Table 4.
Model fitting results of the logistic multilevel analyses for the dichotomous variable "Did someone inform you on what to do in case of an emergency after the cataract operation?"
| Outcome variable (N = 3,871) | Model 1 | Model 2 |
| Probability (π)1 | 0.79 (0.76, 0.81) | 0.84 (0.80, 0.88) |
| OR age (18–74 = reference)1 | - | 0.64 (0.54, 0.75)* |
| OR sex (males = reference)1 | - | 1.09 (0.92, 1.28) |
| OR education (low education = reference)1 | - | 0.91 (0.76, 1.08) |
| OR health (high = reference)1 | - | 0.61 (0.72, 0.85)* |
| OR anaesthesia (injection = reference)1 | 0.98 (0.81, 1.19) | |
| Variance hospitals2 | 0.22 (0.07)* | 0.23 (0.07)* |
| Variance patients2 | 0.98 (0.02)3 | 0.98 (0.02)3 |
| ρ4 | 0.06 | 0.07 |
*p < 0.05
1OR = Odds ratio and 95% confidence intervals added in parentheses
2Standard errors added in parentheses
3By definition, individual level variance is one if the binomial distribution holds
4ρ can be interpreted as the ICC in linear multilevel analyses