Table 2. Pooled Bayesian meta-analysis inference for each exponentiated effect (i.e., RR interpretation).
| Effect | Estimate | 95% credible interval |
|---|---|---|
| Distance between localities (50 km) | 1.06 | (1.03, 1.08) |
| Same locality (yes versus no) | 0.53 | (0.40, 0.68) |
| Date of diagnosis distance (1/2 year) | 1.03 | (1.01, 1.07) |
| Age difference (10 years) | 1.00 | (1.00, 1.01) |
| Age (10 years) | 0.99 | (0.97, 1.01) |
| Household contacts (1 person) | 0.99 | (0.97, 1.01) |
| Sex | ||
| Mixed pair versus both female | 0.96 | (0.91, 1.02) |
| Both male versus both female | 0.91 | (0.82, 1.01) |
| Residence location | ||
| Mixed pair versus both not urban | 1.02 | (0.97, 1.09) |
| Both urban versus both not urban | 1.06 | (0.94, 1.21) |
| Housing | ||
| Mixed pair versus both not homeless | 1.02 | (0.87, 1.20) |
| Both homeless versus both not homeless | 1.11 | (0.83, 1.48) |
| Working status | ||
| Mixed pair versus both unemployed | 1.03 | (0.96, 1.12) |
| Both employed versus both unemployed | 1.13 | (0.96, 1.33) |
| Education | ||
| Mixed pair versus both ≥ secondary | 0.98 | (0.94, 1.02) |
| Both < secondary versus both ≥ secondary | 0.95 | (0.87, 1.03) |
Posterior means and 95% quantile-based credible intervals are presented.
RR, risk ratio.