Table 2.

Reactive IRS compared with reactive case detection or reactive mass drug administration (superiority design)

Outcomes	Anticipated Absolute Effects		Relative Effect (95% CI)	Number of Trials, Participants	Certainty of the Evidence (GRADE)	Comments
	Risk with Reactive IRS (95% CI)	Risk with Standard of Care (95% CI)
Incidence of clinical malaria	30.2 cases/ 1,000 PY (15.0–45.5)	38.9 cases/ 1,000 PY (20.7–57.1)	RR = 0.65 (0.19–1.11)	One trial; 56 clusters (55 analyzed: 28 reactive IRS, 27 no reactive IRS); 18,303 participants (9,464 reactive IRS, 9,352 no reactive IRS)	MODERATE owing to imprecision	There is probably no difference in the incidence of clinical malaria in clusters with reactive IRS plus RACD or rMDA compared with clusters with RACD or rMDA; imprecision resulted from wide confidence intervals
Parasite prevalence	2.92% (2.13–3.99)	4.07% (2.92–5.64)	PR = 0.32 (0.15–0.80)	One trial; 56 clusters (55 analyzed: 28 reactive IRS, 27 no reactive IRS); 4,082 participants in endline survey (2,052 reactive IRS, 2,030 no reactive IRS)	HIGH	The prevalence of malaria in clusters with reactive IRS plus RACD or rMDA was lower than in clusters with RACD or rMDA

GRADE = Grading of Recommendations, Assessment, Development and Evaluation; IRS = indoor residual spraying; rMDA = reactive mass drug administration; PR = prevalence ratio; PY = person-year; RACD = reactive case detection; RR = risk ratio.