Table 4.
Unadjusted and adjusted odds ratios from bivariate and multivariate logistic regression
| Variable | Numbera | Unadjusted OR (95% CI) | P value | Adjusted OR (95% CI) | P value | |
|---|---|---|---|---|---|---|
| Case N = 98 | Control N = 634 | |||||
| Age group (years) | ||||||
| <5 | 9 | 100 | 1 | – | 1 | – |
| 5–14 | 23 | 210 | 1.13 (1.00–1.28) | 0.06 | 1.48 (0.65–3.36) | 0.35 |
| 15–29 | 37 | 150 | 1.22 (1.07–1.38) | 0.002 | 2.14 (0.99–4.63) | 0.05 |
| 30–44 | 14 | 73 | 1.05 (0.89–1.26) | 0.52 | 2.23 (0.87–5.72) | 0.10 |
| 45–59 | 8 | 38 | 1.05 (0.80–1.37) | 0.74 | 3.03 (1.11–8.27) | 0.03 |
| 60+ | 7 | 63 | 0.90 (0.77–1.05) | 0.18 | 1.24 (0.55–2.80) | 0.61 |
| Location slept previous night | ||||||
| In household | 87 | 624 | 1 | – | 1 | – |
| Away from household | 11 | 10 | 2.46 (1.35–4.48) | 0.003 | 5.61 (1.97–16.02) | 0.001 |
| Higher socioeconomic statusb | 98 | 634 | 0.78 (0.62–0.98) | 0.03 | 0.65 (0.47–0.91) | 0.01 |
| Squared term | – | – | 1.10 (1.00–1.19) | 0.03 | 1.13 (1.01–1.27) | 0.04 |
| Predicted travel time to clinic | ||||||
| 0–4 | 52 | 392 | 1 | – | 1 | – |
| 5–14 | 43 | 120 | 2.51 (1.34–4.70) | 0.004 | 2.07 (0.86–5.01) | 0.11 |
| 15–46 | 3 | 122 | 0.58 (0.15–2.33) | 0.44 | 0.05 (0.005–0.59) | 0.02 |
| More than 15 km from Angolan border | 23 | 298 | 1 | – | 1 | – |
| <15 km from Angolan border | 75 | 336 | 4.34 (1.91–9.85) | <0.0001 | 2.86 (1.17–6.97) | 0.02 |
| Enhanced Vegetation Index (EVI)b | ||||||
| 0.11–0.24 | 37 | 443 | 1 | – | 1 | – |
| 0.25–0.34 | 54 | 145 | 5.32 (2.49–11.38) | <0.0001 | 14.62 (3.73–57.27) | <0.0001 |
| 0.35–0.45 | 7 | 46 | 1.70 (0.43–6.71) | <0.0001 | 2.00 (0.23–17.20) | 0.53 |
| Total rainfall in prior month (mm) | ||||||
| 0–19 | 14 | 207 | 1 | – | 1 | – |
| 20–39 | 63 | 171 | 7.25 (3.33–15.79) | <0.0001 | 2.79 (1.01–7.70) | 0.05 |
| 40–67 | 21 | 256 | 2.14 (0.87–5.31) | 0.10 | 0.25 (0.06–0.99) | 0.05 |
| District: Engela | 51 | 277 | 1 | – | 1 | – |
| Oshikukuc | 21 | 74 | 1.54 (0.62–3.81) | 0.94 | 5.06 (1.35–18.97) | 0.02 |
| Outapi | 26 | 283 | 0.50 (0.21–1.16) | 0.11 | 1.13 (0.45–2.87) | 0.79 |
All GEE adjusted for health district (matching variable) and clustering of controls within households. Multivariate GRR adjusted for the interaction effect of gender and travel in Table 3
OR odds ratio, GEE generalized estimating equations, CI confidence interval, QIC quasilikelihood under the independence model criterion, m meters, °C degrees Celsius, mm millimeters, km kilometers
aNumbers restricted to non-missing data in final multivariate model; full bivariate analyses included in Table S3
bSocioeconomic measure is first component of the PCA, included as a continuous measure
cThe higher adjusted odds of malaria observed in Oshikuku is attributed to a lower number of controls recruited per case in this area compared to the other districts