Table 3.
Multivariate analysis of risk factors
| Plasmodium spp. | Sampovloun* | Koh Kong | Preah Vihear | ||||||
| Variables at individual level : (tested/reference) | |||||||||
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| gender (male/female) | 2.31 | 1.56–3.41 | < 0.001 | 1.69 | 1.19–2.38 | 0.003 | 1.31 | 1.10–1.57 | 0.003 |
| age (0–5 years/≥ 40 years) | 0.4 | 0.13–1.23 | 0.109 | 1.53 | 0.59–3.95 | 0.377 | 1.41 | 0.96–2.07 | 0.084 |
| age (5–14 years/> 40 years) | 0.6 | 0.32–1.14 | 0.118 | 3.46 | 1.96–6.11 | < 0.001 | 1.85 | 1.30–2.65 | 0.001 |
| age (15–39 years/> 40 years) | 1.97 | 1.14–3.41 | 0.016 | 2.68 | 1.53–4.67 | 0.001 | 1.63 | 1.14–2.33 | 0.008 |
| bednet (used/not used) | 0.65 | 0.26–1.65 | 0.364 | 1.79 | 0.54–5.95 | 0.341 | 0.6 | 0.46–0.77 | < 0.001 |
| Variables at village level : (increasing distances) | |||||||||
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| distance to forest | 1.21 | 0.85–1.74 | 0.146 | 0.8 | 0.59–1.07 | 0.065 | 0.82 | 0.72–0.94 | 0.003 |
| distance to health center | 0.86 | 0.70–1.07 | 0.085 | 1.05 | 1.04–1.07 | < 0.001 | 1.03 | 0.99–1.08 | 0.076 |
| * Sex-age interaction was observed upon analysis with two ageclasses (tested/reference : age ≥ 15 years/< 15 years) sex × age: OR = 4.37, 95% CI = 1.82–10.47, p = 0.001, likelihood-ratio test A: LR = 11.08, p < 0.001 | |||||||||
| Plasmodium falciparum | Sampovloun | Koh Kong | Preah Vihear | ||||||
| Variables at individual level : (tested/reference) | |||||||||
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| gender (male/female) | 2.75 | 1.61–4.73 | < 0.001 | 1.67 | 1.13–2.46 | 0.01 | 1.43 | 1.17–1.75 | < 0.001 |
| age (0–5 years/> 40 years) | 0.3 | 0.07–1.40 | 0.127 | 1.03 | 0.32–3.31 | 0.965 | 1.19 | 0.78–1.82 | 0.425 |
| age (5–14 years/> 40 years) | 0.3 | 0.12–0.75 | 0.01 | 2.86 | 1.52–5.38 | 0.001 | 1.7 | 1.16–2.51 | 0.007 |
| age (15–39 years/> 40 years) | 1.77 | 0.89–3.50 | 0.101 | 2.51 | 1.36–4.64 | 0.003 | 1.48 | 1.00–2.18 | 0.049 |
| bednet (used/not used) | 0.37 | 0.13–1.05 | 0.06 | 1.79 | 0.47–6.76 | 0.393 | 0.56 | 0.42–0.73 | < 0.001 |
| Variables at village level : (increasing distances) | |||||||||
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| distance to forest | 1.2 | 0.80–1.81 | 0.191 | 0.85 | 0.63–1.15 | 0.15 | 0.82 | 0.68–0.99 | 0.018 |
| distance to health center | 0.87 | 0.69–1.10 | 0.121 | 1.05 | 1.03–1.06 | < 0.001 | 1.05 | 0.99–1.12 | 0.049 |
| Plasmodium vivax | Sampovloun | Koh Kong | Preah Vihear | ||||||
| Variables at individual level : (tested/reference) | |||||||||
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| gender (male/female) | 1.94 | 1.15–3.28 | 0.013 | 1.29 | 0.75–2.21 | 0.357 | 1.06 | 0.76–1.48 | 0.725 |
| age (0–5 years/> 40 years) | 0.91 | 0.22–3.74 | 0.899 | 2.13 | 0.49–9.27 | 0.312 | 2.84 | 1.24–6.49 | 0.014 |
| age (5–14 years/> 40 years) | 1.36 | 0.53–3.46 | 0.521 | 3.8 | 1.42–10.14 | 0.008 | 2.77 | 1.26–6.13 | 0.012 |
| age (15–39 years/> 40 years) | 2.58 | 1.07–6.21 | 0.035 | 2.38 | 0.89–6.41 | 0.086 | 2.17 | 0.97–4.82 | 0.058 |
| bednet (used/not used) | 2.42 | 0.32–18.51 | 0.394 | 1.92 | 0.25–15.08 | 0.534 | 0.77 | 0.47–1.25 | 0.284 |
| Variables at village level : (increasing distances) | |||||||||
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| distance to forest | 1.16 | 0.81–1.66 | 0.212 | 0.84 | 0.64–1.10 | 0.103 | 0.88 | 0.79–0.97 | 0.005 |
| distance to health center | 0.95 | 0.77–1.17 | 0.299 | 1.04 | 1.03–1.05 | < 0.001 | 1.00 | 0.97–1.03 | 0.435 |
The multivariate logistic regression model evaluated potential risk factors of Plasmodium spp., Plasmodium falciparum and Plasmodium vivax infections at two levels: gender, age and bednet use at the individual level (interview of surveyed individuals), distances to evergreen forest and to public health centers at the village level (spatial analysis). For the individual level, odds ratios (OR), 95% confidence intervals (95% CI) and p values are given for the tested versus the reference value of each variable. For the village level, odds ratios (OR), 95% confidence intervals (95% CI) and p values were calculated for increasing distances (in km).