Table 4.
Multivariable Logistic Regression of Elder Mistreatment Incidencea
Independent variable | OR (95% CI) | ||||
---|---|---|---|---|---|
Any mistreatment | Neglect | Emotional abuse | Physical abuse | Financial abuse | |
Age | 1.02 (0.98–1.06) | 1.17 (1.04–1.38)b | 0.97 (0.89–1.04) | 1.02 (0.94–1.10) | 1.02 (0.97–1.07) |
Female sex (reference, male) | 0.63 (0.35–1.15) | 1.19 (0.21–11.55) | 0.47 (0.18–1.21) | 0.37 (0.11–1.17) | 0.73 (0.35–1.55) |
Self-rated poor health (reference, good) | 2.86 (1.35–5.84)c | 1.73 (0.09–17.55) | 3.67 (1.15–11.15)d | 4.21 (1.14–13.70)d | 2.80 (1.16–6.38)e |
Annual household income | 0.96 (0.82–1.13) | 0.91 (0.53–1.49) | 0.90 (0.69–1.17) | 1.17 (0.87–1.61) | 0.88 (0.73–1.06) |
Race/ethnicity (reference, non-Hispanic White) | |||||
Non-Hispanic Black | 2.61 (1.16–5.70)e | 4.55 (0.43–43.04) | 1.35 (0.31–4.63) | 1.15 (0.12–6.04) | 2.80 (1.09–6.91)d |
Hispanic White | 0.74 (0.11–2.87) | 2.97 (0.02–48.62) | 0.42 (0.00–3.71) | 0.60 (0.00–5.11) | 1.33 (0.19–5.47) |
Other race/ethnicityf | 1.32 (0.20–5.15) | 5.36 (0.04–74.33) | 0.46 (0.00–4.88) | 1.02 (0.01–8.80) | 2.35 (0.34–9.62) |
Geocultural context (reference, urban) | |||||
Suburban | 0.80 (0.39–1.56) | 1.14 (0.10–9.80) | 1.03 (0.32–2.98) | 0.64 (0.15–2.12) | 0.70 (0.26–1.69) |
Rural | 0.65 (0.26–1.48) | 1.74 (0.14–18.35) | 0.69 (0.16–2.33) | 0.44 (0.04–2.14) | 0.83 (0.28–2.18) |
Lives with family (reference, lives alone) | 0.74 (0.41–1.38) | 0.91 (0.14–6.45) | 1.32 (0.47–4.24) | 0.46 (0.14–1.51) | 0.64 (0.31–1.32) |
Functional impairment | 0.88 (0.53–1.26) | 0.95 (0.48–1.56) | 0.92 (0.02–1.60) | 0.99 (0.02–1.71) | 0.99 (0.62–1.40) |
Observations, No. | 523 | 515 | 509 | 521 | 516 |
Abbreviation: OR, odds ratio.
Incidence was calculated using Firth bias reduced logistic regression. Model fit was assessed with Tjur r2, with values ranging from 0.03 to 0.19.
P = .01.
P = .004.
P = .03.
P = .02.
Other refers to American Indian Aleut or Eskimo, Asian or Pacific Islander, or self-described other race/ethnicity.