Table 4.
Results of logistic regression models of predictors of death per 30 days at risk of death among 8,879 psychiatric patients with a history of hospitalization
| Noninjury-related deatha | Death from injuryb | |||||||
|---|---|---|---|---|---|---|---|---|
| Variable | B | SE | p | Exp(B) | B | SE | p | Exp(B) |
| Outpatient commitment days per 30 days at risk | −.04 | .01 | <.001 | .96 | −.27 | .04 | <.001 | .76 |
| Treatment days per community care episode | .00 | .00 | .026 | 1.00 | −.02 | .01 | <.001 | .98 |
| Gender (male=2, female=1) | .57 | .08 | <.001 | 1.76 | 1.20 | .25 | <.001 | 3.33 |
| Age | .05 | .09 | .609 | 1.05 | −.17 | .23 | .461 | .84 |
| Age at first contact with the mental | health system | .01 | .09 | .881 | 1.01 | .16 | .23 | .489 | 1.18 |
| Schizophrenia | −.51 | .16 | .001 | .60 | −.06 | .38 | .875 | .94 |
| Major affective disorder | −.64 | .20 | .001 | .53 | −.11 | .44 | .806 | .90 |
| Dementia | .19 | .18 | .285 | 1.21 | −.62 | .54 | .252 | .54 |
| Paranoia and other psychosis | −.18 | .25 | .454 | .83 | .24 | .57 | .672 | 1.27 |
| Lifetime inpatient days | .00 | .00 | .095 | 1.00 | −.00 | .00 | .145 | 1.00 |
| Propensity toward hospitalization and conditional release | .60 | .21 | .005 | 1.82 | −.57 | .58 | .322 | .56 |
| Time from first date in mental health system to end of study in days | .00 | .00 | .904 | 1.00 | .00 | .00 | .421 | 1.00 |
a
When noninjury-related death was the dependent variable, χ2=1,160.06, df=12, p<.001; N for regression=8,647; correct classification=89.3 percent.
b
When death from injury was the dependent variable, χ2= 233.60, df=12, p<.001; N for regression=8,765; correct classification=98.7 percent.