A Successful Multifaceted Trial to Improve Hypertension Control in Primary Care: Why Did it Work?

ABSTRACT Background It is important to understand which components of successful multifaceted interventions are responsible for study outcomes, since some components may be more important contributors to the intervention effect than others. Objective We conducted a mediation analysis to determine which of seven factors had the greatest effect on change in systolic blood pressure (BP) after 6 months in a trial to improve hypertension control. Design The study was a preplanned secondary analysis of a cluster-randomized clinical trial. Eight clinics in an integrated health system were randomized to provide usual care to their patients ( n = 222), and eight were randomized to provide a telemonitoring intervention ( n = 228). Participants Four hundred three of 450 trial participants completing the 6-month follow-up visit were included. Interventions Intervention group participants received home BP telemonitors and transmitted measurements to pharmacists, who adjusted medications and provided advice to improve adherence to medications and lifestyle modification via telephone visits. Main measures Path analytic models estimated indirect effects of the seven potential mediators of intervention effect (defined as the difference between the intervention and usual care groups in change in systolic BP from baseline to 6 months). The potential mediators were change in home BP monitor use, number of BP medication classes, adherence to BP medications, physical activity, salt intake, alcohol use, and weight. Key Results The difference in change in systolic BP was 11.3 mmHg. The multivariable mediation model explained 47 % (5.3 mmHg) of the intervention effect. Nearly all of this was mediated by two factors: an increase in medication treatment intensity (24 %) and increased home BP monitor use (19 %). The other five factors were not significant mediators, although medication adherence and salt intake improved more in the intervention group than in the usual care group. Conclusions Most of the explained intervention effect was attributable to the combination of self-monitoring and medication intensification. High adherence at baseline and the relatively low intensity of resources directed toward lifestyle change may explain why these factors did not contribute to the improvement in BP.