Yapabandara 2001.

Methods	Study design: cRCT Unit of allocation: clusters (villages) Number of units: 8 villages divided equally into 2 arms. On the basis of 1 year's preintervention data the villages were stratified into 4 with high levels of malaria transmission and 4 with lower transmission. Within each strata 2 villages randomly selected for intervention. Outcome assessment/surveillance type: passive case detection. Also 2 mass blood surveys were carried out in July and December during the pre‐ and postintervention years. Blood films were taken, regardless of the presence/ absence of fever, from all the residents of the 8 villages. Length of follow‐up: January 1994 to December 1995 Adjustment for clustering: yes
Participants	Number of participants: 4/8 villages had populations < 500 while the other 4 had populations of 600–1100. Population characteristics: not reported. Withdrawal and loss to follow‐up: not reported.
Interventions	Larvicide: Active ingredient and dosage: pyriproxyfen, S31183 (Adeal 0.5% G) applied at a rate of 0.01 mg active ingredient/L (2 g granules/m³) Formulation: not reported. Manufacturer: not reported. Quality control of the larvicide: community engagement to encourage community to inform about new gem pits so that they could be rapidly treated. Duration of the activity of the larvicide: not reported but assays were conducted to determine if residual activity was present. Method of application: not reported. Frequency of application: 3 applications – 1 in December 1994, 1 between June and July 1995 in the postmonsoon season when river bed pools were formed, and 1 at end of November 1995. Buffer size between clusters: not reported.
Outcomes	Malaria incidence defined as fever/history of fever and parasites detected by blood film. Infection prevalence (slide positivity rates) Number of anophelines
Location profile	Study location: This study was carried out in Kaluganga, which is part of Elahera gem‐mining area situated in Matale District (7°40N, 80°50E) in the dry central zone of Sri Lanka. A cluster of 8 villages with a total area of 23 km² was selected for this study. The numbers of gem pits per village ranged from 311 to 3622. The villages were surrounded by thick jungle. The area was a settlement scheme, which was established about 30 years before the trial was conducted around the rivers, Aban ganga and Kalu ganga. Malaria endemicity: not reported. EIR: not reported. Population proximity/density: treated gem pits and pools up to 1.5 km from villages. Plasmodiumspecies:Plasmodium falciparum and Plasmodium vivax
Vector profile	Primary (and secondary) vector species:An culicifacies (An subpictus and An varuna) Vector behaviour (nature, stability, adult habitat, peak biting times, exophilic/endophilic, exophagic/endophagic, anthropophilic/zoophilic): not reported Aquatic habitats (type, stability and extent (number and size), proximity of aquatic habitats to human habitation): Shallow pits dug by gem miners that filled with water. Breeding of An culicifacies was almost entirely in gem pits but some breeding of An subpictus and most of An varuna was in other sites such as river bed pools and slow‐moving river margins Phenotypic resistance profile: not reported Genotypic resistance profile: not reported Method of mosquito collection: Anopheline populations in the study area were estimated by 7 sampling methods: window exit trap collection; pyrethrum spray sheet collection; indoor HCs; all night or for the first part of the night up to midnight; cattle‐baited hut collection and cattle‐baited net trap collection; and light trap collection. The locations chosen for applying these methods were near the centres of each village to try to avoid interference by immigration of mosquitoes from neighbouring villages. Data reported were only from cattle‐baited huts, partial night landing catches, and all night landing catches.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Stated that they randomized, but unclear how.
Allocation concealment (selection bias)	Unclear risk	Stated that they randomized, but unclear how.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Blinding was not possible for the intervention; however, unlikely to affect the outcome.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported whether blinding was used. Unclear whether slide readers were blinded.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	No cohort established. Movement in and out of study area not documented.
Selective reporting (reporting bias)	Unclear risk	No protocol was published beforehand. Not all the stated entomological outcomes described in the methods were reported.
Incorrect analysis	High risk	Inappropriate analysis, no adjustment for clustering
Other bias	Unclear risk	Recruitment bias: low risk – randomized study where they had selected clusters based on malaria cases before the intervention was introduced to ensure this was equal in both arms. Mass blood surveys and census attempted to include the entire population. Baseline imbalance: low risk – baseline characteristics appeared similar Loss of clusters: low risk – no clusters were lost Comparability with RCTs randomizing participants: low risk – larviciding is expected to have a community wide effect and should be implemented at a community level.

Abbreviations: An: Anopheles; Bti: Bacillus thuringiensis israeliensis; Bs: Bacillus sphaericus; EIR: entomological inoculation rate; ELISA: enzyme‐linked immunosorbent assay; HLC: human landing catches; ITN: insecticide‐treated nets; OR: odds ratio; PCR: polymerase chain reaction; RCT: randomized controlled trial; RDT: rapid diagnostic test; TCU: Ten‐Cell Unit.