Skip to main content
. 2017 Dec 28;11(12):e0006118. doi: 10.1371/journal.pntd.0006118

Table 3. Results of the zero-inflated negative binomial model in a bivariate analysis.

Variables Negative binomial model Zero-inflated model
Incidence Rate ratio* 95% CI P-value Odds ratio* 95% CI P-value
Population density–per km2 1.00 1.00–1.00 0.1079 1.00 1.00–1.00 0.0088
Urban population—% 0.98 0.96–1.01 0.2042 1.01 0.98–1.04 0.3586
Improved sanitation coverage—% 0.99 0.95–1.04 0.7772 1.00 0.98–1.03 0.8170
Improved handwashing coverage—% 1.01 0.96–1.05 0.8074 1.00 0.98–1.03 0.7508
Sanitation and hygiene benchmark 1.02 0.99–1.05 0.1349 1.00 0.97–1.03 0.9080
Improved water sources—% 1.03 1.00–1.05 0.0366 1.01 0.99–1.03 0.3766
Distance from center of the district to nearest hospital—km 1.03 1.00–1.07 0.0359 0.99 0.95–1.02 0.4052
Distance from the center of the district to nearest lake or river—km 0.98 0.98–0.99 0.0004 1.00 0.99–1.01 0.4195
Distance from the center of the district to nearest border of DRC or Kenya—km 0.98 0.97–0.99 < .0001 0.98 0.97–0.99 < .0001
Weighted incidence rate of the 1st order neighbor districts—per 100,000 1.01 1.00–1.02 0.0028 1.05 1.03–1.07 < .0001
Average annual rainfall during 2011–2015 (mm) 1.00 1.00–1.00 0.6943 1.00 1.00–1.00 0.0736

*Obtained by exponentiating the estimates obtained from the model

†Calculation of the benchmark score and weighted incidence rate of the 1st order neighbor districts have been described in the text