Table 3.
Logistic regression analysis: violent caregivers versus all others (n = 254).
| Variable | B | S.E. | Sig. | EXP(B) | 95% C.I. for EXP(B) | |
|---|---|---|---|---|---|---|
| Lower | Upper | |||||
| Patient vulnerability | ||||||
| Age | .564 | .555 | .309 | 1.758 | .593 | 5.214 |
| Gender | −.202 | .972 | .048 | .817 | .422 | .949 |
| Race | 9.569 | 27.669 | .999 | 14.274 | .437 | 8.453 |
| Number of dementia symptoms | 2.470 | 1.050 | .019 | 4.817 | 3.509 | 12.518 |
| Level of functional impairment | 1.231 | .581 | .034 | 2.049 | 1.093 | 4.912 |
| Depression | −.634 | .423 | .134 | .530 | .231 | 1.216 |
| Number of drugs taken | .463 | .538 | .390 | 1.589 | .553 | 4.563 |
| Verbal aggression | 2.129 | 1.450 | .142 | .119 | .007 | 2.841 |
| Violence | 3.185 | 1.228 | .010 | 4.168 | 2.176 | 8.399 |
| Caregiver risk | ||||||
| Age | .355 | .382 | .352 | 1.426 | .675 | 3.013 |
| Sex | −1.497 | 1.471 | .309 | .224 | .013 | 3.998 |
| Race | −8.897 | 27.669 | .999 | 1.010 | .653 | 8.976 |
| Hassle experienced by caregiving | .173 | .234 | .460 | 1.189 | .751 | 1.880 |
| Level of impairment sub-scale | −14.523 | 54.843 | .996 | 4.962 | .655 | 9.877 |
| Cognitive status | .953 | .800 | .233 | 2.594 | .541 | 12.435 |
| Level of social support | .023 | .418 | .956 | 1.023 | .451 | 2.322 |
| Psychiatric symptoms | .249 | .486 | .608 | 1.283 | .495 | 3.326 |
| Depression | .872 | .535 | .103 | 2.391 | .837 | 6.825 |
| Self esteem | −.472 | .514 | .046 | .662 | .591 | .748 |
| Alcoholism | 3.562 | 2.308 | .041 | 3.217 | 2.382 | 4.775 |
| Coping style | −17.491 | 33.673 | .999 | 2.534 | .544 | 5.436 |
| Interaction of risk and vulnerability | ||||||
| Pt dementia ∗ cg depression | 1.728 | .408 | .007 | 5.483 | 3.217 | 7.075 |
| Pt depression ∗ cg hassle ∗ Cg psychiatric symptoms | .162 | .425 | .052 | 6.176 | 4.511 | 9.706 |
| Constant | −38.572 | 26.066 | .139 | .000 | ||
(a) Step wise logistic regression model: variables were considered significant risk factors if they were found significant on step one and if they remained significant when tested against all other significant variables in the model onstep two. All variables found significant on step 1 remained significant at step 2.
(b) Model accounts for 48.3% of the variance. Nagelkerke R square = .483.
(c) Homer-Lemeshow goodness of fit test (X 2 = 9.628, df = 8, P = .365).