Subgroup analyses of the weekend effect on mortality by types of admissions
| Analysis | N | Pooled mean (95% CrI) | Posterior predictive mean (95% CrI) | I2 (95% CrI) |
| All admissions* | 18 | 1.13 (1.09, 1.18) | 1.13 (1.04, 1.22) | 0.19 (0.00, 0.74) |
| Emergency admissions | 32 | 1.11 (1.06, 1.16) | 1.11 (0.94, 1.31) | 0.44 (0.00, 0.90) |
| Elective admissions | 12 | 1.70 (1.08, 2.52) | 1.70 (0.64, 4.11) | 0.44 (0.00, 0.93) |
| Maternity admissions | 6 | 1.06 (0.89, 1.29) | 1.06 (0.75, 1.53) | 0.44 (0.00, 0.96) |
*This analysis focuses on best adjusted studies that include mixed (both emergency and elective admissions within the same study, with or without including maternity admissions); it thus differs from the main Bayesian meta-analysis (pooled mean 1.16, 1.04 to 1.23) which, in addition to studies included in this meta-analysis, also includes individual types or sub-types of admissions provided that they do not overlap with studies that cover mixed types of admissions.
CrI, credible interval; N, number of observations (estimates of the weekend effect from individual studies).