Text message interventions for improving medication adherence in patients with coronary heart disease: a systematic review and individual patient data meta-analysis (Preprint)

<sec> <title>BACKGROUND</title> Following a primary cardiovascular disease (CVD) event, patients with established Coronary Heart Disease (CHD) remain at high risk for secondary CVD events and related mortality. However, this risk is reduced by secondary prevention therapies (comprising medication and lifestyle approaches), with an absolute reduc-tion in CVD events being directly related to risk factor control. Unfortunately, adherence to secondary pre-vention therapies remains poor. Text Message Interventions (TMIs) using short message service (SMS) may have a role in improving cardiovascular medication adherence, particularly owing to their simplicity, accessi-bility and scalability as compared to other approaches. </sec> <sec> <title>OBJECTIVE</title> To determine the effect of TMIs on cardiovascular medication among CHD patients, using a pooled analysis of individual patient data (IPD) from randomised clinical trials (RCTs). </sec> <sec> <title>METHODS</title> Eligible RCTs were identified via systematic review and their corresponding authors were invited to submit IPD for a pooled analysis. Eligible RCTs included participants diagnosed with CHD prior to the intervention and used TMIs to provide participants with educational information. Control groups comprised standard care. The primary outcome was cardiovascular medication adherence (yes/no), and the secondary outcomes were defined as: LDL &lt; 70mg/dL, blood pressure &lt; 140/90, BMI &lt; 25kg/m2, physical activity guideline met, smoking status (current vs never/previous). </sec> <sec> <title>RESULTS</title> Of 21 eligible RCTs, IPD was available from five studies totalling 3488 trial participants (Table 1). IPD analysis found no statistically significant association between the use of TMIs and medication adherence (odds ratio (OR): 1.17, 95% CI: 0.67-2.07, p=0.576); LDL targets (OR: 1.05, 95% CI: 0.89-1.23, p=0.6); blood pressure targets (OR: 0.92, 95% CI: 0.71-1.2, p=0.6); and, smoking status (OR: 0.96, 95% CI: 0.73-1.25, p=0.7). However, there was a statistically significant association between the use of TMIs and BMI targets (OR: 1.39, 95% CI: 1.1-1.75, p=0.006), and meeting physical activity guidelines (OR: 1.94, 95% CI: 1.12-3.34, p=0.017). Interaction analysis revealed that TMIs had a greater effect on adherence in men (OR: 1.12, 95% CI: 0.95-1.31) than in women (OR: 0.72, 95% CI: 0.57-0.93, p=0.015). Prediction analysis revealed that fe-male participants (OR: 0.66, 95% CI: 0.49, 0.88) and retired participants (OR:0.41, 95% CI: 0.31, 0.55) were significantly less likely to adhere to the prescribed medication regimen. Having a partner (OR: 1.31, 95% CI: 1.02, 1.68) and being on more medications (OR: 1.44, 95% CI: 1.34, 1.55) were positively associated with medication adherence. </sec> <sec> <title>CONCLUSIONS</title> While the statistical significance of TMIs effect on medication adherence remains uncertain, this analysis cor-roborated prior studies findings’ on the factors determining adherence, including gender, employment sta-tus, and marital status. Process evaluation with realist approaches might better elucidate the role of these factors. However, heterogeneity in adherence measures between studies was an issue in this and several pri-or metanalyses that should be addressed in future work. </sec>