Achieving high coverage of larval-stage mosquito surveillance: challenges for a community-based mosquito control programme in urban Dar es Salaam, Tanzania

Table 3 Factors associated with habitat detection coverage by CORPs.

Variable	% (n/N)	OR [95%CI]	P
Habitat category	NA	NA	0.053
Natural	66.4 (146/220)	1.00^a	NA
Artificial non-agricultural	66.1(1766/266)	0.60 [0.406,0.909]	0.015
Artificial agricultural	57.9 (44/76)	1.38 [0.607,3.143]	0.441
CORPs familiarity with plot	NA	NA	< 0.001
No evidence of unfamiliarity	75.8(1788/235)	1.00^a	NA
Clear evidence of unfamiliarity	27.6 (168/608)	0.16 [0.130,0.203]	< 0.001
Intervention status	NA	NA	0.978
Non-larviciding	72.4 (775/1070)	1.00^a	NA
Larviciding	62.3(1181/189)	0.99 [0.645,1.548]	0.997

The probability that a wet habitat was detected by the CORPs was modelled with a binary distribution and logit link function using Generalised Estimating Equations (GEE) treating intervention status, CORPs' unfamiliarity with the plots and habitat category as the potential predictors
^a the reference group for the particular variable,
CI; confidence interval,
CORPs; community-owned resource persons
N; the number of wet habitats found during cross-sectional surveys
n; the number of wet habitats found by the CORPs during their routine habitat survey,
NA; Not applicable
OR; Odds ratios,

ISSN: 1475-2875