Table 3.
Plasmodium falciparum parasite prevalence (pfPR) among children 0–15 years old enrolled between October 2011 and Feb 2014 in 12 villages in north-central and northwest regions of Uganda and associations with sex and those characteristics that resulted in P < 0.05 in univariate logistic models
| Characteristics | n positive | Unadjusted | P* | Adjusted | P* | |
|---|---|---|---|---|---|---|
| Weighted pfPR % | Odds ratio (95% CI) | pfPR odds ratio (95% CI)¶ | ||||
| All subjects | 553 | 52.5 | – | – | – | – |
| Sex | ||||||
| Female | 287 | 53.3 | Ref. | |||
| Male | 266 | 51.4 | 0.93 (0.56–1.52) | 0.767 | – | – |
| Mother’s income (Ugandan shillings) | ||||||
| < 30,000 USHS | 242 | 47.9 | Ref. | |||
| ≥ 30,000 USHS | 309 | 57.7 | 1.48 (1.09–2.01) | 0.036 | – | – |
| Distance of home to water source | ||||||
| ≥ 1 km | 338 | 55.7 | Ref. | |||
| < 1 km | 215 | 49.4 | 0.78 (0.20–3.00) | 0.723 | – | – |
| Malaria prevention | ||||||
| Indoor residual spraying (IRS) sub- region | ||||||
| Not an IRS district | 496 | 75.2 | Ref. | Ref. | ||
| IRS district | 57 | 18.4 | 0.07 (0.05–0.11) | < 0.0001 | 0.06 (0.04–0.07) | < 0.0001 |
| Indoor residual spraying (IRS) in house | ||||||
| Never | 506 | 59.0 | Ref. | |||
| In the past year | 47 | 23.7 | 0.22 (0.05–0.86) | 0.061 | – | – |
| Mosquito net used last night | ||||||
| No | 391 | 48.8 | Ref. | – | ||
| Yes | 162 | 64.6 | 1.91 (0.46–7.89) | 0.395 | – | |
| Number of other children in household | ||||||
| 1 | 58 | 67.9 | Ref. | Ref. | ||
| 2 | 81 | 43.6 | 0.37 (0.22–0.60) | 0.70 (0.69–0.73) | ||
| 3 | 98 | 50.9 | 0.49 (0.35–0.68) | 0.49 (0.45–0.52) | ||
| 4 | 110 | 40.6 | 0.32 (0.21–0.49) | 0.34 (0.31–0.37) | ||
| 5 | 88 | 61.6 | 0.76 (0.57–1.00) | 0.24 (0.21–0.26) | ||
| 6+ | 118 | 54.4 | 0.56 (0.24–1.31) | < 0.0001 | 0.17 (0.15–0.19) | < 0.0001 |
| Number of children below 5 years in household | ||||||
| 0 | 130 | 55.5 | Ref. | Ref. | ||
| 1 | 268 | 58.6 | 1.14 (0.75–1.72) | 0.48 (0.30–0.76) | ||
| 2–4 | 155 | 36.5 | 0.46 (0.32–0.66) | 0.0002 | 0.23 (0.12–0.44) | 0.014 |
| Kept goat near or inside house | ||||||
| No | 160 | 60.5 | Ref. | Ref. | ||
| Yes | 393 | 49.8 | 0.65 (0.48–0.87) | 0.021 | 0.42 (0.29–0.62) | 0.0002 |
| Non-malaria fevers | ||||||
| In past 6 months before enrollment | ||||||
| No | 461 | 51.4 | Ref. | |||
| Yes | 92 | 60.6 | 1.45 (0.66–3.21) | 0.381 | ||
| Household members with malaria | ||||||
| Any child positive with malaria | ||||||
| None | 58 | 29.6 | Ref. | |||
| ≥ 1 positive | 495 | 64.1 | 4.24 (1.61–11.2) | 0.019 | ||
| Younger sibling positive with malaria | ||||||
| No | 266 | 46.0 | Ref. | Ref. | ||
| Yes | 287 | 65.1 | 2.20 (1.40–3.45) | 0.009 | 5.39 (2.94–9.90) | 0.0006 |
| Malaria fever | ||||||
| Reports a fever at enrollment | ||||||
| No | 542 | 52.2 | Ref. | |||
| Yes | 12 | 85.5 | 5.38 (2.01–14.4) | 0.010 | 4.80 (1.94–11.9) | 0.0094 |
| Fever count in past 6 months before enrollment | ||||||
| 0 | 132 | 44.7 | Ref. | |||
| 1 | 113 | 42.1 | 0.90 (0.59–1.38) | |||
| 2 | 101 | 54.7 | 1.50 (0.83–2.71) | |||
| 3 | 102 | 66.0 | 2.40 (0.82–7.05) | |||
| 4+ | 105 | 69.4 | 2.80 (1.11–7.05) | 0.0003 | ||
| In past 12 months before enrollment | ||||||
| No | 80 | 38.3 | Ref. | |||
| Yes | 462 | 55.7 | 2.02 (1.25–3.27) | 0.021 | – | |
| Lifetime malaria treatment | ||||||
| Inpatient | ||||||
| Past 12 months | 135 | 65.1 | Ref. | |||
| More than 12 months | 114 | 57.4 | 0.72 (0.25, 2.09) | |||
| Never | 304 | 46.6 | 0.47 (0.13, 1.73) | 0.027 | ||
| Outpatient | ||||||
| Past 12 months | 395 | 53.3 | Ref. | |||
| More than 12 months | 76 | 67.6 | 1.83 (1.15, 2.92) | |||
| Never | 82 | 39.6 | 0.57 (0.41, 0.80) | 0.0057 | ||
* Covariates with several levels were coded with dummy variables for the categories and P is for heterogeneity) in the univariate analysis and using trend coding in the adjusted analyses (P is for trend). ¶ Final adjusted models used forward stepwise regression starting with 14 variables with P < 0.05 in univariate models (IRS district, mother’s income, number of other children in the household, number of malaria fevers in the past 6 months, having a younger sibling with malaria, keeping a goat in the house, inpatient and outpatient treatment for malaria). Mother’s income was estimated in Ugandan shillings (30,000 Ugandan shillings are approximately equal to 10 US dollars). The survey estimates are weighted estimates that account for the differential probabilities in selecting the sample of children. Variance estimation takes the weights into account and accounts for the clustering of the sample of children at the village and household levels. The coefficient of variation of the final weights was 1.25 (defined as standard deviation/mean of the final weights)