Estimating the overall effect of a mass drug administration strategy for malaria in Los Chiles, Costa Rica: one-year target trial emulation with synthetic controls

Abstract Background Mass drug administration (MDA) is a WHO-recommended strategy to interrupt Plasmodium falciparum transmission in areas approaching elimination. After three years without autochthonous cases (2013–2015), local transmission was re-established in Costa Rica, predominantly of P. falciparum , prompting a 2023 MDA in the Los Chiles focus. We describe the intervention and estimate its effect. Methods Two supervised cycles of chloroquine (25 mg/kg per cycle over three days, age-adjusted, seven weeks apart) were offered between April and June 2023 to all residents aged ≥ 1 year without contraindications in three localities (Medio Queso, Coquital and San Gerardo; 4,624 eligible individuals). We emulated a cluster-randomized target trial to estimate the average treatment effect on the treated under an intention-to-treat approach. Control localities from the same focus were selected using explicit criteria (≥ 1 autochthonous case in the preceding six months, ≥ 5 km from treated localities, no overlap in case residence or place of infection; n = 7). Generalized synthetic controls, reweighting control localities to reproduce the 24-month pre-intervention trajectory, provided the counterfactual. The outcome was autochthonous case counts (all Plasmodium species) at 1, 3, 6 and 12 months. Results Coverage (≥ 1 dose) was 93.3% across both cycles (77.3% cycle 1; 51.2% cycle 2); 68.1% completed the full regimen in at least one cycle and 20.4% in both (54 participants reported ≥ 1 adverse event; none serious). Treated localities had higher pre-intervention case counts and testing volumes than control localities. Four autochthonous cases were detected in treated localities in the first 30-day period, and none from the second period through 12 months. The average difference was − 3.72 cases per locality per 30-day period (95% CI − 7.20, − 0.94), or approximately 134 cases averted across the three localities over 12 months (95% CI 34, 259); the offset-based rate ratio was 0.23 (95% CI 0.13, 0.49). Conclusion After a two-cycle chloroquine MDA, no autochthonous cases occurred in treated localities after the first 30 days. Because assignment was not randomized and treated and control localities differed at baseline, residual confounding cannot be excluded, and a concurrent regional decline may have contributed. MDA may help accelerate elimination in very-low-transmission foci with adequate surveillance.