that they ceased to
have any meaning to me. Everybody already knows this stuff - eat
well, exercise, laugh. Reading it again doesn't help. A recipe and a
joke book would have been better.
Six responses suggested that workbook information was useful to have for future reference or
motivation.
I have done goal setting a lot in the past for things in the immediate
future, but never for long term. I will try to work on that in the
future.
Four responses identified that the workbook content was not used to guide the chat group
conversation that week.
We didn't spend much time on the workbook this time. We were
talking about other things. Hope that's okay. I enjoyed that we
didn't stick to the structure when other things were coming up for
people.
There were fewer open-ended responses collected in the weekly Topic-Specific
Suitability Surveys about the NIEFP content (n=8 responses over 8 weeks), but two themes did
emerge: stimulated group discussion and production quality. Specifically, 4 responses indicated
56
that the weekly video clip helped to stimulate the group discussion that week and 2 responses
(from the same participant) reported that it was hard to engage with the content of the videos as
the “lower” production quality was distracting.
Thematic exploration of follow-up interview responses (n=12) concerning content
suitability affirmed the results of the weekly open-ended responses. Two themes were endorsed
by more than 2 participants; conversation starter, specific topics played smaller role in group
discussion than AYA stories. All participants interviewed (n=12) indicated that the PEP/NIEFP
content (workbook or video clip) often played a relatively small role in the discussion each week,
however, 8 interviewees indicated that it was beneficial to have the content to start the discussion
or reinvigorate discussion if it was flagging. Three interviewees, representing both PEP/NIEFP
groups, specifically identified that the content might be better used as a tool for the facilitator to
draw attention to when a specific topic arises in conversation, instead of using it in a prescribed
way where a new topic is identified for each week.
It was not so much the topic, but the people in the video… some
people I connected with much more than others. The themes and
topics were not as helpful, it was the personal experiences that I
connected with the most.
I thought the video was good… it gave a focus point for the chat
room because sometimes you don’t know what to talk about. It
helped always knowing what to expect in the chat that day because
of the video. It helped set the expectations. I liked knowing what
was coming up.
I didn't find the workbook beneficial. Sometimes I would just read
through things towards the end because we really didn’t follow
them… it was the major experiences in peoples lives each week
took over the content… but it is nice to have a topic so that it is not
dead when you first come in… right off the bat someone would
bring something up… maybe in another setting with a different
group it would have been better used.
57
4.3.2 Group Processes
4.3.2.1 Group Cohesiveness
Group cohesiveness scores were aggregated across the Winter/Spring chat-group
sessions. Change in group cohesiveness over time, as an indirect measurement of PEP/NIEFP
impact, is presented graphically in Figure 5. Both psychotherapy models (PEP: slope=0.66,
SD=2.87; NIEFP: slope=1.66, SD=3.49) appeared to increase group cohesiveness across the 10week trajectory; however, the NIEFP groups report a greater increase in group cohesion as
compared with the PEP groups, with a moderate effect size (d=-0.31). The mean group cohesion
pre/post change score (calculated from the slope data) for the PEP groups were 7.92 (SD=34.4)
and 19.89 (SD=41.9) for the NIEFP groups. These change scores suggests a moderate increase in
group cohesion for both the Winter/Spring NIEFP groups as compared to the Winter/Spring PEP
groups.
Figure 5: Overall Change in Group Cohesion for the PEP/NIEFP Groups Over 10 Week
Intervention (Winter/Spring Chat-groups Combined, n=28)
*PEP Model (n=13), NIEFP Model (n=15)
58
4.3.2.2 Group Experiences Inventory
Overall, the distribution of valued group experiences was different between the
PEP/NIEFP groups. Specifically, participants in the PEP groups endorsed all GEI constructs at a
lower level (Median Range=1.00-3.39) as compared to the NIEFP groups (Median Range=3.04.25). We represent these differences graphically in a radar chart (Figure 6) using standardized
variables so that the constructs are on the same scale for the two groups. Two constructs were
statistically different with participants in the NIEFP participants rating Expressing True Feelings
(n=20, U=86.5, p=0.004) and Developing a New Attitude (n=20, U=77.5, p=0.035) as more
important to them in their chat-group experience than the PEP group participants (Table 4).
Expressing True
Feelings
Accessing
Information/Advice
80
70
60
50
40
30
20
10
0
Sexuality
NIEFP
PEP
Developing a New
Attitude
Supportive Contact
Figure 6: Radar Chart of Valued Group Experiences for PEP/NIEFP Intervention Groups
(Winter/Spring Chat-groups Combined, n=20)
Overlapping radar plots of ranked percentile scores (fractional rank percent). Further distance
from the centre indicates greater endorsement of each valued group experience. By examining
shape, readers can perceive the constellation of experiences most salient for each group.
* PEP n=10, NIEFP n=10
59
Table 4: Valued Group Experiences Inventory by PEP/NIEFP Intervention Groups (Winter/Spring Chatgroups Combined, n=20)
PEP (n=10)
Median, 25th-75th Percentile
NIEFP (n=10)
Median, 25th-75th Percentile
MannWhitney U
Expressing True Feelings
3.5 (2.8-3.8)
4.3 (4.0-4.5)
86.5*
Developing a New Attitude
2.9 (2.6-3.5)
3.6 (3.4-4.0)
77.5*
Discussing Sexual Concerns
1.0 (1.0-2.3)
3.0 (1.8-3.3)
75.0
Accessing Resources and Advice
3.3 (2.5-3.8)
3.8 (3.5-4.1)
71.5
Establishing Supportive Contact
3.4 (2.9-3.9)
3.6 (3.4-3.8)
64.5
* Significant at p < 0.05 60
4.4 Study Aim #2 – “Appropriateness of Psychosocial Outcomes”
To explore the Appropriateness of Psychosocial Outcomes study aim, two analyses were
completed; a pre/post LMM analysis for all three initial randomization groups and a LMM
analysis of the two initial randomization groups over 9 months. This is outlined in Figure 7.
Figure 7: Simplified CONSORT Diagram Demonstrating Samples Used for LMM Analysis
of Psychosocial Outcome Data
ITT = Intent-to-treat, LMM = Linear Mixed Model
4.4.1 Pre/Post Analysis of Initial Treatment and Control Groups
The pre/post psychosocial outcome trajectory for the initial randomization groups (PEP,
NIEFP, TAU control), as calculated using LMM of the total score for each measure, is
represented graphically in Figure 8. Table 5 outlines the pre/post change scores and effect sizes
(Cohen’s d) for each psychosocial outcome. Comparing between the two treatment groups, the
NIEFP group improved coping self-efficacy (CBI-B, d=-0.52) and trauma symptoms (PCL-C,
61
d=0.31), more than the PEP group with moderate effect sizes. In the NIEFP group, participants
also maintained their baseline distress-symptom levels while those in the PEP group worsened
slightly with a small effect size (HADS, d=0.26). Comparisons of the groups yielded near zero
effect-size differences for loneliness (UCLA-LS) and post-traumatic growth (PTGI). Comparing
between the PEP group and the TAU control group, there is a small to moderate effect size
across all measures (d = 0.09 - 0.79) that favored the TAU control group. However, comparing
the NIEFP to the TAU control group indicated that participants improved their trauma symptoms
(PCL-C, d=-0.21) more in the NIEFP group with a small effect size while participants in the
TAU control group improved their loneliness (UCLA-LS, d=0.68) more with a moderate effect
size. Chapter 5 discusses the clinical significance of these effect sizes.
62
Table 5: For Initial Randomization of AYA Cancer Participants (n=34), Estimated Pre/Post Mean Change Scores for
Psychosocial Outcomes
PEP (n=8)
M, SD
NIEFP (n=8)
M, SD
Cohen’s d
(PEP vs.
NIEFP)
TAU Control
(n=18)
M, SD
Cohen’s d (PEP
vs. TAU
Control)
Cohen’s d
(NIEFP vs.
TAU Control)
HADS
2.58 (6.82)
0.80(6.82)
0.26
0.71(4.41)
0.33
0.02
UCLA-LS
1.06 (3.54)
0.73 (3.54)
0.09
-1.30 (2.29)
0.79
0.68
PCL-C
2.97 (10.13)
-0.12 (10.13)
0.31
1.67 (6.53)
0.15
-0.21
PTGI
-2.94 (13.69)
-3.18 (13.69)
0.02
-1.93 (8.80)
-0.09
-0.11
CBI-B
-5.55(12.50)
1.0 (12.50)
-0.52
2.28 (8.06)
-0.48
0.14
* HADS = Hospital Anxiety and Depression Scale, UCLA-LS = University of California Los Angeles – Loneliness Scale, PCL-C = Posttraumatic Stress
Checklist – Civilian Version, PTGI = Posttraumatic Growth Inventory, CBI-B = Cancer Behaviour Inventory – Brief Version, PEP = Standard
Psychoeducational Psychotherapy Group, NIEFP = Narrative-Informed Emotion-Focused Psychotherapy Group, TAU Control = Treatment As Usual Control
Group
63
Figure 8: Linear Mixed Models of Pre/Post Change in Psychosocial Outcomes between Initial Randomization of PEP/NIEFP
Intervention Groups and TAU Control (n=34)
*Workbook = PEP Model (n=8), Video = NIEFP Model (n=8), Control = TAU Waitlist Control (n=18)
64
4.4.2 Longitudinal Analysis of Initial Treatment Groups
The 9-month psychosocial outcome trajectory for the initial randomization intervention
groups (PEP/NIEFP), as calculated using LMM of the total score for each measure, is
represented graphically in Figure 9. Slopes and effect size for the difference in slope (Cohen’s d)
for each psychosocial outcome appear in Table 6. Participants in the NIEFP group improved
more on loneliness, coping self-efficacy, and post-traumatic growth over 9 months compared
with the PEP group (UCLA-LS, d=0.97; CBI-B, d=-0.70; PTGI, d=-1.07) with large effect sizes.
Participants in the NIEFP model also improved on trauma symptoms more than the PEP group
with a small effect size (PCL-C, d=0.32). Both PEP/NIEFP groups increased equally in their
distress levels over the 9 months with an effect size close to zero (HADS, d=0.07). Chapter 5
discusses the clinical significance of these results.
Table 6: For Initial Randomization of AYA Cancer Participants, Estimated Mean Slope of
Change for PEP/NIEFP Intervention Groups (n=16) over 9 Months for Psychosocial
Outcomes
PEP (n=8)
Slope, SD
NIEFP (n=8)
Slope, SD
Cohen’s d
HADS
0.58 (1.89)
0.46 (1.33)
0.07
UCLA-LS
0.51 (0.73)
-0.29 (0.90)
0.97
PCL-C
0.81 (2.64)
0.17 (1.07)
0.32
PTGI
-2.42 (4.28)
0.44 (3.86)
-0.70
CBI-B
-1.26 (3.05)
1.42 (1.77)
-1.07
* HADS = Hospital Anxiety and Depression Scale, UCLA-LS = University of California Los Angeles – Loneliness
Scale, PCL-C = Posttraumatic Stress Checklist – Civilian Version, PTGI = Posttraumatic Growth Inventory, CBI-B
= Cancer Behaviour Inventory – Brief Version, PEP = Standard Psychoeducational Psychotherapy Group, NIEFP =
Narrative-Informed Emotion-Focused Psychotherapy Group
65
Figure 9: Linear Mixed Models of Longitudinal Change in Psychosocial Outcomes for Initial Randomization PEP/NIEFP
Intervention Groups over 9 Months (n=16)
*Workbook = PEP Model (n=8), Video = NIEFP Model (n=8)
66
4.5 Study Aim #3 – Feasibility of Study Design and Procedure
4.5.1 Recruitment and Retention
A combined 9.6% of AYAs in Alberta who had been treated for cancer within six months
prior to study initiation responded to our recruitment methods, almost meeting the target
recruitment rate of 10%. When screened for eligibility, 93% of respondents were eligible with
only 3 not meeting inclusion/exclusion criteria (2 = poor English, 1 = too sick as reported by
family members). Of those eligible, 92% consented to be part of the study and completed the
baseline questionnaire with 3 participants declining participation (1 = could not commit, 2 =
incomplete enrolment). After randomization to intervention/control groups, 88% of both the PEP
and NIEFP groups (n=7 for each group), and 94% of the TAU waitlist control group (n=17)
completed the 3-month follow-up questionnaire. For both initial intervention groups 63% (n=5
for each group) filled out the 6-month follow-up questionnaire with 50% of the PEP group (n=4)
and 63% of the NIEFP group (n=5) filling out the final 9-month follow up questionnaire. After
re-randomization of the TAU control group at 3 months, 88% (n=7) of the PEP group and 100%
(n=9) of the NIEFP group completed the 6-month follow-up questionnaire with 63% (n=5) and
100% (n=9) completing the final 9-month follow-up, respectively. In all cases, except for the 9month follow up of the initial PEP group, recruitment and retention percentages were within 5%
of the target value (See Figure 2 for details).
4.5.2 Attendance
The average weekly attendance across all chat groups was 58.6% (SD=8.63%) In real
terms, an average of 5 AYAs (SD=1, Range 3-9) participated in the chat group each week. There
was a lower average weekly attendance in the Spring/Summer session (56.1%) as compared to
the winter session (64.4%). Figure 10 shows the trajectory of weekly attendance over time
between model groups and the overall sample. Unpublished CCC operational (April 2015 –
March 2015) data reported a 70% average weekly attendance, which means the weekly
attendance feasibility target was not met.
67
Figure 10: Trajectory of Weekly Attendance Percentage Overall and by PEP/NIEFP
Intervention Groups (Winter/Spring Chat-groups Combined)
* ITT = Intent to Treat, includes all participants randomized to each group. Low attendance participants (0-2
sessions attended) resulted in no group ever having 100% attendance.
The average number of chat-group sessions attended by each participant throughout the
10-week intervention was 6 (SD=3, Range=0-10). Categorically, 20.6% of study participants
attended 9-10 sessions, 35.3% attended 7-8 sessions, 17.6% attended 5-6 sessions and 26.4%
attended 4 or less. Published CCC data from 2013 found that 44% of their participants attended
9-10 sessions, 25% attended 7-8 session, 12% attended 5-6 sessions and 19% attended 4 or less
(20). Figure 11 provides a graphic comparison of the attendance percentages from these studies.
Figure 12 compares number of sessions attended by PEP/NIEFP group and demonstrates that the
NIEFP groups had stronger attendance overall then the PEP groups. In the follow up interviews
conducted (n=12), all reasons for missing individual chat-group sessions were easily grouped
into 4 primary themes: last-minute scheduling conflicts, childcare responsibilities, vacations, and
general forgetfulness.
68
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
9-10 sessions
7-8 sessions
5-6 sessions
3-4 sessions
2 or less
CCC
44%
25%
12%
7%
12%
AYA
21%
35%
18%
9%
18%
Figure 11: Categorical Attendance Levels of Published CCC data (n=351) and AYA Pilot
Study (n=34) by Number of Sessions Attended
* CCC = CancerChatCanada operational data, AYA = data from this study
Low intervention attendance was defined as those who attended 2 chat-group sessions or
less throughout the 10-week chat group intervention. In total there were 6 participants who had
low attendance (17.6%): 2 participants attended 2 sessions, 4 participants attended no sessions.
Unpublished operational data for CCC (April 2014 – March 2015) found a low attendance rate of
15.0% based on the same criteria, so this feasibility criteria was met. Four of the low attendance
participants were from the PEP groups and 2 from the NIEFP groups; however, one from the
PEP group, was due to a participant death. Comparing the demographic, medical and
psychosocial characteristics of the low-attendance participants with others revealed no
meaningful differences for most variables; however, they did all live in an urban setting, were
treated at the Tom Baker Cancer Centre in Calgary and were within 12 months from initial
diagnosis at intake.
69
40%
35%
30%
25%
20%
15%
10%
5%
0%
9-10 sessions
7-8 sessions
5-6 sessions
3-4 sessions
2 or less
PEP
18%
35%
18%
6%
24%
NIEFP
24%
35%
18%
12%
12%
Figure 12: Categorical Attendance Levels between PEP/NIEFP Intervention Groups by
Number of Sessions Attended (Winter/Spring Chat-groups Combined)
* Standard = PEP Group, Modified = NIEFP Group
Five follow-up interviews were conducted low-attendees (“What was your primary
reason for not continuing with the chat group? Were there any other reasons?”). One participant
identified treatment-related conflicts (chemotherapy on the same day) and one specified a
negative reaction to the online chat-group format, as the primary reason for low attendance.
Three participants identified scheduling conflicts with other activities (sports league,
mindfulness class, childcare) as their primary reason for low attendance, with 2 of them also
indicating that had they known it was not possible switch to another chat group after
randomization, they would not have committed to the conflicting activities. Only two
participants identified secondary reasons for low attendance, with one specifying time of day
(7pm – putting children to bed) and the other indicating technical challenges (unable to
70
download readings and difficulty following the discussions due to connection resetting), as
contributing factors.
4.5.3 System Usability
The overall mean SUS score for the CCC platform was 88.0 (SD=9.7), which correlates
with a high B average using the university grade analog discussed in Chapter 3. Both groups
reported that the CCC platform was satisfactorily usable, with no significant difference
(t(19)=0.794, p=0.44) between the SUS scores of the PEP group (M=89.8, SD=9.9) and the
NIEFP group (M=86.4, SD=9.7). Participant comments from the follow-up interviews (n=12)
supported the SUS rating. Three comments highlighted potentially negative aspects of system
usability: awkward on a cell phone when travelling in a vehicle (cutting in and out as the mobile
device linked from tower to tower), a small cell phone screen made it very difficult to track
conversations, and a “currently typing” function would be useful to help the conversation flow.
Other more positive comments related to system usability included: an ability to participate in a
chat group from anywhere (i.e. another province and continent), and being able to go back and
read transcripts from past chat groups.
There was no significant difference in ease of use of content (t(18)=-0.629, p=0.54)
between the PEP group (M=3.9, SD=1.4) and NIEFP group (M=4.2, SD=0.63). The majority of
participants in both groups agreed or strongly agreed that the workbook chapters or the video
clips were easy to review before each chat group session. However, one PEP participant
disagreed and one strongly disagreed that the workbook was easy to read before each session. In
the NIEFP group no participant disagreed.
Thematic exploration of the qualitative responses concerning model usage (“What was
your experience of using the workbook/video clips”) in the follow-up interviews (n=12) yielded
only 2 relevant “Ease-of-Use” themes: format and engagement preference. Of the 6 PEP group
interviewees, none reported issues in accessing the weekly chapters with 3 participants indicating
their preference for printed copies of the workbook chapters and 3 preferring to view them
digitally. Of the 6 NIEFP group interviewees, none reported technical issues with watching the
weekly video clips, with 3 preferring to watch on a computer and 3 preferring iPads or mobile
devices. For both PEP/NIEFP groups, there was no clear preferred way to engage with the
content.
71
4.5.4 Overall Participant Satisfaction
Overall, participants in both the PEP group (n=10, Median=5.0, 25th Percentile=4.0, 75th
Percentile=5.0) and the NIEFP group (n=10, Median=5.0, 25th Percentile=4.75, 75th
Percentile=5.0) were satisfied with their chat-group experience. Of those who completed the
final post-chat survey (n=20), 70% strongly agreed, 25% agreed and 5% slightly agreed with the
statement that overall, they were satisfied with their chat-group experience. No participants
disagreed with the statement. Overall participant satisfaction did not differ significantly between
PEP/NIEFP groups (U=42.0, p=0.451). Unpublished CCC operational (April 2014 – March
2015) data found that in response to a similar single question, 68% of participants were Very
Satisfied, 27% Somewhat Satisfied and 5% Not at all Satisfied. The participant satisfaction
feasibility target was met.
An open-ended question in the final post-chat survey (week 10) asked, “what was the
best thing about the group for you.” In 15 of the 18 responses to this question, participants
indicated that they felt supported and less alone because of their chat-group experience.
Responses from both PEP and NIEFP groups were very similar as these quotes illustrate:
Talking with other people who are going through what I am
experiencing, thinking that I was alone in some of my thoughts
only to find out that most people have the same feelings.
Being able to hear that the feelings I thought were weird or dark or
strange were not unique… and hearing about others stories.
I never really addressed the emotional side of my cancer
experience. Having a group of people who could talk about the
same problems as me without them feeling bad for me was great. I
am a stronger person from this group.
Feeling support from other group members and learning about
their experiences. It made me feel less alone.
Another open-ended question in the final post-chat survey asked, “what do you wish this
group had done to meet more of your needs.” Feedback was provided by17 participants and of
these, 7 said nothing could be improved, 2 thought it was too short and 2 indicated that they
72
would have liked to meet people face-to-face. Other ideas mentioned were narrowing the age
range, recruiting more men and better introductions at the beginning to get to know each other.
In the follow-up interviews (n=12), questions related to participant satisfaction addressed
the topics of communicating online, intervention format, professional facilitation, confidentiality,
and feeling understood by group members (Appendix C.2). The majority of interviewees (n=11)
found their experience communicating online satisfying despite 4 of them highlighting
challenges which they overcame with more experience and good facilitation. Participants also
talked about the challenges of communicating on-line. These comments included talking about
the technology itself, the social aspects of connecting with others without visual cues (coupled
with following multiple conversations) and personal health.
Technology:
I don’t do chat rooms or anything like that but I know how to use a
computer, so it wasn’t difficult.
It was good… it was certainly challenging… The IT is a little out
of date. I thought there must be a better platform. When we had 3
people it worked well, but when we had 8 it was really difficult…
when you reply to something you don’t want to interrupt another
conversation. A “currently typing” indicator would be useful…
maybe use a platform where you can keep track of the different
conversations.
Social Connection:
It was good communicating online… not the same as a phone call
or face-to-face… a little more disconnect, but in a way people open
up a little bit more.
It is hard to get a sense of a person when just seeing words.
It was easier than I thought. I didn’t know what to expect but
everyone was easy to talk to and open.
I found it hard to relate to people simply because I couldn’t see
them… and the conversation gets all over the place… even
telephone would have worked better for me personally with the
natural pauses.
A little bit of both… it was way more satisfying then I thought it
was going to be because I am a face to face person. I developed a
73
deeper connection than what I thought it was going to be, but it
was different… normally only one person is talking at a time and
the multiple conversations were a little difficult. It wasn’t my most
preferable…
Personal Health:
Probably the biggest challenge was when I found out I had it
(cancer) again. It was a little difficult to chat about it in the
group… but as soon as I said it everyone was so supportive, which
was nice.
All interviewees (n=12) were satisfied with the chat-group format of one, 1.5-hour
session per week for 10 weeks managed by a professional facilitator.
Once a week was awesome (not too much), sometimes a lot can
happen in a week, so it didn’t seem like you were struggling to find
things to say or have too many things to say. It was great timing…
with chemo brain it was hard to remember back to far… I would
love for it to go on for ages. When I realized that it was our last
day and I burst into tears… it has become something that is part of
my routine now, so I wanted it to keep going.
Three themes related to facilitator satisfaction were identified; keeping conversations on
track (n=7), including quieter group members in the discussion (n=6), and summarizing
discussions (n=3).
It was good to have her guide us. If there were multiple
conversations going on, Lisa would slow things down and then
move on to the next topic that people were addressing… she would
pick up on things in a statement that I wouldn’t have. She
definitely had a way of pulling out stuff. She would dig a little
deeper for us and not feel awkward.
She was good that no one’s point or question got missed… people
didn’t feel like they were missed and it helped people feel that they
were listened to.
It was really good having the facilitator, especially at the
beginning. It would have been a lot more small talk instead of the
topics that were brought up… she was definitely really
encouraging and helped to summarize the comments that people
were making overall… added some continuity.
74
All 12 participants indicated that they were satisfied with their freedom to talk about
private things with the group, with 8 reporting this happened very quickly and 4 indicating that it
took 3-4 sessions to open up. Most interviewees (n=10) were satisfied the group understood
them. However, 2 participants, one who had young children and one who had metastatic disease,
were dissatisfied and did not feel the group understood their issues. They both indicated that
even though group members tried to understand their situations, specific aspects of their
experiences were too unique to be understood easily.
4.5.5 Statistical Power Calculation
To explore the sample size necessary for a larger RCT testing the difference in
psychosocial outcomes between model groups, we did a power calculation using the pre/post (3month) standard deviation in PCL-C total score. On the PCL-C scale a 5-10 point change is
reliable (i.e. not due to chance) and a 10-20 point change is clinically meaningful (185).
Therefore, the developers of the PCL-C recommend using 5 points as a minimum threshold for
determining whether an individual has responded to treatment and 10 points as a minimum
threshold for determining whether the improvement is clinically meaningful (179). We
performed a power calculation using the 5-point minimum threshold, the standard deviation
estimates from this pilot study, a Type I error level of 5% and a power level of 80%. Using these
parameters an estimated a sample size of 90 participants (approximately 30 in each of the three
study arms) would be necessary to find a significant difference in pre/post PCL-C outcomes
between PEP/NIEFP groups. However, as the PCL-C pre/post changes scores in this pilot study
were less than 5 points, and there are inherent problems with estimating a population parameter
(SD) from a very small sample, approximately 180 participants (60 participants in each of the
three study arms) would be the recommended sample size. This equates roughly to 7 chat-groups
in each PEP/NIEFP arm.
4.5.6 Matched Qualitative Responses and Psychosocial Outcome Data
The content of the follow-up interviews strongly supported the appropriateness of the
quantitative measures used. The three questions relating to psychosocial outcomes were: 1)
“looking back to before starting the chat, can you identify anything that has changed for you”, 2)
“have you made any practical changes as a result of being in the group”, and 3) “looking into the
future, do you expect anything to be different for you now as a result of the group.” Participant
75
responses to these questions (n=12) were matched to all 5 validated psychosocial outcomes
measures used in this study.
Ten participants indicated feeling less ‘lonely’ and/or ‘isolated’ (UCLA-LS) by their
cancer experience as a result of participation in the study and being able to share their
experiences with other young AYAs:
I learned that other (AYAs) people have the same challenges and
yeah, just learned about the process of cancer and how people are
dealing with it… sometimes it was just nice to chat about it
without making someone else uncomfortable… sometimes when
you are talking to friends and family they are uncomfortable
because they are upset or worried.
It gives me a wake-up call to put things in perspective with how
many people are facing the same things… feeling that you are not
alone, or its so bad after all…
Additionally, 3 participants identified exchanging emails and contact information with other
group participants, 2 had met face-to-face with another participant and 4 had connected on
Facebook with other participants. Only 2 interviewees had not connected at all with other
participants outside of the chat group. Generally, this indicates a widening of their social
networks likely leading to a decrease in feelings of loneliness.
Ten participants reported improvement in areas related to coping self-efficacy (CBI-B)
such as maintaining a positive attitude (n=4), confidence in coping skills (n=6) and ability to
participate in medical/psychosocial care (n=2).
My outlook on life has changed positively… the biggest thing is
speaking to people my own age, it makes a huge difference and
helps develop a sense of normalcy.
I think that in general the view that there is a path forward and that
I can get through this. I am a little more positive about the future.
I am definitely more comfortable with some of the things that are
about to happen. We talked about what happens when the
treatments are done, and I feel better about that process. It is nice
to hear from people who have been there already.
76
Seven participants identified themes related to posttraumatic growth (PTGI) as a result of
the chat groups, including new possibilities (n=2), appreciation for life (n=3), relating to others
(n=5) and spiritual change (n=1).
I don’t feel like a burden anymore… I don’t know how to explain
that really. Before no one else understood what I was feeling so if I
had a bad day I wasn’t supposed to melt down or something. I
don’t feel pressured to pretend that everything is normal
anymore… I thought that I had to over explain myself before, but I
don't feel pressured to justify things anymore because I am not the
only one.
I learnt about other people’s experiences… some more religious
and faith based… opened my eyes to other’s experiences.
Three interviewees made comments about becoming ‘less stressed’ as a result of being able to
express things that were causing them anxiety (HADS).
I think that I am less pent-up angry because I had a way to talk
about my feelings. I live on my own and didn’t have anyone to say
that type of stuff to. Friends can only handle so much emotional
dumping… I am a lot less stressed… had a release that wasn’t on
any of my caregivers.
One participant, who experienced a recurrence during the chat group, reported feeling less
traumatized by the experience because of the support she received in the chat group (PCL-C).
I had a re-occurrence happen during this time… it {chat group}
probably helped me talk {about cancer} more openly… I was able
to process things better with my reoccurrence.
Chapter 5 discusses the implications of these qualitative findings in regards to the
appropriateness of the psychosocial outcomes chosen.
77
Chapter Five: Discussion
5.1 Overview
In this feasibility study we conducted an RCT of two online support group models
(PEP/NIEFP) over ten weeks as compared to a control group. We measured differences in
content sensitivity and group process to explore if these two models created a different support
group experience for participants (study aim #1). We also measured pre/post changes in
important psychosocial outcomes across all three groups (PEP/NIEFP/TAU) and over 9 months
for the two initial intervention groups. We found that this study design is feasible for a larger
RCT, with some modifications, having met almost all of our recruitment, retention, attendance
and participant satisfaction targets. We also found that the online chat platform and PEP/NIEFP
resources were usable and that participants rated the NIEFP content as more suitable than the
standard PEP content. Important group psychotherapy processes (group cohesion and
engagement) also favoured the NIEFP model over the PEP model. Differences in effect sizes
between psychosocial outcomes in the PEP/NIEFP groups were small to moderate over the first
three months with the NIEFP group appearing to improve more in coping self-efficacy and
trauma than the PEP group. However, over 9 months the differences became more pronounced
(moderate to large effect sizes) with the NIEFP model participants appearing to improve more on
coping self-efficacy, trauma, loneliness and posttraumatic growth. We could test none of these
psychosocial outcomes for statistical significance due to the small sample size.
5.2 Study Aim #1 – “Proof of Concept”
Overall, the “proof of concept” data collected through this pilot study led us to three
important conclusions about AYA online group psychotherapy. First, a NIEFP model modifies
important group processes and creates a different chat-group experience for participants than the
PEP model. Second, group cohesion is an important indicator, and might be valuable as either a
process or an outcome measure. Third, listening to and sharing cancer stories is an important
aspect of the group psychotherapy experience for AYAs and may be a central therapeutic
mechanism. Taken together, these results provide a strong “proof of concept” and rational to
move forward with a larger randomized trial comparing these two online intervention models.
78
Our NIEFP model uses storytelling and emotional expression as the primary ingredients.
The psychoeducational group psychotherapy model, though open to emotional expression and
emerging group member issues, provides structured learning objectives and topics for weekly
group discussions. One of the central reasons for developing the new AYA NIEFP model and
examining it in this study was the hypothesis that AYAs would find it more suitable, and that it
would enhance group process and lead to better outcomes. The “proof of concept” results
presented in this thesis support that hypothesis, starting with the reasons participants gave for
their interest in the study (Table 2).
Participants reported that the most important reasons for joining the study were to “be
with people who understand me”, “a place to express feelings” and “learn how others cope”; all
group processes that might be facilitated better by an NIEFP model. Conversely, other reasons
for joining the study directly that are directly addressed in a psycho-educational model, such as
“learn how I can better cope”, “make important changes in my life” and “get information about
cancer”, were less important than the relational aspects of a group therapy experience. For the
AYAs in this study, the opportunity to “be with” others who authentically understand them was
more important than “learning and making changes”. These reasons for participation add
credence to the selection of an NIEFP model, compared with a PEP model in AYA group
therapy.
In addition to the greater topic-specific and overall suitability, two results specifically
highlight our conclusion that the NIEFP theoretical model is sound and improves the AYA chatgroup experience: significantly higher group cohesion in the NIEFP group and significantly
higher endorsement of specific valued group experiences such as “expressing true feelings”. As
discussed earlier, previous research documents that higher group cohesion can improve
psychosocial outcomes (95). The results of this study indicate that group cohesion could be both
a therapeutic mechanism leading to better psychosocial outcomes and an important outcome in
itself. The most common qualitative response regarding the chat-group experience centred
around the feeling of being less “alone” in their cancer journey. As AYAs place a much higher
importance on peer relationships (22, 112), minimizing the profound sense of isolation felt by
AYAs during their cancer experience and creating a “sense of belonging” should be a primary
goal, not just a means to an end. Based on the literature exploring the psychosocial challenges
79
and unmet needs of AYAs, group cohesion may be one of the most important constructs to
measure in AYA group psychotherapy and we recommend that a measure of group cohesion be
included in future RCTs. Our results suggest that AYAs in the NIEFP group felt both a greater
“sense of belonging” during their chat-group experience and less lonely over time up to 9months. This is another strong indicator that a NIEFP model could be more effective with AYAs
than a PEP model.
Exploration of the valued group experiences also seems to corroborate the use of a
NIEFP model with AYAs. Despite the relatively small numbers (n=20), participants in the
NIEFP group endorsed two GEI constructs significantly more than PEP group participants:
Expressing True Feelings (p=0.004) and Developing a New Attitude (p=0.035). These two
valued group experiences best represent the fundamental concepts of NIEFP, which are “reexperiencing” emotion and “re-storying” (reconstructing) a narrative (115). These outcomes
provide some evidence that an NIEFP model impacts an AYAs experience of online group
psychotherapy.
In addition, the valued group experience of “Expressing True Feelings” specifically
relates in a meaningful way to the AYA cancer experience. AYAs report having a hard time
expressing true feelings to their friends and family, and often reference the relief they experience
when they can connect with their AYA cancer patient/survivor peers (22). Psycho-oncology
researchers have found that emotional suppression can predict poorer psychosocial and
physiological outcomes (186) and therefore emotional expression is an important therapeutic
mechanism in group psychotherapy (108). Any intervention that provides AYAs a greater
opportunity to express their true feelings, could significantly improve outcomes. Continual
refinement of psychosocial interventions, so that they create a better environment for AYAs to
share authentically with each other, is an important future direction in AYA psycho-oncology.
This pilot study provides some guidance for future AYA psycho-oncology research by
highlighting the potential differences between psychoeducational and narrative-informed models
of care.
Another lesson gleaned from this pilot study is that personal narratives seem to be more
important than addressing specific topics and themes in AYA group psychotherapy. In follow-up
interviews with NIEFP group participants, 4 of the 6 participants indicated that the personal
80
stories in the video were more meaningful then the topics chosen for each week. One comment
from the EFP group final post-chat survey effectively summarized this general feeling,
It was not so much the topic, but the people in the video. Some
people I connected with much more than others… it was the
personal experiences that I connected with the most… the themes
were not as helpful.
Another participant from the PEP group wrote,
I found we often would start with the topic but very quickly it
would change into what was most pressing on the minds of people
in the chat instead and how we could support one another with
whatever fears were coming up that week.
Overall, for both models, the topics chosen for discussion each week were a good place to start,
but were less important than the opportunity to listen to other participant’s stories, share their
own stories, and encourage each other.
This might indicate that AYAs absorb information important to their psychosocial health
differently than older adults. One finding supporting this idea is that the NIEFP group endorsed
“Accessing Information and Advice” as a valuable chat-group experience to a greater degree
than the PEP group, even though the PEP group specifically provided more formal “information”
to participants. One possible explanation for this is that AYAs interpret what constitutes
“information” differently. Specifically, AYAs may ascribe the same value or higher value to
information gleaned from peers as they do to information provided by health care professionals.
As other studies have shown that narrative information from peers can bias patient decisionmaking (187, 188), this could indeed be the case. Taken together, AYAs increased desire for
connecting with peers and high valuation of personal narratives could be another reason for their
low usage of standard education-focused psycho-oncology programs. As health care systems are
generally not designed to connect patients with each other, the high value placed on peer
narratives could also explain why AYAs consistently highlight lack of peer support as their
largest unmet need.
Overall, this study strengthens the theoretical foundations of NIEFP with AYAs;
however, the implementation of this conceptual model still needs further study. Viewing
personal narratives as a main active ingredient in group psychotherapy with AYAs, rather than
the specific topics or content area’s covered, is a departure from the dominant psychoeducational
81
model used in cancer care. In Psycho-educational models, the information transmitted is equal to
or more important than peer connections, meaning-making and existential processing. In the
AYA context, providing AYA patient stories may be more important than providing information.
Future studies could use a wide variety of personal AYA narratives to stimulate discussion and
convey information more effectively. Many online video tools focus on topics, presenting
peoples’ stories by theme (i.e. “Truth of It” Video Series by the Canadian Partnership Against
Cancer). This study argues that AYAs may benefit more from hearing stories in their entirety.
To conclude, this study supports the use of a NIEFP model in future AYA group psychotherapy
studies.
5.3 Study Aim #2 – Appropriateness of Psychosocial Outcomes
All analysis of the psychosocial outcome measures is strictly exploratory due to small
sample sizes. Our goals for investigating the psychosocial outcomes were to examine trajectories
over time, estimate levels reported by participants and explore the appropriateness of the
measures used. This study documents that the psychosocial constructs measured were
appropriate but the tools used are not sensitive to short-term changes. We also demonstrate that
a longer follow-up period than a simple pre-post assessment may be necessary to evaluate the
differential impact of group psychotherapy models. We suggest that a moderator analysis may
lead to stronger results, and that a Quality of Life measure should be included due to
participants’ many qualitative comments related to this construct. Additionally, we would select
a different measure of global distress for use in a future randomized trial.
The qualitative findings from the 12 follow-up interviews indicated that the outcome
measures selected for this study appeared to measure the psychosocial constructs important to
AYAs (loneliness, distress, anxiety or trauma, self-efficacy). However, 3-month pre/post change
scores of the psychosocial outcome measures showed almost no clinically meaningful change.
Using the PCL-C as an example, while both the PEP and TAU Control groups increased in their
trauma symptoms and the NIEFP group decreased their trauma symptoms from pre-to-post
intervention, all the changes were within only 3 points on the PCL-C scale; 2 points below the
reliability threshold for treatment response and 7 points below the threshold for clinically
82
meaningful change (179). This lack of 3-month sensitivity in outcome measures occurred despite
qualitative responses in follow-up interviews indicating positive changes in participant’s lives.
However, there was a more substantial difference between psychosocial outcomes
measured at the 9-month follow-up, with larger between-group (PEP/NIEFP) effect sizes (Table
6). In particular, the slope of the UCLA-LS (loneliness), PTGI (posttraumatic growth), PCL-C
(trauma) and CBI-B (coping self-efficacy) scales indicated potentially clinically meaningful
differences in improvement over the 9-month trajectory. Therefore, future group-psychotherapy
RCTs to improve AYA psychosocial outcomes may benefit from more sensitive measures over
the short term. For longer-term outcomes, the measures used in this study may be appropriate.
The one exception to this generalization is the Hospital Anxiety and Depression Scale
(HADS). As a measure of distress, the HADS does not seem to be a sensitive measure with the
AYA population despite wide usage in the distress, anxiety and depression literature. The study
sample was “moderately” distressed at baseline and there was a slight trend towards increasing
distress in both PEP/NIEFP groups over 9 months despite improvements in other outcomes and
qualitative comments indicating decreased distress. To shed more light on this potential
discrepancy, we conducted post-hoc analyses of the anxiety and depression sub-scales. These
indicated that the primary contributor to general distress levels was the anxiety subscale, with the
scores being 2 times higher on average then depression scores. The anxiety scores did decrease
slightly over time, which is consistent with the literature that indicates psychosocial interventions
could have a greater impact on anxiety levels than depression levels (49); however, the
depression subscale scores stayed relatively constant over the 9-month period. This may indicate
the depression subscale in particular is non-responsive to change and/or has low sensitivity in the
AYA population. It might also mean that these groups, despite improving participant’s
symptoms on other outcomes, did not improve participants’ depression levels.
Taken together, these results seem to indicate that the constructs measured and most
scales used were appropriate but a longer follow-up period may be necessary to realize fully the
differential impact of the PEP/NIEFP models on outcomes. As participants were enrolled based
on a diagnosis of cancer, and not psychological morbidity, the full preventative impact of the
group psychotherapy interventions may take time to surface. Combined with the strong retention
rate over a 9-month period found in this pilot study, a 12 month follow-up period (the standard in
83
psycho-oncology interventions) is not only be feasible, but may also be the most valid RCT
design to test group psychotherapy efficacy with AYAs.
A moderator analysis built into a future RCT could also help identify who benefits the
most from these interventions. The fact that most psychosocial intervention studies “take all
comers” often means that average improvement over time is small, but for those highly
distressed the improvement might be large. (54). Due to the small AYA oncology population
size, it would not be difficult to restrict enrolment to only those who demonstrate clinical levels
of certain outcomes (i.e. clinical distress, depression, posttraumatic stress). Potentially, a large,
national multi-site randomized trial could restrict enrolment using baseline cut-off scores, but in
group psychotherapy, heterogeneity of group membership can be a positive attribute (126). A
group of clinically distressed patients may not be the best environment for improving
psychosocial outcomes. Therefore, rather than restricting enrolment using clinical cut-offs, we
recommend a moderator analysis to address these issues of group membership and to clarify who
benefits the most from AYA group psychotherapy.
Finally, participants repeatedly mentioned one construct in the qualitative comments and
follow-up interviews that we had not included in this study, Quality of Life (QoL). In retrospect,
we noted two systematic reviews that found psychometrically validated QoL tools for use with
AYAs (159, 160) and one QoL tool designed specifically for use in AYA psycho-oncology(159,
160, 189). With this good selection of well-validated AYA-specific QoL tools, it was an
oversight to have not included this important measure in this pilot study. Overall, the addition of
a QoL measure and a more sensitive distress measure to the group of measures used in this study
should allow for a comprehensive measurement of outcomes in future RCT studies over the
long-term.
5.4 Study Aim #3 – Feasibility of Study Design and Procedure
In this pilot study, we met most of the recruitment, retention, attendance, and usability
targets, and therefore this specific study design (with modifications) is feasible in a larger RCT.
There was high participant satisfaction across PEP/NIEFP groups, which could lead to better
retention regardless of intervention allocation. However, the data collected through this pilot
study also indicated that important modifications can be made to improve the design.
84
Specifically it is important to use the widest geographic area and all recruitment methods
possible to meet recruitment targets because target population is small. It is also important to
offer the intervention during seasons when it is easiest for participants to attend (fall and winter
based on our assessment) and use incentives to increase weekly attendance rates. Providing
multi-modal access to the PEP/NIEFP resources and improving cross-platform compatibility of
the chat-group platform is also important. We also noted that in order to have the statistical
power to demonstrate differential improvement, a minimum of 90 participants would be needed.
However, due to the small pre/post effect sizes and high variability of the estimated parameters,
we would recommend at least 180 participants. Finally, instead of the waitlist control group, we
would also include a minimal intervention control group and rolling recruitment.
The 9.6% response rate for our recruitment procedures, despite seeming low, was
consistent with what we expected based on similar interventions piloted in Alberta (60). In past
research, recruiting AYAs to psycho-oncology studies is generally very difficult despite high
unmet needs (190). These recruitment challenges highlight that a larger RCT cannot rely on a
single recruitment method, such as a cancer-registry mailing, but must use all means including
direct referrals by health-care providers, patients and community-based organizations, social
media, and advertisement. Additionally, it is likely that moving beyond provincial boarders
might be necessary to recruit an appropriately sized sample.
Treatment adherence (attendance rates) in psycho-oncology interventions is important
because it can impact outcomes (21). Overall, average weekly attendance rates in this pilot study
did not meet expected targets (within 5% of CCC operational attendance rates, 56% vs. 70%).
However, this was the only feasibility criterion that this study did not meet, and instead of
indicating that it is not feasible, this points to a specific aspect of the study design (timing) that
we would need to consider in a larger trial. Specifically, the Spring intervention groups that
continued into the summer months (ended July 30th) had a much lower average weekly
attendance than the winter session (51.6% vs. 64.4%). This intuitively makes sense, with summer
being a more difficult time for any weekly commitments, but it is interesting that this carries
over into an online intervention that participants could access regardless of geographic location.
In a future RCT using this study design, intervention windows could take place during the fall,
85
winter and early spring months to increase the treatment adherence (average weekly attendance
rate), or could assess specific dates that might allow higher participation.
Additionally, use of incentives may help increase weekly attendance rates. In this study,
incentives were provided for completing the 4 primary outcome questionnaires (one $25 gift
card), but this could be expanded to chat-group attendance as well. For example, a $25 gift card
could be given to participants who attend more than 7 chat-group sessions in addition to the $25
gift card for completing all the outcome questionnaires. This type of approach does have ethical
implications and would need to be clearly articulated to institutional ethics review committees
and participants themselves.
As was expected, there were no serious usability issues for either the CCC platform or the
therapeutic-model components. However, we learned three important lessons from follow-up
interviews that specifically relate to the PEP/NIEFP materials and chat-group platform used in
this pilot study. First, in a future RCT, providing the option to receive both a printed or/and a
digital copy of the workbook, may be important. In the pilot study, we assumed that AYAs
would prefer a digital copy and therefore we offered no hard copy. However, three out of six
participants interviewed indicated a preference for a printed copy. Secondly, NIEFP participant
interviews indicated that they were viewing the videos on many different devices from desktop
and laptop computers to iPads and smartphones. Therefore, it is important that a cross-platform
compatible video hosting service is used for the video content. Finally, some participants
reported challenges with participation in the chat group when using a smartphone or tablet and
steps could be taken to ensure the cross-platform compatibility of the CCC platform itself.
Overall, it will be important to provide model materials in all potential modalities and platforms
because there were no clear preferences in the participants interviewed.
Most participants were highly satisfied with their chat-group experience across
PEP/NIEFP groups. This is a good indicator that AYAs would be willing to participate
regardless of study arm randomization, and indicates that participation in PEP/NIEFP groups
should not be differentially impact completion of the study protocol and psychosocial outcome
questionnaires. The high participant satisfaction across groups strengthens our conclusion that
using a randomized design to test these two online group psychotherapy models is feasible.
86
Some of the pilot studies reviewed for this study reported participant satisfaction as an
indicator of study efficacy or platform usability (154, 191); however, we have concluded that
satisfaction should not be linked to either of these outcomes. It is a reasonable indicator of
randomized study-design feasibility, but was unrelated to longer-term outcomes of the
intervention. This distinction is clear from the results of this pilot study, as participant
satisfaction did not appear to differentiate PEP/NIEFP groups as did psychosocial outcomes and
process measures. The quantitative measures used in this study showed trends towards
differences in most measures across PEP/NIEFP groups despite equal satisfaction levels.
Because the amount of psychosocial trauma experienced during cancer diagnosis and
treatment is a good indicator of future psychosocial problems, such as distress and depression
(192), the PCL-C slope was the parameter used to estimate the sample size necessary for a larger
RCT. However, using the PCL-C slope and standard deviation calculated from a sample this
small might only provide a 40% chance of predicting the planned power necessary for an RCT
(193). For this reason, the minimum sample size of 90 participants determined by the power
calculation should be interpreted cautiously; not as the sample size necessary for a larger RCT,
but as a measure of feasibility. Specifically, the sample size of 90 participants is feasible for an
online intervention spanning multiple provinces and multiple years. Recruiting double that
sample, 180 participants, would also likely be feasible as a similar unpublished study, comparing
peer-facilitated and professional-facilitated online chat groups for young women with breast
cancer, was able to recruit over 200 participants in two years. Therefore, and with caution (161,
162), we conducted this power calculation to demonstrate the feasibility of our study design.
In addition to the changes already discussed, two primary modifications to the study
design will be necessary for a larger RCT. Firstly, a minimal intervention control group, instead
of a TAU control, could be used. This is important because it would allow for a comparison
group over the full study trajectory (9-12 months). One suggestion for this minimal intervention
could be a 2-hour online PEP chat group, but careful weighing of additional options would be
important before proceeding.
Second, ongoing rolling recruitment, instead of one recruitment period, will allow the
study to span a number of years and increase the ability to recruit the required number of
participants. This rolling recruitment model, combined with a minimal intervention control,
87
would allow for a new round of chat groups to begin whenever another 21-27 participants enrol
in the study (7-9 participants in each study arm). Overall, employing a minimal intervention
control, continuously recruiting across a wide geographic area using all recruitment methods
possible, ensuring cross-platform compatibility of all study materials and platforms and
providing incentives to increase treatment adherence, will allow for the highest possible chance
of success in a future RCT study of online chat-group interventions for AYAs. Such a trial can
help to establish the best group psychotherapy intervention model to improve the lives of AYA
cancer patients.
5.5 Limitations
The limitations of this study include the small sample size that precludes most statistical
tests, and lack of a longer longitudinal comparison group to control for natural changes in
psychosocial outcomes over time. However, this pilot study was effective to test the integrity of
the study design and gain a greater understanding of the therapeutic models employed. The study
provided insights into the online chat-group experience for AYAs and the potential differences
between the PEP and NIEFP therapeutic models. Therefore, the small sample size and lack of
longitudinal comparison group do not diminish the importance of this study.
Due to the small sample size, we focused on the total scores rather than subscales in this
pilot study. A limitation of this strategy became apparent in post-hoc analysis of the HADS
anxiety and depression subscales that provided more information about specificity and
sensitivity. Similarly, further subscale analysis might provide valuable process information.
Another limitation is that rigorous qualitative methodology was not used in the collection
and analysis of open-ended participant comments and follow-up interviews. Two qualitative
reviewers used their consensus to select the quotes and themes included in this paper, but no
specific qualitative methodology directed this data collection or analysis. The lack of rigour
limits the generalizability of these results; however, the purpose of collecting comments was to
help to improve the design of a larger study and generate hypotheses concerning PEP/NIEFP
models, not predict the experiences of AYAs in a larger study. Future studies would benefit from
rigorous qualitative data collection and analysis to further understand how AYA-specific
interventions can be more effective.
88
Finally, there were limitations concerning the measures used to collect data. In retrospect,
we could have used better test-construction methods to improve the sensitivity of some of the
“Proof of Concept” measures used to explore differences in the PEP/NIEFP models. For
example, the “feeling” item developed for this study to explore a participant’s emotional reaction
to the PEP/NIEFP resource content, while interesting, did not provide information that could be
easily analyzed and interpreted. A validated, quantitative measure of emotional expression would
be more appropriate to assess participants’ reactions to the weekly PEP/NIEFP resources and
their emotional valence after each chat-group session. In general, more specific measures of
emotion could be collected so that differences in the PEP/NIEFP group experience can be
quantified.
5.6 Future Directions
A pilot study is designed with future directions in mind, and this has been a common
theme throughout this discussion. However, we have yet to discuss some potential future
directions. To be effective with AYAs, more exploration of our intervention process using
NIEFP theoretical constructs could occur. A next step in the analysis of this pilot study data,
which could directly address this need, is an in-depth qualitative analysis of chat-group
transcripts. This could shed light on other important group psychotherapy processes, such as
levels of emotional expression, and differences in discussion content between the PEP/NIEFP
models. This study indicates that NIEFP with AYAs shows promise, and future studies using
both face-to-face and online platforms could be designed to further understand and test this
model.
Another future direction could involve comparative studies with other psycho-oncology
interventions or with a modified, AYA-specific version of the NuCare manual. Using an AYAspecific NuCare manual could provide greater insight into the difference between group
psychotherapy models. Also, comparing an online-chat NIEFP group intervention to a face-toface group psychotherapy arm, could provide a different opportunity to compare the efficacy of
group formats for AYAs. This could provide evidence toward a routine standard of care for
interventions offered to AYAs. These studies could also include an economic evaluation to add
further health-care-policy relevant data.
89
Finally, an important future direction for AYA psycho-oncology is to design and/or
validate psychosocial outcome measures for use with the AYA population. Specifically, outcome
measures that are sensitive to changes in the short-term could improve knowledge about what
symptoms can improve quickly for AYAs. Careful design and evaluation of study-specific
measures is also required to evaluate effectively psychosocial interventions for AYAs.
Overall, the field of AYA psycho-oncology has many potential directions, but online
interventions may be one of the most promising. However, not all online interventions provide
equal value to AYAs, and RCT’s exploring these interventions are necessary. Online NIEFP
interventions show potential and this study demonstrates that it is important to explore these in
AYA psycho-oncology research. Based on this pilot study, we conclude that RCT studies
comparing different models of online group psychotherapy are feasible. Specifically future RCTs
should include a long follow-up period (12-months), and choose appropriate and sensitive
psychosocial outcome measures, (including Quality of Life). They should also use all possible
recruitment methods, enhance retention and attendance through incentives, offer interventions
during the Fall and Winter months, and use a minimal intervention control group with rolling
recruitment. This pilot study laid the groundwork for a larger randomized study and provided
valuable insights into the design of online psychosocial interventions for AYAs with cancer.
AYAs prefer and potentially benefit more from narrative-informed emotion-focused online
group interventions compared with psychoeducational online group interventions.
90
APPENDIX A: RECRUITMENT DOCUMENTS A.1. Letters of Initial Contact
Michael Lang
__________
__________
__________
January 14th, 2014
Hello,
Please allow me to introduce myself and explain the reason I am contacting you. I am a student researcher at
the University of Calgary and the CancerBridges Survivor Network Coordinator with CancerControl Alberta.
In my work, I repeatedly hear from younger men and women who have been diagnosed with cancer that they
want more age-specific information and support, including the opportunity to connect with other young adults
who have gone through a cancer experience. However, it can be difficult to connect with other young adult
cancer patients/survivors because cancer is less common in this age group (which is defined as 18-39 years
old).
When you were first diagnosed with cancer, you automatically became part of the Alberta Cancer Registry. If
there is a good reason, and with many protections around your privacy, the Cancer Registry allows people like
myself to send information to those treated for cancer in Alberta. I have enclosed a brochure about the Registry
if you are interested to understand the details. I am contacting you to let you know about a pilot project for
younger women and men with cancer that you may be interested in.
In this study, Speaking Their Language: A Randomized Study of Synchronous Online Chat Groups for Young
Adults with Cancer, we are examining the benefits of online discussion groups, which use either a video or a
workbook to stimulate conversation. The discussion groups are facilitated by a professional counsellor, occur
once a week for 10 weeks and are 90 minutes in duration. Participants will be randomly allocated to either the
Winter (February-April, 2014) or Spring (April-June, 2014) session with half of the participants in each
session using the video and the other half using a workbook. There will be questionnaires (10-15min) filled out
at different time-points throughout the study.
We are now enrolling participants for both winter and spring sessions. If you are interested in enrolling in the
study, you can register directly online on our secure website: CancerChatCanada.ca. Click register at the top of
the page and when asked the question “How did you first learn about our online support groups?” choose
“Other” and write “Alberta Young Adult Group” in the adjacent text box. Within two days of registration I
will contact you by phone to give you more details about the groups and complete the enrolment process.
If you would like more information, please me directly at __________ or via email at __________. I am
supervised by Dr. Janine Giese-Davis in the Psychosocial Oncology Department at the Tom Baker Cancer
Centre and her email is _______________. Thank you for your time.
Sincerely,
Mike Lang
91
92
A.2. Follow up Letter
Michael Lang
__________
__________
__________
January 29th, 2014
Hello,
A few weeks ago you received a letter about the opportunity to participate in a study called, Speaking Their
Language: A Randomized Study of Synchronous Online Chat Groups for Young Adults with Cancer,
examining the benefits of online discussion groups based on a video or workbook.
We are still enrolling participants for the groups starting this February (Winter Session) and May (Spring
Session). If you are interested in participating please register as soon as possible online at our secure website:
CancerChatCanada.ca. Click register at the top of the page and when asked the question “How did you first
learn about our online support groups?” choose “Other” and write “Alberta Young Adult Group” in the
adjacent text box. Within two days of registration I will contact you by phone to give you more details about
the groups and complete the enrolment process.
All of the participants, randomly selected, will join a 90-minute discussion group led by a professional
counsellor once a week for 10 weeks in either the winter or spring session. One half of the participants will use
the video as a supplement to the online chat group with the other half using a workbook.
If you would like more information, please me directly at __________ or via email at __________. I am
supervised by Dr. Janine Giese-Davis in the Psychosocial Oncology Department at the Tom Baker Cancer
Centre and her email is __________. Thank you for your time.
Sincerely,
Mike Lang
93
A.3. Recruitment Poster
94
A.4. CancerChatCanada Online Registration Form
95
A.5. Intake Phone Interview Script
Intake Phone Interview Script (Template from CancerChatCanada)
Purpose Use this script when calling a young adult who have registered for the study on the CancerChatCanada website. There are three possible outcomes for this enrolment process: Scenario 1: The subject agrees to be enrolled Scenario 2: The subject wants to remain on the waiting list Scenario 3: The subject declines and does not want to remain on a waiting list Materials and Preparation 1. Pre-­‐registration Call List
2. Informed Consent (Attachment: Participant Information and Consent Form.doc)
3. Data Entry Form (Attachment: Database Registration Form.xls)
Review the available information from the database registration form. Write the date and time of the call on the database registration form. START HERE IF, THE PERSON IS PREREGISTERED (RECEIVED A REGISTRY LETTER AND THEN REGISTERED ON THE WEBSITE). 1. If you are put into voice mail, say the following:
VOICEMAIL
Hi, my name is Mike Lang and I’m calling for __________. I’m calling regarding a study you registered
for online a few weeks ago. Please give me a call at your earliest convenience at 1-­‐888-­‐998-­‐8148. If I’m
not available you can leave a message and I will reply as soon as possible. Thank you and Good day!
If someone answers the telephone, ask for the person on the list and leave the above message with the person who answered the phone if the woman is not available. If she is available, begin the telephone script with an introduction: Hi, my name is Mike Lang and I’m calling from the Tom Baker Cancer Centre on behalf of CancerChatCanada. You registered online a few weeks ago for a research study we are conducting in Alberta and I wondered if you would have a few minutes to talk. (How are you?) If no, ask if there would be a more convenient time for you to call back and make a note of this on the call list or registration form. 2. Describe the study
I am a young adult cancer survivor myself and currently a MSc. student at the University of Calgary. I
am studying support and education resources for young adults with cancer. I know from personal
experience and from other research that many young adults experience challenges during and after
their treatment ends, and I am trying to learn more about the types of emotional and educational
support that would most benefit young adults. If you decide to enrol in this research project you be
96
will “randomized,” roll the dice if you will, to determine the group you will be assigned to. This study has two different programs, each lasting 10 weeks: one uses a work-­‐book in conjunction with an online real-­‐time chat group, the other uses a patient-­‐created video in conjunction with an online real-­‐
time chat group. There will be two sessions, a winter session and a spring session and you will be randomly assigned to one of those sessions. Those who are assigned to the spring session will still be asked to fill out all surveys during the 10 weeks they are waiting to start their group. No matter what group you are assigned to, at the end of the study you will be sent a DVD copy of the patient-­‐created video to keep. 3. Assess Interest
Ask the person if she is interested in taking part in the study:
Now that you know a little bit more about it, does this sound like something you might be interested
in trying?
If no, go to Step 8.
If yes, continue.
Scenario 1 Great! So if you are assigned to the winter session, these will begin meeting sometime the week of February 3rd in the evenings and if you are assigned to the spring session you will begin meeting sometime the week of April 21st in the evenings. The groups meet for 90 minutes each week for 10 weeks. In general, do Mondays, Wednesdays or Thursday evenings work for you at around 7:30pm? No-­‐Skip to 8 Scenario 2] 4. Determine Eligibility
Okay, then I will just need to ask a few questions to confirm your eligibility for the study and then
review an Informed Consent Form that is part of our research ethics guidelines.
Okay, you are (repeat person’s name), and you are ____ years old, is that right? Could I also get your date of birth? And what is the approximate month and year of your last active treatment, which could be surgery, radiation or chemotherapy? Okay, that concludes the eligibility information we needed. Need to get mailing address for Coffee Card mailing… 5. Collect emergency contact information
Great, one other thing we do when enrolling participants in our groups is to get an emergency contact
name and phone number. We will only be using this information if your group leader is unable to
contact you for any reason and is concerned.
6. Review Informed Consent
The final thing for today is to review our Informed Consent Document. This I will cover briefly but if
you are in front of your computer I can send you a copy of the document and you can review it with
me.
7. Interview Close
97
Okay, that takes care of the enrolment. In the next few minutes I will be sending you an email to confirm everything and let you know about next steps but the first step is to: 1) Log-­‐in to CancerChatCanada.ca right now, using the login name and password you have
created and fill out the final enrollment form to complete the enrollment process. This will just
ask you a few more questions about you and your cancer experience so far.
2) After you have done this, look for an email from me in a few days, which will have a link to the
first questionnaire. Please fill that out as soon as possible.
3) And then on Wednesday, January 29th you will get an email from me to confirm which group
you were randomized to, the Winter or Spring session.
Does this make sense to you? Please feel free to give us a call if you have any questions on the Informed Consent. Thank you so much for your time today, and I hope you enjoy taking part in the online chat groups. Please login to CancerChat.ca right now and fill out the final enrollment for before the chat group starts so that you can participate. 8. Record reason for decline
If the young adult is not interested in enrolling in the study, record the reason for declining.
Scenario 3 Okay, that’s fine, thank you for taking the time to talk with me. So that we can better understand why this type of research might not be of interest, would it be okay if I asked why you are not interested at this time? Write down the reason the person declined the study. If they is declining due to a time constraint, ask them if they would like to be informed about the next group when we begin enrolling: Scenario 2 We will have new groups starting up every other month and the next group will be at a different time of day and quite possibly a different day of the week. Would you like to remain on our waiting list and be contacted when a new group starts? 9. If yes, ask about their general availability for participating.
Great, in general, what might your availability be like in a few months time? Would you prefer an
evening versus a morning or afternoon group, etc.
10. Close the interview
Thank you for your time today, and we will be in touch soon!
98
A.6. Participant Information and Consent Form
Participant Information & Consent Form
TITLE: Speaking Thier Language: A randomized pilot study of sychronous online chat groups for young
adults with cancer.
SPONSOR:
Funding for this study has been provided by the Psychosocial Oncology Research Training Program
and CancerChatCanada.
INVESTIGATORS:
Dr. Janine Giese-Davis
CancerControl Alberta -- Tom Baker Cancer Centre, Holy Cross Site
Department of Psychosocial Resources
Telephone: __________
Email: __________
Michael Lang
MSc. Student, Department of Community Health Sciences, Faculty of Medicine
University of Calgary
Telephone: __________
Email: __________
This consent form is only part of the process of informed consent. It should give you the basic idea
of what the research is about and what your participation will involve. If you would like more detail
about something mentioned here, or information not included here, please ask. Take the time to
read this carefully and to understand any accompanying information. You will receive a copy of this form.
Emergency Contact Number: (24 hours/7 days a week)
For emergencies only: Call the centre nearest you and ask for Patient and Family Counseling Services
Edmonton Crisis Response Team - __________
Calgary Crisis Response Team – __________
BACKGROUND
You are invited to participate in this research project because a diagnosis of cancer brings many
challenges. We are doing this study as part of our efforts to help young adults who have been diagnosed
with cancer and may be experiencing some cancer related distress. Many young adults talk about the
importance of connecting with other people their age who also have cancer but this can be difficult to do
because there are relatively few young adults diagnosed with cancer. This study will provide an easy way
for you to connect with other people your age who ‘get it’, using the established online
CancerChatCanada program (www.cancerchatcanada.com).
If you are taking part in this study, you will join a professionally facilitated, live internet chat group that will
take place once a week for 10 consecutive weeks. Each chat group will be 1.5 hours long, completely
anonymous and password protected. There will be approximately 6-9 young adult cancer survivors in
each group. Materials on various topics, such as managing fear and uncertainty and relationships, will be
distributed to help guide the chat group discussion.
Each young adult that is enrolled will be randomly assigned to participate in the 10-week chat group in
99
either the Winter Session (January – March, 2014) or the Spring Session (April – June, 2014). In addition
to participating in the chat group, regular questionnaires will be filled out by all participants about their
chat group experiences and other feelings about their cancer experience. At the end of the study,
participants who complete all the questionnaires will receive a thank you gift card. More details about the
study procedures are provided below.
There will be approximately 36 young adult cancer survivors enrolled in this study from all over in the
province of Alberta. Taking part in this study is voluntary. You can decide not to participate or to withdraw
from the study at any time.
WHAT IS THE PURPOSE OF THIS STUDY?
This project is a demonstration study. Before proceeding to a full-scale program, a demonstration study is
often carried out first to test the design of a study and its acceptability to participants.
The design of a study is how the program is implemented, how the data will be collected, whether the
data can provide useful information and whether it will be practical to proceed to a larger program.
WHAT WOULD I HAVE TO DO?
If you agree to take part in this study, you will be asked to participate in a 1.5 hr online chat group once a
week for 10 weeks and fill out a series of questionnaires online. The questionnaires will be completed at
the following times:
1. Prior to enrollment into the online chat group.
2. One week prior to the start of the winter session (week 0).
3. After each 1.5 hour chat group (10 weeks during either the winter or spring session).
4. At the end of the Winter Session (12 weeks from start).
5. At the end of the Spring Session (24 weeks from start).
6. At the end of the Study period (36 weeks from start).
You will be prompted by email and sent a private link to access the questionnaires. It is important that you
complete each set of questionnaires on time. The surveys taken at week 0, 12, 24, 36 will take
approximately 15-20 minutes. The weekly post-chat group survey will take approximately 5 minutes.
After you have completed the first set of forms you will be enrolled in the online chat group. You will
participate in a 1.5 hour online group with a CancerControl Alberta professional, once a week for 10
weeks during either the Winter session (February-April, 2014) or the Spring session (April-June, 2014).
You will be emailed by the researcher and provided with instructions for logging onto the online group.
After the end of your online chat group session, you may also be contacted for a short 10-15 minute
follow up interview.
If you have any questions about the research procedures, please feel free to contact the student
researcher, Mike Lang. His telephone number is listed on the first page of this consent form.
WHAT ARE THE RISKS?
There are no known medical risks to participating in this study. However, it may be distressing to
participate in the chat groups as they may address difficult topics such as fear and uncertainty,
relationships and personal identity. If you experience increased distress over the course of the study, you
may request to speak with the counselor individually, and she can help you to identify an appropriate
course of action to help decrease your feelings of distress.
100
In the online support groups, there may be risks associated with the electronic transfer of information. As
with any online interaction, it is possible that third parties could view the electronic information through
interception or misdirection of network transmissions. In addition, information could be directly viewed on
a participant’s computer or from printed copies of the session transcripts. However, no personal
information could be linked to potential viewings. The following measures will be taken to ensure that
confidentiality in the online support groups is maintained:
•
Your privacy on the online support groups will be protected through the utilization of security
features that include the use of study ID numbers, initials only, file encryption and password
protection. Your online registration with CancerChatCanada.ca is on a Canadian, secure,
encrypted, password-protected site.
•
Other questionnaire data, identified only by your initials, will be collected through a Canadian
online survey company, Fluid Surveys, which is compliant with all Canadian privacy laws and
does not store any data on servers in the United States. Many medical researchers in Canada
use this secure site. All questionnaire data once collected will be stored on password protected
and encrypted Alberta Health Services servers.
•
Electronic transcripts of the online groups will be stored within the BC Cancer Agency until the
end of the study, at which time they will be transferred to the Tom Baker Cancer Centre – Holy
Cross Site through an encrypted and password protected file transfer service. This data will be
associated with initials only, and will not include any identifying information.
•
Participants will be made aware of the need for physical privacy during online sessions and the
need to protect any printed copies of session transcripts.
These measures safeguard the privacy of participants to the best of our ability. However, we cannot
guarantee absolute privacy of electronic data, and the confidentiality risks involved in this study are the
same as with any other online exchanges.
WILL I BENEFIT IF I TAKE PART?
If you agree to participate in this study there may or may not be a direct benefit to you. You have been
asked to participate because you have been identified as having had a cancer diagnosis between the age
of 18-39 and while the CancerChatCanada chat groups may be beneficial for you there is no guarantee
that this research will help you. The information we get from this study may help us to provide better
treatments in the future for young adults with cancer.
The potential benefits of the study in general include the investigators learning how to conduct this type of
online chat group and which types of supplementary materials work best. Individual participants may
learn new coping skills and experience enhanced mood and well-being and feel supported by peers but
we cannot be sure this will happen.
DO I HAVE TO PARTICIPATE?
No, you do not have to participate in this study. If you would like to participate in an online chat group with
CancerChatCanada but would not want to participate in this particular study, you can sign up anytime for
another one of their chat groups at www.cancerchatcanada.com.
Participation in this study is voluntary and you may withdraw from the study at any time without
jeopardizing your health care. You may withdraw from the study at any time by phoning or emailing the
study coordinator, Mike Lang (contact information provide on the first page of this document), and asking
to be removed from the study.
It is also possible that you may be withdrawn from the study by the investigators. This would only occur if
101
the professional chat group facilitator asses that your health care needs would be best met through a
different treatment modality or if you do not meet all of the eligibility criteria below:
•
You are between the age of 18-39 and have been diagnosed with any form of cancer
•
You fully understand the project and give your informed consent to participate as demonstrated
by signing this consent form
•
You are familiar with computer usage and have access to the Internet in a private area where you
will feel comfortable entering personal information
•
You have the ability to read and write in English
•
You can commit to the weekly time requirement
•
You do not have a history of serious psychiatric disorder
WILL I BE PAID FOR PARTICIPATING, OR DO I HAVE TO PAY FOR ANYTHING?
You will not be paid for participating in this study. As this is an at home intervention, there are no costs to
you for participating, and no re-imbursement will be provided. However, those who complete all the
surveys in the study will be sent a $25 Starbucks/Tim Hortons gift card as a thank you for participating.
You do not waive any of your legal rights for compensation by signing this form.
WILL MY RECORDS BE KEPT PRIVATE?
As a participant in this study your confidentiality will be respected. The following measures will be taken:
•
No information that discloses your identity will be released or published without your specific
consent.
•
Your identity will not be used in any reports about the study.
•
Records that leave the Tom Baker Cancer Centre – Holy Cross Site will be identified by a study
code number only. Your birth date will also be provided if requested by the sponsor or
responsible regulatory agency.
•
All identifying and medical treatment data will be stored within Canada.
•
All information associated with this study will be kept behind locked doors, in a locked filing
cabinet or in computer files secured by a password.
Your rights to privacy are protected and guaranteed by federal and provincial laws that require
safeguards to insure that your privacy is respected and also give you the right of access to the
information about you that has been provided to the sponsor and, if need be, an opportunity to correct
any errors in this information.
IF I SUFFER A RESEARCH-REALTED INJURY, WILL I BE COMPENSATED?
In the event that you suffer injury as a result of participating in this research, CancerChatCanada, the
University of Calgary, Alberta Health Services or the Researchers will provide no compensation to you.
You still have all your legal rights. Nothing said in this consent form alters your right to seek damages.
102
SIGNATURES
Your signature on this form indicates that you have understood to your satisfaction the information
regarding your participation in the research project and agree to participate as a participant. In no way
does this waive your legal rights nor release the investigators or involved institutions from their legal and
professional responsibilities. You are free to withdraw from the study at any time without jeopardizing your
health care. If you have further questions concerning matters related to this research, please contact:
Dr. Janine Giese-Davis __________
Or
Mike Lang __________
If you have any questions concerning your rights as a possible participant in this research, please contact
the Chair, Conjoint Health Research Ethics Board, University of Calgary at __________.
Participant’s Name
Signature and Date
Investigator/Delegate’s Name
Signature and Date
Witness’ Name
Signature and Date
The University of Calgary Conjoint Health Research Ethics Board has approved this research study.
A signed copy of this consent form has been given to you to keep for your records and reference.
103
A.7. CancerChat Canada Enrolment Questionnaire
104
105
106
APPENDIX B: LINKS TO MODELS B.1. Workbook Links
Week #1 – N/A
Week #2 - http://cancerchatcanada.ca/files/manuals/CopingandMindfulness1.pdf
Week #3 - http://cancerchatcanada.ca/files/manuals/RelaxationandImagery.pdf
Week #4 - http://cancerchatcanada.ca/files/manuals/WaysofThinking1.pdf
Week #5 - http://cancerchatcanada.ca/files/manuals/Communication1.pdf
Week #6 - http://cancerchatcanada.ca/files/manuals/ProblemSolving1.pdf
Week #7 - http://cancerchatcanada.ca/files/manuals/SocialSupport.pdf
Week #8 - http://cancerchatcanada.ca/files/manuals/HealthyLifestyle.pdf
Week #9 - http://cancerchatcanada.ca/files/manuals/GoalSetting2.pdf
Week #10 - N/A
B.2. Video Links and Discussion Questions
Week #1 - N/A
Week #2 - http://youtu.be/WFC4NZ9BGAI
Discussion Questions: N/A
Week #3 - http://youtu.be/_6CbGOTWnTk
Discussion Questions:
1.
2.
3.
What did you most identify with in the stories that were shared?
What transitions are you currently going through? Are these transitions positive or negative? Why?
What can be done to help manage the abrupt transitions of a cancer experience or life in general?
Week #4 - http://youtu.be/wqe9phqVgF0
Discussion Questions:
1.
2.
What did you most identify with in the stories that were shared?
Is it possible for fear and uncertainty to be a both a positive and negative influence in our lives? If so, how?
107
3.
What actions or activities can we engage in to move forward in a healthy way despite the fear and
uncertainty that we may be facing?
Week #5 - http://youtu.be/3-0ec7hRlVg
Discussion Questions:
1.
2.
3.
What did you most identify with in the stories that were shared?
Is it possible to be surrounded by people but still feel isolated? If so, why or how does this occur? (Potential
other question: Is being isolated a negative thing? If so, why?)
What actions or activities can we engage in to decrease feelings of isolation that we may face in our cancer
experience?
Week #6 - http://youtu.be/uQxkOTgwIYk
Discussion Questions:
1.
2.
3.
What did you most identify with in the stories that were shared?
If there has been noticeable changes in the relationships in your life, what has changed and why do you
think that occurred?
What can we do to keep our relationships strong or create new relationships despite what is happening in
our lives?
Week #7 - http://youtu.be/a4mM9zx4l6Y
Discussion Questions:
1.
2.
3.
What piece of Alston's 2 min story did you most connect with?
Do you think it is easy for yourself, and others, to only see the cancer part of you?
What can we do to allow cancer to be apart of or lives but not let it become our identity?
Week #8 - http://youtu.be/-ByS6NrviMY
Discussion Questions:
1.
2.
3.
Do you think that times of silence or reflection is an important part of life?
Is it difficult for you to create times of silence or reflecting in your life? Why?
What is something we can to help us reflect, refocus and rebuild our lives in a meaningful way despite
cancer?
Week #9 - http://youtu.be/-ntLjLIO0JM
Discussion Questions:
1.
2.
3.
What does "Reintrajectorization" (or Survivrship as it is also called) mean to you?
Do you want to make changes in your life because of your cancer diagnosis? If so, have you thought about
when or how you are going to start these changes?
What does "living well with, through and beyond cancer" mean to you?
Week #10 - N/A
108
APPENDIX C: FOLLOW UP INTERVIEW SCRIPTS
C.1. Low Attendance Interview Script
Hi_____
You recently signed up to be apart of a online chat group for young adults with cancer. We
noticed that you did not attend many of those chat group sessions. I am calling to ask about the
reasons that you were not able to participate in the chat sessions.
What was the main reason that you did not attend the chat group sessions?
Was there any other factors that contributed to you not attending the chat group sessions?
Thank you for your time and for agreeing to be apart of the study.
C.2. Follow-Up Interview Script
Hi_____
You recently completed an online chat group for young adults with cancer and you were
randomly selected for a follow-up phone interview so that we can get some insight into your chat
group experience. Do you have about 10mins now to talk with me?
I will start with some general questions about your online group experience:
1. Why did you choose to join the online chat group? Was there anything about your
particular circumstance that made online appealing (can we hear more than ‘seeking
connection, understanding…)?
2. What was it like communicating online? Was it satisfying to chat or did you have
challenges?
3. How did the online group program work for you? (timing, length of sessions,
frequency)
4. What was your experience of the workbook/video clips sent out to you? Did any
topics stand out for you as especially helpful or not?
We would also like to understand more your perceptions and feelings about the chat sessions.
1. What were you hoping or expecting to get from these chat sessions? Did it happen?
109
2. Were you able to talk about private, important things?
3. Did you feel understood by the others in the group?
4. What did you learn (if anything) from the discussions you had in the group?
5. Was it helpful to have a facilitator facilitate the chats? Was it ever hindering? (probe
more about facilitator style, unstructured format; what the facilitator did, leadership
behaviours etc)
6. Have you remained in contact with any group members? Or sought other support online?
(whenever possible, elicit discussion to compare/contrast our synchronous OSG with
other options)
I have a few final questions about the impact of the online chat group in your life.
1. Looking back to before you started this program, can you identify anything that has
changed for you (e.g. quality of life)
2. Have you made any practical (i.e. behavioural, concrete) changes as a result of being in
the group?
3. As you look ahead to the future, do you expect anything to be different for you now?
Is there anything else you think I should know?
Thank you for your time and for agreeing to be apart of the study.
110
APPENDIX D: RESEARCH STUDY LOGIC MODEL Study Aims
1. Proof of Concept
Measure
2.5.1 Topic-Specific
Suitability
2.5.1 Overall Suitability
2. Appropriateness
of Psychosocial
Outcomes
Data Analysis
a) “Use of Model” and “Feeling about Model” items compared
between model groups using proportions and ranges (no statistical tests
performed)
b) “Relevance of Model” items combined and ranked by median to
identify most relevant topics from each model group (3.7.2)
2.5.2 Group Cohesiveness
Scale (GCS)
Week 1,4,7 and 9; 7-items on 5
point Likert-type scale (3.6.2.3)
2.5.2 Group Experiences
Inventory (GEI)
Once, immediately post
intervention; 25 items using 6-point
Likert-type scale (3.6.2.4)
Baseline, 3, 6, 9 months; 14 items
on 4 point Likert-type scale
(3.6.3.1)
Baseline, 3, 6, 9 months; 20 items
on 4 point Likert-type scale
(3.6.3.2)
Baseline, 3, 6, 9 months; 17 items
on 5 point Likert-type scale
(3.6.3.3)
Baseline, 3, 6, 9 months; 21 items
on 6 point Likert-type scale
(3.6.3.4) Baseline, 3, 6, 9 months; 14 items
on 9 point Likert-type scale
(3.6.3.5)
On-going; Response rates to
recruitment methods (3.6.4.1)
2.6.1 Hospital Anxiety and
Depression Scale (HADS)
2.6.2 UCLA Loneliness
Scale (UCLA-LS)
2.6.3 Post-traumatic Stress
Disorder Checklist –
Civilian (PCL-C)
2.6.4 Post-traumatic
Growth Inventory (PTGI) 2.6.5 Cancer Behaviour
Inventory – Brief (CBI-B)
3. Feasibility of
Study Design and
Procedures
Data Collection
Week 2-9; 3 items on 4 point
Likert-type (“Relevance”), 1
qualitative “Use” question, 1
qualitative “feeling” question and 1
open-ended qualitative response
(3.6.2.1)
Once, immediately post
intervention; 5-items using 6-point
Likert-type scale (3.6.2.2)
2.7.1 Recruitment and
Retention
2.7.2 Attendance
Attendance - On-going; attendance
tracking in facilitator notes
(3.6.4.2)
Low Attendance - On-going;
withdrawal numbers and Primary
and Secondary reason for
discontinuing intervention (exit
interview) (3.6.4.2)
2.7.3 Usability 2.7.4 Participant
Satisfaction
System Usability Scale (SUS)Once, immediately postintervention (3.6.4.3) Single Item Ease of Use- Once,
immediately post-intervention; 6point Likert-type scale (3.6.4.3) Once, immediately post
intervention (single Item of Overall
Satisfaction) and qualitative
feedback from post-intervention
interviews (3.6.4.4)
111
a) Mann-Whitney U Test comparison between model groups
b) Qualitative responses from both Topic-Specific Suitability and
follow-up interviews were combined and explored for common themes
related to model suitability (3.7.2)
a) Comparison of mean slope between model groups
b) Comparison of pre/post mean change scores between model groups
(3.7.2)
a) Mann-Whitney U Test comparison of GEI subscales between model
groups to characterize differences in group experiences. (3.7.2)
a) LMM of initial randomization (standard, modified, control) from
baseline to 3 months and change scores
b) LMM of initial intervention groups (standard, modified) from
baseline to 9 months and change scores
c) Qualitative responses explored for common themes related to the
psychosocial outcomes (3.7.3)
d) Power calculation for PCL-C measure (3.7.4)
a) Proportion of AYAs interested in the study (target 10%)
b) Proportion of AYA respondents eligible to participate (target 95%)
c) Proportion of AYA respondents consenting to participate (target
95%)
d) Proportion of consented AYAs who completed the 3, 6 and 9 month
psychosocial outcome questionnaires (target 80%, 65% and 65%)
Feasibility achieve if all proportions within 5% of target. (3.7.4)
a) Mean weekly attendance across groups
b) Mean number of sessions attended by participant
c) Convert attendance to categorical variable
Feasibility achieved if within 5% of CCC operational attendance
levels. (3.7.4)
a) Proportion of study withdrawals compared to operational CCC data
b) Primary and secondary reason for withdrawal summarized by theme
using withdrawal interview qualitative data (3.7.4)
Feasibility achieve if study dropout proportion is within 5% of CCC.
a) Student’s t-test comparison between model groups (3.7.4) a) Mann-Whitney U Test comparison between model groups
b) Qualitative responses explored for common themes related to
system usability (3.7.4)
a) Proportion of highly satisfied participants
b) Comparison between model groups using Mann-Whitney U Test for
non-parametric data
c) Qualitative responses to relevant questions (Online Communication,
Privacy/Confidentiality, Understanding of Group, Facilitator) explored
for common themes related to overall participant satisfaction (3.7.4)
APPENDIX E: ONLINE SURVEYS E.1. Weekly Post Chat Surveys
E.1.1. Standard Survey - Week 2,3,5,6,8
112
113
E.1.2. Extended Survey - Week 1,4,7,9 (Additional to Standard Survey)
114
115
116
117
E.2. Final Post Chat Survey – Week 10
118
119
120
121
122
123
124
125
E.3. Psychosocial Outcomes Questionnaire – Baseline, 3, 6, 9 Months
126
127
128
129
130
131
132
133
134
135
References
1.
NCI. Closing the Gap: Research and Care Imperatives for Adolescents and Young Adults
with Cancer. NIH Pub. 2006;06(6067).
2.
CCS. Canadian Cancer Statistics 2012. 2012 May 2012.
3.
Lawrence L. Survival lags for adolescents, young adults with cancer. Hem/Onc Today.
2011;12(21).
4.
Bleyer A. Latest Estimates of Survival Rates of the 24 Most Common Cancers in
Adolescent and Young Adult Americans. Journal of Adolescent and Young Adult Oncology.
2011;1(1).
5.
Mery LM, H.; Barr, R.; Greenberg, M.; and Franco, E.L. . Delays in diagnosis and
treatment among children and adolescents with cancer in Canada. Pediatric Blood and Cancer.
2008;51:468-74
6.
Zebrack B, Isaacson S. Psychosocial care of adolescent and young adult patients with
cancer and survivors. Journal of Clinical Oncology. 2012;30(11):1221-6.
7.
Bleyer A. The Adolescent and Young Adult Gap in Cancer Care and Outcome. Curr
Probl Pediatr Adolesc Health Care. 2005;35:182-217.
8.
Bleyer AB, R.; Hayes-Lattin, B. et. al. The distinctive biology of cancer in adolescents
and young adults. . National Review of Cancer. 2008;8:288-98.
9.
Ram R, Wolach O, Vidal L, Gafter-Gvili A, Shpilberg O, Raanani P. Adolescents and
young adults with acute lymphoblastic leukemia have a better outcome when treated with
pediatric-inspired regimens: systematic review and meta-analysis. Am J Hematol. 2012
May;87(5):472-8.
10.
Bultz BDaH, J.C. Emotional Distress in patients with cancer: The sixth vital sign.
Community Oncology. 2006;3(5):311-4.
11.
Kim J-H, Yoon S, Won W-Y, Lee C, Lee C-U, Song KY, et al. Age-specific influences
of emotional distress on performance status in cancer patients. Psycho-Oncology. 2013
Oct;22(10):2220-6.
12.
Seitz DCM, Besier T, Goldbeck L. Psychosocial interventions for adolescent cancer
patients: a systematic review of the literature. Psycho-Oncology. 2009;18(7):683-90.
13.
Keegan TH, Lichtensztajn DY, Kato I, Kent EE, Wu XC, West MM, et al. Unmet
adolescent and young adult cancer survivors information and service needs: a population-based
cancer registry study. Journal of cancer survivorship : research and practice. 2012 Sep;6(3):23950.
14.
Tonorezos ES, Oeffinger KC. Research challenges in adolescent and young adult cancer
survivor research. Cancer. 2011;117(S10):2295-300.
15.
Rabin C, Simpson N, Morrow K, Pinto B. Behavioral and psychosocial program needs of
young adult cancer survivors. Qualitative health research. 2011 Jun;21(6):796-806.
16.
Zebrack BJ, Block R, Hayes-Lattin B, Embry L, Aguilar C, Meeske KA, et al.
Psychosocial service use and unmet need among recently diagnosed adolescent and young adult
cancer patients. Cancer. 2013;119(1):201-14.
17.
Zebrack B, Bleyer A, Albritton K, Medearis S, Tang J. Assessing the health care needs of
adolescent and young adult cancer patients and survivors. Cancer. 2006;107(12):2915-23.
136
18.
Griffiths KM, Calear AL, Banfield M. Systematic review on Internet Support Groups
(ISGs) and depression (1): Do ISGs reduce depressive symptoms? J Med Internet Res.
2009;11(3):e40.
19.
Hoey LM, Ieropoli SC, White VM, Jefford M. Systematic review of peer-support
programs for people with cancer. Patient education and counseling. 2008 Mar;70(3):315-37.
20.
Stephen JR, A.; MacGregor, K.; McLeod, D.; Speca, M.; Taylor–Brown, J.;, Fergus KC,
K.; Turner, J.; Sellick, S.; Mackenzie, G. Evaluation of CancerChatCanada: a program of online
support for Canadians affected by cancer. Curr Oncol. 2013;20:39-47.
21.
Sansom-Daly UMP, Michelle; Wakefield, Claire E.; Bryant, Richard A.; Cohn, Richard
J. A Systematic Review of Psychological Interventions for Adolescents and Young Adults
Living With Chronic Illness. Health Psychology. 2012;31(3):380-93.
22.
D'Agostino NM, Penney A, Zebrack B. Providing developmentally appropriate
psychosocial care to adolescent and young adult cancer survivors. Cancer. 2011;117(S10):232934.
23.
Zebrack BJ. Information and service needs for young adult cancer patients. Support Care
Cancer. 2009;17:349-57.
24.
Hong Y, Peña-Purcell NC, Ory MG. Outcomes of online support and resources for cancer
survivors: A systematic literature review. Patient education and counseling. 2012;86(3):288-96.
25.
Watson M, Dunn J, Holland JC. Review of the history and development in the field of
psychosocial oncology. International Review of Psychiatry. 2014;26(1):128-35.
26.
NCCN. Clinical Practice Guidelines for Supportive Care. Distress Management. Fort
Washington, PA: National Comprehensive Cancer Network, 2002.
27.
NHMRC. Australian Clinical Practice Guidelines for the Psychosocial care of Adults
with Cancer. Canberra: National Health and Medical Research Council, 2003.
28.
NICE. Guidance on Cancer Services Improving Supportive and Palliative Care for Adults
with Cancer. London: National Institute of Health and Clinical Excellence. 2004.
29.
IOM. Cancer Care for the Whole Patient: Meeting Psychosocial Health Needs. National
Academy Press. 2007.
30.
Peabody C. The care of the patient. Journal of the American Medical Association.
1927;88:10.
31.
Holland JMD, Weiss TBA. The New Standard of Quality Cancer Care: Integrating the
Psychosocial Aspects in Routine Cancer From Diagnosis Through Survivorship. Cancer Journal
November/December. 2008;14(6):425-8. English.
32.
Balogh EP, Ganz PA, Murphy SB, Nass SJ, Ferrell BR, Stovall E. Patient-Centered
Cancer Treatment Planning: Improving the Quality of Oncology Care. Summary of an Institute
of Medicine Workshop. The oncologist. 2011;16(12):1800-5.
33.
Andersen B, Thornton L, Shapiro C, Farrar W, Mundy B, Yang H-C, et al.
Biobehavioral, Immune, and Health Benefits following Recurrence for Psychological
Intervention Participants. Clinical Cancer Research. 2010;16(12):3270-8.
34.
Spiegel D, Giese-Davis J. Depression and Cancer: Mechanisms and Disease
Progression. Biological Psychiatry. 2003;54(3):269-82.
35.
Spiegel D, Kraemer H, Bloom J, Gottheil E. Effect of Psychosocial Treatment on
Survival of Patients with Metastatic Breast Cancer. The Lancet. 1989;334(68):888-91.
137
36.
Meyer F, Fortin A, Gelinas M, Nabid A, Brochet F, Tetu B, et al. Health-Related Quality
of Life As a Survival Predictor for Patients With Localized Head and Neck Cancer Treated With
Radiation Therapy. Journal of Clinical Oncology. 2009;27(18):2970-6.
37.
Movsas B, Moughan J, Sarna L, Langer C, Werner-Wasik M, Nicolaou N, et al. Quality
of life supersedes the classic prognosticators for long-term survival in locally advanced nonsmall-cell lung cancer: an analysis of RTOG 9801. Journal of Clinical Oncology.
2009;27(34):5816-22.
38.
Carlson LE, Speca M, Faris P, Patel KD. One year pre–post intervention follow-up of
psychological, immune, endocrine and blood pressure outcomes of mindfulness-based stress
reduction (MBSR) in breast and prostate cancer outpatients. Brain, Behavior, and Immunity.
2007;21(8):1038-49.
39.
Carlson LE, Speca M, Patel KD, Goodey E. Mindfulness-based stress reduction in
relation to quality of life, mood, symptoms of stress and levels of cortisol,
dehydroepiandrosterone sulfate (DHEAS) and melatonin in breast and prostate cancer
outpatients. Psychoneuroendocrinology. 2004;29(4):448-74.
40.
Witek-Janusek L, Albuquerque K, Chroniak KR, Chroniak C, Durazo-Arvizu R,
Mathews HL. Effect of mindfulness based stress reduction on immune function, quality of life
and coping in women newly diagnosed with early stage breast cancer. Brain, Behavior, and
Immunity. 2008;22(6):969-81.
41.
Andersen BL, Farrar WB, Golden-Kreutz DM, Glaser R, Emery CF, Crespin TR, et al.
Psychological, behavioral, and immune changes after a psychological intervention: a clinical
trial. J Clin Oncol. 2004;22(17):3570-80.
42.
Fawzy FI, Fawzy NW, Arndt LA, Pasnau RO. CRitical review of psychosocial
interventions in cancer care. Archives of General Psychiatry. 1995;52(2):100-13.
43.
Tamagawa R, Garland S, Vaska M, Carlson LE. Who benefits from psychosocial
interventions in oncology? A systematic review of psychological moderators of treatment
outcome. J Behav Med. 2012;35:658.
44.
Carlson L, Garland S. Impact of mindfulness-based stress reduction (MBSR) on sleep,
mood, stress and fatigue symptoms in cancer outpatients. IntJ Behav Med. 2005;12(4):278-85.
45.
Tatrow K, Montgomery G. Cognitive Behavioral Therapy Techniques for Distress and
Pain in Breast Cancer Patients: A Meta-Analysis. J Behav Med. 2006;29(1):17-27.
46.
Rowland J, Meyerowitz B, Crespi C, Leedham B, Desmond K, Belin T, et al. Addressing
intimacy and partner communication after breast cancer: a randomized controlled group
intervention. Breast Cancer Res Treat. 2009;118(1):99-111.
47.
Serra D, Parris CR, Carper E, Homel P, Fleishman SB, Harrison LB, et al. Outcomes of
guided imagery in patients receiving radiation therapy for breast cancer. Clinical Journal of
Oncology Nursing. 2012:617.
48.
Owen JE, Goldstein MS, Lee JH, Breen N, Rowland JH. Use of health-related and
cancer-specific support groups among adult cancer survivors. Cancer. 2007;109(12):2580-9.
49.
Sheard TM, P. . The effect of psychological interventions on anxiety and depression in
cancer patients: results of two meta-analyses. Br J Cancer. 1999;80:315-33.
50.
Lepore SLC, J.C. Psychological interventions for distress in cancer patients: a review of
reviews. Behav Med. 2006;32:85-92.
51.
Jacobsen PB. Clinical practice guidelines for the psychosocial care of cancer survivors.
Cancer. 2009;115(S18):4419-29.
138
52.
Waxler-Morrison N, Doll R, Hislop TG. The Use of Qualitative Methods to Strengthen
Psychosocial Research on Cancer. Journal of Psychosocial Oncology. 1995;13(1-2):177-91.
53.
Creswell J, W. Research Design: qualitative, quantitative and mixed methods approaches.
Sage Publications Inc. 2009 (3rd Edition).
54.
Heron-Speirs HA, Baken DM, Harvey ST. Moderators of Psycho-Oncology Therapy
Effectiveness: Meta-Analysis of Socio-Demographic and Medical Patient Characteristics.
Clinical Psychology: Science and Practice. 2012;19(4):402-16.
55.
Zimmermann T, Heinrichs N, Baucom D. “Does one size fit all?” moderators in
psychosocial interventions for breast cancer patients: A meta-analysis. Annals of Behavioral
Medicine. 2007;34(3):225-39.
56.
Schneider S, Moyer A, Knapp Oliver S, Sohl S, Cannella D, Targhetta V. Preintervention distress moderates the efficacy of psychosocial treatment for cancer patients: a
meta-analysis. Journal of Behavioral Medicine. 2010;33(1):1-14.
57.
Heron Speirs H, Harvey S, Baken D. Moderators of psycho-oncology therapy
effectiveness: meta-analysis of therapy characteristics. Journal of Psychosocial Oncology.
2013;31(6):617-41.
58.
Faller H, Schuler M, Richard M, Heckl U, Weis J, Küffner R. Effects of PsychoOncologic Interventions on Emotional Distress and Quality of Life in Adult Patients With
Cancer: Systematic Review and Meta-Analysis. Journal of Clinical Oncology. 2013;31(6):78293.
59.
Rains SA, Young V. A Meta-Analysis of Research on Formal Computer-Mediated
Support Groups: Examining Group Characteristics and Health Outcomes. Human
Communication Research. 2009;35(3):309-36.
60.
Zernicke KA, Campbell TS, Speca MP, McCabe-Ruff KM, Flowers SM, Carlson LE. A
Randomized Wait-List Controlled Trial of Feasibility and Efficacy of an Online MindfulnessBased Cancer Recovery Program: The eTherapy for Cancer Applying Mindfulness Trial.
Psychosomatic Medicine. 2014;76(4):257-67.
61.
Eysenbach G, Powell J, Englesakis M, Rizo C, Stern A. Health related virtual
communities and electronic support groups: systematic review of the effects of online peer to
peer interactions. BMJ (Clinical research ed). 2004;328(7449):1166.
62.
Klemm P, Bunnell D, Cullen M, Soneji R, Gibbons P, Holecek A. Online cancer support
groups: a review of the research literature. Computers, informatics, nursing : CIN.
2003;21(3):136-42.
63.
Schnur JB, Montgomery GH. E-counseling in psychosocial cancer care: a survey of
practice, attitudes, and training among providers. Telemedicine journal and e-health : the official
journal of the American Telemedicine Association. 2012 May;18(4):305-8.
64.
Bumpous JM, Gregg JL, Head BA, Keeney C, Pfeifer MP, Scharfenberger JA, et al.
Development of a Telehealth Intervention for Head and Neck Cancer Patients. Telemedicine and
e-Health. 2009;15:44+.
65.
Lounsberry JJ, MacRae H, Angen M, Hoeber M, Carlson LE. Feasibility study of a
telehealth delivered, psychoeducational support group for allogeneic hematopoietic stem cell
transplant patients. Psycho-Oncology. 2010;19(7):777-81.
66.
Lieberman MA, Golant M, Giese-Davis J, Winzlenberg A, Benjamin H, Humphreys K, et
al. Electronic support groups for breast carcinoma. Cancer. 2003;97(4):920-5.
139
67.
Bleyer W. Cancer in older adolescents and young adults: epidemiology, diagnosis,
treatment, survival, and importance of clinical trials. Medical & Pediatric Oncology.
2002;38(1):1 - 10.
68.
Bleyer AOL, M.; Barr, R.; Ries, L.A.G. (eds):. Cancer Epidemiology in Older
Adolescents and Young Adults 15 to 29 Years of Age, Including SEER Incidence and Survival:
1975-2000. National Cancer Institute. 2006;NIH Pub. No. 06-5767.
69.
NCCN. Adolescent and Young Adult Oncology Clinical Practice Guidelines - Version
1.2012.
70.
Bleyer A. How NCCN Guidelines Can Help Young Adults and Older Adolescents With
Cancer and the Professionals Who Care for Them. JNCCN. 2012;10(9):1065-71.
71.
Adolescents and Young Adults with Cancer: Towards Better Outcomes in Canada.
Cancer. 2011;117(S10):i-ii.
72.
Lang M, David, V, Giese-Davis J. The Age Conundrum: A scoping review of “younger
age” or “adolescent and young adult” as a risk factor for clinical distress, depression or anxiety
in cancer. Under Review by the Journal of Adolescent and Young Adult Oncology. 2015.
73.
Evan EE, Zeltzer LK. Psychosocial dimensions of cancer in adolescents and young
adults. Cancer. 2006;107(S7):1663-71.
74.
Zebrack BJ. Psychological, social, and behavioral issues for young adults with cancer.
Cancer. 2011;117(S10):2289-94.
75.
Roberts CST, M.E.; Knowles, A.M. Psychosocial issues of adolescents with cancer. . Soc
Work Health Care 1998;27:3-18.
76.
Sutcliffe SB. A review of Canadian health care and cancer care systems. Cancer.
2011;117(S10):2241-4.
77.
Bellizzi KM, Smith A, Schmidt S, Keegan THM, Zebrack B, Lynch CF, et al. Positive
and negative psychosocial impact of being diagnosed with cancer as an adolescent or young
adult. Cancer. 2012;118(20):5155-62.
78.
Kwak M, Zebrack BJ, Meeske KA, Embry L, Aguilar C, Block R, et al. Prevalence and
predictors of post-traumatic stress symptoms in adolescent and young adult cancer survivors: a
1-year follow-up study. Psycho-Oncology. 2012;22(8):1798-1806.
79.
Eiser CH, J.J; Vance, Y.H. Examining the psychological consequences of surviving
childhood cancer: Systematic review as a research method in pediatric psychology. J Pediatr
Psychol. 2000;25:449-60.
80.
Burns DSR, S.L.; Haase, J.E. Exploring the feasibility of a therapeutic music video
intervention in adolescents and young adults during stem-cell transplantation. Cancer Nurs.
2009;32:E8-E16.
81.
Thompson CM, Crook B, Love B, Macpherson CF, Johnson R. Understanding how
adolescents and young adults with cancer talk about needs in online and face-to-face support
groups. Journal of Health Psychology. Published online April 27, 2015.
82.
Kato PC, S.; Bradlyn, A., et. al. A video game improves behavioral outcomes in
adolescents and young adults with cancer: A randomized trial. Pediatrics. 2008;122:305-17.
83.
Santacroce SJA, K.; Kadan-Lottick, N., et. al. Feasibility and preliminary outcomes from
a pilot study of coping skills training for adolescent-young adult survivors of childhood cancer
and their parents. Journal of pediatric oncology nursing. 2009;27:10-20.
84.
Gafni A. Treatment of cancer in adolescents and young adults. Cancer.
2011;117(S10):2258-61.
140
85.
Stephen J, Collie K, McLeod D, Rojubally A, Fergus K, Speca M, et al. Talking with
text: Communication in therapist-led, live chat cancer support groups. Social Science &
Medicine. 2014;104(0):178-86.
86.
Stewart M, Barnfather A, Magill-Evans J, Ray L, Letourneau N. Brief report: An online
support intervention: Perceptions of adolescents with physical disabilities. Journal of
Adolescence. 2011;34(4):795-800.
87.
van der Zanden RK, J.; Gerrits, R: Cuijpers, P. Effectiveness of an Online Group Course
for Depression in Adolescents and Young Adults: A Randomized Trial. Journal of Medical
Internet Research. 2012;14(3).
88.
Klemm P, Hardie T. Depression in Internet and face-to-face cancer support groups: a
pilot study. Oncology Nursing Forum. 2002;29(4):E45-51.
89.
Leykin Y, Thekdi SM, Shumay DM, Muñoz RF, Riba M, Dunn LB. Internet
Interventions for Improving Psychological Well-Being in Psycho-Oncology: Review and
Recommendations. Psycho-oncology. 2012;21(9):1016-25.
90.
Houston TK, Cooper LA, Ford DE. Internet support groups for depression: a 1-year
prospective cohort study. The American journal of psychiatry. 2002;159(12):2062-8.
91.
Kent EE, Smith AW, Keegan THM, Lynch CF, Wu X-C, Hamilton AS, et al. Talking
About Cancer and Meeting Peer Survivors: Social Information Needs of Adolescents and Young
Adults Diagnosed with Cancer. Journal of Adolescent and Young Adult Oncology.
2013;2(2):44-52.
92.
Trevino KM, Fasciano K, Block S, Prigerson HG. Correlates of social support in young
adults with advanced cancer. Support Care Cancer. 2013;21(2):421-9.
93.
Rabin C, Simpson N, Morrow K, Pinto B. Intervention Format and Delivery Preferences
Among Young Adult Cancer Survivors. IntJ Behav Med. 2013;20(2):304-10.
94.
Sally H. Barlow GMB, Addie J. Fuhriman. The History of Group Practice: A Century of
Knowledge. The Handbook of Group Research and Practice. SAGE Publications, Inc. Thousand
Oaks, CA: SAGE Publications, Inc. 39-65 p.
95.
Yalom ID. The Theory and practice of group psychotherapy Basic Books. 1970;1.
96.
Lieberman MA, Yalom ID, Miles MB. Encounter groups: first facts: Basic Books; 1973.
97.
Antoni MH, Antoni MH, Lehman JM, Klibourn KM, Boyers AE. Cognitive-behavioral
stress management intervention decreases the prevalence of depression and enhances benefit
finding among women under treatment for early-stage breast cancer. Health psychology.
2001;20(1):20-32.
98.
Classen C, Butler LD, Koopman C, et al. Supportive-expressive group therapy and
distress in patients with metastatic breast cancer: A randomized clinical intervention trial.
Archives of General Psychiatry. 2001;58(5):494-501.
99.
Goodwin PJ, Leszcz M, Ennis M, Koopmans J, Vincent L, Guther H, et al. The Effect of
Group Psychosocial Support on Survival in Metastatic Breast Cancer. New England Journal of
Medicine. 2001;345(24):1719-26.
100. Spiegel D, Bloom JR, Yalom I. Group support for patients with metastatic cancer: A
randomized prospective outcome study. Archives of General Psychiatry. 1981;38(5):527-33.
101. Koopman C, Giese-Davis J, Palesh O, Neri E. Effects of supportive-expressive group
therapy on pain in women with metastatic breast cancer. Health psychology. 2009;28(5):579.
102. Spiegel D, Bloom JR. Group therapy and hypnosis reduce metastatic breast carcinoma
pain : Psychosom. Med., 45 (1983) 333–339. Pain. 1985;22(2):215-6.
141
103. Giese-Davis J, Koopman C, Butler LD, Classen C. Change in emotion-regulation strategy
for women with metastatic breast cancer following supportive-expressive group therapy. Journal
of consulting and clinical psychology. 2002;70(4):916-25.
104. Antoni MH, Lechner S, Diaz A, Vargas S, Holley H, Phillips K, et al. Cognitive
behavioral stress management effects on psychosocial and physiological adaptation in women
undergoing treatment for breast cancer. Brain, Behavior, and Immunity. 2009;23(5):580-91.
105. Mustafa M, Carson-Stevens A, Gillespie D, Edwards AGK. Psychological interventions
for women with metastatic breast cancer. Cochrane database of systematic reviews.
2013;6:CD004253.
106. Antoni M. Psychosocial intervention effects on adaptation, disease course and
biobehavioral processes in cancer. Brain, Behavior, and Immunity. 2013:S88-98
107. Angus L. Toward an integrative understanding of narrative and emotion processes in
Emotion-focused therapy of depression: Implications for theory, research and practice.
Psychotherapy Research. 2012;22(4):367-80.
108. Giese-Davis J, Koopman C, Butler LD, Classen C, Cordova M, Fobair P, et al. Change in
Emotion-Regulation Strategy for Women With Metastatic Breast Cancer Following SupportiveExpressive Group Therapy. Journal of Consulting & Clinical Psychology. 2002;70(4):916-25.
109. Tamagawa R, Li Y, Gravity T, Piemme K, DiMiceli S, Collie K, et al. Deconstructing
therapeutic mechanisms in cancer support groups: do we express more emotion when we tell
stories or talk directly to each other? J Behav Med. 2015;38(1):171-82.
110. Kissane DW, Grabsch B, Clarke DM, Christie G, Clifton D, Gold S, et al. Supportiveexpressive group therapy: The transformation of existential ambivalence into creative living
while enhancing adherence to anti-cancer therapies. Psycho-Oncology. 2004;13(11):755-68.
111. Brown NW. Psychoeducational Groups. The Handbook of Group Research and Practice.
SAGE Publications, Inc. Thousand Oaks, CA: SAGE Publications, Inc. 511-31 p.
112. Docherty SL, Kayle M, Maslow GR, Santacroce SJ. The Adolescent and Young Adult
with Cancer: A Developmental Life Course Perspective. Seminars in Oncology Nursing. 2012.
113. Edgar L, Rosberger Z, Nowlis D. Coping with cancer during the first year after diagnosis.
Assessment and intervention. Cancer. 1992;69(3):817-28.
114. Greenberg LS. Emotion–focused therapy. Clinical Psychology & Psychotherapy.
2004;11(1):3-16.
115. Angus LE, Greenberg, Leslie S. Working with narrative in emotion-focused therapy:
Changing stories, healing lives.: American Psychological Association; 2011.
116. Spiegel D, Classen C. Group Therapy for Cancer Patients: A Research-Based Handbook
of Psychosocial Care: Basic Books; 2008.
117. Greenberg LS. Emotion and change processes in psychotherapy. . In M Lewis & J M
Haviland (Eds), Handbook of Emotions. New York: Guilford Press. 1993;499-519.
118. Foa EB, Davidson, J., & Rothbaum, B. O. . Treatment of Post-Traumatic Stress Disorder.
. In Gabbard G O (Ed), Treatment of Psychiatric Disorders: The DSM-IV. Washington D.C.:
American Psychiatric Press.1994.
119. Janine Giese-Davis SD, Sandra Seohton, David Spiegel. Emotional expression and
diurnal cortisol slope in women with metastatic breast cancer in supportive-expressive group
therapy: A preliminary study. Bilogical Psychology. 2006;73:190-8.
142
120. Foa EB, Riggs DS, Massie ED, Yarczower M. The impact of fear activation and anger on
the efficacy of exposure treatment for posttraumatic stress disorder. Behavior Therapy. 1995
//Summer;26(3):487-99.
121. Media Metrix: 3 Mo. Avg. Home & Work Internet Usage. Media Metrix. Q4 2011.
122. Kandler LSL, M.; Sellick, S.M. Understanding the young adult cancer patient experience:
An examination of viewer responses to a patient-led film initiative. APOS 2012 PsychoOncology. 2012;21(S1):Abstract P1-5.
123. Bollinger S, Kreuter MW. Real-time moment-to-moment emotional responses to
narrative and informational breast cancer videos in African American women. Health education
research. 2012;27(3):537-43.
124. Andersen BL, Shelby RA, Golden-Kreutz DM. RCT of a Psychological Intervention for
Patients With Cancer: I. Mechanisms of Change. Journal of Consulting & Clinical Psychology.
2007;75(6):927-38.
125. Lancaster GA, Dodd S, Williamson PR. Design and analysis of pilot studies:
recommendations for good practice. Journal of Evaluation in Clinical Practice. 2004;10(2):30712.
126. Randy H. Magen EM. 19 Groups and individual Change. The Handbook of Group
Research and Practice. SAGE Publications, Inc. Thousand Oaks, CA: SAGE Publications, Inc.
351-63 p.
127. Burlingame GM, Fuhriman A, Johnson JE. COHESION IN GROUP
PSYCHOTHERAPY. Psychotherapy: Theory/Research/Practice/Training Winter.
2001;38(4):373-9.
128. Yalom ID, Rand K. COmpatibility and cohesiveness in therapy groups. Archives of
General Psychiatry. 1966;15(3):267-75.
129. Beeber LS, Schmitt, M.H. Cohesiveness in groups: a concept in search of a definition.
Advances in nursing science. 1986;8(2):11.
130. Wongpakaran T, Wongpakaran N, Intachote-Sakamoto R, Boripuntakul T. The Group
Cohesiveness Scale (GCS) for Psychiatric Inpatients. Perspectives in Psychiatric Care.
2013;49(1):58-64.
131. Burlingame GM, McClendon DT, Alonso J. Cohesion in Group Therapy. Psychotherapy.
2011;48(1):34-42.
132. Giese-Davis J, Kronenwetter, C., Golant, M., Cordova, M., Twirbutt, S., Chang, V.,
Spiegel, D. Methods to illustrate multi-faceted dimensions of group therapy and support. (In
Preparation). .
133. Stanton AL. Psychosocial concerns and interventions for cancer survivors. J Clin Oncol.
2006;24(32):5132-7.
134. Carlson LE, Bultz BD. Cancer distress screening: Needs, models, and methods. Journal
of Psychosomatic Research. 2003 11;55(5):403-9.
135. Boyes AW, Girgis A, D'Este C, Zucca AC. Flourishing or floundering? Prevalence and
correlates of anxiety and depression among a population-based sample of adult cancer survivors
6 months after diagnosis. Journal of Affective Disorders. 2011;135(1–3):184-92.
136. Zabora J, BrintzenhofeSzoc K, Curbow B, Hooker C, Piantadosi S. The prevalence of
psychological distress by cancer site. Psycho-Oncology. 2001;10(1):19-28.
137. Carlson LEA, M.; Cullum, J.; et al. High levels of untreated distress and fatigue in cancer
patients. Br J Cancer. 2004;90:2297-304.
143
138. Noyes R, Kathol RG, Debelius-Enemark P, Williams J, Mutgi A, Suelzer MT, et al.
Distress Associated with Cancer as Measured by the illness Distress Scale. Psychosomatics.
1990 8//;31(3):321-30.
139. Elder GH, Jr. The Life Course as Developmental Theory. Child Development.
1998;69(1):1-12.
140. Friedman GF, V.; Zernitsky-Shurka, E. The Experience of Loneliness Among young
Adult Cancer Patients. Journal of Psychosocial Oncology. 1989;7(3).
141. Ozer EJB, Suzanne R.; Lipsey, Tami L.; Weiss, Daniel S. Predictors of Posttraumatic
Stress Disorder and Symptoms in Adults: A Meta-Analysis. Psychological Bulletin.
2003;129(1):52–73.
142. Smith MY, Redd WH, Peyser C, Vogl D. Post-traumatic stress disorder in cancer: a
review. Psycho-Oncology. 1999;8(6):521-37.
143. Andrykowski MA, Cordova MJ, Studts JL, Miller TW. Posttraumatic Stress Disorder
After Treatment for Breast Cancer: Prevalence of Diagnosis and Use of the PTSD ChecklistCivilian Version (PCL-C) as a Screening Instrument. Journal of Consulting & Clinical
Psychology. 1998;66(3):586-90.
144. Palesh O, Koopman C. Breast cancer: Post-traumatic stress disorder[mdash]prevalent and
persistent. Nat Rev Clin Oncol. 2013;10(5):252-4.
145. Cordova MJ, Andrykowski MA, Kenady DE, McGrath PC, Sloan DA, Redd WH.
Frequency and Correlates of Posttraumatic-Stress-Disorder-Like Symptoms After Treatment for
Breast Cancer. Journal of Consulting & Clinical Psychology. 1995;63(6):981-6.
146. Levine EG, Eckhardt J, Targ E. Change in post-traumatic stress symptoms following
psychosocial treatment for breast cancer. Psycho-Oncology. 2005;14(8):618-35.
147. Cordova MJ, Studts JL, Hann DM, Jacobsen PB, Andrykowski MA. Symptom structure
of PTSD following breast cancer. Journal of traumatic stress. 2000;13(2):301-19.
148. Koutrouli N, Anagnostopoulos F, Potamianos G. Posttraumatic Stress Disorder and
Posttraumatic Growth in Breast Cancer Patients: A Systematic Review. Women & Health.
2012;52(5):503-16.
149. Taku K, Cann A, Calhoun LG, Tedeschi RG. The factor structure of the posttraumatic
growth inventory: a comparison of five models using confirmatory factor analysis. Journal of
traumatic stress. 2008;21(2):158-64.
150. Zebrack B, Kwak M, Salsman J, Cousino M, Meeske K, Aguilar C, et al. The relationship
between posttraumatic stress and posttraumatic growth among adolescent and young adult
(AYA) cancer patients. Psycho-Oncology. 2015;24(2):162-8.
151. Merluzzi TVS, Mary Ann Martinez Assessment of Self-Efficacy and Coping With
Cancer: Development and Validation of the Cancer Behavior Inventory. Health Psychology.
1997;16(2):163-70.
152. Mystakidou K, Tsilika E, Parpa E, Gogou P, Theodorakis P, Vlahos L. Self-efficacy
beliefs and levels of anxiety in advanced cancer patients. European Journal of Cancer Care.
2010;19(2):205-11.
153. Bandura A. Self-efficacy: Toward a unifying theory of behavioral change. Psychological
review. 1977;84(2):191-215.
154. Fergus KD, McLeod D, Carter W, Warner E, Gardner SL, Granek L, et al. Development
and pilot testing of an online intervention to support young couples' coping and adjustment to
breast cancer. Eur J Cancer Care (Engl). 2014;23(4):481-92.
144
155. David N, Schlenker P, Prudlo U, Larbig W. Internet-based program for coping with
cancer: a randomized controlled trial with hematologic cancer patients. Psychooncology.
2013;22(5):1064-72.
156. Kortum PT, Bangor A. Usability Ratings for Everyday Products Measured With the
System Usability Scale. International Journal of Human-Computer Interaction. 2012;29(2):6776.
157. Brooke J. SUS: A “quick and dirty” usability scale. In Usability evaluation in industry,
Edited by: Jordan, PW, Thomas B, Weerdmeester, BA and McClelland IL. 1996;UK: Taylor &
Francis:189–94.
158. Bangor A, Kortum PT, Miller JT. An Empirical Evaluation of the System Usability
Scale. International Journal of Human-Computer Interaction. 2008;24(6):574-94.
159. Clinton-McHarg T, Carey M, Sanson-Fisher R, Shakeshaft A, Rainbird K. Measuring the
psychosocial health of adolescent and young adult (AYA) cancer survivors: a critical review.
Health and quality of life outcomes. 2010;8:25.
160. Wakefield CE, Patterson P, McDonald FEJ, Wilson HL, Davis E, Sansom-Daly UM.
Assessment of psychosocial outcomes in adolescents and young adults with cancer: a systematic
review of available instruments. Clinical Oncology in Adolescents and Young Adults.
2013;3:13+.
161. Kraemer H, Mintz J, Noda A, Tinklenberg J, Yesavage JA. CAution regarding the use of
pilot studies to guide power calculations for study proposals. Archives of General Psychiatry.
2006;63(5):484-9.
162. Leon AC, Davis LL, Kraemer HC. The role and interpretation of pilot studies in clinical
research. Journal of Psychiatric Research. 2011;45(5):626-9.
163. Hertzog MA. Considerations in determining sample size for pilot studies. Research in
Nursing & Health. 2008;31(2):180-91.
164. Owen J, Bantum EC, Criswell K, Bazzo J, Gorlick A, Stanton A. Representativeness of
two sampling procedures for an internet intervention targeting cancer-related distress: a
comparison of convenience and registry samples. J Behav Med. 2014;37(4):630-41.
165. Suresh K. An overview of randomization techniques: An unbiased assessment of
outcome in clinical research. Journal of human reproductive sciences. 2011 Jan;4(1):8-11.
166. Cancerchatcanada.ca. 2013.
167. Stephen JE, Christie G, Flood K, Golant M, Rahn M, Rennie H, et al. Facilitating online
support groups for cancer patients: the learning experience of psycho-oncology clinicians.
Psychooncology. 2011;20(8):832-40.
168. Rosberger Z, Edgar L, Collet J-P, Fournier MA. Patterns of Coping in Women
Completing Treatment for Breast Cancer: A Randomized Controlled Trial of Nucare, a Brief
Psychoeducational Workshop. Journal of Psychosocial Oncology. 2002;20(3):19-37.
169. Kelly PK, Meza J, Pyke-Grimm KA, Stewart JL. Feasibility, acceptability, and usability
of web-based data collection in parents of children with cancer. Oncology Nursing Forum.
2011:428+.
170. Oken MMMDa, Creech RHMDb, Tormey DCMDPDc, Horton JMDCBd, Davis TEMDc,
McFadden ETe, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group.
American Journal of Clinical Oncology. 1982;5(6):649-56.
171. Giese-Davis J. Group Experiences Questionnaire. Unpublished Self-Report Scale. (1995).
145
172. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatrica
Scandinavica. 1983;67(6):361-70.
173. Ziegler L, Hill K, Neilly L, Bennett MI, Higginson IJ, Murray SA, et al. Identifying
Psychological Distress at Key Stages of the Cancer Illness Trajectory: A Systematic Review of
Validated Self-Report Measures. Journal of Pain and Symptom Management. 2011;41(3):61936.
174. Snaith RP. The Hospital Anxiety And Depression Scale. Health and quality of life
outcomes. 2003;1:29.
175. Herrmann C. International experiences with the Hospital Anxiety and Depression ScaleA review of validation data and clinical results. Journal of Psychosomatic Research. 1997
1//;42(1):17-41.
176. Mitchell AJ, Meader N, Symonds P. Diagnostic validity of the Hospital Anxiety and
Depression Scale (HADS) in cancer and palliative settings: A meta-analysis. Journal of Affective
Disorders. 2010;126(3):335-48.
177. Russell DW. UCLA Loneliness Scale (Version 3): Reliability, Validity, and Factor
Structure. Journal of Personality Assessment. 1996;66(1):20.
178. Smith MY, Redd W, DuHamel K, Vickberg SJ, Ricketts P. Validation of the PTSD
Checklist–Civilian Version in Survivors of Bone Marrow Transplantation. Journal of traumatic
stress. 1999;12(3):485.
179. Wilkins KC, Lang AJ, Norman SB. Synthesis of the psychometric properties of the PTSD
checklist (PCL) military, civilian, and specific versions. Depression & Anxiety. 2011;28(7):596606.
180. Shakespeare-Finch J, Enders T. Corroborating evidence of posttraumatic growth. Journal
of traumatic stress. 2008;21(4):421-4.
181. Tedeschi RG, Calhoun LG. The Posttraumatic Growth Inventory: measuring the positive
legacy of trauma. Journal of traumatic stress. 1996 Jul;9(3):455-71.
182. Love C, Sabiston CM. Exploring the links between physical activity and posttraumatic
growth in young adult cancer survivors. Psycho-Oncology. 2011;20(3):278-86.
183. McMillen JC, Fisher RH. The Perceived Benefit Scales: measuring perceived positive
life changes after negative events. Social Work Research. 1998;22(3):173-86.
184. Heitzmann CM, T.; Jean-Pierre, P.; Roscoe, J.; Kirsh K.; Passik, S. Assessing selfefficacy for coping with cancer: development and psychometric analysis of the brief version of
the Cancer Behavior Inventory (CBI-B). Psycho-Oncology. 2011;20:302–12.
185. Monson CM, 2; Gradus, Jaimie L.1,2; Young-Xu, Yinong3; Schnurr, Paula P.3,4; Price,
Jennifer L.5; Schumm, Jeremiah A.6,7. Change in Posttraumatic Stress Disorder Symptoms: Do
Clinicians and Patients Agree? Psychological Assessment. 2008;20(2):131–8.
186. Giese-Davis J, Conrad A, Nouriani B, Spiegel D. Exploring emotion-regulation and
autonomic physiology in metastatic breast cancer patients: Repression, suppression, and restraint
of hostility. Personality and Individual Differences. 2008;44(1):226-37.
187. Winterbottom A, Bekker HL, Conner M, Mooney A. Does narrative information bias
individual's decision making? A systematic review. Social science & medicine (1982).
2008;67(12):2079-88.
188. Høybye MT, Johansen C, Tjørnhøj-Thomsen T. Online interaction. Effects of storytelling
in an internet breast cancer support group. Psycho-Oncology. 2005;14(3):211-20.
146
189. Zebrack B, Chesler M. A psychometric analysis of the Quality of Life-Cancer Survivors
(QOL-CS) in survivors of childhood cancer. Qual Life Res. 2001;10(4):319-29.
190. Hendricks-Ferguson VL, Cherven BO, Burns DS, Docherty SL, Phillips-Salimi CR, Roll
L, et al. Recruitment Strategies and Rates of a Multi-Site Behavioral Intervention for
Adolescents and Young Adults With Cancer. Journal of Pediatric Health Care. 2013;27(6):43442.
191. Rabin C, Dunsiger S, Ness KK, Marcus BH. Internet-Based Physical Activity
Intervention Targeting Young Adult Cancer Survivors. J Adolesc Young Adult Oncol.
2011;1(4):188-94.
192. Golden-Kreutz DM, Andersen BL. Depressive symptoms after breast cancer surgery:
Relationships with global, cancer-related, and life event stress. Psycho-Oncology.
2004;13(3):211-20.
193. Browne RH. On the use of a pilot sample for sample size determination. Statistics in
Medicine. 1995;14(17):1933-40.
147