Subscribe to E-Bulletin Donate to CEGA

Evaluating the Efficacy of Mobile Phone Messaging in Improving ART Adherence in Kenya

Development Challenge

People living with HIV face a number of challenges, particularly due to HIV’s unique ability to weaken the immune system, making individuals more susceptible to infectious diseases, thereby increasing their vulnerability to mortality and morbidity.  As a result of a weakened immune system, people living with HIV may not actively or consistently participate in the labor force, potentially restricting economic growth in sub-Saharan Africa by as much as 1.5 percent annually.[1] The increasing availability of cost-effective antiretroviral therapy (ART) has greatly improved the ability to manage, care, and treat HIV. But imperfect adherence to ART regimens can result in drug resistance, rendering first-line treatments ineffective.  Treatment interruptions can arise for a host of behavioral, economic, and institutional reasons.  The importance of reminders and behavioral nudges in the developing country context is not well understood.

Context

Nyanza Province, bears the double burden of being one of the poorest provinces in Kenya, while also having one of the highest HIV prevalence rates of 13.9 percent (of the adult population).[2]  In the rural areas of the province, where access to comprehensive information, medication, health resources or trained health personnel may be difficult, the HIV prevalence rate has remained high. But even in these resource-limited settings, mobile phone coverage has been expanding rapidly. This study seeks to examine whether sending short message service (SMS) reminders through mobile phones can improve ART adherence for HIV-positive patients.

Evaluation Strategy

The study sought to reach individuals in the early phase of ART initiation, targeting individuals over age 18 who had started ART in the preceding 3 months. Participation did not depend on whether recipients had a cell phone or not, instead enrolled participants were given a mobile phone, phone credits, and recharging fees at the beginning of the study.  

The total number of participants included in the study was 431, with 289 randomly assigned to one of the four intervention groups, while the remainder were assigned into a comparison group. The four intervention groups received either short messages, intended to be a simple reminder, or long messages, intended to provide additional support, at varying frequencies:

  Daily Reminder Weekly Reminder
Short Reminder: "This is your reminder" Group 1 (70) Group 3 (73)
Long Reminder: "This is your reminder. Be strong and courageous, we care about you" Group 2 (72) Group 4 (74)

To determine whether patients actually took the medication, data was collected by monitoring whether pill bottles were opened.  Using a medication event monitoring system (MEMS), clinic staff could monitor whether dosages were being taken.  The phone number and MEMS cap number was recorded by researchers at the beginning of the study, and when participants returned to the clinic each month, the caps were scanned for data. Data were collected over the 48 weeks following treatment assignment; this follow-up period was divided into 12-week segments for analysis. The outcomes that were measured included whether or not adherence was greater than 90 percent in the 12-week segments (i.e. whether the ratio of bottle-openings measured by MEMS to the number of prescribed dosages was greater than 0.9), as well as whether or not there were any lapses in treatment beyond 48 hours (i.e. MEMS did not detect a bottle opening in more than 48 hours).

Results and Policy Implications

Participants who received weekly SMS reminders were significantly more likely to adhere to their ART regimen than those who did not receive reminders. Weekly reminders also reduced the frequency of interruptions in treatment. However, daily reminders did not have an overall impact on improving adherence.  It is possible that patients may have felt overstimulated by daily messages, but further research is needed to disentangle the mechanism for why weekly reminders may have been more effective than daily ones. Finally, the researchers determined that the length of the message – short or long - had no impact on adherence. 

As a cost-effective method to improve adherence and reduce interruptions of ART, the use of SMS reminders has the potential to be an important tool in optimizing behavior, especially in low-resource settings. However, more research is needed to fully understand the effects of timing and the role of message content, as well as studies that explore whether mobile phones can improve behavior outside of adherence reminders, such as appointment reminders, treatment side-effects, and other communications channels between healthcare workers and patients.

[1] Bell, C, Devarajan, S., and Gersbach, H., (2003), “The Long-Run Economic Costs of AIDS.” World Bank Policy Research Working Paper (no.3152).

[2] Kenyan National AIDS Control Council, (2012), The Kenya AIDS Epidemic: Update 2011. Nairobi, Kenya