Article Text
PDF
Abstract
Introduction The Building on Existing Tools to Improve Cancer and Chronic Disease Prevention and Screening in Primary Care (BETTER) programme trains allied health professionals working in primary care settings to develop personalised chronic disease ‘prevention prescriptions’ with patients. However, maintenance of health behaviour changes is difficult without ongoing support. Sustainable options to enhance the BETTER programme and ensure accessibility to underserved populations are needed. We designed the BETTER Women programme, which uses a digital app to match patients with a trained peer health coach (PHC) who provides ongoing support for health behaviour change after receipt of a BETTER prevention prescription in primary care.
Methods and analysis We will conduct a type 1 hybrid implementation-effectiveness patient-randomised trial. Interested women aged 40–68 years will be recruited from three large, sociodemographically distinct primary care clinics (urban, suburban and rural). Patients will be randomised 1:1 to intervention or wait-list control after receipt of their BETTER prevention prescription. We will aim to recruit 204 patients per group (408 total). Effectiveness will be assessed by the primary outcome of targeted behaviours achieved for each participant at 6 months, consisting of three cancer screening tests (cervical, breast and colorectal) and four behavioural determinants of cancer and chronic disease (diet, smoking, alcohol use and physical activity). Data will be collected through patient survey and clinical chart review, measured at 3, 6 and 12 months. Implementation outcomes will be assessed through patient surveys and interviews with patients, peer health coaches and healthcare providers. An embedded economic evaluation will examine cost per quality-adjusted life-year and per additional health behavioural targets achieved.
Ethics and dissemination This study has been approved by Women’s College Hospital Research Ethics Board (REB), the Royal Victoria Regional Health Centre REB and the University of Toronto REB. All participants will provide informed consent prior to enrolment. Participation is voluntary and withdrawal will have no impact on the usual care received from their primary care provider. The results of this trial will be published in peer-reviewed journals and shared via conference presentations. Deidentified datasets will be shared on request, after publication of results.
Trial registration number NCT04746859.
- Primary Health Care
- Chronic Disease
- Clinical Trial
- Randomized Controlled Trial
- Health Services
- Implementation Science
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- Primary Health Care
- Chronic Disease
- Clinical Trial
- Randomized Controlled Trial
- Health Services
- Implementation Science
STRENGTHS AND LIMITATIONS OF THIS STUDY
This work will test an intervention that adds support from community-based peer health coaches liaising with health care providers embedded in primary care to extend the evidence-based Building on Existing Tools to Improve Cancer and Chronic Disease Prevention and Screening in Primary Care (BETTER) Programme.
A major challenge in conducting this trial will be participant recruitment, as it is challenging to recruit trial participants for primary care-based interventions, particularly in the COVID-19 pandemic context.
Given the focus on lifestyle and behaviours, some key variables will rely on self-report.
Introduction
Over one in five Canadian adults have at least one chronic condition, such as cardiovascular (CV) disease, cancer, chronic renal disease or diabetes.1 Many of these conditions share common risk factors, which include diet, physical activity, tobacco use and alcohol consumption.1–3 Primary care clinicians are well situated to identify patients who may benefit from addressing such lifestyle and behavioural factors. Modifying these factors to promote health, in addition to screening for chronic diseases and cancers, are key components of primary care and essential to the sustainability of our healthcare system.
The BETTER (Building on Existing Tools to Improve Cancer and Chronic Disease Prevention and Screening in Primary Care) approach is an evidence-based and proactive approach to prevention and screening for chronic diseases.4 5 With the BETTER approach, allied health professionals (eg, nurses, dietitians) working in primary care settings are trained to be BETTER Prevention Practitioners: prevention and screening experts who conduct prevention visits after review of each patient’s completed BETTER Health Survey. Briefly, during the prevention visit, the prevention practitioner uses motivational interviewing principles and shared decision-making to develop a personalised BETTER Prevention Prescription. Prevention prescriptions include screening and prevention recommendations, tailored to the patient’s individual risk factors, based on personal medical history and family history. The prevention prescription also includes co-developed personalised goals, documented on a goals sheet—often related to diet, physical activity or cancer screening—to help patients reduce their risk of cancers and chronic diseases. A randomised controlled trial (RCT) of BETTER demonstrated that patients aged 40–65 years receiving the intervention had a substantial increase in prevention and screening actions at 6 months.5
Given how difficult it is to sustain changes to health behaviours, it is unrealistic to fund health professionals to offer the frequency of encounters needed to ensure goals are met and sustained. Fortunately, some evidence indicates that health coaches can improve and extend the quality and value of primary care.6 Specifically, evidence suggests that coaches can support self-management efforts by linking to community-based, culturally tailored resources that reduce health inequalities.7 This approach is particularly promising for patients from structurally disadvantaged communities (eg, ethno-racial groups) because a community-based health worker or lay coach may be better able to deliver proactive, targeted care.4 There is also evidence that community supports and relationships can help overcome challenges for people with chronic diseases in rural and remote areas.8 Taken together, these suggest that the outcomes of the BETTER programme could be enhanced and/or sustained, including for systemically marginalised groups and rural communities, by the implementation of peer health coaches (PHCs) who provide continued support to patients after they receive a prevention prescription.
The objectives of this type 1 effectiveness-implementation trial are two-fold. First, we will assess whether adding a 6-month long PHC digitally mediated intervention to the core BETTER programme, specifically for women 40–68 years old, leads to greater achievement of evidence-based screening and preventive health behaviours 6 months after receipt of the core BETTER programme. Second, we will examine the process of implementation to assess fidelity, mechanism(s) of action, sustainability and cost-effectiveness of the BETTER Women programme.
Methods and analysis
This protocol was completed in accordance with the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) statement and checklist. This statement serves as confirmation that the methods and manuscript of this protocol were prepared in adherence with SPIRIT.
Study design
This is a three-site, wait-list-controlled, type 1 hybrid effectiveness-implementation patient-randomised trial, with blinded outcome collection. All patients will receive the BETTER visit and personalised prevention prescription (https://www.better-program.ca/). Patients will be allocated to one of two study arms: (1) the intervention group will receive the PHC intervention shortly after their visit with a BETTER Prevention Practitioner (intervention group) or (2) the control group will receive usual care after their visit with a BETTER Prevention Practitioner and will be offered peer health coaching 6 months later. An embedded process and economic evaluation will be conducted during the trial. The trial has been registered on ClinicalTrials.gov (NCT04746859).
Since first developing the study protocol, the evaluation team has made several modifications to increase recruitment rates and adapt to changes resulting from the COVID-19 pandemic. Informed by the CONSERVE statement,9 we have summarised the study’s modifications in table 1. We report here the current version of the protocol (V.4.5, approved on 7 July 2023).
Summary of modifications to the study protocol
Study settings
The study takes place in Ontario, Canada, where physician services deemed medically necessary are provided to patients at no cost. Primary care is intended to be each patient’s main point of entry into the healthcare system and, ideally, serves as the patient’s medical home.10 While most primary care practices in Ontario are physician led, many are staffed by an interdisciplinary team that may also include nurse practitioners, nurses, dietitians and/or pharmacists.
Patient recruitment will occur in three multidisciplinary primary care clinics:
Women’s College Hospital Family Practice Health Centre (WCH FPHC) is a large urban, academic, hospital-based clinic in Toronto, Ontario and includes the Women’s College Academic Family Health Team (WCAFHT).
Summerville Family Health Team (Summerville FHT) is a medium-sized, suburban clinic in the Peel Region of the Greater Toronto Area and serves a large South Asian population.
Barrie and Community FHT (Barrie FHT) is a large clinic in Barrie, a small city in southern Ontario, with a large proportion of patients from rural areas.
These clinics offer comprehensive, longitudinal care to diverse populations with rosters ranging from 11 000 to 43 500 patients. Recruitment began at WCH FPHC in June 2021 and January 2022 for the two other clinics. Recruitment will continue at all sites until March 2024.
Eligibility criteria
Participants will be eligible women (self-identified) from the three primary care sites described above. We focus on women because middle-aged females are more likely than males to report three or more chronic diseases11; to visit a family physician for both physical and mental health concerns12 and to enrol in clinical trials of preventive interventions,13 including in the initial BETTER trial.
Patients are eligible to participate in the trial if they:
Identify as a woman.
Are 40–68 years of age.
Have a valid number for the Ontario health insurance plan (OHIP), which is available to all Ontario residents and is the means through which health services are accessed free of charge.
Have at least one behavioural risk factor (ie, smoking, alcohol use, diet or exercise) not at target or is overdue for screening for cervical, breast and/or colorectal cancer.
Have medical records accessible through their primary care clinic for the previous 3+ years.
Patients will be excluded if they are unable to give informed consent in English; unable or unwilling to communicate with the study team to collect outcomes in English via both email and phone; intend to spend >3 months out of town during the 6-month period when the coaching intervention is planned or have medical comorbidities at baseline that might interfere with their ability to follow through on the intervention. Prior to randomisation, we confirm the last exclusion criterion with the patient’s own primary care clinician who may request, for any reason, that a patient be excluded.
Intervention
All interested patients complete a BETTER visit with a Prevention Practitioner affiliated with their primary care clinic prior to randomisation to immediate or delayed coaching.
Prevention practitioners
All prevention practitioners will receive training to perform their role from the BETTER Institute or a trainer certified by the BETTER Institute (https://www.better-program.ca/institute/). They will receive robust PHC training from the Canadian Cancer Society (CCS) regarding their role and methods of providing behavioural support. They will also receive training from the research team to support delivery and evaluation of BETTER Women.
Peer Health Coaches
To be eligible to be a PHC, individuals must identify as a woman; be 40–68 years old; have experience with cancer screening or achieving at least one of the evidence-based healthcare behaviours under evaluation; be comfortable using computers and the Internet; be able to attend the PHC training programme; commit to volunteering 3–4 hours per week and be able to read, write and speak English. There was purposeful recruitment of PHCs who self-identified as South Asian and able to speak one of the following South Asian languages: Punjabi, Hindi, Tamil, Bengali or Urdu as the Summerville FHT suburban site serves a high proportion of South Asian patients who are multilingual. Individuals who do not complete the PHC screening process or training programme will be ineligible. All PHC applicants will be screened and undergo an interview process with staff from CCS who will administer the volunteer PHC programme. Applicants admitted to the PHC training programme will receive orientation to CCS and its organisational policies and procedures, as well as orientation to the BETTER Women programme.
The training programme consists of 24 hours of facilitated e-learning delivered over the course of 5 weeks involving self-study as well as synchronous and asynchronous group activities and role play and is meant to equip PHCs with the knowledge, skills and confidence necessary to use techniques of motivational interviewing brief action planning and coping planning. It was developed based on a set of 54 competencies that we developed and verified for the PHC role (Samadi et al; manuscript submitted for publication). It also teaches PHCs to be aware of the social determinants of health and factors that influence behaviour change and how to engage in antioppressive practice to support patients with achievement of existing goals, goal setting and planning to achieve goals. PHCs will support patients to achieve the personalised health goals they developed with the prevention practitioner and complete the overdue evidence-based preventive healthcare tasks identified by their prevention practitioner during their BETTER visit.14 Additional antioppression training, specifically designed for South Asian PHCs was offered to promote further cultural competency in working with diverse South Asian patients based on differing migration histories, ethnic origins and social identities (eg, gender, class, caste and religion). Details regarding the PHC training programme are outlined in Bender et al (manuscript in progress) and the antioppression training for South Asian PHCs in Banerjee et al (manuscript submitted for publication).
PHC-patient matching
Patients will be connected and supported by their PHC through a customised version of the NexJ Connected Wellness digital app designed for this study, herein hereafter referred to as digital app. The digital app is a PHIPA-compliant digital platform that will enable PHCs to communicate with patients virtually, monitor their progress towards their behaviour change goals and manage their matches. This platform can be accessed via mobile applications on a phone or tablet, or on a desktop browser. Patients will be matched with a PHC on the digital app based on a set of criteria including age, stage in life, ethnicity, geography and prevention-related needs (eg, risk factors). The digital app will present each patient with a shortlist of relevant and available PHCs and the patient will view the profiles of the PHCs and choose the PHC that is the most suitable for them. At the suburban site, South Asian women who prefer to communicate in their native language during the coaching sessions will be matched with a PHC who can speak that language (ie, Urdu, Hindi, Tamil, Bengali or Punjabi). PHCs may be matched to, and coach more than one patient, based on their availability.
Coaching process and activities
With consent from the patient, the selected PHC will receive, through the digital app, a lay summary and copies of the documents generated through the encounter with the prevention practitioner (ie, goals sheet and prevention prescription) to review in preparation for their first meeting. The PHC will call the woman, orient her to the PHC programme and begin the process of building rapport by exploring her goals, barriers to achieving those goals and any competing priorities. The PHC and woman will establish a schedule of regular coaching sessions for the 6-month intervention period and preferred modes/channels for communication such as phone, email or text messaging via phone or digital app. It is anticipated that PHCs spend, on average, 10–15 min per week communicating with each woman for ten weeks, tapering down toward monthly follow-ups until 6 months. The exact schedule will depend on the needs of the patient and their progress towards their goals, but the expectation is for 12–16 total encounters over the 6-month period.
The PHC will work collaboratively with the woman to help her follow through on a stepwise, personalised plan to achieve her chronic disease prevention and cancer screening goals. PHCs will also provide social support, navigate to relevant information and community resources and advise about options for self-monitoring (eg, applications that may be helpful to track health behaviours). The PHC will have access, via digital app, to evidence-based guidelines and various templates for notes to support and document their interactions and the participant’s progress toward their health goals. The digital app will also include a library of community resources to which PHCs can refer and recommend to their matches.
PHCs will have a clear scope of practice regarding expectations, boundaries and triage mechanisms to ensure that women participating in the programme are appropriately referred to a cancer information specialist (CIS), their prevention practitioner or primary care team when necessary. PHCs will use the support of CISs, who are regulated health professionals, in terms of providing evidence-based information and/or responding to more challenging barriers or issues faced by participants.
PHCs will be supervised and supported by CCS senior specialists in cancer prevention (including one designated for the suburban site who identifies as South Asian) in conjunction with the referring prevention practitioner and each site’s primary care team. Summary reports outlining support that the PHC provided to each patient and their progress will be securely shared electronically with prevention practitioners at the end of the 6-month peer health coaching intervention. CCS will lead monthly debriefing meetings of PHCs in each region and a ‘community of practice’ session to supervise and support PHCs in their roles and to help PHCs share experiences and promising approaches in supporting patients’ behaviour change. They will also receive one-on-one case supervision, mentorship and education from the prevention practitioners and the project team.
Outcomes
The primary outcome is a count of evidence-based screening and preventive health behavioural targets at 6 months after their prevention visit in comparison to baseline. Specifically, the primary outcome includes seven measures for each participant: the status (up to date or out of date) of three cancer screening tests (cervical, breast, colorectal) and the status (meeting evidence-based target or not) of four behavioural determinants of cancer and chronic disease (diet, smoking, alcohol use, physical activity). The outcome is binary if a patient meets at least one more target at 6 months than they met at baseline, their outcome will be considered positive.
Secondary outcomes include the 3-month and 6-month statuses of the three cancer screening tests (breast, cervical and colorectal screening) and four behavioural determinants of cancer and chronic disease (diet, smoking, alcohol and physical activity). For each behavioural determinant, we will measure the status, management and improvements through the course of the intervention. We expect more rapid changes to these health behaviours in response to the coaching and want to relate the behavioural changes to the timing where the coaching is more or less intense. For each behavioural determinant, we will also assess whether the risk factor was actively managed after the prevention visit (eg, newly treated or engaging with a dietician) and whether there was meaningful improvement achieved (ie, change in outcome in the direction desired).
To understand the effects on physiological measures of chronic disease management, there are a total 15 measures that will be assessed. We will assess whether patients are up to date for measurement of glycaemic control (haemoglobin A1c), hypertension (blood pressure), obesity (body mass index (BMI) or waist circumference) and cholesterol (specifically low-density lipoprotein (LDL)); whether new action was taken to improve these values (if not at target); and whether the actual values for A1c, blood pressure, LDL cholesterol and BMI were at-target based on each patient’s risk factors and diagnoses.
In addition, patients will report the achievement of personal health goals that were co-developed during their prevention visit. We will categorise and report the types of goals set by the type of targeted behavioural health risk factor and whether the goal matched the evidence-based target or was less than the evidence-based target (eg, targeting 90 min of exercise per week rather than evidence-based target of 150 min). Increased targeted behaviours from baseline and goal achievements for each patient will be accounted for.
Finally, the tertiary outcomes to be assessed are an estimation of 10-year CV risk measured using QRISK2,15 health-related quality of life and quality-adjusted life-year (QALY), measured using European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L),16 various economic and health services utilisation indicators such as intervention cost per patient, total healthcare costs, the number of outpatient visits, the number of emergency room visits and finally, the duration of the intervention effect which will be computed for intervention group patients only.
Participant timeline
After the prevention visit, the research assistant will send potential participants an introductory email summarising details of the intervention and the research requirements. The email includes a link to the electronic baseline survey hosted on the REDCap platform (https://www.project-redcap.org). After sending the baseline survey, the research assistant will phone the patient to answer any potential questions regarding the survey or the intervention. The research assistant may also send additional reminders to the patient, including two emails and three phone calls, before the baseline survey expires. Following completion of the baseline survey, participants will be randomised through REDCap. Intervention arm participants will then be contacted by CCS staff and directed to sign up on the digital app to facilitate matching and communication with a PHC (see below). Control arm participants will sign up on the digital app and be matched with a PHC 6 months post-enrolment if so desired. The participant timeline and flow of activities have been outlined in figure 1.
CONSORT diagram outlining participant flow through the BETTER Women trial. BETTER, Building on Existing Tools to Improve Cancer and Chronic Disease Prevention and Screening in Primary Care; CONSORT, Consolidated Standards of Reporting Trials.
Sample size
Data from the initial BETTER trial indicated that, on average, 46% of women between the ages of 40 and 65 who received the prevention practitioner intervention met at least one additional target.5 We consider a 30% increase (RR=1.33) to 60% as the minimally clinically important difference. To detect this difference with 80% power at a two-sided type I error of 5%, a sample size of 173 per group is required. To account for a possible attrition of 15%, we will aim to recruit 204 patients per group (408 total).
Recruitment
To identify potentially eligible patients, we will search each clinic’s roster based on the demographic inclusion criteria for patients and at least one of the behavioural risk factors (eg, current smoker) or screening risk factors (eg, overdue for mammography). Information for patients who appear to meet the eligibility criteria will be sent to their primary care provider to confirm eligibility. One week later, all patients who are flagged as eligible will be sent a programme invitation email via their primary care practice’s electronic medical record (EMR) platform.
The programme invitation email provides a brief description of the programme and is signed by the patient’s primary care provider unless their provider has not consented to having their signature included. In such cases, the invitation is signed by the research team. The invitation email also links to the study’s information letter and the electronic consent form (online supplemental file 1). If the patient consents to participate, they are directed to the BETTER Health Survey which collects a detailed health history to support development of the prevention prescription. If the patient declines, there is no further contact from the primary care team before a period of 1 year has passed, at which point, the patient may be invited to the programme again. Patients may also use the consent form to request further information from the primary care team or research assistant.
Supplemental material
Patients may also enter the programme by self-referral or referral by their primary care provider. Promotional materials (eg, poster and handouts) will be distributed to all three sites, and are available in translation at Summerville FHT, a site which serves a large South Asian population. Virtual events to share information with patients about the study will also be held at each of the sites to provide interested patients with an opportunity to learn more about the programme. In cases of self-referral, the research assistant will confirm eligibility with the patient before they are sent the programme invitation email. If at any point a patient is deemed ineligible, the research assistant will notify them that they do not currently meet the study criteria and they will not be sent the programme invitation email. At the Summerville FHT site, the prevention practitioner also makes phone calls to potentially eligible women to discuss recommended preventive care and make referrals to BETTER Women, as appropriate.
Assignment of intervention
Allocation
The research assistant will conduct randomisation for each patient using REDCap. The allocation file will provide 1:1 allocation, stratified by centre, following a block-randomised, computer-generated list of randomly permuted, variable block sizes to ensure concealment. Remaining blinded to the group allocation, the research assistant will securely share the results of randomisation with CCS staff who will contact patients and provide next steps according to their assigned group.
Blinding
Participants cannot be blinded, but those extracting data from EMRs and analysts will be blinded to group allocation. EMR extractors will complete a study form when extracting data to document their knowledge of group allocation during outcome assessment, and any deviations from the blinding protocol will be reported.
Data collection
Research staff will manually review patients’ EMRs and documents generated from prevention visits to extract data regarding cancer screening, achievement of health goals and physiological indicators of health. These data will be stored in REDCap and secure hospital-based servers. Data regarding participant demographic information, the four behavioural risk factors for cancers and chronic diseases, health-related quality of life and actions taken to manage physiological indicators of health will be collected from self-reported electronic surveys administered via REDCap at baseline as well as 3 months and 6 months post-enrolment. For the intervention group only, the electronic surveys and EMR data extraction will also occur at 12 months post-enrolment. Links to surveys will be sent via email and will expire 2 weeks later, except for the 6-month and 12-month surveys which expire after 4 weeks. Participants will receive a CAD$10 gift card for each completed survey. The data collection timeline is summarised in table 2.
Data collection timeline and instruments
For patients who consent to data linkage, health services utilisation data will be provided by the Institute for Clinical Evaluative Sciences (ICES) —a provincial not-for-profit research institute that collects and analyses population-based health and social data to support research and policy (https://www.ices.on.ca/). Data linkage will occur at the end of the study period.
Data management
Patient data from each primary care site will be shared securely with the trial team at WCH. CCS will also oversee secure transfer of patients’ data collection forms, goal sheets and prevention prescriptions among the primary care sites, the digital app and WCH. Each patient’s study identifier (which will be contained in a master linking template, managed by the research team) will be associated with relevant baseline data from the patient survey, the EMR, the prevention prescription and their personalised goals.
Data monitoring is conducted through monthly data quality checks by the WCH research team. Collected data are verified for accuracy, completeness and consistency with source data, and further validated to ensure quality prior to analysis.
Statistical analysis of trial data will be conducted by blinded analysts at the Applied Health Research Centre (AHRC) which is integrated with the Li Ka Shing Knowledge Institute of St. Michael’s Hospital in Toronto, Canada. AHRC will also coordinate linkage to ICES, using provincial health insurance numbers, for administrative database outcomes.
Statistical methods
Baseline data and participant demographic characteristics will be reported by group and then stratified by clinic and by rural/not rural and South Asian/other at the individual level. We will present baseline means, proportions, ranges and appropriate measures of variation in a table. In addition, types of goals set at baseline by participating women will be categorised and described. We will follow intention-to-treat principles in the analysis. The primary outcome will be compared with a χ2 test and both absolute and relative treatment effects will be estimated with 95% confidence intervals (CIs). If more than 5% of patients are missing from the primary outcome assessment, an inverse probability weighted analysis will be conducted to account for the potential bias arising from the improper subgroup. For each secondary outcome, we will model treatment effects using logistic regression for dichotomous outcomes and linear regression for continuous outcomes. Treatment effects will be expressed as (adjusted) odds ratios (ORs) or (adjusted) mean differences along with 95% CIs accordingly.
We will define an outcome variable—duration of effect—that reflects progress between 6 and 12 months. This variable will be modelled off the primary outcome but based only on the status of the four health behaviours since cancer screening status may not have the opportunity to be met at 6 months and then unmet at 12 months. Individual targets will be classified as newly met since 6 months, still met, or no longer met and analysed descriptively. If more targets are met at 12 months, patients will be said to have improved. If the same number of targets are met, patients will be said to be unchanged. Finally, if fewer targets are met, compared with 6 months, patients will be said to have worsened. This ordinal variable will be summarised. To address whether there has been meaningful worsening overall, we will consider the proportion who worsened. We consider more than 25% of patients worsening in the 6–12 months window clinically meaningful. We will conduct a one sample one-sided test of proportions against this value. That is, 
.
Sensitivity and subgroup analyses
A sensitivity analysis will assume those with a missing 6-month assessment did not satisfy the primary outcome. Additionally, to account for the possibility that some participants may decrease (rather than increase or maintain) the number of target screening actions and behaviours achieved, we will perform another sensitivity analysis whereby we convert the outcome from binary to ordinal (targets achieved increased or stayed the same or decreased). We will conduct an ordinal logistic regression to evaluate any differences in the distribution of these three categories between the intervention and control groups. Additionally, we will descriptively report the proportion of patients in each group who decreased and maintained the number of targets achieved. We will also explore a subgroup effect, acknowledging that such models will be exploratory and underpowered. These secondary analyses will examine the study site, age, neighbourhood income quintile (lowest vs other), ethnicity (South Asian vs other) and baseline value for the outcome in a logistic regression (or log-binomial) model.17 Subgroup effects will be explored by including treatment by subgroup interaction terms in the logistic model.
Economic analysis
We will conduct a health economic evaluation alongside the trial to compare total costs and health outcomes of intervention and waitlist control groups. Costs and outcomes will be assessed over the course of the first 6 months of each participant’s enrolment in the study. Data on services used, health utility and the effectiveness of the intervention will be obtained from the concurrent trial and health administrative databases housed at ICES. All costs associated with delivering the intervention will be included, including healthcare utilisation costs by patients within each study arm. Unit costs for all resources will be obtained from Canadian sources. All costs will be calculated from the perspective of Canada’s publicly funded healthcare system and expressed in 2024 Canadian Dollars. Responses to the EQ-5D-5L will be scored using Canadian preference weights18 and QALYs will be estimated using the total area under the curve method.19 The statistical analysis will be conducted in accordance with current guidelines for cost-effectiveness analysis alongside RCTs.20 The incremental cost and incremental outcome will be estimated using the generalised estimating equations that explicitly allow for the modelling of normal and non-normal distributional forms of repeated measure data.21 We will estimate the following incremental cost-effectiveness ratios (ICERs): cost per one QALY gained and cost per one additional evidence-based screening and preventive health behavioural targets achieved. The ICERs will be obtained through the difference in the mean costs of the two strategies divided by the difference in the mean value of each clinical outcome as denoted by the coefficient of the intervention indicator variable. Uncertainty in the analysis will be addressed by estimating 95% CIs using a non-parametric bootstrapping method. For this study, we will obtain 5000 estimates of costs and outcomes for each strategy. Results from the bootstrapping exercise will also be used to depict cost-effectiveness acceptability curves, which represents the probability of the intervention being cost-effective over a range of potential threshold values that the health system may be willing to pay for an additional unit of effect.22
We will also estimate a 3-year budgetary impact of implementing the intervention in Ontario from the Ontario Ministry of Health’s perspective. The results will be expressed in terms of absolute annual budget impact and budget impact per person per year.
Intervention implementation
This complex intervention will include an embedded process evaluation, informed by the UK Medical Research Council guidance on process evaluations of complex interventions.23 This multimethod process is intended to assess acceptability and fidelity of the programme as implemented, explore mechanism(s) of action, effect modifiers and factors that might affect sustainability of the programme and future attempts at spread and scale.
Outcomes
We will assess acceptability of the programme by determining the proportion of eligible women referred to BETTER Women who consent to the trial (recruitment rate) and go on to enrol in the trial (enrolment rate). For enrolled (ie, randomised) participants, we will also record the proportion for whom valid primary outcome data is obtained (retention rate). These data will be recorded for each partnering primary care site. We will assess each participant’s satisfaction with their PHC via an adapted scale24 and participants’ perception of the usability of digital app (the software platform used to connect participants with PHCs) via the System Usability Scale.25
Additional programme data to reflect engagement and participation of PHCs and participants will be tracked by PHCs. Specifically, each PHC will record the type/nature, length and mode of communication for each interaction with their matched participant, the number of referrals to community-based supports (eg, local groups, Smokers’ Helpline (https://www.smokershelpline.ca)), whether goals and/or action plans were refined or added, as well as interactions with the participant’s primary care team and/or prevention practitioner. We will aggregate this data to report on the total number (adherence) and frequency of coaching contacts, referrals, etc for each PHC, site and the programme overall. Engagement of the PHCs with the PHC Community of Practice, including participation in individual and group case supervision sessions and continuing education sessions, will also be monitored. Engagement and participation of these two groups have implications for sustainability of the programme.
Additional implementation outcomes will include qualitative exploration of fidelity, that is, whether the intervention was delivered as intended; whether, how and why any adaptations were made; feasibility and sustainability of the programme, as well as perceived or potential effect-modifying factors. We will also conduct quantitative exploratory analyses of adherence and whether the number of coaching contacts in the treatment group is associated with improvement. Mechanism of action will be explored by examining constructs related to the Health Action Process Approach (HAPA) from electronic surveys administered to patients.26
Recruitment
All patients in the intervention arm will be asked to indicate their interest in participating in a research interview on their 3-month and 6-month surveys. Each woman will be interviewed only once. A member of the research team will follow up with interested women via email to provide more information and invite them to schedule an interview (consent for the interview is included in consent for the study). Of those interested, we will seek maximum variation by clinical site, rural status, ethnicity, whether goals were or were not achieved and perceived satisfaction with their PHC and the programme. Once scheduled, patients will participate in a semi-structured phone interview that lasts 30–45 min.
All PHCs will be eligible to participate in a one-on-one semistructured interview with a member of the research team. Periodically, each site’s CCS coordinator will send an email invitation for a research interview to all PHCs who have coached at least one participant. Interested PHCs will be provided with a more detailed information letter and a consent form (online supplemental file 2). Those who consent will be scheduled for a 45–60 min interview.
Supplemental material
Similarly, all prevention practitioners will be eligible for a one-on-one semistructured interview with a member of the research team. Periodically, a research team representative will email the prevention practitioners at each site to invite them to participate in an interview. Other site staff who are directly involved with administration of the programme may also be invited to participate in an interview. Interested staff will be provided with a more detailed information letter and a consent form (online supplemental file 3) and those who consent will be scheduled for a 45–60 min interview.
Supplemental material
We will attempt to interview approximately one-quarter of the patients in the intervention group and at least one staff member and PHC from each primary care site but sample size will be informed by data saturation for each type of participant, from each of the primary care sites. All interview participants will receive a CAD$50 grocery store gift card as compensation for their time.
Data collection and analyses
Qualitative interviews
Qualitative interviews will be semistructured audio conversations conducted via Microsoft Teams. Interviews with patients and PHCs will be informed by the theoretical domains framework (TDF). The TDF is a commonly used technique in implementation science to unpack determinants of desired behaviour change.27 28 This represents a theory-informed, comprehensive but applied approach to understanding barriers and facilitators to programme participation, health behaviour change and delivery of the intervention as planned. Topics covered in the patient interview guide include their perceptions of the value of the PHC model and suggestions for strategies to improve the intervention. Topics covered in the PHC interview guide include the most critical activities to sustain the PHCs in their roles, opportunities for improvement of the programme and reflection on the perceived effects of being a coach on their own health behaviours.
Interviews of prevention practitioners will be informed by both the TDF and normalisation process theory (NPT).29 NPT helps to unpack how the day-to-day tasks of the programme do or do not align with the prevention practitioners’ other roles and their understanding of their role within the programme. Topics covered in the prevention practitioner interview guide include reflection on interactions between PHCs and prevention practitioners, how revised workflows can be routinised as it relates to programme activities and considerations for spread and scale of the programme. Interview guides for all three participant groups are available as online supplemental file 4.
Supplemental material
Interviews will be audio recorded, transcribed verbatim and analysed using the TDF (and NPT for the prevention practitioners). Thematic analysis will be informed by the approach of Braun and Clarke.30
Exploratory analyses
Adherence
We will explore if the number of coaching contacts in the treatment group is associated with improvement by examining the distribution of number of coaching calls for participants with a successful versus unsuccessful primary outcome and by conducting a per-protocol analysis, considering those who received at least 12 coaching contacts as ‘adherent’. We will also explore whether (1) number of goals set or (2) number of targets reached at baseline or (3) new personalised goals set during the coaching are associated with the effect size. Finally, we will examine variation in effects across different coaches with PHC identifiers added as a random effect.
Mechanism of action will be explored by examining constructs related to HAPA.26 This is a theory of behaviour change that considers two phases: a motivational phase and a volitional phase. In the motivational phase, individuals form an intention to perform a behaviour, based on how confident they are in being able to enact the behaviour (self-efficacy), what they believe may happen if they engage in the behaviour (outcome expectancies), and/or what they believe may happen if they do not engage (risk perceptions). The volitional phase describes how these intentions are translated into behaviour. This phase consists of action planning to specify when, where and how to perform the behaviour, coping planning to circumvent anticipated barriers to ensure the behaviour is performed and action control involving being aware of standards, self-monitoring progress and putting in effort.31 Self-efficacy plays a role in both phases, as do barriers and resources and social support. We will assess these constructs in the research surveys described above for the four health behaviours in the primary outcome. We will compare scores for the HAPA variables between arms to determine whether the intervention acted as expected on these variables.
Finally, we will explore whether sustainability of effects (ie, outcomes for the intervention group at 12 months) vary by targeted subgroup, health goal and/or coach. This could help inform refinement in the recruitment and selection criteria of patients, coaches and primary care sites.
Data synthesis and triangulation
We will compare quantitative and qualitative data to explore consistencies and contradictions across the data. Comparison will also allow us to use the data from interviews to explain quantitative findings. We will triangulate data that explore key questions regarding the embedded process evaluation, including implementation, mechanism of action, sustainability and effects of the intervention. For example, we will compare and integrate findings regarding implementation from PHC interviews, participant interviews and programme data. Similarly, regarding mechanism of action, we will compare and integrate findings from participant interviews and quantitative data regarding self-efficacy, risk perception, outcome expectancies, social support and completion of personalised goals. The synthesis of findings from different sources will be used to highlight key mechanisms, implementation difficulties and outcomes of the trial. This analysis is intended to clarify the core intervention components necessary to sustain the effects over time and/or to spread the programme in additional primary care sites.
Patient and public involvement
None.
Ethics and dissemination
Ethical considerations
This study was approved by the Women’s College Hospital Research Ethics Board (No. 2020-0079-E; initial approval October 2020; latest approval of the most recent protocol, V.4.5, was 7 July 2023) for the WCH site, by the Royal Victoria Regional Health Centre Research Ethics Board (No. R20-025, initial approval 4 August 2021; latest approval 13 March 2023) for Barrie FHT site and the University of Toronto Research Ethics Board (No. 29184, initial approval 16 November 2021; latest approval 9 May 2023) for the Summerville FHT site.
These REBs will provide trial oversight and there will not be a data monitoring committee because this is a pragmatic trial with no foreseeable risks of major harms. All protocol modifications will be submitted to the WCH, RVH and UofT REBs for approval and reflected on ClinicalTrials.gov. Participants will be informed of any protocol amendments that affect their participation in the study and the informed consent process will be repeated. The WCH REB will be consulted about whether participants need to be reconsented following any major protocol amendments that do not directly affect participants.
All participants will provide informed consent before they are enrolled, and any data collected. They will be informed that they are free to withdraw from the study at any point and that, for patients, withdrawal will have no impact on the usual care received from their primary care provider. Data from participants at each clinic will be shared securely with the evaluation team. WCH has completed a data transfer agreement with each primary care site, enabling the secure transfer of deidentified extracted EMR data. Each patient’s study identifier (which will be contained in a master linking template, managed by the research team) will be associated with relevant baseline data from the surveys, the EMR data and the BETTER programme documents.
The evaluation team has received input from the project sponsors with regards to intervention design to ensure feasibility of implementation and long-term sustainability. However, the PIs, and not the sponsor or funder, have ultimate authority over the evaluation activities, including design, analysis and publication/dissemination of study data.
Dissemination and results to date
The results of this trial will be published in peer-reviewed journals and shared via conference presentations after all data is collected, cleaned and analysed. Deidentified datasets will be shared on request, after publication of results. The results will be updated on ClinicalTrials.gov within 1 year of the primary completion date. We will also report the study results, in more appropriate formats (eg, reports, lay summaries, targeted seminars) to all stakeholders, that is, the funders, study sites and participants. The findings may also be shared via social media channels (eg, @WCHospital or @WCHResearch on X) and on the websites of the BETTER Institute (https://www.better-program.ca/evidence/the-better-women-program/) and the CCS (www.cancer.ca/BETTERWomen).
At the time of manuscript submission, 243 trial participants have enrolled and been allocated to intervention or waitlist control. 187 participants have completed follow-up (101 from intervention arm; 86 from waitlist control arm) and the balance are receiving the intervention or completing follow-up assessments.
Discussion
This trial will be the first to test an intervention that adds support from PHCs to enhance the evidence-based BETTER Programme.
We anticipate that the major challenge in conducting this trial will be participant recruitment. It is well established that it is challenging to recruit trial participants for primary care-based interventions.32 A study of primary care trials in the UK found that only 29% of eligible trials achieved their target sample size within the originally planned time frame and 35% required more than 50% more time than originally planned.33 The additional challenge of launching during the COVID-19 pandemic has both delayed launch of the trial due to significant fluctuations in healthcare providers’ capacities for additional programming and, we suspect, negatively impacted recruitment due to reductions in the number of patients willing or able to re-engage with the healthcare system for preventive care.34 35 We have taken steps to optimise our recruitment strategies, including extending our project timeline and hope to achieve our target sample size within the revised project timeline.
Given the focus on lifestyle and behaviours, it is unavoidable that some key variables will rely on self-report. While some risk factors, like regular cancer screening, will be objectively measured through medical records or health insurance databases, others like diet and alcohol consumption cannot be objectively measured without intensive and intrusive data collection methods. As such, we rely on self-reports to balance feasibility and validity of the research. We also recognise the possibility that those who are most likely to be interested and able to participate in this intervention may not be those who are most in need of support to address behavioural risk factors for chronic diseases and cancer. While this study is not explicitly focused on addressing populations at greater risk of developing chronic diseases and cancer, our planned subgroup analyses and process evaluation will provide some insights into the feasibility of delivering this intervention to patients who identify as South Asian and those residing in rural areas.
A key strength of the BETTER Women programme is that it builds on previous BETTER studies,4 5 36 which show that patients can increase their actions for chronic disease prevention and cancer screening after a prevention visit. This pragmatic randomised trial is intended to determine whether the benefits of BETTER can be amplified and/or extended by the addition of a peer health coaching intervention. Additionally, given that the pandemic necessitated virtual delivery of both prevention visits and coaching interactions, the study will offer useful insights into the challenges and feasibility of remotely delivering primary care-linked interventions for chronic disease prevention and cancer screening.
A further strength is the development of a comprehensive competency-based PHC training programme. Also, we have leveraged and customised an existing digital app to facilitate the matching of patients with PHCs, to enable PHCs to monitor behaviour change progress and to enable PHCs to communicate with their matches to support them in achieving their behaviour change goals. Similar versions of the NexJ digital app have been shown to be effective in other populations and health conditions including to help prostate cancer patients and caregivers navigate the cancer journey with support from a trained peer navigator.37
Ethics statements
Patient consent for publication
Acknowledgments
We would like to acknowledge the rest of the BETTER Women project team. We would also like to thank the study participants, peer health coaches and staff at the study sites for their time, dedication and contributions to this study. The work described here has been presented at the 2023 Cancer and Primary Care Research International conference in Oxford, UK.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
NK and CW are joint first authors.
Contributors JA, SF, EGrunfeld, RH, AL and NMI conceived the project. NK, CW, JA, ATB, JLB, SF, LG, EGoulbourne, EGrunfeld, RH, KS, AL and NMI conceptualised the design of the project. NK and CW wrote the first draft of the manuscript. NK, CW, AS, SA, FB and AR contributed to the plan for acquisition of data. NK, CW, AS, AR, AL and NMI contributed to the plan for data analysis and interpretation. AL acted as guarantor. All authors provided substantive edits to the project concept and the manuscript, and all authors approved the final version.
Funding This study is jointly funded by the Public Health Agency of Canada (project #: 1920-HQ-000094), the Canadian Cancer Society, the Peter Gilgan Foundation and Women’s College Hospital Foundation.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.