Using Mobile Phone Text Messaging for Malaria Surveillance in Rural Kenya

Journal Article
  • Sophie Githinji
  • Samwel Kigen
  • Dorothy Memusi
  • Andrew Nyandigisi
  • Andrew Wamari
  • Alex Muturi
  • George Jagoe
  • RenĂ© Ziegler
  • Robert W. Snow
  • Dejan Zurovac
Malaria Journal
March 19, 2014, 13:107

Abstract

Background

Effective surveillance systems are required to track malaria testing and treatment practices. A 26-week study “SMS for Life” was piloted in five rural districts of Kenya to examine whether SMS reported surveillance data could ensure real-time visibility of accurate data and their use by district managers to impact on malaria case-management.

Methods

Health workers from 87 public health facilities used their personal mobile phones to send a weekly structured SMS text message reporting the counts of four basic surveillance data elements to a web-based system accessed by district managers. Longitudinal monitoring of SMS reported data through the web-based system and two rounds of cross-sectional health facility surveys were done to validate accuracy of data.

Results

Mean response rates were 96% with 87% of facilities reporting on time. Fifty-eight per cent of surveillance data parameters were accurately reported. Overall mean testing rates were 37% with minor weekly variations ranging from 32 to 45%. Overall test positivity rate was 24% (weekly range: 17-37%). Ratio of anti-malarial treatments to test positive cases was 1.7:1 (weekly range: 1.3:1–2.2:1). District specific trends showed fluctuating patterns in testing rates without notable improvement over time but the ratio of anti-malarial treatments to test positive cases improved over short periods of time in three out of five districts.

Conclusions

The study demonstrated the feasibility of using simple mobile phone text messages to transmit timely surveillance data from peripheral health facilities to higher levels. However, accuracy of data reported was suboptimal. Future work should focus on improving quality of SMS reported surveillance data.

Printer Friendly VersionPDF