Table 3.
Multivariable Cox Hazard Proportion Analysis for the Overall Incidence of Glaucoma by Different Modeling
| Model 1 | Model 2 | Model 3 | Model 4 | |||||
|---|---|---|---|---|---|---|---|---|
| Variables | Adjusted HR (95% CI) | P Value | Adjusted HR (95% CI) | P Value | Adjusted HR (95% CI) | P Value | Adjusted HR (95% CI) | P Value |
| Group | ||||||||
| Control | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| Chronic renal disease | 2.1 (1.76–2.52) | <0.001 | 1.63 (1.34–1.98) | <0.001 | 1.61 (1.33–1.96) | <0.001 | 1.63 (1.34–1.98) | <0.001 |
| Hypertension | ||||||||
| No | 1.00 | 1.00 | 1.00 | |||||
| Yes | 1.48 (1.23–1.78) | <0.001 | 1.46 (1.22–1.75) | <0.001 | 1.48 (1.23–1.78) | <0.001 | ||
| Diabetes mellitus | ||||||||
| No | 1.00 | 1.00 | 1.00 | |||||
| Yes | 1.51 (1.25–1.83) | <0.001 | 1.51 (1.26–1.81) | <0.001 | 1.52 (1.26–1.83) | <0.001 | ||
| Hyperlipidemia | ||||||||
| No | 1.00 | 1.00 | ||||||
| Yes | 1.01 (0.83–1.24) | 0.900 | 1.00 (0.83–1.23) | 0.940 | ||||
| Stroke | ||||||||
| No | 1.00 | 1.00 | ||||||
| Yes | 0.82 (0.61–1.12) | 0.218 | 0.83 (0.61–1.13) | 0.241 | ||||
| Age group (year) | ||||||||
| <50 | 1.00 | 1.00 | 1.00 | 1.00 | ||||
| 50–59 | 2.47 (1.85–3.30) | <0.001 | 2.09 (1.56–2.80) | <0.001 | 2.09 (1.56–2.81) | <0.001 | 2.10 (1.56–2.81) | <0.001 |
| 60–69 | 4.01 (3.09–5.19) | <0.001 | 3.08 (2.35–4.04) | <0.001 | 3.07 (2.35–4.03) | <0.001 | 3.11 (2.37–4.07) | <0.001 |
| 70–79 | 4.78 (3.68–6.22) | <0.001 | 3.54 (2.68–4.69) | <0.001 | 3.49 (2.64–4.61) | <0.001 | 3.56 (2.69–4.71) | <0.001 |
| ≥80 | 2.36 (1.55–3.58) | <0.001 | 1.80 (1.17–2.76) | 0.007 | 1.75 (1.14–2.69) | 0.010 | 1.79 (1.16–2.75) | 0.008 |
| Sex | ||||||||
| Male | 1.00 | 1.00 | 1.00 | |||||
| Female | 1.16 (0.99–1.35) | 0.062 | 1.16 (0.99–1.35) | 0.065 | 1.15 (0.99–1.35) | 0.069 | ||
| Residence | ||||||||
| Seoul (metropolitan) | 1.00 | 1.00 | 1.00 | |||||
| Second area | 0.82 (0.59–1.14) | 0.230 | 0.83 (0.60–1.15) | 0.263 | 0.82 (0.59–1.14) | 0.247 | ||
| Third area | 0.66 (0.48–0.92) | 0.015 | 0.66 (0.48–0.92) | 0.015 | 0.66 (0.48–0.92) | 0.015 | ||
| Fourth area | 0.93 (0.77–1.11) | 0.425 | 0.93 (0.77–1.11) | 0.422 | 0.93 (0.77–1.11) | 0.423 | ||
| Household income | ||||||||
| 0–30% | 1.00 | |||||||
| 30–70% | 1.08 (0.87–1.34) | 0.477 | ||||||
| 70–100% | 1.03 (0.84–1.26) | 0.775 | ||||||
| Model criterion value | AUC: 0.668; AIC: 12451.1 | AUC: 0.696; AIC: 12405.6 | AUC: 0.696; AIC: 12401.7 | AUC: 0.697; AIC: 12407.7 | ||||
AUC, area under the curve; AIC, Akaike's Information Criterion; the model with the largest AUC or lowest AIC value being considered the best.
Model 1 adjusted for age and sex. Model 2 adjusted for confounding factors which were significant in univariate analysis. Model 3 adjusted for the confounding factor which were selected by the best subset selection method. It is a method that finds the lowest Akaike information criterion (AIC) value among all possible combinations of independent variables. Model 4 adjusted for all independent variables.