Feasibility of Systems Support Mapping to guide patient-driven health self-management in colorectal cancer survivors

Abstract

Objective:

To evaluate feasibility of System Support Mapping (MAP), a systems thinking activity that involves creating a diagram of existing self-management activities (e.g., symptom management, health behaviors) to facilitate autonomous engagement in optimal self-management.

Design:

One-arm pilot study of MAP in colorectal cancer survivors (NCT03520283).

Main Outcome Measures:

Feasibility of recruitment and retention (primary outcome), acceptability, and outcome variability over time.

Results:

We enrolled 24 of 66 cancer survivors approached (36%) and 20 completed follow-up (83%). Key reasons for declining participation included: not interested (n=18), did not perceive a need (n=9), and emotional distress/overwhelmed (n=7). Most participants reported that MAP was acceptable (e.g., 80% liked MAP quite a bit/very much). Exploratory analyses revealed a −4.68 point reduction in fatigue from before to 2 weeks after MAP exceeding a minimally important difference (d=−0.68). There were also improvements in patient autonomy (d=0.63), self-efficacy (for managing symptoms: d=0.56, for managing chronic disease: d=0.44), psychological stress (d=−0.45), anxiety (d=−0.34), sleep disturbance (d=−0.29) and pain (d=−0.32). Qualitative feedback enhanced interpretation of results.

Conclusions:

MAP feasibility in colorectal cancer survivors was mixed, predominantly because many patients did not perceive a need for this approach. MAP was acceptable among participants and showed promise for improving health outcomes.

Introduction

Approximately 65% of the over 149,000 people diagnosed with colorectal cancer in the United States (US) each year will survive for at least 5 years post diagnosis, and there are more than 1.5 million colorectal cancer survivors (defined as people living in the US with a history of colorectal cancer; American Cancer Society, 2019, 2021). As cancer survivors are expected to live longer now than in the past, there is a shift in perspective from cancer as a terminal to a chronic illness (American Cancer Society, 2019). This shift increases the need for cancer survivors to become more actively engaged in self-management to optimize their own health outcomes, in partnership with health care providers (McCorkle et al., 2011; National Research Council, 2005). Self-management includes “tasks that individuals undertake to deal with the medical, role, and emotional management of their health condition(s)” (McCorkle et al., 2011). Thus, this definition of self-management considers health comprehensively as encompassing physical, social, and mental well-being. The Institute of Medicine and other professional associations recommend that patients receive comprehensive survivorship care planning that addresses follow-up care needs, including multiple complex self-management activities (e.g., psychosocial support, symptom management, healthy behaviors;National Research Council, 2005; van de Poll-Franse et al., 2017) when transitioning from the end of active treatment to extended survival. In addition, more recent recommendations note that patients prefer more detailed care planning information than health care providers (van de Poll-Franse et al., 2017).

Initial survivorship care planning efforts that primarily focused on information delivery demonstrated limited efficacy for improving patient-reported health outcomes (van de Poll-Franse et al., 2017). More recent studies based on the Chronic Care Model (CCM) suggest that survivorship care planning is more efficacious when it includes self-management support, which is a key component of a health system that delivers high-quality chronic illness care (Bodenheimer et al., 2002; Kvale et al., 2016; Reb et al., 2017). A particularly strong survivorship care planning study that resulted in improved self-reported health outcomes at a 3-month follow-up (i.e., improvements in social, emotional, and physical role functioning, self-reported health, bodily pain, a trend toward improving self-efficacy), included one intervention session with a mental health professional within one year of completing active cancer treatment (beyond the most stressful acute transition time period). This intervention focused on creating a patient-owned care plan through the process of eliciting the patient narrative, goal setting, action planning, and other motivational interviewing techniques (Kvale et al., 2016). Another similar pilot study of a single session with an advanced practice nurse that included goal setting showed preliminary support for improving patient-reported health outcomes two months after the intervention (e.g., depression, anxiety, self-efficacy, physical functioning, pain, general health, quality of life) (Reb et al., 2017). Thus, there is evidence that interventions to improve cancer survivors’ engagement in self-management improve health outcomes, and even one session may lead to such improvements.

Improving the quality of cancer care starts with actively engaged patients (Committee on Improving the Quality of Cancer Care, Institute of Medicine, 2013). Yet, how to actively engage patients is not an explicit focus of existing interventions. Self-Determination Theory provides more in depth guidance for the self-management support component of the CCM needed to result in the activated patient and productive interactions (Deci & Ryan, 2008; Sohl, Birdee, et al., 2016). Self-determination Theory is consistent with CCM in emphasizing support for patients’ partnership in healthcare decisions and further highlights the importance of facilitating patients’ autonomous, or self-motivated, engagement in self-management in order to produce more sustainable improvement in health outcomes (Ng et al., 2012). Patient-driven goals are more likely to be achieved than goals that are extrinsically motivated (e.g., by being told what to change) (Mann et al., 2013). According to Self-determination Theory, awareness of automatic behaviors is a necessary first step in facilitating patients’ autonomous motivation for engaging in self-management (Brown et al., 2007; Brown & Ryan, 2003). One of the prior survivorship care planning studies described elicited survivors’ narratives, suggesting some implied consideration of broader life context when selecting goals (Kvale et al., 2016). Therefore, more explicitly enhancing survivors’ awareness and selection of what self-management activities would best fit within their life context (e.g., consistent with their values and their consideration of tradeoffs) could strengthen the sustainability of results of prior studies.

System Support Mapping (MAP) is a systems thinking approach that facilitates awareness of how existing self-management behaviors fit within the broader context of a cancer survivor’s life. Systems thinking is an approach to studying interconnections and wholes, in this case unpacking complex relationships between behaviors (e.g., self-management), survivors’ objectives, and supports/constraints in their broader contextual life system (Frerichs et al., 2016). MAP is a structured activity to facilitate survivors’ creation of a visual diagram that will tangibly illustrate and create awareness for how an existing automatic self-management behavior relates to the broader context of tradeoffs with other behaviors and achieving most-valued outcomes. MAP thus enables patients to clearly identify any discrepancies and autonomously select intervention targets for self-management (e.g., increases autonomy). Highlighting discrepancies between current and desired states is used as a technique to generate patient-driven motivation for behavior change (Deci & Ryan, 2008; Hettema et al., 2005). Selecting self-management behaviors that are tied to broader life values may also increase the perceived importance of self-management goals and thus patient engagement. Thinking through tradeoffs with existing behaviors may lead to defining more realistic goals that fit within one’s life context and increase confidence in the ability to sustainably achieve them (e.g., increase self-efficacy). Increasing self-efficacy for managing symptoms is related to reductions in symptoms and improvement in other health outcomes (e.g., psychological stress, symptoms) (McCorkle et al., 2011).

MAP has primarily been applied to health and social service systems in the community setting (Calancie et al., 2020) with the exception of previous work that took an individual-focused approach to understanding self-management and health equity in women with migraine (Befus et al., 2018). It is novel to apply an individual-focused MAP to guide patient engagement in self-management during the transition from active cancer treatment to extended survivorship and investigate if participants find the MAP process therapeutic. This study provides pilot data on study feasibility (primary outcome), intervention acceptability (secondary outcome), and outcome variability (secondary outcome). We conducted a one-arm feasibility study of MAP among colorectal cancer survivors who were within two years of completing active treatment. We evaluated recruitment and retention (feasibility); intervention completion and acceptability (intervention acceptability); and variability and changes in proposed outcomes (i.e., measures of patient engagement, health outcomes) to guide future study planning. We determined a priori that >50% of eligible patients would need to enroll and that >70% of participants would need to complete the intervention and be retained in the study for a future study to be considered feasible.

Materials and Methods

Participants

We identified participants through collaboration with healthcare providers and posting the study on an institutional research website. Eligibility criteria required for participation were as follows: Adults ≥18 years of age; diagnosis of stage I-III colorectal cancer; within 2 years of completing active treatment for colorectal cancer; cognitively able to complete interventions as judged by the study team; and able to understand, read and write English. We extended eligibility criteria from 1 to 2 years of completing active treatment to increase feasibility of recruitment.

Procedures

This trial was approved by the local Institutional Review Board and registered with clinicaltrials.gov (NCT03520283). Research staff approached potential participants who completed informed consent either in person or by telephone regarding their interest in study participation from July 2018 through December 2019. The trial was stopped because an adequate number of participants were recruited to meet the primary study aim. Participants were told that the mapping activity was designed to help them plan health goals related to areas of their lives impacted by cancer/cancer treatment. The MAP intervention was scheduled to coordinate with an already scheduled clinic visit or at another time and clinic location, as preferred by the participant. Participants were asked to complete questionnaires at baseline before the MAP intervention and two weeks following the intervention either remotely via the Research Electronic Data Capture system (REDCap) or on paper forms. We also asked them to complete a semi-structured follow-up interview by telephone after all questionnaire data collection was complete. The interviewer was a research associate experienced in qualitative research methodology who was part of an institutional qualitative research shared resource and independent of other study personnel. Research associates working in this shared resource undergo initial mentored training and annual continuing education in qualitative research methodology. These trainings include learning skills such as how to ensure consistent application of interview questions (e.g., not to share personal perspectives). Participants were compensated up to $50 depending on the number of study assessments completed.

Measures

We evaluated the primary outcome, feasibility of the MAP intervention, by enrollment (enrolled/approached) and retention (completed assessments/enrolled) proportions.

Intervention acceptability was measured by participant ratings on 5 items (how much participants liked the intervention; how helpful the intervention was; whether participants would continue to use what they learned; whether participants thought the interventionist was competent; and whether participants thought the interventionist was sensitive). All items were scored on a scale of 0 (not at all) to 4 (very much). Relatedness, or feeling of personal connection with, the interventionist was also assessed by the 7-item HEAL Patient-Provider Connection (Greco et al., 2016) measure at follow-up, which uses a scale for each from 1 (not at all) to 5 (very much) and is converted to a standardized T-score.

Proximal outcomes assessed included autonomy, measured with the Index of Autonomous Functioning Authorship/Self-Congruence Subscale (e.g., “My decisions represent my most important values and feelings”), mean values ranged from 0-5 and higher values indicated more autonomy (Weinstein et al., 2012). Self-efficacy for managing cancer was assessed using the Self-efficacy to Manage Chronic Disease Scale (e.g., “how confident are you that you can keep the fatigue caused by your disease from interfering with the things you want to do?”) with ratings ranging from 0 (not at all confident) to 10 (totally confident) and was also assessed with the Patient-Reported Outcomes Measurement Information System (PROMIS®) self-efficacy for managing symptoms short form (e.g., “I can manage my symptoms during my daily activities”) rated on a standardized scale ranging from 0-100 with higher values indicating more self-efficacy (Gruber-Baldini et al., 2017; Moore et al., 2016; Ritter & Lorig, 2014).

Health outcomes assessed were psychological stress and symptoms. Psychological stress was assessed with the 4-item Perceived Stress Scale (Cohen et al., 1983); (e.g., “In the last month, how often have you felt that you were unable to control the important things in your life?”). Response options for the PSS range from 0 (never) to 4 (very often) and higher total scores (range 0-16) indicate more stress. The following symptoms were assessed with the PROMIS Profile 29 version 2.0 (Cella et al., 2010) such that higher values indicate more of each construct: physical function, anxiety, depression, fatigue, sleep disturbance, ability to participate in social roles, pain interference and pain intensity. We utilized the HealthMeasures Scoring Service to calculate standardized T-scores for each construct (approximate mean = 50, standard deviation =10; https://www.healthmeasures.net/).

Sociodemographic variables assessed were age, gender, race/ethnicity, marital status, education level, rural-urban residence, health literacy, and financial toxicity. Rural-urban residence was determined by classifying zip codes as described by the Federal Office of Rural Health Policy (Health Resources and Services Administration, 2017). This definition of a rural population accepts all non-Metro counties as rural and uses an additional method of determining rurality (Rural-Urban Commuting Area codes). Health literacy was defined by responses to the item, “How confident are you in filling out medical forms by yourself?” (Wallace et al., 2006). Financial toxicity was assessed by the COmprehensive Score for financial Toxicity (COST) (e.g., “I am able to meet my monthly expenses) with mean scores ranging from 0-4 and lower scores indicate worse financial toxicity (de Souza et al., 2017) and two other items that asked, “How difficult is it for you (and your family) to pay your monthly bills?” and “How many times have you received income from AFDC, TANF, Work First, WIC or food stamps as an adult?”

Clinical factors including cancer type, disease stage, recurrence status, type of treatment, body mass index, and comorbidities were abstracted from medical charts. For comorbidities, thirteen common comorbidities were listed with an option to also list an “other” comorbidity; participants selected all that applied. Semi-structured interviews were conducted as a qualitative assessment of acceptability. Example open-ended prompts included, “Tell me about your experience completing the mapping activity” and “What suggestions do you have for improving the mapping activity?”

System Support Mapping (MAP) Intervention

MAP was delivered individually with outpatients by a doctorally trained nurse with experience using the intervention (DB - conducted 1 intervention) or a trained interventionist who had a master’s degree in education, completed an accredited Health and Wellness Coach training, and experience conducting qualitative research in a medical setting (NP - conducted all other interventions). This interventionist and potential back-up interventionist (who did not conduct any interventions) were trained by leading mock interventions using a script. There were also follow-up meetings after reviewing recordings of the first three participants, and there was ongoing correspondence throughout the study between the trainer (DB) and primary interventionist (NP). In addition, the interventionists self-documented each session with a checklist to support consistency of delivery.

The interventionists guided participants in creating visual diagrams of their self-management activities within their broader life context. MAP aimed to help participants tangibly see complex self-management activities on paper to make them more actionable. Each participant was provided with a piece of poster-sized paper with five concentric circles on it and six pads of differently-colored sticky notes used to for each different ring. Participants began by considering, “What is most important to you in your life that you’re not getting enough of or doing enough of since your experience with cancer or its treatment?” to identify desired valued outcomes. After a few participants completed the intervention, we clarified our questions to more explicitly ask participants to focus on their current life experience rather than reflecting upon what was experienced during treatment. Participants were asked to systematically describe the following topics in each ring: (1) self-identity; (2) current life impacts of cancer and its treatment; (3) existing self-management activities; (4) facilitators and needs for successfully completing self-management; and (5) outcomes and tradeoffs of self-management (Figure 1). The interventionist facilitated discussing and writing responses for each topic on the designated-colored sticky note that was then placed in the respective different ring. We asked participants to reflect aloud, discuss any discrepancies between current and desired outcomes of self-management behaviors (e.g., going for a walk, taking medication), and brainstorm about what they would need to reconcile them. We then had participants identify one goal that was important and achievable in the next two weeks and set action steps (Williams & Carey, 2003). The MAP intervention sessions lasted a median of 100 minutes (range: 60-185 minutes).

Figure 1.

Key for guiding the System Support Mapping (MAP) intervention

Analyses

The target sample size for this study of 24 was based on the number of participants needed for determining a reliable estimate of feasibility measures (e.g., variance) to inform the design of an efficacy trial (Julious, 2005). We determined a priori that if the proportion recruited (from those eligible) was below 50% and intervention completion and retention proportions were below 70%, a larger study may not be feasible. Our criteria for feasibility targets were based on recruitment proportions from our prior study in patients with colorectal cancer (Sohl et al., 2016). The number of participants interviewed was determined by the number recruited to participate.

Enrolled participants were compared to those who declined using chi-square tests and t-tests. We used one-sample z-tests of proportions to compare recruitment and retention proportions to chosen feasibility criteria. Proportions were summarized using point estimates and 95% confidence intervals constructed with the exact binomial method. We used descriptive statistics (means, medians, proportions) to summarize participant characteristics and ratings of acceptability. In exploratory analyses, we compared participant characteristics by recruitment (those who agreed to participate versus those who declined participation) and retention status (those who remained in study versus those who did not remain in study) using Fisher’s exact tests and t-tests. In addition, we summarized outcomes of interest at both assessment time points. For PROMIS outcomes, we described variability at each time point (standard deviations) and compared T-scores to minimally important differences (MID)—the recommended MIDs for PROMIS measures of anxiety, depression, fatigue, sleep disturbance, and pain interference assessed by the PROMIS Profile 29 are 3.0 with a range of 2.0-4.0 (Kroenke et al., 2019). We also computed Cohen’s d effect sizes for the PROMIS measures and other outcomes. We used the following guideline for interpreting Cohen’s d: 0.2, small effect; 0.5, medium effect; and 0.8, large effect (J. Cohen, 1988).

An independent professional service transcribed the audio files verbatim into a word processing document. Transcripts were compared to the original recordings to ensure accuracy and de-identified. A codebook was developed from the concepts originally presented in the interview guide, and also incorporated new or emergent concepts presented by the participants during discussions. Two study team members independently coded all transcripts and resolved any incongruent interpretations of the codebook and coding discrepancies. We summarized intervention goals and conducted a qualitative content analysis of follow-up interviews (Bengtsson, 2016). We adopted a sequential QUAN → qual design with a complementarity function such that the qualitative results enhanced understanding of the quantitative results to refine the protocol for future study (Palinkas et al., 2011; Plano-Clark & Ivankova, 2016).

Results

Study Feasibility

Figure 2 displays the study flow including the number of participants screened, reasons for ineligibility, number of eligible patients approached, number and reasons for declining participation, and number enrolled. Regarding feasibility of recruitment, we enrolled 24 of the 66 patients approached (recruitment proportion = 0.36; 95% Confidence Interval [CI]: 0.25-0.49; observed proportion significantly lower than target of 0.50, p<0.05). Key reasons for declining participation included not interested (n=18), did not perceive a need (n=9), emotional distress/overwhelmed (n=7), and the study requests were too demanding (n=3). Participants enrolled did not significantly differ based on age, gender, race or ethnicity from those who declined participation (Please see Table 1 for a complete description of participant characteristics; p-values comparing enrolled to declined on age, gender, race, ethnicity all >0.05). Those enrolled (n=24) had a mean age of 60.91 years (SD=15.70), 63% were female, and 84% identified as White. They were a mean of 9 months post completion of cancer treatment (range: 1-21 months) and travelled a mean of 34 minutes for treatment (range: 10-150 minutes). In addition, 21% of participants were rural residents, 13% were not health literate, and 27% found it very or somewhat difficult to pay monthly bills.

Figure 2.

Study Flow Diagram

Table 1.

Baseline characteristics of study participants and those who declined participation

Characteristic	Enrolled N (%) (N=24)	Declined N (%) (N=42)
Gender
Men	9 (37%)	20 (48%)
Women	15 (63%)	22 (52%)
Race
White	20 (84%)	37 (88%)
Black or African American	2 (8%)	5 (12%)
Other	2 (8%)	0 (0%)
Ethnicity
Hispanic or Latino/a	0 (0%)	0 (0%)
Marital Status
Married or Living with Partner	13 (56%)
Separated/Divorced	2 (9%)
Widowed/Single	8 (35%)
Education Level
Less than High School/High School Graduate	3 (13%)
Technical School/Some College	9 (39%)
College Graduate/Post Graduate Degree	11 (48%)
Difficulty Paying Monthly Bills
Very/somewhat difficult	6 (27%)
Not very difficult	11 (50%)
Not at all difficult	5 (23%)
Confident in Filling out Medical Forms (Health Literacy)
Extremely/Quite a bit	20 (87%)
Somewhat/A little bit/Not at all	3 (13%)
Times Received Income from AFDC, TANF, Work First, WIC or food stamps
Never	19 (83%)
Four times or less	3 (13%)
More than four times	1 (4%)
Rural Residence – defined by FORHP (Yes)	5 (21%)
Use the Internet or Email (Yes)	22 (96%)
Cancer Type
Colon	17 (71%)
Rectum	7 (29%)
Summary Disease Stage at Diagnosis
I	5 (21%)
II	7 (29%)
III	12 (50%)
Underwent Surgery (Yes)	23 (96%)
Received Radiation (Yes)	9 (38%)
Received Chemotherapy (Yes)	17 (71%)
	Mean (SD)	Mean (SD)
Age (Years)	60.91 (15.70)	66.9 (11.29)
BMI (kg/m2)	29.25 (6.52)
Financial Toxicity (Range 0-4)	2.25 (0.82)
	Median (Range)
Number of Comorbidities (Range 0-14)	1 (0-5)
Months Since Last Cancer Treatment	9 (1-21)
Minutes Travelled to Clinic	34 (10-150)

Note. Rural residence was defined by the Federal Office of Rural Health Policy (FORHP). Financial toxicity was assessed by the COmprehensive Score for financial Toxicity (COST) and lower scores indicate worse financial toxicity.

Once enrolled, 20 participants completed the 2-week follow-up assessment (retention proportion = 0.83; CI: 0.63-0.95; the observed proportion is not significantly different from our target retention proportion of 0.70). We found no significant differences in any of the characteristics displayed in Table 1 between the four participants who dropped out and the 20 who remained in for the second assessment. Two participants experienced tearfulness during and probably related to the intervention that resolved, which were documented as mild (grade 1) adverse events.

Acceptability of the MAP Intervention

Once enrolled, 21 of 24 participants completed the MAP intervention. Participants assessed at follow-up (n=20) liked the intervention (80% quite a bit/very much; 20% a little bit/somewhat), found it helpful (75% quite a bit/very much; 25% a little bit/somewhat), would continue to use what they learned (90% quite a bit/very much; 10% a little bit/somewhat), and found the interventionist to be competent (100% quite a bit/very much) and sensitive (100% quite a bit/very much). No participants selected the option “not at all” for any of the acceptability indices. Participants also rated the interventionist as highly relatable (M=66.2, SD=7.5).

Qualitative feedback from interviews (n=19) related to the intervention process enhanced understanding of acceptability results. Participants spoke about their successes and challenges associated with the MAP intervention, their MAP activity experience, interactions with study staff, personal expectations of the program, overall opinion, and suggestions for improvement. Successful experiences resulting from the MAP intervention were related to having the ability to self-reflect and their comfort with open-ended questions. Thus, participants who had cognitive difficulties or who preferred concrete thinking reported challenges with MAP. One participant clearly described the MAP activity experience in terms of both ease of completing the intervention and engagement of thinking required:

“Again, it was pretty simple as far as the steps to complete it. It was a little difficult as far as like actually thinking about it and considering the questions and what she was asking. Essentially, things I just hadn’t thought about or concentrated on or realized. It was easy and difficult.”

(Participant 01)

Feedback on interactions with the study staff were consistently positive. Some participants said they did not completely understand what to expect based on the study description, however, most expressed that their expectations were met or exceeded. Overall, participants reported that the program was useful for clarifying their current experience. For example, one participant said that it was:

“Useful to learn about what you’re going through and to face your—or to—to concretely put into words what you’re going through, your emotions and thoughts, that it was useful in that way.”

(Participant 03)

Suggestions for improvements included considering a more comfortable environment than a clinical setting, providing guidelines on what experts recommend that they do, and follow-up sessions to support progress.

Intervention goals selected by participants (n=21) were related to physical activity (n=8), enhancing social connections (n=3), finding more quiet time (n=3), improving eating habits (n=2), doing cognitive exercises (n=2), other personal development (n=2), and preventive care (n=1) as described with examples in Table 2. Both participants who identified as Black or African American selected goals related to enhancing social connections. A majority of participants reported that they fully (n=12) or partially (n=3) met their goal behaviors and some did not meet their goals (n=4; n=2 did not complete the interview). A partially met goal indicates that participants changed their behavior toward their goal, but had not yet met the specified target (e.g., started exercising regularly although not to the selected target of daily). Participants reported that the experience of considering the holistic context of their lives through the MAP intervention helped them feel accomplished in what they were already doing and more optimistic about achieving their goals. In a participant’s words:

Table 2.

Participant two-week goal topics and status

Participant Numbers	Two-Week Goal Topics	Example Goal	Goal fully met	Goal partially met	Goal Not met
05, 07, 09, 11, 16, 17, 22, 23	Physical activity (e.g., walking, going to the gym, yoga)	To regain his physical strength and stamina after treatment to be able to accompany his wife to her family’s home in another state by walking every other day in his neighborhood ¼ mile and to work up to ½ mile over the next 2 weeks.	N=6	N=1	N=1
08, 12, 14	Enhancing social connections (e.g., reconnect with friends, attend a party)	He will check graduation schedules and his brother’s work schedule and find a date that they can go fishing in June.	N=1	N=1	N=1
01, 04, 10	Finding more quiet time (e.g., to pray, meditate).	To reduce anxiety by taking 15 minutes each morning to go out on porch swing and pray before family wakes up.	N=2		N=1
02*, 13	Improving eating habits	To have better eating habits by preparing small, healthy meals to take to work to be able to eat more frequently.	N=1
03, 19	Other personal development (e.g., seek financial guidance, ask for help to complete projects)	To work more on self-preservation, specifically financial issues, she will contact her attorneys on Wednesday to learn about financial obligations.	N=1	N=1
20, 24*	Cognitive exercises	To improve his recall for remembering the steps to use computer programs he will go on the computer more frequently=1 x day, 4 days per week M, T, W, TH.			N=1
15	Preventive care	She will email through the online portal her primary care physician’s office to ask for a referral to a gynecologist.	N=1

^*The goal status for participants #02 and #24 was not determined because they did not complete the follow-up interview.

Note. Participant #s 06, 18 and 21 did not complete the MAP intervention, therefore those participant numbers are omitted from this table. A partially met goal indicates that participants changed their behavior toward their goal, yet had not yet met the specified target (e.g., started exercising regularly although not to the selected target of daily).

“It was interesting. It was a lot more detailed than I expected. It was very good to see the entire map all at once. It gives you a better idea of what areas you need to work on and what other areas you’ve done better than you expected. I felt optimistic afterwards that I would be able to reach my goal.”

(Participant 16)

Participants found their goals personally important and some reported changing their behaviors because they felt accountable after telling someone they would do it. Some participants highlighted the involvement of others when engaging in the goal process or stated that they wanted more support from others. Of the four participants who did not meet their goals, two reported cognitive difficulties. One participant said he did not reach out to his friends as planned because he didn’t have any energy and didn’t feel well. He specifically said:

“I’m exhausted. I’m just tired, and I sit, and I might read. I might go to sleep. I don’t know. I just don’t feel so great. I have to go to the bathroom a lot. It’s bad.”

(Participant 08)

This same participant also reported during the interview that he had felt depressed and had recently seen a physician who prescribed antidepressants (depression was a baseline condition). Another participant said that she did not go to the gym as planned because she was not feeling well and had trouble breathing outside in the recent extreme heat (Participant 17). She said the goal set was important to her and she was confident that she would still achieve it. Most other participants also reported that they were motivated and very confident in their ability to continue to be successful in achieving their goal. The theme of a new awareness of issues that may otherwise go unnoticed also emerged as a result of MAP. This awareness led to confidence about their need for changes in self-management. For example:

“I think that the program is helpful. It helps to bring to the forefront what issues might be causing some discomfort or procrastination in your life. It was enlightening. Can lead to self-improvement. Maybe lead to a little more peace

(Participant 07).”

Outcome Summary

A quantitative summary of outcomes and changes in outcomes over assessment time points is presented in Table 3. Exploratory analyses revealed small to medium-sized changes in proximal outcomes of patient autonomy (d = 0.63) and self-efficacy for managing symptoms (d = 0.56), and self-efficacy for managing chronic disease (d = 0.44) from before to two-weeks after the intervention. There was a reduction in fatigue from before to after the intervention and this change of −4.68 exceeded the MID (d = −0.68) (Kroenke et al., 2019). The effect sizes for improvements in psychological stress (d = −0.45), anxiety (d = −0.34), sleep disturbance (d = −0.29) and pain intensity (d = −0.32) were also small to medium sized. The changes in anxiety and sleep disturbance did not exceeded the MID. Participants also qualitatively reported perceived benefits of the MAP intervention such as finding it helpful for managing stress, energy levels, sleep patterns, and pain in the context of their daily lives. For example, one participant said:

Table 3.

Description of Study Outcomes

	Baseline Mean (SD) N=24	Follow-Up Mean (SD) N=20	Difference Mean (SD) N=20	Cohen's d for Change
Proximal Outcomes (Possible Range)
Index of Autonomous Functioning – Autonomy (Range: 0-5)	4.4 (0.6)	4.7 (0.4)	0.4 (0.6)	0.6
Self-efficacy for Managing Chronic Disease Scale (Range: 0-10)	7.7 (2.1)	8.6 (1.5)	0.6 (1.2)	0.4
PROMIS Self-efficacy for Managing Symptoms (Range: 0-100)	49.8 (8.4), n=23	55.2 (8.6)	4.3 (7.4)	0.6
Health Outcomes (Possible Range)
PSS Psychological Stress (Range 0-16)	8.1 (1.9)	6.7 (2.9)	−1.0 (2.2)	−0.5
PROMIS Physical Function (Range 0-100)	46.0 (8.5)	47.3 (10.4)	0.2 (6.5)	0.0
PROMIS Anxiety (Range 0-100)	50.3 (8.8), n=23	46.4 (8.5)	−2.3 (6.7)	−0.3
PROMIS Depression (Range 0-100)	47.1 (7.7)	45.3 (6.1)	−0.1 (5.7)	−0.0
PROMIS Fatigue (Range 0-100)	51.9 (10.2)	46.2 (9.4)	−4.7 (6.9)	−0.7
PROMIS Sleep Disturbance (Range 0-100)	52.1 (7.8)	49.3 (9.3)	−1.8 (6.3)	−0.3
PROMIS Social Roles (Range 0-100)	50.1 (8.3)	52.0 (10.6), n=18	1.6 (8.9), n=18	0.2
PROMIS Pain Interference (Range 0-100)	52.4 (9.5)	51.4 (10.0)	−0.9 (7.2)	−0.1
PROMIS Pain Intensity (Range 0-10)	3.0 (2.8)	2.5 (2.6)	−0.6 (1.7)	−0.3

Note. Higher values indicate more of each construct for all assessments. PSS = Perceived Stress Scale. Patient-Reported Outcomes Measurement Information System (PROMIS) values reported are standardized scores such that mean values generally center around 50 with standard deviations of 10. All health outcomes are from the PROMIS Profile 29 version 2.0.

“It was wonderful. I really enjoyed it. Like I said, as helpful in my day-to-day pain, with my pain management, and things that I could do to help me throughout the day, or whenever I’m feeling bad, or when I’m in a lot of pain and stuff. It has helped me out really, really good. I really enjoyed it. It was really, really, really helpful.”

(Participant 12)

Discussion

Study Feasibility

Overall, results for our primary outcome of feasibility were mixed. The recruitment rate of 36% was significantly less than expected (50%), whereas the retention rate of 83% was in line with what was proposed (70%). A majority of survivors declined to enroll because they were not interested or did not perceive a need, which may be because stages I-III colorectal cancer is a highly treatable disease (National Cancer Institute, 2021) and many patients experience a similar global quality of life to the general population after completion of treatment (Arndt et al., 2004). The high proportion of declines may also have been partially due to initial challenges in describing the MAP intervention. Over time, we found it helpful to refer to the intervention as “mind mapping” and to provide anecdotes describing feedback from other participants. The present study required staying at the clinic or finding another time to meet in person for at least an hour to complete the study intervention, which may have discouraged some patients from enrolling since some declined participation due to feeling distressed/overwhelmed or finding the study requests too demanding. Our criterion for a target recruitment proportion was based on recruitment proportions from our other studies in patients with colorectal cancer who were exclusively enrolled and the intervention was implemented during already-scheduled patient visits, resulting in greater acceptance. This important distinction in study design may have affected feasibility.

After realizing our recruitment proportion was lower than expected, we adapted our recruitment strategy to approach participants about their willingness to participate at their clinic visit instead of initiating contact by phone and extended our eligibility to include survivors within two years of completing treatment. Other studies of self-management programs in the context of cancer survivorship care planning have considered recruitment rates closer to 40% as feasible, which is similar to the rate achieved in this study (Kvale et al., 2016; Song et al., 2020). Although, initial survivorship care planning studies that focused on providing cancer treatment summaries aimed to reach participants immediately following treatment, more recent studies with an expanded focus on self-management have extended that time period to within a year when survivors transition stress may be reduced (Kvale et al., 2016). In addition, care plans are intended to be updated over time to accommodate dynamic health needs, and health self-management is relevant to ongoing survivorship follow-up care, thus it may be beneficial to provide care planning components incrementally at different time points (Earle, 2006; McCorkle et al., 2011; National Comprehensive Cancer Network (NCCN), 2019; van de Poll-Franse et al., 2017). Thus, extending our eligibility window to within two years post treatment is consistent with viewing survivorship care planning as an ongoing process and allowed participants to draw upon what they have learned from personal experience with health self-management during the MAP intervention. Once participants were enrolled, it was feasible for them to complete follow-up assessments.

Acceptability of the MAP Intervention

Participants generally completed the MAP intervention and found it to be acceptable. A majority of acceptability ratings were positive and participants found the interventionist highly relatable. The small number of participants who identified as Black or African American or Other had consistently positive feedback, supporting that MAP may be appropriate for subgroups of cancer survivors more likely to experience health disparities related to self-management (e.g., lower screening surveillance; (Rolnick et al., 2005)). The intention of MAP, a systems thinking activity, is to provide a way to work with the complexities of actual self-management behaviors and challenges faced versus what patients may be instructed to do in a healthcare system that does not always account for the social and structural determinants of equity and health (Befus et al., 2018). Therefore, future research could focus more on the potential for MAP to also identify how individual-level health self-management interacts with contextual factors associated with health disparities (e.g., lack of transportation, food insecurity; (Hastert et al., 2021)).

Since most challenges in achieving goals were due to cognitive challenges or depressed mood, it is important for future research on MAP to use a more consistent method to screen participants and exclude those who are not cognitively able to participate (e.g., have dementia) or who need to first manage clinical levels of depression. Participants found the interventionist to be highly competent, sensitive, and supportive. Some participant suggestions for improving the one-session MAP intervention were to consider delivery in a more comfortable environment, provide more information on guidelines for recommended health behaviors, and provide additional follow-up to support progress. Therefore, future studies could incorporate these suggestions, while maintaining the feasibility of retention to a brief intervention by potentially implementing MAP or follow-up sessions remotely (e.g., telephone coaching (Damschroder et al., 2017; Hawkes et al., 2013)) and allow participation from the comfort of home. Remote implementation may also increase feasibility of recruitment and potentially facilitate enrollment of historically underserved patients who commonly experience transportation challenges.

A majority of participants were successful in achieving the goals they set. It is also notable that autonomy increased after the MAP intervention, which asked participants to select their own goals rather than be prescribed a goal by the study interventionist. Goal topics that emerged (i.e., physical activity, social connections, quiet time, eating habits, cognitive exercises, other personal development, preventive care) were similar to those offered as options in other interventions that have provided a selection of topics for goals (eat wisely, be physically active, manage stress (Damschroder et al., 2017)). Additional topics were related to social connections, preventive care, cognitive exercises and other personal development. The topic of social connections may be of particular importance to survivors who identify as Black or African American. Another study further suggests that social support plays a stronger role in buffering the impact of cancer on mental health in African American than White survivors (Matthews et al., 2012). Offering options for goal topics in a future study could be informed by these goals that emerged.

Outcome Summary

Our main interest in conducting analyses of secondary quantitative outcomes was to estimate variances for use in planning future studies. Exploratory analyses revealed clinically significant reductions in fatigue from before to 2 weeks after the intervention. This is a particularly promising result because fatigue is one of the most common side-effects experienced among cancer survivors (Arndt et al., 2004; Berger et al., 2015). It is possible that fatigue may naturally decline in some survivors over time as it did in another study (Hawkes et al., 2013), thus it will be important to further evaluate this intervention in a randomized trial with a control group. Results also suggested possible improvements in psychological stress, anxiety, sleep disturbance and pain intensity. The MAP intervention’s focus on engaging cancer survivors in self-management was similar to components of certain survivorship care planning interventions that also improved patient-reported health outcomes (Kvale et al., 2016; Reb et al., 2017). In addition, the current intervention holistically considered individual-level contextual factors such as values and tradeoffs with existing behaviors, which could more sustainably integrate self-management into survivors’ existing life context and ultimately facilitate translation of the intervention into practice (Brownson et al., 2012; Riley, 2017).

The consideration of life context may have contributed to changes in proximal outcomes relevant in Self-determination Theory (i.e., autonomy, self-efficacy), which increases the robustness of results (Ng et al., 2012). Quantitative and qualitative findings support that the MAP intervention facilitated participants’ autonomy and that participants felt competent in managing symptoms, which ultimately may have led to improvements in health outcomes that participants indicated would be sustainable. Raised awareness of issues that may otherwise go unnoticed was highlighted in qualitative feedback as a precursor to understanding what self-management behaviors were important. The type of awareness referred to in the context of Self-determination Theory as a precursor autonomously motivated self-management is mindfulness (Brown et al., 2007; Brown & Ryan, 2003). Mindfulness practices are another method for raising awareness that is increasingly recognized as an important component of self-management interventions (Hawkes et al., 2013; Sohl, Birdee, et al., 2016).

Study Limitations

Limitations to our findings here include that this was a one-arm study with a small sample size, a short follow-up period, and self-reported psychological measures. However, the self-reported measures used were psychometrically strong, and some allowed for interpretation of minimally important differences. We also had a low participation proportion, so it is possible that those who agreed to participate were the ones that were most likely to benefit from this type of exercise; if so, the intervention may not be as helpful to all survivors. Further, our study had limited representation of racial and ethnic minorities, and thus, results may not be widely generalizable. Yet, the two participants who identified as Black or African American or Other in this study reported favorable feedback and prior research does support that vulnerable women (e.g., occupying lower socioeconomic status) with migraines were able to successfully complete MAP (Befus et al., 2018).

Conclusions

In summary, we found that recruitment to the MAP intervention to facilitate self-management in colorectal cancer survivors was more challenging than expected and did not meet pre-established feasibility targets. However, it was feasible to retain participants for two week post-intervention assessments. The intervention was highly acceptable to most of the participants and showed promise for improving patient health outcomes. MAP facilitates a holistic awareness of life context that theoretically could have contributed to autonomous motivation and confidence for engaging in self-management. These proximal outcomes may have led to the improvements in health outcomes that participants indicated would be sustainable. It is novel to apply MAP individually to guide patient engagement in self-management and results of this pilot study support the need for further investigation, including exploring other recruitment strategies, identifying subgroups of patients more likely to engage with and benefit from MAP, adding longer-term follow-up assessment, and including comparison of a MAP intervention group to a control group.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.