Table 1.
Univariable analysis of the association between CHIKV seroprevalence and environmental variables. The models were adjusted for clustering of individuals in villages and rural communities (random intercept).
| Environmental Variables | OR (95%CI) | LRT p-Value | AIC |
|---|---|---|---|
| EVI_max (per 0.1 increase) | 1.54 (1.16; 2.02) | 0.002 | 1354 |
| EVI_mean (per 0.1 increase) | 2.23 (1.44; 3.47) | <0.001 | 1351 |
| EVI_sd (per 0.01 increase) | 1.14 (1.03; 1.27) | 0.010 | 1357 |
| NDVI_max (per 0.1 increase) | 1.90 (1.42; 2.57) | <0.001 | 1348 |
| NDVI_mean (per 0.1 increase) | 1.85 (1.35; 2.52) | <0.001 | 1350 |
| NDVI_sd (per 0.01 increase) | 1.16 (1.03; 1.30) | 0.012 | 1357 |
| MIR_max (per 0.1 increase) | 0.68 (0.35; 1.35) | 0.258 | 1362 |
| MIR_mean (per 0.1 increase) | 0.22 (0.10; 0.51) | <0.001 | 1352 |
| MIR_sd (per 0.01 increase) | 1.02 (0.82; 1.28) | 0.829 | 1363 |
| Distance to water bodies (km) | 1.01 (1.00; 1.03) | 0.048 | 1360 |
| Distance to rivers (km) | 1.02 (0.99; 1.05) | 0.202 | 1361 |
| Population density per km2 (log-transformed) | 0.76 (0.69; 0.84) | <0.001 | 1340 |
| Slope (degree) | 1.12 (0.98; 1.29) | 0.089 | 1360 |
| Altitude (meters) | 1.00 (1.00; 1.00) | 0.937 | 1363 |
| Forest area (proportion, per 0.1 increase) | 1.06 (0.99; 1.13) | 0.081 | 1360 |
| Distance to forest (km) | 0.86 (0.76; 0.98) | 0.023 | 1358 |
| Accessibility (travel time to city per hour increase) | 1.00 (1.00; 1.00) | 0.985 | 1363 |
| Random Intercepts: | |||
| Village only | NA | 0.003 | 1362 |
| Rural community only | NA | 0.094 | 1368 |
| Village and rural | NA | <0.001 | 1361 |
NDVI: Normalized Difference Vegetation Index; EVI: Enhanced Vegetation Index; MIR: mid-infrared band; OR: Odds Ratio, CI: confidence interval; LRT: likelihood ratio test; AIC: Akaike information criterion.