Table 3.
Multilevel mixed-effects logistic regression models estimating the association between malaria camps and RDT+ and/or PCR+ malaria infections.
| Malaria by RDT only | Malaria by PCR only | Malaria by RDT or PCR | ||||
|---|---|---|---|---|---|---|
| Outcome | Crude OR (95% CI) | Adjusted OR (95% CI)1 | Crude OR (95% CI) | Adjusted OR (95% CI)1 | Crude OR (95% CI) | Adjusted OR (95% CI)1 |
|
| ||||||
| Malaria at baseline by study arm | ||||||
| Arm B Delayed MC | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
| Arm A New MC | 2.19 (0.15, 32.47) | 2.27 (0.18, 29.35) | 4.81 (0.98, 23.76) | 4.96 (1.04, 23.58) | 4.49 (0.96, 20.98) | 4.64 (1.03, 21.00) |
| Arm C Old MC | 0.34 (0.01, 15.32) | 0.36 (0.01, 12.56) | 0.49 (0.06, 3.72) | 0.46 (0.06, 3.37) | 0.48 (0.07, 3.48) | 0.46 (0.07, 3.19) |
| Malaria by Follow-up | ||||||
| Baseline | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
| Follow-up 3 | 1.10 (0.45, 2.67) | 1.10 (0.47, 2.60) | 0.39 (0.23, 0.64) | 0.38 (0.23, 0.63) | 0.46 (0.28, 0.74) | 0.45 (0.28, 0.72) |
| Malaria at follow-up 3 by study arm | ||||||
| Arm B Delayed MC | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
| Arm A New MC | 0.14 (0.03, 0.62) | 0.17 (0.04, 0.70) | 0.35 (0.17, 0.73) | 0.36 (0.17, 0.73) | 0.32 (0.16, 0.65) | 0.33 (0.16, 0.65) |
| Arm C Old MC | --2 | --2 | 0.67 (0.19, 2.43) | 0.70 (0.20, 2.50) | 0.57 (0.17, 2.07) | 0.61 (0.18, 2.12) |
| Model fit statistics | ||||||
| Log likelihood | −273.48 | −266.64 | −924.31 | −902.74 | −945.57 | −924.48 |
| Model p-value | 0.033 | 0.028 | <0.001 | <0.001 | <0.001 | <0.001 |
| Akaike’s Information Criteria (AIC) | 560.96 | 557.28 | 1864.61 | 1831.48 | 1883.43 | 1874.96 |
| Bayesian Information Criteria (BIC) | 605.31 | 633.18 | 1915.92 | 1914.71 | 1934.66 | 1958.31 |
RDT = rapid diagnostic test; PCR = polymerase chain reaction; OR = odds ratio; CI = confidence interval; MC = malaria camp
1
Adjusted for age, gender, education, and occupation
2
No RDT malaria at follow-up 3; ORs small and unstable