Trends in malaria cases, hospital admissions and deaths following scale-up of anti-malarial interventions, 2000-2010, Rwanda.

BACKGROUND: To control malaria, the Rwandan government and its partners distributed insecticide-treated nets (ITN) and made artemisinin-based combination therapy (ACT) widely available from 2005 onwards. The impact of these interventions on malaria cases, admissions and deaths was assessed using data from district hospitals and household surveys. METHODS: District records of ITN and ACT distribution were reviewed. Malaria and non-malaria indictors in 30 district hospitals were ascertained from surveillance records. Trends in cases, admissions and deaths for 2000 to 2010 were assessed by segmented log-linear regression, adjusting the effect size for time trends during the pre-intervention period, 2000-2005. Changes were estimated by comparing trends in post-intervention (2006-2010) with that of pre-intervention (2000-2005) period. All-cause deaths in children under-five in household surveys of 2005 and 2010 were also reviewed to corroborate with the trends of deaths observed in hospitals. RESULTS: The proportion of the population potentially protected by ITN increased from nearly zero in 2005 to 38% in 2006, and 76% in 2010; no major health facility stock-outs of ACT were recorded following their introduction in 2006. In district hospitals, after falling during 2006-2008, confirmed malaria cases increased in 2009 coinciding with decreased potential ITN coverage and declined again in 2010 following an ITN distribution campaign. For all age groups, from the pre-intervention period, microscopically confirmed cases declined by 72%, (95% Confidence Interval [CI], 12-91%) in 2010, slide positivity rate declined 58%, (CI, 47%-68%), malaria inpatient cases declined 76% (CI, 49%-88%); and malaria deaths declined 47% (CI, 47%-81%). In children below five years of age, malaria inpatients decreased 82% (CI, 61%-92%) and malaria hospital deaths decreased 77% (CI, 40%-91%). Concurrently, outpatient cases, admissions and deaths due to non-malaria diseases in all age groups either increased or remained unchanged. Rainfall and temperature remained favourable for malaria transmission. The annual all-cause mortality in children under-five in household surveys declined from 152 per 1,000 live births during 2001-2005, to 76 per 1,000 live births in 2006-2010 (55% decline). The five-year cumulative number of all-cause deaths in hospital declined 28% (8,051 to 5,801) during the same period. CONCLUSIONS: A greater than 50% decline in confirmed malaria cases, admissions and deaths at district hospitals in Rwanda since 2005 followed a marked increase in ITN coverage and use of ACT. The decline occurred among both children under-five and in those five years and above, while hospital utilization increased and suitable conditions for malaria transmission persisted. Declines in malaria indicators in children under 5 years were more striking than in the older age groups. The resurgence in cases associated with decreased ITN coverage in 2009 highlights the need for sustained high levels of anti-malarial interventions in Rwanda and other malaria endemic countries.

Impact Evaluation of Malaria Control Interventions on Morbidity and All-Cause Child Mortality in Rwanda, 2000-2010.

The impressive decline in child mortality that occurred in Rwanda from 1996-2000 to 2006-2010 coincided with a period of rapid increase of malaria control interventions such as indoor residual spraying (IRS); insecticide-treated net (ITN) distribution and use, and improved malaria case management. The impact of these interventions was examined through ecological correlation analysis, and robust decomposition analysis of contextual factors on all-cause child mortality. Child mortality fell 61% during the evaluation period and prevalence of severe anemia in children 6-23 months declined 71% between 2005 and 2010. These changes in malaria morbidity and mortality occurred concurrently with a substantial increase in vector control activities. ITN use increased among children under five, from 4% to 70%. The IRS program began in 2007 and covered 1.3 million people in the highest burden districts by 2010. At the same time, diagnosis and treatment with an effective antimalarial expanded nationally, and included making services available to children under the age of 5 at the community level. The percentage of children under 5 who sought care for a fever increased from 26% in 2000 to 48% in 2010. Multivariable models of the change in child mortality between 2000 and 2010 using nationally representative data reveal the importance of increasing ITN ownership in explaining the observed mortality declines. Taken as a whole, the evidence supports the conclusion that malaria control interventions contributed to the observed decline in child mortality in Rwanda from 2000 to 2010, even in a context of improving socioeconomic, maternal, and child health conditions.