Table 2. Parameter estimates based on bivariate logistic regression models for M. perstans microfilaremia in school-aged children in Uganda (2000–2003).
Data source | Covariate | M perstans parasitemia | ||
---|---|---|---|---|
ORa | 95% CIb | P-value (AICc) | ||
School survey | Age (5–9 years) | 1.00 | ||
10–14 years | 1.57 | 1.26, 1.96 | ||
15–19 years | 2.48 | 1.77, 3.47 | <0.001 | |
Sex (female) | 1.00 | |||
Male | 0.96 | 0.81, 1.14 | 0.660 | |
Season (dry) | 1.00 | |||
Wet | 1.52 | 0.39, 5.87 | 0.546 | |
Sattelite imagery | Land surface temperature (LST) | |||
Day (wet season) | 0.79 | 0.66, 0.93 | 0.005 (3,699) | |
Day (dry season) | 0.79 | 0.66, 0.93 | 0.005 (3,699) | |
Day (annual) | 0.78 | 0.65, 0.91 | 0.003 (3,698) | |
Night (wet season) | 1.09 | 0.85, 1.40 | 0.498 (3,706) | |
Night (dry season) | 1.12 | 0.82, 1.52 | 0.473 (3,706) | |
Night (annual) | 1.10 | 0.81, 1.45 | 0.521 (3,707) | |
Diurnal_diff (wet season)* | 0.63 | 0.51, 0.78 | <0.001 (3,688) | |
Diurnal_diff (dry season) | 0.65 | 0.53, 0.79 | <0.001 (3,689) | |
Diurnal_diff (annual) | 0.63 | 0.51, 0.77 | <0.001 (3,687) | |
Normalized difference vegetation index (NDVI) | ||||
Wet season* | 1.16 | 1.09, 1.23 | <0.001 (3,683) | |
Dry season | 1.15 | 1.08, 1.21 | <0.001 (3,684) | |
Annual | 1.15 | 1.08, 1.22 | <0.001 (3,684) | |
Land cover (%) | ||||
Forest cover* | 1.06 | 1.02, 1.10 | 0.005 (3,697) | |
Open vegetation | 0.97 | 0.95, 0.99 | 0.019 (3,709) | |
Cropland | 0.98 | 0.95, 1.03 | 0.201 (3,715) | |
GIS maps | Altitude | 0.99 | 0.99, 1.00 | 0.013 |
Livestock | ||||
Cattle density* | 1.04 | 1.02, 1.06 | <0.001 | |
Sheep density | 1.07 | 0.99, 1.16 | 0.105 | |
Goat density | 0.99 | 0.98, 1.01 | 0.523 | |
Pig density | 1.02 | 0.96, 1.07 | 0.593 | |
Chicken density | 1.00 | 0.99, 1.00 | 0.673 |
*Chosen for further multivariate modeling. Models include a school-level random effect to account for clustering at school level.
aOR, odds ratios
bCI, 95% confidence interval
cAIC, Akaike information criterion.