Malaria intervention scale-up in Africa : effectiveness predictions for health programme planning tools, based on dynamic transmission modelling

Korenromp, Eline and Mahiané, Guy and Hamilton, Matthew and Pretorius, Carel and Cibulskis, Richard and Lauer, Jeremy and Smith, Thomas A. and Briët, Olivier J.T. (2016) Malaria intervention scale-up in Africa : effectiveness predictions for health programme planning tools, based on dynamic transmission modelling. Malaria Journal, 15. 417. ISSN 1475-2875 (https://doi.org/10.1186/s12936-016-1461-9)

[thumbnail of Korenromp-etal-MJ2016-Malaria-intervention-scale-up-Africa-effectiveness-predictions-health-programme-planning-tools-based-dynamic-transmission]
Preview
Text. Filename: Korenromp_etal_MJ2016_Malaria_intervention_scale_up_Africa_effectiveness_predictions_health_programme_planning_tools_based_dynamic_transmission.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo

Download (1MB)| Preview

Abstract

Background: Scale-up of malaria prevention and treatment needs to continue to further important gains made in the past decade, but national strategies and budget allocations are not always evidence-based. Statistical models were developed summarizing dynamically simulated relations between increases in coverage and intervention impact, to inform a malaria module in the Spectrum health programme planning tool. Methods: The dynamic Plasmodium falciparum transmission model OpenMalaria was used to simulate health effects of scale-up of insecticide-treated net (ITN) usage, indoor residual spraying (IRS), management of uncomplicated malaria cases (CM) and seasonal malaria chemoprophylaxis (SMC) over a 10-year horizon, over a range of settings with stable endemic malaria. Generalized linear regression models (GLMs) were used to summarize determinants of impact across a range of sub-Sahara African settings. Results: Selected (best) GLMs explained 94-97 % of variation in simulated post-intervention parasite infection prevalence, 86-97 % of variation in case incidence (three age groups, three 3-year horizons), and 74-95 % of variation in malaria mortality. For any given effective population coverage, CM and ITNs were predicted to avert most prevalent infections, cases and deaths, with lower impacts for IRS, and impacts of SMC limited to young children reached. Proportional impacts were larger at lower endemicity, and (except for SMC) largest in low-endemic settings with little seasonality. Incremental health impacts for a given coverage increase started to diminish noticeably at above ~40 % coverage, while in high-endemic settings, CM and ITNs acted in synergy by lowering endemicity. Vector control and CM, by reducing endemicity and acquired immunity, entail a partial rebound in malaria mortality among people above 5 years of age from around 5-7 years following scale-up. SMC does not reduce endemicity, but slightly shifts malaria to older ages by reducing immunity in child cohorts reached. Conclusion: Health improvements following malaria intervention scale-up vary with endemicity, seasonality, age and time. Statistical models can emulate epidemiological dynamics and inform strategic planning and target setting for malaria control .