The purpose of this study is to assess whether twice-daily text message reminders over a six-month period to health workers in Malawi about diagnosis and treatment of malaria, pneumonia, and diarrhea improve case management of these diseases.

Background Mobile health, or m-health includes the use of portable devices, such as mobile phones, patient monitoring devices, personal digital assistants, and other wireless devices to support medical and public health practice. Mobile phones are widespread in Asia and Africa, and dissemination of health-related messages via mobile phones is an inexpensive way to reach geographically dispersed recipients. Evidence from developing countries, is scarce, particularly for improving health service delivery. Only one rigorously designed study, a randomized controlled trial in Kenya, assessed the use of mobile phone text messaging to improve health worker case management practices, showing that sending daily text message reminders to health workers over six months led to a 24-percentage-point improvement in correct management of uncomplicated malaria.

Malaria is endemic throughout Malawi and poses a significant burden for the health system. Malawi changed its malaria case management guidelines in 2011 to require diagnostic confirmation with microscopy or a malaria rapid diagnostic test (RDT) prior to treatment. A health facility survey in 2011 prior to the rollout of RDTs found that only 67% of patients with malaria were correctly treated, primarily due to a missed diagnosis of malaria. Although more recent assessments are lacking, it is likely that Malawi faces similar problems documented in other sub-Saharan African countries: non-systematic diagnostic testing of febrile patients for malaria, poor adherence to negative test results, failure to administer the first dose of antimalarial therapy during the consultation, and gaps in patient counseling. Furthermore, additional assessments have pointed to deficiencies in quality of care for other common illnesses in Malawi, such as pneumonia and diarrhea, which are similar to some of the widespread quality problems found in other developing countries.

Study design and objectives

This study is a trial of text message reminders to health workers in Malawi to improve case management of malaria and other common illnesses. The study will employ a cluster-randomized, controlled trial design with pre- and post-intervention measures, with health facilities as the cluster and unit of randomization. In several districts of Malawi, health facilities will be randomized to one of three arms:

1. Text messages to health facility-based health workers on the correct management of malaria (patients of all ages)
2. Text messages to health facility-based health workers on the correct management of malaria (patients of all ages) and other common illnesses, per Integrated Management of Childhood Illness guidelines (patients < 5 years)
3. Control health facilities where health workers receive no study intervention (only routine supervision and supports from supervisors and district health management team, which will occur in text message arms as well)

Justification

This study will add to the small but growing body of literature on the potential for cell phones and text messages to improve malaria case management in different settings. In addition to providing evidence from a rigorously designed study on the effectiveness of text message reminders to health workers to improve malaria case management, this proposed study expands on the promising initial evidence base in the following ways:

• Assesses the effectiveness of text message reminders in a setting with malaria diagnostic testing
• Expands the intervention to include patients of all ages, not just children < 5 years
• Includes messages on common non-malarial illnesses, including pneumonia and diarrhea

Methodology All health workers providing clinical care or working in the pharmacy dispensing drugs at facilities randomized to arms 1 or 2 will receive text messages. Before the intervention, data collection via a cross-sectional health facility survey will occur in the three study arms and will include patient exit interviews with a focused history, physical examination, and blood smear; health worker interviews; and a brief facility assessment. Results from this survey will be used to measure the baseline levels of case-management quality for malaria and other diseases and to pinpoint performance deficiencies to target with the text message intervention. Once preliminary baseline results are available, a workshop will be held with key stakeholders, including National Malaria Control Programme staff, researchers, technology specialists, health education specialists, and health workers, to discuss the results and design an appropriate intervention strategy, including message content and timing.

Text message reminders on case management of malaria (arm 1) and of malaria and other common illnesses (arm 2) will be sent to health workers twice a day in the intervention groups for six months. At the end of the six-month text message intervention, approximately one year after the baseline survey, follow-up data will be collected with another cross-sectional health facility survey. At this time, in-depth interviews will also be conducted with selected health workers in intervention facilities to better understand their reactions to the text messages and to help determine mechanisms of action of the text message reminders. A second follow-up health facility survey will be conducted six months after the end of the intervention to assess the extent that changes in performance are maintained over time. Cost data on the intervention will also be collected for a cost-effectiveness analysis.

(Health workers will receive the text message reminders. These health workers (and others working in the sampled outpatient department) will be eligible for interviews on the day of the team's visit. In addition, patients visiting the study health facilities will be eligible for exit interviews, which is the primary way that the outcomes will be assessed.)

Inclusion Criteria:

• For health workers to receive text messages: Providing outpatient care or dispensing drugs at health facility randomized to one of two intervention arms.
• For health workers to be interviewed during baseline, endline, and follow-up surveys: working in the outpatient department sampled by the team during their visit.
• For patients to be interviewed during baseline, endline, and follow-up surveys: visiting facility to see a clinician for the first time for current illness