A Tailored Electronic Survivorship Care Plan for Prostate Cancer Survivors: A Multicenter Randomized Controlled Trial

Abstract

Purpose:

Although the provision of a survivorship care plan (SCP) has been recommended after prostate cancer (PCa) treatment, there have been no randomized controlled trials to examine their impact. The objective of this study was to evaluate the effect of a tailored PCa-SCP intervention provided to early-stage PCa survivors.

Materials and Methods:

A prospective, parallel 1:1 randomized controlled trial was conducted at 3 sites across Canada. Early-stage PCa survivors were randomized (n = 189; 64% response rate) to receive PCa-SCP intervention or usual care. Assessments were performed at baseline, 6 months, and 12 months. The primary outcome was change in patient activation (Patient Activation Measure-13); secondary outcomes included satisfaction with information, self-management support, prostate-specific quality of life, cancer worry, health care utilization, and health behaviors.

Results:

For the primary patient activation analyses, no significant group-by-time differences were detected. There were significant between-group differences in (1) satisfaction with information from baseline to 6 months (mean [95% CI]: 7.13 [5.53-8.71]; effect size [ES], 0.34; P < .001) and baseline to 12 months (4.91 [3.06, 7.06]; ES, 0.19; P < .001); (2) social integration and support from baseline to 6 months (mean [95% CI]: 0.16 [0.06-0.26]; ES, 0.25; P = .002); and (3) skill and technique acquisition from baseline to 6 months (mean [95% CI]: 0.12 [−0.02 to 0.24]; ES, 0.33; P = .05). No other group-by-time differences were detected.

Conclusions:

The PCa-SCP intervention did not have a significant effect on patient activation but did increase satisfaction with information and some aspects of self-management support. Although the benefits of the provision of an SCP alone remain unclear, the PCa-SCP intervention is likely a valuable tool that can be built upon as part of a comprehensive approach to enhance the quality of care for PCa survivors.

Trial Registration:

ClinicalTrials.gov Identifier: NCT03017456

Prostate cancer (PCa) survivors are a rapidly growing population with diverse and unique needs.^1,2 Providing tailored patient-centered information and support that is responsive to needs during the transition to post-treatment care is critical to the long-term health of PCa survivors and may help address barriers to follow-up care.^3,4

JU Insight

• Study Need and Importance

• Prostate cancer (PCa) survivors face unique challenges during the transition from treatment to survivorship. Survivorship care plans (SCPs) aim to enhance patient-centered follow-up, yet their effectiveness remains unclear, especially for PCa survivors. This study assessed the impact of an electronic, tailored SCP delivered by trained oncology nurses to early-stage PCa survivors after completion of primary treatment.

• What We Found

• The SCP intervention improved patients’ satisfaction with information received and aspects of self-management, including social integration and skill acquisition. However, it did not significantly enhance patient activation, the primary outcome. Secondary measures, such as quality of life, cancer worry, health care utilization, and health behaviors, showed no notable differences between the SCP and usual care groups. The Table outlines key study findings.

• Limitations

• The study had limitations, including significant ceiling effects in patient activation scores, limiting sensitivity to detect changes. The sample was predominantly well educated, White, and English speaking, reducing generalizability. Additionally, SCP usage post intervention was not tracked, which may have affected the results. The unblinded design could have introduced response bias, and the study’s urban settings with robust supportive care services might have minimized the intervention’s incremental benefits.

• Interpretation for Patient Care

• Although SCPs alone may not significantly alter PCa survivors’ patient activation or broad health outcomes, they enhance satisfaction with information and short-term self-management. While it is unlikely that SCP alone can address all of survivors’ educational and informational needs, SCPs should be seen as an important tool within a larger multifaceted comprehensive approach to follow-up care. Future studies should involve underrepresented populations and refine measures to better evaluate SCP impacts.

  Table.

  Key Findings

  Outcome	Time point	UC (n = 93)		PCa-SCP (n = 96)		Effect size (95% CI)	Change in between-group difference over timea (95% CI)	P valueb
  		n (available)	Mean (SE)	N (available)	Mean (SE)
  PAM-13	Baseline	85	66.0 (1.70)	92	67.5 (1.67)	—	—	—
  	6 mo	81	65.5 (1.72)	82	69.4 (1.71)	0.31 (0.001-0.62)	2.40 (−0.93 to 5.72)	.16
  	12 mo	83	65.9 (1.71)	70	68.5 (1.75)	0.19 (−0.13 to 0.51)	1.06 (−2.36 to 4.48)	.5
  EORTC QLQ-INFO25	Baseline	93	58.4 (0.09)	95	56.2 (0.76)	—	—	—
  	6 mo	89	57.3 (1.39)	86	62.2 (1.83)	0.34 (0.04-0.64)	7.13 (5.53-8.71)	< .001
  	12 mo	86	59.3 (1.77)	77	61.9 (1.55)	0.19 (−0.12 to 0.50)	4.91 (3.06-6.76)	< .001
  heiQ
  Social integration and support	Baseline	91	3.21 (0.06)	96	3.16 (0.05)	—	—	—
  	6 mo	88	3.13 (0.02)	81	3.25 (0.05)	0.25 (−0.05 to 0.55)	0.16 (0.06 to −0.26)	.002
  	12 mo	89	3.12 (0.02)	77	3.19 (0.04)	0.15 (−0.16 to 0.46)	0.12 (−0.01 to 0.24)	.069
  Skill and technique acquisition	Baseline	93	2.94 (0.06)	96	2.97 (0.06)	—	—	—
  	6 mo	89	2.95 (0.06)	86	3.10 (0.06)	0.33 (0.032-0.63)	0.12 (−0.02 to 0.24)	.050
  	12 mo	89	2.99 (0.06)	77	3.07 (0.06)	0.18 (−0.13 to 0.49)	0.05 (−0.07 to 0.17)	.4

  Abbreviations: EORTC QLQ-INFO25, European Organization for Research and Treatment of Cancer Quality of Life Information Module; heiQ, Health Education Impact Questionnaire; PAM, Patient Activation Measure; PCa-SCP, Prostate Cancer Survivorship Care Plan; UC, usual care.

  Bold text indicates statistically significant value.

  ^a

  Estimated change in the outcomes between the UC and PCa-SCP groups, relative to the baseline difference between the groups.

  ^b

  Interaction P value: The interaction P value tests whether the change in mean scores over time differs significantly between the 2 groups.

The provision of a survivorship care plan (SCP), which includes a detailed treatment summary and tailored follow-up care plan, is one strategy first proposed by the Institute of Medicine in 2006 to address the needs of cancer survivors.⁵ Since that time, there has been a growing body of research on SCPs. Findings from nonrandomized controlled trials (non-RCTs) suggest that patients and clinicians are satisfied and support the added value of SCPs and believe that SCPs may improve survivorship knowledge, satisfaction with care, and communication with providers.⁶ However, results from RCTs provide little evidence to support that SCPs are superior to usual care on patient-reported outcomes or health care delivery,⁷ and 1 study on endometrial cancer survivors reported higher patient concerns, emotional impact, and symptoms in those receiving an SCP compared with the usual care arm.⁸ Across these studies, there is wide heterogeneity in terms of the SCP format and delivery models,⁷ and limitations have been highlighted including timing of SCP delivery and failure to identify the most appropriate outcomes.^7,9 Furthermore, the focus to date has been primarily on the SCP document in isolation, with few details regarding the care models and therapeutic approaches used in its delivery. Barriers to implementation have also been noted, and there is a need for creative solutions and approaches to improve efficiency and ensure sustainability of SCPs in overburdened oncology programs including electronic solutions for care plan development.¹⁰

Although organizations such as the American Cancer Society have recommended the provision of SCPs to PCa survivors,³ there have been no RCTs to examine their impact. To address this, we conducted a phase 3 RCT to evaluate the effect of a personalized PCa-SCP intervention in early-stage PCa survivors after completion of primary treatment.

PATIENTS AND METHODS

We conducted a 2-arm parallel 1:1 RCT at 3 sites across Canada: (1) Princess Margaret Cancer Centre, (2) Vancouver Prostate Centre, and (3) McGill University Health Centre. The trial adhered to relevant reporting guidelines,¹¹ was registered (ClinicalTrials.gov identifier: NCT03017456), and was approved by the research ethics boards of the University Health Network, University of British Columbia, and McGill University Health Centre. All patients provided written informed consent.

Participants were eligible if they had histologically confirmed localized PCa (T1-T3N0M0), were older than 18 years, and had completed treatment within 1 to 6 months. Patients who did not receive treatment and were followed by active surveillance and those unable to speak and read in English or French were excluded from the study.

Consenting participants completed the baseline questionnaire and were randomly assigned (1:1) to either the SCP intervention combined with usual follow-up care (PCa-SCP) or usual follow-up care (UC) using a computer-generated randomization process (random blocks known only to the statistician). Participants randomized to the PCa-SCP arm were scheduled for their intervention after randomization, and this was delivered at the first follow-up appointment with their oncologist/urologist and was typically 2 to 4 weeks from baseline assessment. This study was open label, and neither the study staff nor the participants were masked to the intervention allocation. After allocation, participants in the PCa-SCP arm were scheduled for the PCa-SCP appointment.

Experimental Arms

Intervention

The PCa-SCP intervention comprised of 2 components: a 30-minute oncology nurse-led face-to-face transition appointment and the provision of a tailored PCa-specific SCP. The content of the PCa-SCP was established using a modified Delphi process that included a multidisciplinary expert panel (n = 17). The final SCP contained the following sections: (1) Patient Information; (2) Disease Background; (3) Treatment Summary; (4) Care Plan (including subsections Cancer Surveillance Plan, Persistent Treatment Effects Monitoring and Management Plan, and Wellness Plan); and (5) Contact Information. The platform for the PCa-SCP was developed by the Techna Institute at University Health Network and interfaced with existing clinical databases to autopopulate demographic, diagnostic, and treatment data. The SCP also contained prepopulated checkboxes and drop-down lists used by the nurse to tailor the SCP to participant needs. Based on established follow-up guidelines, system logic/algorithms were built to automatically prepopulate the follow-up recommendations section of the SCP based on disease and treatment details that were pulled from the clinical databases. These could be changed if the clinician recommended a different follow-up protocol. In addition, algorithms were created that prepopulated potential side effects related to each treatment. If a side effect was reported, resources were then suggested and linked to the SCP with live links. In some cases, referrals were made and noted in the SCP. The resources related to each side effect were built into the system logic and tailored to local resources so the nurses could easily select them. Those side effects not reported would not appear in the final document. Before launching the trial, the final SCP was piloted with 5 to 10 patients per site. See Supplementary File 1 (https://www.jurology.com) for screenshots of the SCP.

The nurse-led transition appointment drew upon a well-established person-centered nursing framework¹² and used a customized approach to tailor the intervention to each individual’s unique needs and circumstances. The final tailored PCa-SCP was then emailed to the participant as a pdf. The document included live links to the prescribed resources. A synoptic copy was also sent to their oncologist/urologist and primary care physician (if applicable).

Study nurses participated in a formal 2-day training program that included (1) PCa treatment and survivorship overview, (2) an introduction to the study and SCP recommendations, (3) principles of adult education and expanding patient education skills, (4) an introduction to and hands-on training with the PCa-SCP system, and (5) motivational communication training, which is an evidence-based counseling approach that is effective in increasing health-related behaviors.¹³ The nurses then had the opportunity to practice delivering the SCP to standardized patients using the system and received feedback from the team. After this training, the study nurses had access to the study team and Motivational Interviewing Network of Trainers–certified trainer for additional support as needed and all felt confident with the system and how to deliver the intervention before beginning.

Usual Care

Participants in both the intervention and control arms received follow-up care according to usual practice of the participating institutions with their urologist and/or radiation oncologist. Appointments typically occur every 3 to 6 months (depending on treatment received), and the primary focus of these appointments is on screening for disease recurrence. Patients in the UC arm had access to the participating institution’s patient education materials and supportive care programs on demand or by referral, which varied by site. SCPs and dedicated transition appointments were not provided as part of usual practice at any of the participating sites.

Outcomes

The outcome measures were chosen based on recent consensus and recommendations on the evaluation of SCPs and were available in English and French. Stakeholders have highlighted how the cancer care system limits capacity for self-management and sense of control and suggest that SCPs have the potential to empower survivors to engage with their providers and manage their own health, thus shifting control toward survivors. Other important outcomes identified include meeting survivors’ educational and information needs and mitigating fear and anxiety.^6,9,14 Outcomes measures were completed at consent (baseline T0), 6 months (T1), and 12 months (T2) after baseline.

The primary outcome was patient activation assessed using the Patient Activation Measure (PAM-13)¹⁵ measured at 12 months after intervention. The PAM-13 assesses knowledge, skills, beliefs, and confidence for managing health and health care. The overall score is transformed to a 0 to 100 scale (with higher scores indicating greater activation). The total score can be used to assign to 4 levels of patient activation: level 1 (PAM score <47): disengaged and passive recipients of care; level 2 (PAM score 47.1-55.1): becoming aware but lacking confidence to self-manage; level 3 (PAM score 55.2-67): taking action and gaining control; and level 4 (PAM score >67.1): maintaining behaviors and pushing further.¹⁵ As per developer’s instructions, observations with a PAM score of 100 (38 observations across 3 time points) were excluded from the analysis (Note: sensitivity analyses revealed no significant change in outcome).

The secondary outcomes included:

1. Satisfaction with information: The 25-item European Organization for Research and Treatment of Cancer Quality of Life Information Module (EORTC QLQ-INFO25) is a reliable measure of patients’ perceptions of received information and can be used to evaluate information-based interventions and programs.¹⁶ A global score is calculated with higher scores indicating higher satisfaction with information received (max value 100).

2. Self-management support: Five subscales of the Health Education Impact Questionnaire (heiQ),¹⁷ which have been identified as key dimensions of empowerment in the cancer care context, were used to measure self-management support: Social Integration and Support, Health Service Navigation, Constructive Attitudes and Approaches, Skill and Technique Acquisition, and Emotional Distress.¹⁸ The subscales are scored by summing item responses (1 indicates strongly disagree; 2, disagree; 3, agree; 4, strongly agree) and then dividing by the number of scale items. Higher scores indicate higher levels of empowerment; scores can vary from 1.00 to 4.00.

3. Quality of life: The Expanded Prostate Cancer Index Composite (EPIC-26) short form was used to assess PC-specific quality of life.¹⁹ The measure contains 26 items and 5 domains: Urinary Incontinence, Urinary Irritative/Obstructive, Bowel, Sexual, and Hormonal. Response options for each Expanded Prostate Cancer Index Composite item form a Likert scale, and multi-item scale scores are transformed linearly to a 0 to 100 scale, with higher scores representing better health-related quality of life.

4. Cancer worry: The 6-item Assessment of Survivors Concerns²⁰ measure was included to assess cancer worry (3 items) and general health worry (3 items). All items are rated on a 4-point Likert scale, with 1 = not at all, 2 = a little bit, 3 = somewhat, and 4 = very much, and scores in each subscale are totaled with higher scores indicating greater worry (range 3-12).

5. Supportive care service utilization (past 4 weeks): The Health and Social Services Utilization Inventory modified for use in cancer populations²¹ was used to measure service use across 4 categories: health professionals, institutional/hospital programs, nursing and homemaker support programs, other community and social support program/resources.

6. Health behaviors and health screening: Participants reported on current diet, physical activity, alcohol, cigarette use, as well as visits to the family doctor (past 3 months) and general health and cancer screening tests.

Sample Size and Statistical Analysis

This study was designed to have 85% power to detect a difference of 8 points between groups on the primary outcome (PAM-13 total score), equivalent to a half SD difference, which represents a moderate effect and is a universally applicable distribution method to identify minimally important difference when anchor-based calculations are unavailable,^15,22 with a 1-sided type I error rate of 0.025. Given these specifications and assuming attrition of up to 20%, the required number of patients was 180 (or 90 patients per arm).

Baseline demographic and clinical variables were tabulated for the study groups using median (IQR) for continuous variables and N (%) for categorical variables. Primary and secondary outcomes measured at baseline, 6 months, and 12 months were treated as correlated outcomes, and mixed-model regression analysis was conducted on available data.²³ An unstructured variance-covariance matrix was used to model within-patient correlations. Visit (baseline, 6-month, and 12-month), treatment (UC, PCa-SCP), and visit-by-treatment interactions were included as fixed effects and center was treated as the random effect in the mixed-effects model analysis. Center was treated as the random effect because the primary goal was to estimate the treatment effect and to minimize the number of estimated parameters. The difference in outcomes between the UC and PCa-SCP groups at each time point was calculated as the difference in estimated least square means (between-group differences). The treatment effect at 6 and 12 months was estimated as the change in least square means (Δ_LSmeans) from baseline, with corresponding 95% CIs and P values, to assess whether the difference between the study groups varied significantly over time (ie, the interaction effect).

The ES was calculated as the mean difference in the outcome score between the groups, divided by the pooled SD. Student t test, χ² test, or Fisher exact test is used as appropriate to report significance between groups at 6-month and 12-month visits on supportive care service utilization, health behaviors, and health screening. Statistical significance is considered as P ≤ .05. Statistical analysis was performed using SAS version 9.4 or R version 4.2.2.

RESULTS

A total of 298 eligible men were approached, and 188 (63%) consented, completed the baseline assessment (T0), and were randomized. Of those randomly assigned, 176 of 189 (93%) completed the 6-month (T1) and 166 of 189 (88%) completed the 12-month (T2) follow-up assessments (Figure).

Figure.

CONSORT flow diagram. MUHC indicates McGill University Health Centre; PCa-SCP, Prostate Cancer Survivorship Care Plan; PM, Princess Margaret Cancer Centre; VPC, Vancouver Prostate Centre.

Table 1 provides demographic and clinical characteristics of the 2 study groups.

Table 1.

Baseline Demographics (Usual Care vs Prostate Cancer Survivorship Care Plan)

Covariate	UC (n = 92)a No. (%) or median (IQR)	PCa-SCP (n = 96) No. (%) or median (IQR)
Age, y	64.5 (60-70)	65 (60-70)
Current marital status
Married/common law	72 (78)	81 (84)
Divorced/single/widowed/other	20 (22)	15 (16)
Living arrangements
Live with spouse/partner	75 (82)	83 (87)
Live alone/family or friend/other	17 (18)	13 (13)
Ethnic group
White/Anglo/European Canadian	79 (86)	76 (79)
Other	10 (11)	20 (21)
Missing	3 (3)	0 (0)
Education
Less than or completed high school	21 (23)	28 (29)
Completed technical school	14 (15)	7 (7)
Completed college or university	28 (30)	37 (39)
Professional or postgraduate degree	27 (29)	24 (25)
Missing	2 (2)	0 (0)
Language spoken at home—English, yes	73 (79)	80 (83)
Employed full time, yes	39 (42)	38 (40)
Employed part time, yes	7 (8)	6 (6)
Retired, yes	42 (46)	49 (51)

Abbreviations: PCa-SCP, Prostate Cancer Survivorship Care Plan; UC, usual care.

^aBaseline Demographic Questionnaire was not completed by “02-0227C” from the Vancouver site, thus making the sample size 188 for this analysis.

The PAM score entered into the mixed-effects regression model ranged from 29.7 to 91.6. In the primary PAM-13 analyses, while the mean scores between groups differed at 6 months (P = .047), the change in scores over time between the groups was not significantly different (P = .16; Table 2). At 12 months, no significant differences were observed between groups (P = .2) or changed over time (P = .5).

Table 2.

Primary and Secondary Outcomes

Outcome	Time point	UC (n = 93)		PCa-SCP (n = 96)		Between-group difference (95% CI)a	P valueb	Effect size (95% CI)	Change in between-group difference over time (95% CI)c	P valued
		n (available)	Mean (SE)	N (available)	Mean (SE)
PAM-13	Baseline	85	66.0 (1.70)	92	67.5 (1.67)	1.50 (−2.25 to 5.22)	—	—	—	—
	6 mo	81	65.5 (1.72)	82	69.4 (1.71)	3.90 (0.05-7.71)	.047	0.31 (0.001-0.62)	2.40 (−0.93 to 5.72)	.16
	12 mo	83	65.9 (1.71)	70	68.5 (1.75)	2.54 (−1.36 to 6.44)	.2	0.19 (−0.13 to 0.51)	1.06 (−2.36 to 4.48)	.5
EORTC QLQ-INFO25	Baseline	93	58.4 (0.09)	95	56.2 (0.76)	−2.21 (−3.57 to −0.09	—	—	—	—
	6 mo	89	57.3 (1.39)	86	62.2 (1.83)	4.91 (2.85-6.97)	< .001	0.34 (0.04-0.64)	7.13 (5.53-8.71)	< .001
	12 mo	86	59.3 (1.77)	77	61.9 (1.55)	2.69 (2.15-3.25)	< .001	0.19 (−0.12 to 0.50)	4.91 (3.06-6.76)	< .001
heiQ
Social integration and support	Baseline	91	3.21 (0.06)	96	3.16 (0.05)	−0.04 (−0.27 to 0.17)	—	—	—	—
	6 mo	88	3.13 (0.02)	81	3.25 (0.05)	0.12 (−0.03 to 0.26)	.12	0.25 (−0.05 to 0.55)	0.16 (0.06 to −0.26)	.002
	12 mo	89	3.12 (0.02)	77	3.19 (0.04)	0.07 (−0.02 to 0.16)	.11	0.15 (−0.16 to 0.46)	0.12 (−0.01 to 0.24)	.069
Health service navigation	Baseline	93	3.25 (0.05)	96	3.32 (0.045)	0.07 (−0.05 to 0.20)	—	—	—	—
	6 mo	89	3.25 (0.05)	86	3.34 (0.047)	0.10 (−0.04 to 0.22)	.17	0.22 (−0.08 to 0.52)	0.02 (−0.09 to 0.13)	.7
	12 mo	89	3.20 (0.05)	77	3.29 (0.048)	0.09 (−0.04 to 0.22)	.18	0.2 (−0.11 to 0.51)	0.01 (−0.10 to 0.13)	.8
Constructive attitudes and approaches	Baseline	93	3.26 (0.49)	96	3.32 (0.51)	0.06 (−0.07 to 0.20)	—	—	—	—
	6 mo	89	3.23 (0.05)	86	3.36 (0.054)	0.13 (−0.01 to 0.28)	.07	0.27 (−0.03 to 0.57)	0.07 (−0.06 to 0.19)	.3
	12 mo	89	3.27 (0.05)	77	3.35 (0.056)	0.08 (−0.07 to 0.22)	.3	0.16 (−0.14 to 0.47)	0.01 (−0.11 to 0.14)	.8
Skill and technique acquisition	Baseline	93	2.94 (0.06)	96	2.97 (0.06)	0.03 (−0.11 to 0.16)	—	—	—	—
	6 mo	89	2.95 (0.06)	86	3.10 (0.06)	0.15 (0.01-0.28)	.03	0.33 (0.032-0.63)	0.12 (−0.02 to 0.24)	.050
	12 mo	89	2.99 (0.06)	77	3.07 (0.06)	0.08 (−0.06 to 0.22)	.3	0.18 (−0.13 to 0.49)	0.05 (−0.07 to 0.17)	.4
Emotional distress	Baseline	93	2.92 (0.08)	96	3.12 (0.08)	0.19 (0.03-0.36)	—	—	—	—
	6 mo	89	2.90 (0.08)	85	3.13 (0.08)	0.23 (0.06-0.40)	.007	0.38 (0.08-0.68)	0.04 (−0.07 to 0.16)	.5
	12 mo	89	2.94 (0.08)	77	3.12 (0.08)	0.18 (0.01-0.35)	.039	0.3 (−0.007 to 0.61)	−0.01 (−0.13 to 0.10)	.8
EPIC-26
Urinary incontinence	Baseline	87	46.6 (3.00)	84	52.0 (3.04)	5.46 (−2.94 to 13.8)	—	—	—	—
	6 mo	80	60.3 (3.05)	73	62.8 (3.13)	2.49 (−6.11 to 11)	.6	0.09 (−0.23 to 0.41)	−2.97 (−9.28 to 3.34)	.3
	12 mo	79	65.8 (3.05)	70	69.2 (3.16)	3.43 (−5.21 to 12.1)	.4	0.13 (−0.20 to 0.45)	−2.03 (−8.39 to 4.33)	.5
Urinary irritative/obstructive	Baseline	89	77.3 (1.73)	91	80.1 (1.72)	2.74 (−1.29 to 6.79)	—	—	—	—
	6 mo	88	86.9 (1.74)	84	89.2 (1.75)	2.32 (−1.79 to 6.44)	.3	0.19 (−0.11 to 0.49)	−0.43 (−4.57 to 3.71)	.8
	12 mo	85	89.0 (1.75)	75	91.2 (1.81)	2.19 (−2.03 to 6.41)	.3	0.2 (−0.11 to −0.51)	−0.56 (−4.81 to 3.69)	.7
Bowel	Baseline	93	92.3 (1.67)	93	93.6 (1.66)	1.30 (−2.49 to 5.08)	—	—	—	—
	6 mo	88	97.2 (1.69)	86	98.3 (1.70)	1.13 (−2.75 to 5.01)	.6	0.09 (−0.21 to 0.39)	−0.17 (−3.93 to 3.59)	.9
	12 mo	87	96.7 (1.70)	75	98.4 (1.75)	1.66 (−2.32 to 5.63)	.4	0.15 (−0.16 to 0.46)	0.36 (−3.50 to 4.22)	.8
Sexual	Baseline	89	23.1 (2.51)	89	31.7 (2.48)	8.58 (1.63-15.5)	—	—	—	—
	6 mo	88	26.3 (2.51)	84	33.8 (2.52)	7.51 (0.49-14.5)	.036	0.32 (0.02-0.62)	−1.07 (−6.09 to 3.95)	.6
	12 mo	87	28.2 (2.52)	73	37.9 (2.59)	9.76 (2.64-16.9)	.007	0.39 (0.08-0.70)	1.18 (−3.98 to 6.34)	.6
Hormonal	Baseline	92	83.4 (1.83)	91	85.6 (1.82)	2.24 (−2.50 to 6.97)	—	—	—	—
	6 mo	88	84.5 (1.85)	84	88.5 (1.85)	4.02 (−0.79 to 8.83)	.10	0.19 (−0.11 to 0.49)	1.78 (−1.87 to 5.44)	.3
	12 mo	86	84.5 (1.85)	74	90.4 (1.90)	5.94 (1.05-10.8)	.017	0.35 (0.04-0.66)	3.70 (−0.06 to 7.46)	.054
ASC
Cancer worry scale	Baseline	91	6.99 (0.27)	95	6.31 (0.27)	−0.68 (−1.35 to 0.01)	—	—	—	—
	6 mo	88	6.64 (0.27)	84	6.14 (0.27)	−0.50 (−1.18 to 0.19)	.15	−0.23 (−0.53 to 0.07)	0.18 (−0.32 to 0.68)	.5
	12 mo	84	6.78 (0.28)	75	6.06 (0.28)	−0.71 (−1.39 to −0.03)	.044	−0.31 (−0.62 to 0.003)	−0.03 (−0.55 to 0.48)	.9
Health worry scale	Baseline	84	6.67 (0.23)	92	6.37 (0.22)	−0.30 (−0.92 to 0.33)	—	—	—	—
	6 mo	85	6.33 (0.23)	82	5.82 (0.23)	−0.50 (−1.14 to 0.13)	.12	−0.24 (−0.55 to 0.05)	−0.21 (−0.72 to 0.30)	.43
	12 mo	79	6.22 (0.23)	73	5.80 (0.23)	−0.42 (−1.07 to 0.23)	.2	−0.19 (−0.52 to 0.12)	−0.12 (−0.65 to 0.41)	.7

Abbreviations: ASC, Assessment of Survivors Concerns; EPIC, Expanded Prostate Cancer Index Composite; EORTC QLQ-INFO25, European Organization for Research and Treatment of Cancer Quality of Life Information Module; heiQ, Health Education Impact Questionnaire; PAM, Patient Activation Measure; PCa-SCP, Prostate Cancer Survivorship Care Plan; UC, usual care.

The P values presented in the table were calculated using mixed-effects models in SAS (reference for mixed model methodology provided in the article).

Bold text indicates statistically significant value.

^aThis column represents the estimated difference in the outcomes between the UC and PCa-SCP groups at each specific time point.

^bThis P value tests whether the between-group differences are statistically significant at 6 and 12 months.

^c This column represents the estimated change in the outcomes between the UC and PCa-SCP groups at 6 and 12 months, relative to the baseline difference between the groups.

^dInteraction P value: The interaction P value tests whether the change in mean scores over time differs significantly between the 2 groups.

There was a significant increase in the EORTC QLQ-INFO25 total score in the PCa-SCP arm compared with the UC arm at 6 and 12 months (P < .001). Between groups, differences were significant (P < .001) between T0 and T1 (mean [95% CI]: 7.13 [5.53-8.71]) and T0 and T2 (4.91 [3.06-6.76]).

There was also significant between-group difference from T0 to T1 (mean [95% CI]: 0.16 [0.06-0.26], P = .002) on the heiQ Social Integration and Support subscale and borderline significance on the heiQ Skill and Technique Acquisition subscale from T0 to T1 (mean [95% CI]: 0.12 [−0.02 to 0.24], P = .05).

No group-by-time differences were detected on any of the other secondary measures (Table 2). Outcome data for the secondary measures across time points are provided in Supplementary File 2 (https://www.jurology.com). In addition, there were no differences on health behaviors (see Supplementary File 3, https://www.jurology.com) or health care utilization (see Supplementary File 4, https://www.jurology.com) between UC and PCa-SCP.

DISCUSSION

The results from this multicenter trial suggests that the provision of a prostate-specific electronic SCP during a dedicated nurse-led appointment resulted in higher scores on patient perception of information receipt and self-management skills, including social integration and support and skill and technique. However, there were no significant group-by-time differences on any other measures including patient activation, which was our primary outcome. This study was unique in that it used an electronic platform to pull and tailor the information provided to the patient based on algorithms and assessed needs, was delivered during a dedicated appointment with an oncology nurse who was trained in health coaching, and used outcomes that are considered relevant and proximal to SCPs by stakeholders. It was also the first clinical trial of an SCP developed and tailored for PCa survivors.

Although studies on SCPs have suggested that patients and health care providers find value in SCP,⁶ identifying measures that capture this has been difficult. Birken et al⁹ noted that most RCTs have included outcomes such as health-related quality of life, which SCPs likely have limited direct impact on and have encouraged the use of more proximal outcomes such as meeting informational needs and improving survivors’ capacity for self-management. We hypothesized that the PCa-SCP intervention would increase patient activation, given the SCP delivery was designed to take a collaborative and person-centered approach and focused on increasing knowledge and promoting self-management by connecting with appropriate and relevant resources. While the between-group ES at the 6-month time point would be considered moderate and encouraging (0.31), there was no between-group difference over time on the patient activation scores. The PAM-13 has not been used in other SCP studies, but it has been used in other supportive care interventions with PCa survivors.²⁴ In the current study, there were unanticipated significant ceiling effects, with mean scores aligning with level 3, and ∼80% of participants scoring in > level 3 at each time point, which likely contributed to the nonsignificant findings on this outcome. Future studies will need to carefully consider the primary outcome measure to include when evaluating the impact of SCPs. Given the ceiling effects we encountered with this population on the PAM-13, other measures may be more appropriate.

We did find a significant difference on the EORTC QLQ-INFO25, which suggests that the PCa-SCP positively affected participants’ perception of the information received. This was also found in a sizable pragmatic cluster trial of automatically generated SCPs in endometrial cancer survivors.⁸ Meeting patients’ informational needs is a fundamental aspect of patient-centered care and a key indicator of quality cancer care.²⁵ In addition, we found a significant difference on heiQ Social Integration and Support at the 6-month point. Individuals with high scores on the Social Integration and Support subscale have high levels of social interaction and confidence to seek support.¹⁷ We found a similar difference in the heiQ Skill and Technique Acquisition scores, which reflect highly developed skills in symptom relief and techniques to manage health.¹⁷ The PCa-SCP provided pragmatic information regarding the diagnosis and treatments received, an outline of follow-up care, tailored information and resources on symptom management, and whom to contact. This likely helped participants feel more knowledgeable regarding their cancer history and future follow-up care and, in the short term, more prepared to manage the challenges and to know where to seek support as needed.

While the study built upon previous recommendations regarding SCP research and we had good study retention between both study arms, it is important to consider the limitations of this study. To begin, there is not established minimally important clinical difference for the PAM-13 score and we based our sample size calculation on a half SD difference (8 points). While an ES on the PAM at 6 months of 0.31 and 0.19 is considered important for behavioral and psychological interventions, respectively,²⁶ we would have required a much larger sample to detect a difference. Second, we did not track the use of the SCP after receipt. Likely, some participants never referred back to the SCP, which has been documented in other studies,⁷ in which case the intervention impact would likely be suboptimal. Future research should track use of the tool provided. In addition, the study was conducted at 3 urban tertiary centers across Canada with well-developed supportive care services, which may have already met the needs of this population. The study participants were also predominantly White, English speaking, and well educated, which may help to explain the elevated PAM-13 scores²⁷ and reduces the generalizability of the results. Future research should strive to include underrepresented and disadvantaged populations or those with low scores at baseline, who may potentially benefit more from these interventions.²⁸ In addition, the study was not blinded because of the nature of the intervention, which may have affected participants’ responses to the questionnaires. Finally, although the missing data in our study did not seem to introduce bias based on baseline comparisons, missing data may still limit the generalizability of our findings.

CONCLUSIONS

The negative results of this RCT on patient activation should not end the discussion about the potential benefits of SCPs in PCa and other cancer survivors. While it is unlikely that SCP alone can address all of survivors’ educational and informational needs, SCPs should be seen as an important tool within a larger multifaceted comprehensive approach to follow-up care and can be used as a platform to facilitate discussion and reinforce key messages regarding health information, assessment, and care planning.^4,29

Supplementary Material

juro-213-407-s001.pdf ^{(1.7MB, pdf)}

juro-213-407-s002.pdf ^{(360.2KB, pdf)}

juro-213-407-s003.pdf ^{(525.5KB, pdf)}

juro-213-407-s004.pdf ^{(850.1KB, pdf)}