Skip to main content

Table 3 Multivariate logistic regression models for associations between abuse during the life course and mental ill-health (n = 135)

From: Elder abuse and life-course victimization in hospitalized older adults in Sweden: prevalence and associations with mental ill-health

 

PHQ-9 ≥ 10

Depression or

anxiety

Anti-depressant medication

≥ 9 symptom

diagnoses

 

n

OR

95% CI

OR

95% CI

OR

95% CI

OR

95% CI

Age at abuse

 No abusive experiences

80

1

 

1

 

1

 

1

 

 Abuse only < age 65

31

1.9

0.6–6.1

2.2

0.7–6.5

3.3

0.8–14.4

0.8

0.3–2.5

 Abuse only ≥ age 65

10

0.9

0.1–10.1

1.0

0.1–9.3

3.6

0.3–44.0

0.5

0.0–4.4

 Abuse both < and ≥ age 65

14

10.4**

2.1–51.0

2.8

0.7–11.4

6.6*

1.1–39.2

4.0*

1.0–16.1

Sex

 Male

62

1

 

1

 

1

 

1

 

 Female

73

0.6

0.2–1.7

3.0*

1.1–8.7

3.2

0.7–14.1

3.2*

1.2–8.7

Age

 65–74

31

1

 

1

 

1

 

1

 

 75–84

55

1.7

0.4–6.7

0.5

0.2–1.6

1.4

0.3–6.9

1.0

0.3–3.1

  ≥85

48

2.3

0.5–10.2

0.5

0.1–2.0

0.5

0.1–3.1

0.8

0.2–2.9

ADL level

 No help

62

1

 

1

 

1

 

1

 

 I-ADL

46

2.0

0.6–6.7

0.8

0.3–2.6

3.8

0.8–17.3

3.7*

1.2–11.0

 B-ADL

25

1.1

0.3–4.9

2.4

0.7–8.5

7.8*

1.3–47.0

5.9**

1.6–21.2

Marital status

 Non-married

67

1

 

1

 

1

 

1

 

 Married

68

3.2*

1.0–9.9

0.9

0.4–2.5

2.4

0.6–9.0

1.9

0.7–5.0

  1. Area under the ROC for each respective model: PHQ-9: 0.78 (95% CI 0.68–0.88); Depression or anxiety diagnoses 0.73 (95% CI 0.62–0.83); Anti-depressant medication 0.81 (95% CI 0.71–0.92); Symptom diagnoses 0.76 (95% CI 0.67–0.86). Missing cases PHQ-9 = 31; Depression or anxiety diagnoses = 15; Anti-depressant medication n = 16; Symptom diagnoses n = 12
  2. OR Odds ratio, 95% CI 95% confidence interval, I-ADL need for support for instrumental ADL, B-ADL need for support for basic ADL
  3. * = p < 0.05,
  4. ** = p < 0.01