Table 1 Evaluation criteria for alternative sets of adjusted malaria incidence estimates
Criteria | Description | Evaluation method |
|---|---|---|
(a) Overall malaria incidence | Overall adjusted malaria incidence for Ifanadiana should be similar to overall malaria incidence in populations with optimal health care access in the district, to avoid under- or overestimation | Ratio of adjusted malaria estimates to malaria in optimal access areas between 0.7 and 1.3 |
(b) Distance decay | Adjusted malaria incidence estimates should remove the distance decay observed in unadjusted malaria incidence, to limit bias due to geographic access to health care | Ratio of incidence less than 5 km from a health center to incidence greater than 5 km away 0.7–1.3 |
(c) Financial access | Differences in adjusted malaria incidence between health centers according to fee-exemption status should be minimal, to limit bias due to financial access to health care | Ratio of fee-exemption to no fee-exemption in adjusted malaria estimates between 0.7 and 1.3 |
((d) Geographic heterogeneity | The geographic distribution of adjusted malaria incidence should be similar to the geographic distribution of fever reported in the IHOPE cohort study during the high transmission season | Percent of hotspot cluster area overlap between the two datasets during high malaria transmission season (SaTScan) |