The Interventions Supporting Long-term Adherence aNd Decreasing cardiovascular events (ISLAND) trial was a three-arm, single-blinded pragmatic randomized trial of theory-informed interventions that sought to improve adherence to evidence-based medications and completion of cardiac rehabilitation among patients discharged post-heart attack. The trial randomized over 2700 patients across nine cardiac centres in Ontario, Canada, finding no effects on medication adherence, but meaningful increases in cardiac rehabilitation.  Upon acceptance of the ISLAND trial, the editors of The BMJ asked for our reflections and so we asked ourselves: How did we get here? And what now?
First, we developed and implemented a single-centre pilot randomized controlled trial. In that study, we rapidly developed and tested a reminder-letter urging survivors of myocardial infarction to adhere to their prescribed medications. The trial demonstrated the general feasibility of our approach, but did not find changes in patient behaviour.  We realized that our intervention did not adequately address the key underlying determinants of our targeted behaviours. One great challenge in preparing for ISLAND was in utilizing a theory-informed approach to the intervention components, while incorporating user-centred design principles to the intervention implementation. [3,4] If we are to build interventions that people use and find useful, and if we are to learn as much as possible from them, further methodological development in combining theory-based and user-centred design approaches are likely needed.
Our approach in ISLAND was built not only on theory, user-input, and experience from a pilot trial, but also on insights from leaders carrying out similar, hospital-funded programs. However, for many conditions, multiple (oft-competing) local implementations of interventions (“letting a thousand flowers bloom”) will not be as efficient as a centralized approach. For ISLAND, it was simply not feasible for each participating cardiac centre to hire staff to carry out these post-discharge interventions. It’s also not equitable that patients covered by the same health insurance should get different care depending on the fund-raising abilities of the hospital foundation where the ambulance happened to take them. Our principle for ISLAND was to develop and test an intervention that could be operationalized consistently and equitably for all patients in the health-system. We believe the success of ISLAND could act as an impetus for discussions about which programs should be run locally, and which should seek more efficient approaches to central implementation. Centralization need not limit innovation or contextualization; an implementation laboratory approach could be used to embed quality improvement trials within a system-wide intervention. 
The key to allowing ISLAND to be centrally organized was ensuring that all interventions were delivered on behalf of the cardiology team of the hospital where the patient was treated for their heart attack. In effect, the research team conducting the intervention (and the program evaluation) acted as an agent of the cardiology team. In ISLAND, our limited resources meant that we could not have study staff in each cardiac centre, and our pragmatic approach meant that we wanted the evaluation to reflect a real-world implementation. This trial of a low-risk, low-burden, quality improvement intervention met criteria for an altered approach to consent, wherein participants had multiple opportunities to opt-out and only formally consented during telephone-based outcome collection at the end of the study. The lack of formal complaints validated the approach taken; the burden of consent should not be greater than the burden of the intervention tested in a quality improvement study. We believe that ISLAND thus offers a key example for other low-risk, low-burden quality improvement trials.
We’re proud of what we’ve accomplished over the last 10 years… but we have so much more to learn. In the coming years, we will strive to build on the foundations learned with ISLAND in order to benefit patients and health systems around the world.
Noah Ivers, clinician scientist, Department of Family and Community Medicine, Women’s College Hospital, Toronto, Canada.
J-D Schwalm, associate professor and scientist, Population Health Research Institute, Hamilton Health Sciences, McMaster University, Canada.
Competing interests: Please see research paper.
1 Ivers, Schwalm et al. Interventions Supporting Long-term Adherence aNd Decreasing cardiovascular events (ISLAND): Pragmatic randomized trial. https://www.bmj.com/content/369/bmj.m1731
2 Schwalm et al. Cluster randomized controlled trial of Delayed Educational Reminders for Long-term Medication Adherence in ST-Elevation Myocardial Infarction (DERLA-STEMI). Am Heart J. 2015 Nov;170(5):903-13.
3 Presseau et al. Identifying determinants of medication adherence following myocardial infarction using the Theoretical Domains Framework and the Health Action Process Approach. Psychol Health. 2017 Oct;32(10):1176-1194.
4 Witteman et al. Negotiating Tensions Between Theory and Design in the Development of Mailings for People Recovering From Acute Coronary Syndrome. JMIR Hum Factors. 2017 Mar 1;4(1):e6.
5 Grimshaw et al. Reinvigorating stagnant science: implementation laboratories and a meta-laboratory to efficiently advance the science of audit and feedback. BMJ Qual Saf. 2019 May;28(5):416-423.