Skip to main content

Table 2 Potential confounders, factors and covariates expected to have the largest effect on observed insecticide susceptibility

From: Developing global maps of insecticide resistance risk to improve vector control



Sampling bias (spatial)

The dataset was not generated using a single systematic sampling design; the data are highly clustered in geographical space

Sampling bias (temporal)

The dataset did not come from a time series that sampled the same locations at regular intervals; each time period incorporates a different set of sites and much higher data volumes are available for more recent years


The full dataset is linked to 74 malaria vector species and species complexes, however, over half of the bioassay records are linked to members of the An. gambiae species complex


Within each insecticide class, different insecticides were tested (6 carbamates, 5 organochlorines, 16 organophosphates, and 8 pyrethroids)

Protocol variation

Corrected mortality values were derived from a mixture of WHO bioassays (using 9 updated protocols) and CDC bottle assays

Exposure dose and duration

The exposure dose and duration used in the bioassays varied although the majority of bioassays used standard doses and times

Generation tested

Population samples were maintained in the laboratory for differing periods, however, only results from bioassays using F0 and F1 generations were included