Table 2. Results of statistical model estimating effects of geographic barriers on CHW consultation rates.
Incidence rate ratios (IRR) of variables included in the model. Following Wasserstein et al. [48], we have chosen to provide 95% confidence intervals and not to dichotomize these continuous measures into thresholds of significance (e.g. p-values). CHW: community health worker, PHC: primary health center.
| Mean IRR | 95% CI | |
|---|---|---|
| Geographic Variables | ||
| CHW Geographic Dispersion (km) | 0.719 | 0.638–0.811 |
| Distance to PHC (km) | 1.019 | 0.988–1.052 |
| Confounding Variables | ||
| Wealth Score | 0.686 | 0.533–0.882 |
| CHW Age (year) | 1.000 | 0.991–1.015 |
| CHW Gender- All Men (compared to all women) |
1.428 | 1.062–1.919 |
| CHW Gender—Mixed (compared to all women) |
1.514 | 1.115–1.984 |
| Proactive Care | 1.471 | 1.297–1.668 |
| Model Structure | ||
| Zero-inflation Intercept | 0.026 | 0.020–0.033 |