Improving Adherence to Antiretroviral Therapy with Triggered Real-time Text Message Reminders
Lora L Sabin,Mary Bachman DeSilva,Christopher J Gill,Li Zhong,Taryn Vian,Wubin Xie,Feng Cheng,Keyi Xu,Guanghua Lan,Jessica E Haberer,David R Bangsberg,Yongzhen Li,Hongyan Lu,Allen L Gifford
DOI: https://doi.org/10.1097/qai.0000000000000651
2015-01-01
Journal of Acquired Immune Deficiency Syndromes
Abstract:BACKGROUND:Real-time adherence monitoring is now possible through medication storage devices equipped with cellular technology. We assessed the effect of triggered cell phone reminders and counseling using objective adherence data on antiretroviral therapy (ART) adherence among Chinese HIV-infected patients.METHODS:We provided ART patients in Nanning, China, with a medication device (Wisepill) to monitor their ART adherence electronically. After 3 months, we randomized subjects within optimal (≥95%) and suboptimal (<95%) adherence strata to intervention vs. control arms. In months 4-9, intervention subjects received individualized reminders triggered by late dose taking (no device opening by 30 minutes past dose time) and counseling using device-generated data. Controls received no reminders or data-informed counseling. We compared postintervention proportions achieving optimal adherence, mean adherence, and clinical outcomes.RESULTS:Of 120 subjects enrolled, 116 (96.7%) completed the trial. Preintervention optimal adherence was similar in intervention vs. control arms (63.5% vs. 58.9%, respectively; P = 0.60). In the last intervention month, 87.3% vs. 51.8% achieved optimal adherence [risk ratio (RR): 1.7, 95% confidence interval (CI): 1.3 to 2.2] and mean adherence was 96.2% vs. 89.1% (P = 0.003). Among preintervention suboptimal adherers, 78.3% vs. 33.3% (RR: 2.4, CI: 1.2 to 4.5) achieved optimal adherence and mean adherence was 93.3% vs. 84.7% (P = 0.039). Proportions were 92.5% and 62.9% among optimal adherers, respectively (RR: 1.5, CI: 1.1 to 1.9) and mean adherence was 97.8% vs. 91.7% (P = 0.028). Postintervention clinical outcomes were not significant.CONCLUSIONS:Real-time reminders significantly improved ART adherence in this population. This approach seems promising for managing HIV and other chronic diseases and warrants further investigation and adaptation in other settings.