Abstract
BACKGROUND: We developed and tested the effectiveness of a tailored health information technology driven intervention: "Talking Prescriptions" (Talking Rx) to improve medication adherence in a resource challenged environment. METHODS: We conducted a parallel, randomized, controlled, assessor-blinded trial at the Aga Khan University (AKU), Karachi, Pakistan. Adults with diagnosis of cerebrovascular accident (CVA) or coronary artery disease (CAD) diagnosed least one month before enrollment, on anti-platelets and statins, with access to a mobile phone were enrolled. The intervention group received a) Daily Interactive Voice Response (IVR) call services regarding specific statin and antiplatelet b) Daily tailored medication reminders for statin and antiplatelet and c) Weekly lifestyle modification messages for a period of 3 months. We assessed Medication adherence to statin and antiplatelets by a validated version of the 8-item Morisky Medication Adherence scale 8 (MMAS-8) at 3 months by a blinded assessment officer. Analysis was conducted by intention-to-treat principle (ITT). RESULTS: Between April 2015 and December 2015, 197 participants (99 in intervention and 98 in the usual care group) enrolled in the Talking Rx Study. The dropout rate was 9.6%. Baseline group characteristics were similar. At baseline, the mean MMAS-8 was 6.68 (SD = 1.28) in the intervention group and 6.77 (SD = 1.36) in usual care group. At end of follow-up, the mean MMAS-8 increased to 7.41(0.78) in the intervention group compared with 7.38 (0.99) in usual care group with mean difference of 0.03 (S.D 0.13) (95% C.I [-0.23, 0.29]), which was not statistically significant. (P-Value = 0.40) CVA patients showed a relatively greater magnitude of adherence via the MMAS-8 at the end of follow up where the mean MMAS-8 increased to 7.29 (S.D 0.82) in the intervention group as compared to 7.07(S.D 1.24) in usual care group with mean difference of 0.22 (SD = 0.22) 95% C.I (-0.20, 0.65) with (P-value = 0.15). Around 84% of those on intervention arm used the service, calling at least 3 times and listening to their prescriptions for an average of 8 minutes. No user was excluded due to technologic reasons. CONCLUSION: The use of a phone based medication adherence program was feasible in LMIC settings with high volume clinics and low patient literacy. In this early study, with limited follow up, the program did not achieve any statistically significant differences in adherence behavior as self-reported by the MMAS-8 Scale. TRIAL REGISTRATION: Clinical Trials.gov NCT02354040.