A Hope-based Intervention to Address Disrupted Goal Pursuits and Quality of Life Among Young Adult Cancer Survivors

Abstract

Over 70,000 US young adults are diagnosed with cancer annually, disrupting important life transitions and goal pursuits. Hope is a positive psychology construct associated with better quality of life (QOL) that focuses on goal-oriented thinking. We developed and tested Achieving Wellness After Kancer in Early life (AWAKE), a scalable eight-week app-based program consisting of educational videos, mood/activity tracking, and telephone-based coaching to promote hope and QOL in young adult cancer survivors (YACS, 18–40 years old). A two-arm RCT was used to examine the feasibility, acceptability, and potential efficacy of AWAKE (n=38) versus attention control (AC; n=18) among YACS within two years of completing treatment and recruited from two NCI-designated cancer centers. Outcomes including hope (Trait Hope Scale), QOL (36-Item Short Form Health Survey; Functional Assessment of Cancer Therapy-General), depressive symptoms (Patient Health Questionnaire-9), and substance use were assessed at baseline, 8 weeks, and 6 months. Participants were an average of 32.55 (SD=5.45) years old; 75.0% were female, and 80.4% White. The most common cancers were breast (28.6%), melanoma (16.1%), and leukemia/lymphoma (12.5%). High retention, engagement, and satisfaction rates were documented in both conditions; AWAKE versus AC participants rated video content as more relevant (p=.007) and reported greater likelihood of talking positively about the program (p=.005). Many efficacy change scores showed positive trends in AWAKE versus AC. Reorienting to one’s goal pursuits after cancer diagnosis and treatment is critical and may be supported through hope-based interventions. Findings suggest that the AWAKE warrants subsequent research testing its efficacy, effectiveness, and scalability.

INTRODUCTION

In 2018, there will be an estimated 1,735,350 new cancer cases diagnosed and 609,640 cancer deaths in the US [1]. As new patients continue to be diagnosed with cancer and survival continues to improve, the number of survivors in the US will increase [1]. Thus, it is critical to address the long-term needs of cancer survivors. Particularly relevant to this study, a recent systematic review [2] documented the impact of cancer on individual life goals. In many cases, it results in shifts or disturbances in goals across life domains. Negative consequences such as fewer life goals, achievement-related and leisure goals, and long-term goals may result, which is related to poorer psychological outcomes in cancer patients [2].

In 2018, 70,000 new cancer cases were estimated to be diagnosed among adolescents and young adults between ages 15–39 [3]. Over 60% of adolescents and young adults will suffer adverse physical, psychosocial, and/or behavioral outcomes months or years after completion of cancer treatment [3–6]. This period, particularly young adulthood, is a critical time for negotiating several life transitions and establishing and pursuing important life goals [2]. This group frequently suffers developmental, social, mental health, and educational setbacks as a result of their cancer treatment [7]. Not surprisingly, the aforementioned review documented that, compared to older patients, young adult cancer survivors (YACS) report more life goal disturbances [2], which are related to negative psychological outcomes [2]. One recent study of YACS indicated several concerns about employment and career, intimate relationships and building a family, financial concerns, and maintaining one’s health – all important life goals [8, 9].

In recent years, adolescents and young adults have received increased attention as a population that experiences disparities in cancer care [4–7, 10–17]. Prior reports highlighted the need to enhance both quality of life (QOL) and care quality for this group and call for providers and hospital systems to address the specific health and psychosocial needs of adolescent and young adults who have faced cancer [4–7, 10–17]. A recent review indicated that individuals with cancer may believe that their existential needs are not being met [18]. This review [18], as well as the larger community of cancer researchers and practitioners [7, 16, 19], call for developing health promotion strategies to address the psychosocial needs and support improved QOL among adolescents and young adults with cancer. However, few resources currently address these needs.

Positive psychology is the scientific study of the strengths that enable individuals and communities to thrive [20]. Indeed, the aforementioned systematic review highlighted research indicating that some cancer survivors experience positive changes in life goals, including having more intrinsic goals (e.g., social, transcendental, and health-related goals) over time compared with healthy peers [2]. Goal revision and reprioritization may be related to positive psychological outcomes [2], leading to more positive affect [21].

Despite the relevance of positive psychology, limited intervention research has been conducted in this area. A recent systematic review [22] identified only 16 studies of positive psychology interventions for breast cancer patients, representing five foundational approaches: mindfulness- based approaches, expression of positive emotions, spiritual interventions, meaning-making interventions, and hope therapy. These specific interventions promoted some positive changes (e.g., enhanced QOL, well- being, benefit-finding, or hope); however, the long-term impact of these interventions and other methodological issues (e.g., small sample sizes) undermine the existing research on positive psychology interventions for cancer survivors.

Hope is a subconstruct within the broader domain of QOL [23] that may be particularly relevant and important for YACS. There are several ways to define hope based on both cultural and scientific perspectives. Particularly relevant to goal pursuits, according to Snyder and colleagues [24], the construct of hope involves: 1) goals, which are targets of mental or physical action; 2) the perception of having both the routes to reach one’s goals (pathways thinking); and 3) the motivation (agency) to use those routes. Pathways thinking reflects the ability to generate multiple plausible routes to goals and to choose new routes when an obstacle blocks the attainment of a desired goal. Agency reflects the ability to initiate and sustain movement and motivation toward goals, particularly in the face of obstacles. Pathways thinking and agency are distinct, but highly correlated factors, and together, they iterate and drive engagement and disengagement throughout goal pursuit. Compared to lower hope individuals, those with higher hope more clearly conceptualize their goals, more often see one major pathway but also numerous alternative pathways and goals, are more prepared for setbacks, bounce back from losses more quickly, and establish goals in a greater number of life domains (e.g., academic, relationships, spirituality, physical health) [25].

Hope is related to promoting and maintaining good health as well as preventing and treating illness [24]. Particularly relevant, higher hope is related to better adjustment in coping with cancer, specifically breast cancer [26], as well as better pain tolerance [27]. Moreover, higher hope – in groups including young adults and cancer survivors – is related to lower rates of binge drinking and smoking, more frequent exercising, and better nutrition [28–30] – critical factors in preventing cancer recurrence and increasing QOL [31]. Thus, hope is a relevant and critical psychological construct for cancer survivors, particularly those in young adulthood.

Hope has been used as a basis for intervention. As mentioned above [22], one hope-based intervention study focused on cancer survivors was conducted by Rustoen et al. [32]. The intervention consisted of eight weekly two-hour sessions provided to 195 participants in a single-group longitudinal study. Results indicated an increase in hope levels immediately after therapy, but reductions in hope thereafter (three and twelve months post-treatment). Despite the limitations of the Rustoen et al. study [32], other intervention research leveraging Snyder’s definition of hope has shown promising results in other populations. One example is an eight-session group treatment emphasizing building goal-pursuit skills [33]. Findings from a randomized, wait-list control trial using a community sample indicated that this intervention demonstrated efficacy in increasing hope, life meaning, and self-esteem as well as reductions in symptoms of depression and anxiety [33]. Moreover, a brief one-session hope-based intervention targeting hope and pain tolerance resulted in increased hope and pain tolerance among 172 young adults [27]. These results suggest that hope-based interventions can increase some psychological strengths, reduce some symptoms of psychopathology, and enhance coping with physical symptoms.

Integrating positive psychology – and hope in particular – into YACS patient care might be strategic in the process of enhancing mental health [33], coping [24, 26, 27, 33], health behaviors [28–30], and overall QOL [31, 33]. Thus, the current proposal involved the development of Achieving Wellness After Kancer in Early life (AWAKE), a scalable eight-week app-based program consisting of educational videos, mood/activity tracking, and telephone-based coaching to promote goal-oriented thinking and QOL in YACS. Using a two-arm pilot RCT, we examined the feasibility, acceptability, and potential efficacy of AWAKE versus an attention control (AC) among YACS within 2 years of treatment completion and recruited from two NCI-designated cancer centers. The primary outcome was changes in hope, with secondary outcomes of changes in QOL, depressive symptoms, and substance use.

METHODS

Study Design

This study was approved by Institutional Review Boards at Emory University and the University of Kentucky. The current study leveraged a two-arm pilot RCT to determine feasibility and acceptability of the AWAKE intervention approach and estimate effect sizes for a subsequent efficacy trial. AWAKE was compared to an AC condition. Both involved two primary components: 1) smartphone app-based communication/content; and 2) coaching. Table 1 presents the components involved in the AWAKE and the AC conditions.

Table 1.

Components of the AWAKE vs. attention control (AC) conditions

Component	AWAKE	AC
Phone-based functioning	8 weekly modules presented via AWAKE app.	8 weekly modules presented via text messaging.
	Included daily health behavior and mood tracking capability.	No daily health behavior and mood tracking capability.
Module content	Session 1: Hope: Goals, Willpower, and Waypower: Introduction to the program, what it entails, overview of hope and its components, and the relevance of hope to cancer survivors.	Session 1: How to Save Money: Introduction to budget, regular savings account, and retirement accounts including CD’s.
	Session 2: Get and Stay Motivated! How thoughts impact emotions, energy, and motivation to pursue goals during challenges.	Session 2: How to Get a Raise: Overview of challenges and helpful tips for getting a raise.
	Session 3: Setting Good Goals: How to set goals, establishing concrete goals with specific endpoints, using an approach rather than avoidance approach, and creating subgoals.	Session 3: How to Budget in Excel: How to create a monthly budget tracking in excel, focusing on five sections - salary, savings goals, expenses, due dates of expenses, and surplus.
	Session 4: No “Stinkin’ Thinkin’”! Using positive self talk and avoiding cognitive distortions (e.g., all-or-nothing thinking, overgeneralization, magnification).	Session 4: 9 Investing Terms You Should Know: A brief introduction to corporate financing and investing.
	Session 5: Waypower: How to Reach Goals! Using waypower to maintain focus on goals, enhancing ability to create multiple pathways to reach goals, and using visualization to enhance ability to reach goals.	Session 5: Smart Ways to Spend $100: Different options to spend extra money at the end of the month.
	Session 6: Physical Willpower: Health Matters! Maintaining motivation, re-evaluating goals throughout the goal pursuit process, and the importance of physical health and well-being in aiding in the goal pursuit process.	Session 6: Where to Invest (and Where to Skimp) In Your Home: Three “Do’s and Don’ts” when shopping for your home.
	Session 7: Dealing with Roadblocks: Staying motivated, figuring out other pathways, and sometimes changing or shifting goals when they are blocked or entail significant obstacles.	Session 7: 12 Ways to be Better with Money Today: Ways to be better with money and finding ways to help track spending habits.
	Session 8: Staying on Track! Staying on track, regardless of mistakes, setbacks, or lapses in motivation.	Session 8: 10 Questions for Healthy Budget: How to best prioritize and allocate income to obtain a healthy budget is discussed.
Homework	Session 1: Hope: Goals, Willpower, and Waypower: Consider the goals that were impacted by having cancer and how they have adjusted/adapted; list five goals (short- and long-term) across different life domains; begin tracking health behaviors and mood.	N/A
	Session 2: Get and Stay Motivated! Prioritize the goals on the list created in Session 1 and select one to focus on for homework assignments.
	Session 3: Setting Good Goals: Revisit list of goals made in Session 1 and specifically the goal selected in Session 2 and ensure that they are concrete goals with specific endpoints, use an approach rather than avoidance approach, and include subgoals.
	Session 4: No “Stinkin’ Thinkin’”! Self-monitor for negative self-talk and identifying cognitive distortions and their impact.
	Session 5: Waypower: How to Reach Goals! Using the goal identified in Session 2, create a diagram of that goal including steps to the goal to help visualize the path and the accomplishments along the way.
	Session 6: Physical Willpower: Health Matters! Consider a time that re-evaluating a goal changed your perspective on the goal; notice how physical activity, alcohol use, tobacco use, and sleep affect mood and willpower using daily health behavior and mood tracking.
	Session 7: Dealing with Roadblocks: Consider obstacles that have been involved in prior goal pursuits and how they were overcome or worked around; in preparation for the final session, summarize experience of the program and major take-aways from the program.
	Session 8: Staying on Track! Use the program and its content to assist in future goal pursuits; ensure enjoyment of goal pursuit process; celebrate accomplishments; and accept and work around challenges and setbacks.
Coaching	Weekly contacts via phone.	Weekly contacts via phone.
	Coverage of 1) the previous week’s content and homework; 2) psychoeducation regarding a new hope-related skill; 3) ways of applying these skills; and 4) the next week’s homework assignment.	Coverage of the week’s module.

AWAKE Intervention

The AWAKE intervention was based on an empirically-supported protocol [33], augmented with additional evidence-based strategies [27], and specifically targets the needs of YACS. As indicated in Table 1, AWAKE’s eight weekly modules covered a range of topics, with homework assignments that aligned with the weekly topics. The app also had a function to monitor their daily mood and health behaviors, which they could customize to fit their health behavior goals (e.g., substance use, physical activity). In addition, the coaches (trained Masters-level female staff) called each intervention participant weekly, adapting an empirically-supported protocol [33]. Each session involved discussing the previous week’s content and homework, psychoeducation regarding a new hope-related skill, ways of applying skills to participants’ lives, and the next week’s homework assignment. The coach sent text messages to participants twice per week encouraging completion of the weekly module and providing them support in their goal pursuits and other sources of support related to that week’s homework.

Attention Control (AC) Condition

The AC group received educational materials regarding personal finance via text each week for eight weeks and weekly opportunities for coaching calls (Table 1). They did not have homework assignments to complete; however, the educational video content made suggestions for applying the content. The AC participants did not have the ability to track their health behaviors and mood.

Procedures and Participants

In order to maximize the diversity of participants (e.g., race/ethnicity, rural/urban), participants were recruited from the Emory University Winship Cancer Institute and the University of Kentucky Markey Cancer Center. Recruitment started at Winship in January 2017 and ended at Markey in January 2018. Potential participants were identified via medical record review. Inclusion criteria were: 1) 18–40 years old; 2) speak English; 3) within two years of cancer treatment completion; 4) functioning smartphone; and 5) willing to complete study activities (including scheduling an in-person appointment with survivorship clinic). Exclusion criteria included: 1) cancer recurrence since treatment completion; 2) diagnosis of a central nervous system cancer (to ensure requisite mental/emotional functioning to engage in the program); 3) prior diagnosis of alcohol or drug dependency, psychosis, bipolar disorder, or major depressive disorder; and 4) in hospice.

At Winship, research staff called potential participants using phone numbers recorded in the medical record; if potential participants were interested, staff verified eligibility and obtained consent. At Markey, staff sent information via mail and email to potential participants using contact information from the medical record; potential participants then contacted the research team via phone or email to indicate interest in participating the study. Eligibility was then verified via an online mini-survey. Eligible participants were then directed via email to the online consent along with the baseline assessment. Enrolled participants, after completion of baseline survey and orientation call, were assigned a study ID number and randomized to a study arm per a pre-determined blocked random number sequence, with stratification of sex (male vs. female) and age (18–30 vs. 31–39 years old). We randomized participants in a 2:1 ratio for the AWAKE intervention versus AC condition in order to be able to examine associations between engagement in the AWAKE intervention in relation to outcomes.

Figure 1 shows a diagram portrays recruitment and retention figures. We officially enrolled 63 participants (41 at Winship; 22 at Markey; 44 in the AWAKE group; 19 in the AC group). Excluding withdrawals (n=7, most of whom withdrew due to time constraints), our final sample size was N=56.

Figure 1.

Flow diagram of participant progress through phases of randomized controlled trial (RCT)

Measures/Data Collection

Participants completed 20-minute online surveys at baseline, eight weeks (end-of-treatment [EOT]), and six months (follow-up [FU]) in order to assess immediate effects (at EOT) and longer-term effects (at FU).

Feasibility and Acceptability

In terms of feasibility, we recorded participation and retention rates. Regarding acceptability, we assessed use of and satisfaction with intervention components (see Table 3).

Table 3.

Process evaluation outcomes at EOT (8 weeks) among participants in the AWAKE vs. attention control (AC) conditions

Variable	AWAKE, N=38	AC, N=18	P
	Mean (SD) or N (%)	Mean (SD) or N (%)
Retention
Number retained at EOT (N, %)	36 (94.7)	18 (100.0)	.289
Number retained at FU (N, %)	33 (86.8)	16 (94.4)	.391
Participation
Number of coaching sessions completed (N, %)	6.16 (2.83)	6.94 (1.4)	.272
Number of weekly videos watched (N, %)	6.79 (2.00)	--	--
Number of weekly homework assignments completed (N, %)	4.50 (2.31)	--	--
Number of weeks completed at least one intervention component (N, %)	6.92 (1.92)	--	--
Participant Assessments	N=36	N=18
How much of the video content did you watch and listen to?a (M, SD)	3.64 (0.93)	3.72 (0.57)	.730
How relevant was the material in the videos to you? b (M, SD)	3.14 (1.00)	2.33 (0.97)	.007
Would you recommend keeping the videos in the program? (N, %)			.071
No	4 (11.4)	6 (33.3)
Yes	31 (88.6)	12 (66.7)
Would you recommend keeping the coaching in the program? (N, %)			.589
No	2 (5.7)	2 (11.8)
Yes	33 (94.3)	15 (88.2)
Would you recommend keeping behavior and mood tracking in the program? (N, %)			--
No	5 (14.7)	--
Yes	29 (85.3)	--
Would you recommend keeping the homework in the program? (N, %)			--
No	6 (17.6)	--
Yes	28 (82.4)	--
Did you talk to your friends or family about the program? (N, %)			.005
No	5 (13.9)	6 (35.3)
Yes, I said good things.	26 (72.2)	4 (23.5)
Yes, I said bad things.	1 (2.8)	0 (0.0)
Yes, I didn’t say anything good or bad.	4 (11.1)	7 (41.2)
Overall, were you satisfied with the program? b (M, SD)	3.14 (1.17)	2.59 (0.87)	.092
Would you recommend participating in this program to friends who are survivors? (N, %)			.999
No	4 (11.1)	2 (11.8)
Yes	32 (88.9)	15 (88.2)

^aOn a scale of 0=none to 4=most or all.

^bOn a scale of 0=not at all to 4=very. (Option to indicate no engagement, N=1.)

Efficacy Outcomes

Primary Outcome.

Our primary efficacy outcome was hope as measured by the Adult Trait Hope Scale [34]. This scale is a psychometrically sound measure [35, 36] and includes four agency items (e.g., “I energetically pursue my goals”), four pathways items (e.g., “I can think of many ways to get out of a jam”), and four distracter items. Each item is responded to on a scale of 1=definitely false to 8=definitely true. This measure has a range of 8 to 64 (4 to 32 for each subscale), with higher scores indicating higher hope symptoms. Cronbach’s alpha for the total scale and for the agency and pathways subscales at baseline were .95, .91, and .89, respectively.

Secondary Outcomes.

Our secondary efficacy outcomes were QOL, depressive symptoms, and substance use. Regarding QOL, we included two strategically chosen measures of QOL: 1) the RAND Medical Outcome Study 36-Item Short Form Health Survey (SF-36) [37] and 2) the Functional Assessment of Cancer Therapy – General (FACT-G) [38]. The SF-36 is standardized to the general population and includes nine subscales (and an additional one-item measure). Each subscale has a possible range of 0 to 100 (with a mean of 50 and a standard deviation of 10); higher scores on each subscale indicate higher functioning. This measure allows for comparisons in relation to overall QOL and the physical and mental dimensions of QOL between our sample and the general population. Cronbach’s alpha for each subscale were as follows: physical functioning: .89, role limitations due to physical health: .82, role limitations due to emotional problems: .89, energy/fatigue: .84, emotional well-being: .79, social functioning: .85, pain: .94, and general health: .83. One item also asks participants to rate their health relative to their health status one year prior.

The FACT-G is a QOL measure specific to cancer patients that provides insight into cancer specific factors in four dimensions. Each item is assessed on a scale from 0=not at all to 4=very much. Possible scores range from 0 to 28, with higher scores indicating higher functioning. Cronbach’s alphas for each of the well-being subscales were as follows: physical: .84, social/family: .84, emotional: .81, and functional: .89.

Depressive symptoms were assessed using the Patient Health Questionnaire – 9 item (PHQ-9) [39]. Each item has response options of 0=not at all to 3=nearly every day. This measure has a range of 0 to 27, with higher scores indicating higher depressive symptoms. Cronbach’s alpha for this measure at baseline was .77.

Regarding substance use, we assessed number of days in the past 30 days participants smoked cigarettes, used alcohol, and used marijuana [40]. (No participants reported cigarette use, so this outcome was not reported in the results.)

Covariates

We assessed sociodemographics, cancer-related factors (i.e., cancer diagnosis, time since diagnosis), and other health-related factors.

Data Analysis

All analyses were conducted using SPSS 25.0. Participant characteristics were summarized using descriptive statistics. We then conducted bivariate analyses (i.e., ANOVAs, Chi-Square tests) to compare AWAKE versus AC participants to determine if randomization resulted in any significant differences between groups.

Feasibility and acceptability measures were summarized using descriptive statistics as well as bivariate analyses. Primary and secondary outcomes were examined using change scores; baseline scores were subtracted from EOT scores and FU scores, respectively.

Allowing for 20% attrition from baseline to FU data collection (yielding 15 in the AC group), our sample size was able to yield valuable information regarding feasibility and acceptability. Additionally, this sample size allowed for estimating effect sizes to inform the design of the subsequent efficacy trial. This sample size allowed us to detect a two-point change hope scores, a 10-point change in SF-36 scores, a 15-point change in FACT-G scores, and a two-point change in PHQ-9 scores.

RESULTS

Participant Characteristics

Participants were an average 32.55 (SD=5.45) years old; 75.0% were female, 80.4% White, 7.1% Hispanic, and 67.9% married or living with their partner (Table 2). The most common cancer types were breast (28.6%), melanoma (16.1%), and leukemia/lymphoma (12.5%), and average time since diagnosis was 2.07 (SD=0.77) years. Only two differences between the AWAKE and AC conditions were found; specifically, more AC than AWAKE participants were married or living with a partner (p=.031) and AWAKE participants had higher scores on role limitations due to physical health compared to AC participants (p=.042), indicating higher functioning.

Table 2.

Comparison of participants in the AWAKE vs. attention control (AC) conditions

Variable	Total, N=56	AWAKE, N=38	AC, N=18	P
	Mean (SD) or N (%)	Mean (SD) or N (%)	Mean (SD) or N (%)
Sociodemographics
Age (M, SD)	32.55 (5.45)	32.63 (5.87)	32.39 (4.60)	.878
Sex (N, %)				.999
Male	14 (25.0)	10 (26.3)	4 (22.2)
Female	42 (75.0)	28 (73.7)	14 (77.8)
Race (N, %)				.314
White	45 (80.4)	32 (84.2)	13 (72.2)
Black	7 (12.5)	3 (7.9)	4 (22.2)
Other	4 (7.1)	3 (7.9)	1 (5.6)
Hispanic (N, %)				.294
No	52 (92.9)	34 (89.5)	18 (100.0)
Yes	4 (7.1)	4 (10.5)	0 (0.0)
Employment (N, %)				.815
Full- or part-time	36 (64.3)	24 (63.2)	12 (66.7)
Student	8 (14.3)	5 (13.2)	3 (16.7)
Other	12 (21.4)	9 (23.7)	3 (16.7)
Parental Education (N, %)				.779
< Bachelors	29 (51.8)	19 (50.0)	10 (17.9)
≥ Bachelors	27 (48.2)	19 (50.0)	8 (44.4)
Household income (N, %)				.563
< $2,400 per month	23 (41.1)	17 (44.7)	6 (33.3)
≥ $2,400 per month	33 (58.9)	21 (55.3)	12 (66.7)
Relationship status (N, %)				.031
Married or living with partner	38 (67.9)	22 (57.9)	16 (88.9)
Other	18 (32.1)	16 (42.1)	2 (11.1)
Children (N, %)				.999
No	25 (44.6)	17 (44.7)	8 (44.4)
Yes	31 (55.4)	21 (55.3)	10 (55.6)
Insurance (N, %)				.333
Other	40 (72.7)	25 (67.6)	15 (83.3)
Medicare/Medicaid	12 (21.8)	9 (24.3)	3 (16.7)
None	3 (5.5)	3 (8.1)	0 (0.0)
Cancer-Related Factors				.432
Cancer type (N, %)
Breast	16 (28.6)	10 (26.3)	6 (33.3)
Melanoma	9 (16.1)	6 (15.8)	3 (16.7)
Leukemia/Lymphoma	7 (12.5)	5 (13.2)	2 (11.1)
Sarcoma	5 (8.9)	5 (13.2)	0 (0.0)
Colorectal	3 (5.4)	1 (2.6)	2 (11.1)
Testicular	3 (5.4)	1 (2.6)	2 (11.1)
Cervical	2 (3.6)	2 (5.3)	0 (0.0)
Other	11 (19.6)	8 (21.1)	3 (16.7)
Time since diagnosis (M, SD)	2.07 (0.77)	2.11 (0.75)	2.00 (0.84)	.244
Clinic site (N, %)				.999
Winship	36 (64.3)	24 (63.2)	12 (66.7)
Markey	20 (35.7)	14 (36.8)	6 (33.3)
Psychosocial Factors
Hope (M, SD)	51.02 (10.86)	49.97 (12.33)	53.06 (7.10)	.346
Agency	25.18 (5.63)	24.82 (6.31)	25.88 (4.09)	.532
Pathways	25.84 (5.61)	25.15 (6.44)	27.18 (3.24)	.230
SF-36 (M, SD)		or
Physical functioning	82.30 (20.63)	81.67 (21.31)	83.53 (19.82)	.766
Role limitations due to physical health	70.50 (36.66)	78.03 (33.52)	55.88 (39.06)	.042
Role limitations due to emotional problems	76.00 (38.71)	80.81 (37.30)	66.67 (40.82)	.225
Energy/fatigue	48.70 (21.85)	51.97 (23.42)	42.35 (17.33)	.142
Emotional well-being	70.64 (16.57)	73.70 (15.72)	64.71 (17.04)	.225
Social functioning	53.25 (21.99)	56.44 (21.22)	47.06 (22.76)	.155
Pain	34.20 (23.35)	35.30 (22.94)	32.06 (24.69)	.646
General health	60.60 (22.58)	62.88 (21.32)	56.18 (24.91)	.325
Health change	72.50 (25.88)	73.48 (24.16)	70.59 (29.53)	.712
FACT-G (M, SD)
Physical well-being	21.82 (5.50)	22.00 (5.53)	21.47 (5.58)	.751
Social/family well-being	21.08 (5.34)	20.92 (5.64)	21.36 (4.94)	.810
Emotional well-being	18.84 (4.55)	19.21 (4.72)	18.12 (4.24)	.426
Functional well-being	20.94 (6.00)	21.39 (5.94)	20.06 (6.20)	.462
Depressive symptoms (M, SD)	6.02 (4.20)	5.36 (3.77)	7.29 (4.79)	.125
Health Behaviors
Days of alcohol use, past 30 days (M, SD)	2.13 (3.70)	1.92 (3.27)	2.61 (4.54)	.519
Days of marijuana use, past 30 days (M, SD)	0.77 (4.14)	0.79 (4.87)	0.72 (1.96)	.955

Feasibility and Acceptability

Table 3 presents data regarding feasibility and acceptability. No differences were found in relation to retention at EOT or FU (exceeding 85%) or number of coaching calls completed (averages exceeding 6 of 8 possible calls). Although participants in the two conditions did not indicate different amounts of videos watched, AWAKE participants rated the content of the videos as more relevant, relative to AC participants (p=.007). The proportion who recommended keeping the videos trended toward being greater among AWAKE versus AC participants (p=.071). In addition, large proportions of AWAKE and AC participants recommended keeping the coaching in the program (94.3% and 88.2%, respectively). Over 80% of AWAKE participants also recommended keeping the behavior/mood tracking and homework in the program. Compared to AC participants, AWAKE participants were more likely to have talked with friends or family about the program and comment positively about it (p=.005). High proportions of participants in both groups (>88%) indicated that they would recommend the program to friends who are cancer survivors.

Primary and Secondary Outcomes

Table 4 shows differences in change scores across primary and secondary outcomes at EOT and FU. While few significant differences were found, many change scores trended in the anticipated direction (i.e., greater improvements among AWAKE vs. AC participants). Particularly noteworthy trends included greater improvements from baseline to FU in overall hope (p=.083) and pathways thinking (p=.052), as well as general health (per the SF-36, p=.038); greater improvements from baseline to EOT were also found in social/family well-being (per the FACT-G, p=.078) and depressive symptoms (per the PHQ-9, p=.061). One trend was identified in the unexpected direction – days of marijuana use increased in AWAKE participants but decreased in AC participants (p=.053).

Table 4.

Participant characteristics and bivariate analyses comparing psychosocial outcomes and health behaviors from baseline to end-of-treatment (EOT; 8 week) and to follow-up (FU; 6 months) among AWAKE vs. attention control (AC) conditions

Variable	EOT		P	FU		P
	AWAKE, N=36	AC, N=18		AWAKE, N=33	AC, N=16
	M (SD)	M (SD)		M (SD)	M (SD)
Psychosocial Factors
Hope	0.69 (13.46)	−2.65 (7.40)	.902	3.73 (9.88)	−1.18 (7.88)	.083
Agency	0.00 (6.97)	−1.00 (3.71)	.584	1.61 (5.04)	−0.24 (3.88)	.195
Pathways	0.70 (7.03)	−1.65 (4.11)	.212	272 (5.45)	−0.94 (4.49)	.052
SF-36
Physical functioning	−1.21 (23.05)	5.88 (14.17)	.253	3.79 (20.69)	6.18 (11.80)	.662
Role limitations due to physical health	28.79 (15.14)	22.79 (20.37)	.245	31.44 (20.99)	30.88 (25.43)	.934
Role limitations due to emotional problems	0.00 (41.67)	−3.92 (30.92)	.734	5.05 (43.40)	0.00 (44.10)	.700
Energy/fatigue	3.64 (17.60)	0.88 (17.76)	.564	2.73 (4.41)	17.59 (19.68)	.759
Emotional well-being	1.33 (13.40)	2.59 (13.34)	.755	2.67 (12.95)	4.71 (14.44)	.614
Social functioning	28.79 (15.14)	22.79 (20.37)	.245	31.44 (20.99)	30.88 (25.43)	.934
Pain	43.26 (24.94)	45.74 (23.89)	.737	43.41 (23.28)	41.62 (18.18)	.783
General health	0.61 (10.21)	2.94 (16.62)	.527	7.12 (16.30)	−3.53 (17.66)	.038
Health change	−2.27 (23.69)	0.00 (21.65)	.742	−6.06 (30.64)	1.47(25.72)	.390
FACT-G
Physical well-being	1.24 (4.04)	1.41 (4.03)	.889	1.79 (3.92)	0.88 (4.46)	.464
Social/family well-being	0.71 (3.35)	−2.15 (5.87)	.078	0.38 (4.18)	1.00 (5.76)	.719
Emotional well-being	−0.64 (2.93)	−0.47 (4.08)	.869	−0.03 (3.40)	−0.41 (3.52)	.712
Functional well-being	0.91 (4.20)	−0.41 (3.50)	272	0.76 (4.74)	0.76 (4.53)	.996
Depressive Symptoms	−1.15 (3.24)	−0.88 (4.36)	.061	−0.58 (3.98)	−1.18 (4.22)	.623
Health Behaviors
Days of drinking, past 30 days (SD)	5.92 (22.36)	1.17 (3.00)	.375	13.00 (33.98)	6.44 (23.26)	.463
Days of marijuana use, past 30 days (SD)	0.11 (0.367)	−0.33 (0.97)	.053	0.34 (3.48)	−0.29 (1.26)	.293

Engagement in AWAKE in Relation to Outcomes

To examine correlations between engagement in the AWAKE program and outcomes, first, we examined overall participation rates. Of the 38 AWAKE participants, 8 (21.1%) engaged in <75% (<6 of 8) weeks of program in relation to any component. Thus, rather than examining subgroup differences, we examined correlations between change scores and program engagement (per number of videos watched, coaching calls completed, homework completed, and completion of any one of these components) among AWAKE participants.

At FU, two outcomes were related to higher engagement in the program: 1) greater increases in SF-36 physical functioning (related to higher rates of completion of any component: r=.39, p=.020); and 2) greater reductions in number of days of alcohol use (related to more videos watched: r=−.62, p<.001; more coaching calls completed: r=−.62, p<.001; and higher rates of completion of any component: r=−.68, p<.001). At EOT, three outcomes were related to higher engagement in the program: 1) greater increases in SF-36 pain-related functioning (related to more videos watched: r=.63, p<.001; more coaching calls completed: r=.40, p=.022; and higher rates of completion of any component: r=.59, p<.001); 2) greater increases in FACT-G physical well-being (related to more videos watched: r=.34, p=.050; and higher rates of completion of any component: r=.39, p=.024); and 3) greater reductions in number of days of alcohol use (related to more videos watched: r=−.49, p=.002; more coaching calls completed: r=−.40, p=.013; and higher rates of completion of any component: r=−.52, p=.001).

DISCUSSION

This study leveraged an empirically-supported intervention protocol focusing on hope [33], a critical aspect of QOL [23] and spiritual well-being [41–43], to address disrupted goal pursuits and QOL among YACS. This intervention used the conceptualization of hope by Snyder [24] focused on goal-oriented thinking, which also is a cognitive skill highly relevant to coping with cancer [26], particularly among young adults [2, 44]. The main goal of this study was to examine feasibility and acceptability of the AWAKE intervention as well as to obtain estimates of efficacy of the AWAKE intervention versus an AC.

Findings indicated feasibility and acceptability of the program. Regarding feasibility, we successfully recruited over 50% of participants who were reached and met eligibility for the current study. We were also able to maintain high retention (exceeding 85%) and adherence (80% exceeding 75% retention). In terms of acceptability, a vast majority of participants recommended keeping the videos and coaching in the program and indicated that they would recommend the program (>88%); on average, they also indicated being highly satisfied with the program. However, these dimensions of feasibility and acceptability were indicated in both the AWAKE and the AC conditions. In part, this may be due to the nature and content of the AC condition – specifically, its focus on finances, which is highly relevant to young adults, as indicated by the data collected in this study. However, compared to the AC condition, participants in the AWAKE condition rated the content of the videos as more relevant on average, and the proportion of participants recommending keeping the videos trended toward being higher. Compared to AC participants, AWAKE participants were more likely to have talked with friends or family about the program and comment positively about it. Finally, the vast majority of AWAKE participants (80%) recommended keeping the behavior/mood tracking and homework in the program. These results indicate that the AWAKE intervention is acceptable, even when comparing it to a relatively relevant and useful AC condition.

Regarding efficacy outcomes, the primary outcome was hope, given that hope is a highly relevant intervention target related to better mental and physical health and functioning [24, 33], spiritual well-being [41–43], better adjustment in coping with cancer [26], better pain tolerance [27], lower rates of substance use [28, 30], and better overall QOL [31, 33]. We also examined secondary outcomes of QOL, depressive symptoms, and substance use. Despite a small sample size, this study indicated anticipated trends in primary and secondary efficacy outcomes, particularly in relation to FU assessments of changes in hope and its components. Engagement with the AWAKE intervention was also associated with better outcomes among those in the AWAKE program.

Particularly related to the potential impact of this intervention, this approach is highly scalable given its app-based nature and the brief interactions with coaches via telephone. The Rustoen et al. study [32] tested a hope-focused intervention consisting of eight weekly two-hour sessions provided to 195 participants and indicated only immediate effects on hope. The Cheavens et al. study [33] involved an eight-session group treatment that showed efficacy in increasing hope and in reducing symptoms of depression and anxiety [33]. Another brief one-session intervention study documented increases in hope and pain tolerance among young adults [27], Current findings, along with findings of these other studies, suggest that hope is a highly relevant and potentially impactful intervention target, particularly for YACS [2], and could be addressed through a scalable intervention such as AWAKE.

This research has implications for future research and practice. In terms of research, these findings indicate the promise of pursuing this type of intervention in an adequately powered efficacy trial with long-term outcomes being documented. Doing so is particularly important given the methodological challenges in previously published positive psychology intervention research focusing on cancer survivors (e.g., sample sizes, measurement of long-term effects) [22]. Moreover, the development of scalable interventions that can be delivered at no or minimal cost and in ways that fit within the lives of YACS is critical in promoting uptake and use of such programs in addressing the cancer-related sequelae faced by this population. Within this context, it is also important to note that hope, as operationalized in this study, is defined as being able to set appropriate goals across life domains, find pathways to reach those goals, and be able to maintain motivation to reach them, even in the face of barriers [24]. Thus, despite the cultural and spiritual influences that may impact “hope” as the term is used more broadly , the operationalization of hope by Snyder [24] and used in this study transcends the ways in which hope may be differentially valued or perceived across cultures or religious groups. In practice, determining ways that the spirit of positive psychology can be integrated into cancer care could help support survivors as they re-engage with goal pursuits across life domains (e.g., relational, occupational) and may have to renegotiate blocked or disrupted goals (e.g., reproduction).

Limitations

Study limitations include small sample size, limiting the extent to which we could detect differences between groups, particularly effects of the AWAKE intervention versus AC. As an additional note, we used a strong AC condition, as we aimed to have sufficient relevance of the content to engage participants in order to control for time spent in coaching and attending to intervention content. This may have impacted our ability to detect differences/effects. Moreover, we were underpowered to examine the associations between engagement in the AWAKE intervention in relation to outcomes in order to establish whether there was a dose-response relationship. Fortunately, this was in part due to high engagement in the AWAKE intervention. The sample was also limited in terms of generalizability and diversity (e.g., racial, ethnic), as participants were recruited from two NCI-designated cancer centers in the South; this has implications for future scalability. In addition, there was inherent difficulty ascertaining differences between those enrolled and retained versus those not enrolled or retained. However, we included a broad range of cancer types and those within two years of treatment completion, underscoring the diversity of the sample in terms of cancer-related factors, not to mention the sociodemographic diversity of the sample.

Conclusions

This study provides preliminary data suggesting that an app-based positive psychology intervention, specifically targeting hope, is feasible and acceptable, with outcomes indicating trends such that participants in the AWAKE versus AC conditions demonstrated larger improvements in psychological, functional, and substance use outcomes. Moreover, this study highlights the potential utility of positive psychology constructs, particularly hope, in supporting YACS cope with cancer-related sequelae and re-establish goals across life domains after the cancer experience, a particularly relevant challenge facing many cancer survivors, particularly in young adulthood [2].