Developing a community-led SMS reporting tool for the rapid assessment of lymphatic filariasis morbidity burden : case studies from Malawi and Ghana

Stanton, Michelle C. and Mkwanda, Square Z. and Debrah, Alexander Y. and Batsa, Linda and Biritwum, Nana Kwadwo and Hoerauf, Achim and Cliffe, Matthew and Best, Abigail and Molineux, Andrew and Kelly-Hope, Louise A. (2015) Developing a community-led SMS reporting tool for the rapid assessment of lymphatic filariasis morbidity burden : case studies from Malawi and Ghana. BMC Infectious Diseases, 15: 214. ISSN 1471-2334

Abstract

BACKGROUND: Lymphoedema and hydrocoele are the two most common clinical manifestations of lymphatic filariasis (LF). In order to effectively target morbidity management strategies, more information is rapidly needed on morbidity burden across all endemic countries. The purpose of this study was to develop and test an SMS tool (MeasureSMS) which enables trained community-based health workers to report basic information on all cases they identified. METHODS: The tool was trialled in Chikwawa district, Malawi and Ahanta West district, Ghana in 2014. Salaried health surveillance assistants (HSAs) identified and reported cases in Malawi whereas volunteer community health workers (CHWs) were used in Ghana. Health workers were trained in recognising lymphoedema and hydrocoeles and submitting individual case data using MeasureSMS, after which they undertook a LF morbidity survey. After the reporting period, a random sample of reported cases was visited by a physician to verify the health workers' diagnoses. The proportion of correctly diagnosed cases i.e. the positive predictive value (PPV) was then calculated. RESULTS: HSAs in Malawi successfully reported 256 unique cases by SMS from 107 communities (166 hydrocoele, 88 lymphoedema, 2 with both), resulting in an estimated adult prevalence of 17.7 per 10,000 and 33.0 per 10,000 for lymphoedema and hydrocoele respectively. In Ghana, despite being less experienced in using SMS, CHWs successfully reported 360 unique cases by SMS from 33 communities (169 hydrocoele, 185 lymphoedema, 6 with both), resulting in an estimated adult prevalence of 76.9 per 10,000 and 70.5 per 10,000 adults for lymphoedema and hydrocoele respectively. The verification exercise resulted in a PPV for lymphoedema and hydrocoele diagnosis of 90 % (n = 42, 95 % CI 76.5 - 96.9) and 92 % (n = 49, 95 % CI 79.5 - 97.4) in Malawi and 94 % (n = 34, 95 % CI 78.9 %-99.0 %) and 47 % (n = 59, 35.1 %-61.7 %) in Ghana, indicating that non-invasive methods for diagnosing hydrocoeles needed to be further emphasised. CONCLUSIONS: The study concludes that given the appropriate education and tools, community-based health workers are exceptionally well-placed to participate in quantifying LF morbidity burden, and other NTDs with observable symptoms. This concept has the potential to enable national programmes to more effectively monitor their community impact in an efficient, timely and cost-effective way.