Evaluation of seasonal malaria chemoprevention in two areas of intense seasonal malaria transmission: Secondary analysis of a household-randomised, placebo-controlled trial in Houndé District, Burkina Faso and Bougouni District, Mali

Abstract

Background: Seasonal malaria chemoprevention (SMC) is now widely deployed in the Sahel, including several countries that are major contributors to the global burden of malaria. Consequently, it is important to understand whether SMC continues to provide a high level of protection and how SMC might be improved. SMC was evaluated using data from a large, household-randomised trial in Houndé, Burkina Faso and Bougouni, Mali.

Methods and findings: The parent trial evaluated monthly SMC plus either azithromycin (AZ) or placebo, administered as directly observed therapy 4 times per year between August and November (2014-2016). In July 2014, 19,578 children aged 3-59 months were randomised by household to study group. Children who remained within the age range 3-59 months in August each year, plus children born into study households or who moved into the study area, received study drugs in 2015 and 2016. These analyses focus on the approximately 10,000 children (5,000 per country) under observation each year in the SMC plus placebo group. Despite high coverage and high adherence to SMC, the incidence of hospitalisations or deaths due to malaria and uncomplicated clinical malaria remained high in the study areas (overall incidence rates 12.5 [95% confidence interval (CI): 11.2, 14.1] and 871.1 [95% CI: 852.3, 890.6] cases per 1,000 person-years, respectively) and peaked in July each year, before SMC delivery began in August. The incidence rate ratio comparing SMC within the past 28 days with SMC more than 35 days ago-adjusted for age, country, and household clustering-was 0.13 (95% CI: 0.08, 0.20), P < 0.001 for malaria hospitalisations and deaths from malaria and 0.21 (95% CI 0.20, 0.23), P < 0.001 for uncomplicated malaria, indicating protective efficacy of 87.4% (95% CI: 79.6%, 92.2%) and 78.3% (95% CI: 76.8%, 79.6%), respectively. The prevalence of malaria parasitaemia at weekly surveys during the rainy season and at the end of the transmission season was several times higher in children who missed the SMC course preceding the survey contact, and the smallest prevalence ratio observed was 2.98 (95% CI: 1.95, 4.54), P < 0.001. The frequency of molecular markers of sulfadoxine-pyrimethamine (SP) and amodiaquine (AQ) resistance did not increase markedly over the study period either amongst study children or amongst school-age children resident in the study areas. After 3 years of SMC deployment, the day 28 PCR-unadjusted adequate clinical and parasitological response rate of the SP + AQ regimen in children with asymptomatic malaria was 98.3% (95% CI: 88.6%, 99.8%) in Burkina Faso and 96.1% (95% CI: 91.5%, 98.2%) in Mali. Key limitations of this study are the potential overdiagnosis of uncomplicated malaria by rapid diagnostic tests and the potential for residual confounding from factors related to adherence to the monthly SMC schedule.

Conclusion: Despite strong evidence that SMC is providing a high level of protection, the burden of malaria remains substantial in the 2 study areas. These results emphasise the need for continuing support of SMC programmes. A fifth monthly SMC course is needed to adequately cover the whole transmission season in the study areas and in settings with similar epidemiology.

Trial registration: The AZ-SMC trial in which these data were collected was registered at clinicaltrials.gov: NCT02211729.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1. Incidence of malaria hospital admissions and deaths from malaria and clinical malaria by age.

Incidence of malaria hospitalisations and deaths from malaria (left panel) and uncomplicated clinical malaria (right panel) by age group over the study period. Malaria hospitalisations and deaths from malaria were defined as hospital admission with a diagnosis of malaria and blood-slide–or RDT-confirmed P. falciparum infection or deaths for which malaria was listed as the primary diagnosis. Clinical malaria was defined as attendance at study health facility with a history of fever or measured temperature ≥37.5 °C, with malaria infection confirmed by RDT. Incidence rates are presented per 1,000 person-years and include repeat events in the same child, provided the healthcare contact occurred more than 7 days apart. Vertical bars show 95% CIs. CI, confidence interval; RDT, rapid diagnostic test.

Fig 2. The incidence of malaria hospitalisations and deaths from malaria and uncomplicated clinical malaria over the study period by month of year.

Incidence of malaria hospitalisations and deaths from malaria (a) and clinical malaria (b) by calendar month over the study period. (c) shows the incidence of uncomplicated malaria during the period when SMC was delivered (shown by vertical dashed lines) amongst children who had received SMC within the previous 28 days or who had not received recent SMC (no SMC in the previous 35 days). Incidence rates are presented as per 1,000 person-months at risk rather than per 1,000 person-years and include repeat events in the same child, provided the healthcare contact occurred more than 7 days apart. Vertical bars show 95% CIs. The analysis of children with ‘no recent SMC’ was restricted to children who received 3 courses of SMC during that intervention year (i.e., this excludes children who missed SMC on more than one occasion). Malaria hospitalisations and deaths from malaria were defined as hospital admission with a diagnosis of malaria and blood-slide–or RDT-confirmed P. falciparum infection or deaths for which malaria was listed as the primary diagnosis. Clinical malaria was defined as attendance at study health facility with a history of fever or measured temperature ≥37.5 °C, with malaria infection confirmed by RDT. CI, confidence interval; RDT, rapid diagnostic test; SMC, seasonal malaria chemoprevention.

Fig 3. The incidence of malaria hospitalisations and deaths from malaria and episodes of uncomplicated clinical malaria by time since the most recent SMC treatment.

Incidence of malaria hospitalisations and deaths from malaria (a) and clinical malaria (b) according to time since the most recent SMC course. Incidence rates are presented per 1,000 person-years at risk. Vertical bars show 95% CIs. ‘No SMC this year’ indicates children who had not yet received their first course of SMC in that year of the study (but who subsequently received SMC). Malaria hospitalisations and deaths from malaria were defined as hospital admission with a diagnosis of malaria and blood-slide–or RDT-confirmed P. falciparum infection or deaths for which malaria was listed as the primary diagnosis. Clinical malaria was defined as attendance at study health facility with a history of fever or measured temperature ≥37.5 °C, with malaria infection confirmed by RDT. CI, confidence interval; RDT, rapid diagnostic test; SMC, seasonal malaria chemoprevention.