Table 3.
Association of total hospitalization events, hospitalized days, and time to first hospitalization with depressive affect for incident hemodialysis patients in 2006
| Dependent Variable per Model | Rate/Hazard Ratio for Level 2 versus Level 1 (95% CL) | P Value |
|---|---|---|
| Poisson regression with logarithmic link function and offset with exposure time | ||
| Total hospitalization events | ||
| Model 1 | 1.25 (1.16 to 1.35) | <0.001 |
| Model 2 | 1.10 (1.01 to 1.21) | 0.03 |
| Model 3 | 1.13 (1.02 to 1.25) | 0.02 |
| Total hospitalized days | ||
| Model 1 | 1.31 (1.2 to 1.43) | <0.001 |
| Model 2 | 1.15 (1.03 to 1.27) | 0.01 |
| Model 3 | 1.20 (1.07 to 1.35) | 0.002 |
| Cox proportional hazards regression for time to first hospitalization | ||
| Time to first hospitalization | ||
| Model 1 | 1.18 (1.12 to 1.25) | <0.001 |
| Model 2 | 1.10 (1.03 to 1.18) | 0.003 |
| Model 3 | 1.12 (1.04 to 1.20) | 0.003 |
Rate ratios are given for Poisson regression; hazard ratios are given for Cox regression. Category for depressive affect: level 1=less likely depressed, score 1–2; level 2=potentially depressed, score>2–6. Model 1 was unadjusted. Model 2 was adjusted for physical composite score, mental composite score with mental health items removed, and three remaining mental health items. Model 3 was additionally adjusted for hemodialysis vintage≤90 days (incidence) versus >90 days (prevalence), age, sex, race, diebetes, albumin, creatinine, hemoglobin, calcium, phosphorus, and transferrin saturation. 95% CL, 95% confidence limit.