Associations between average crime exposure, change in crime exposure and mental health outcomes
| Total sample (n=112 251) | Stayers (n=80 958) | Past movers† (n=15 940) | Recent movers‡ (n=15 353) | |||||
|---|---|---|---|---|---|---|---|---|
Average crime exposure ( ) | Change in crime exposure ( ) | Average crime exposure ( ) | Change in crime exposure ( ) | Average crime exposure ( ) | Change in crime exposure ( ) | Average crime exposure ( ) | Change in crime exposure ( ) | |
| Self-reported mental health | ||||||||
| Model 1§ | 3.44 (2.87–4.13) | 1.05 (1.04–1.07) | 3.12 (2.50–3.90) | 1.09 (1.00–1.18) | 3.71 (2.96–4.65) | 1.02 (1.00–1.04) | 4.55 (3.36–6.16) | 1.06 (1.03–1.08) |
| Model 2¶ | 1.79 (1.60–2.00) | 1.04 (1.02–1.06) | 1.57 (1.38–1.77) | 1.03 (0.95–1.11) | 2.00 (1.64–2.45) | 1.03 (1.00–1.06) | 2.28 (1.71–3.02) | 1.05 (1.02–1.08) |
| Model 3** | 1.51 (1.35–1.68) | 1.04 (1.02–1.06) | 1.40 (1.24–1.57) | 1.02 (0.94–1.11) | 1.56 (1.27–1.91) | 1.02 (0.99–1.05) | 1.77 (1.33–2.36) | 1.04 (1.01–1.07) |
| Antidepressant prescriptions | ||||||||
| Model 1§ | 1.98 (1.85–2.12) | 1.03 (1.01–1.05) | 1.89 (1.76–2.03) | 1.05 (1.02–1.08) | 1.90 (1.65–2.20) | 1.00 (0.99–1.02) | 2.44 (2.09–2.86) | 1.03 (1.01–1.06) |
| Model 2¶ | 1.37 (1.30–1.45) | 1.02 (1.00–1.04) | 1.32 (1.24–1.40) | 1.01 (0.98–1.04) | 1.29 (1.11–1.50) | 1.00 (0.98–1.03) | 1.55 (1.32–1.83) | 1.02 (0.99–1.05) |
| Model 3** | 1.27 (1.20–1.34) | 1.01 (0.99–1.03) | 1.25 (1.17–1.34) | 1.01 (0.98–1.04) | 1.16 (0.99–1.36) | 1.00 (0.98–1.03) | 1.32 (1.12–1.56) | 1.01 (0.98–1.04) |
| Antipsychotic prescriptions | ||||||||
| Model 1§ | 3.30 (2.76–3.96) | 1.06 (1.03–1.10) | 3.16 (2.48–4.01) | 1.16 (1.04–1.29) | 2.98 (1.95–4.56) | 1.01 (0.98–1.05) | 4.70 (3.11–7.10) | 1.06 (1.03–1.09) |
| Model 2¶ | 1.42 (1.22–1.67) | 1.07 (1.02–1.12) | 1.34 (1.08–1.67) | 1.10 (0.98–1.24) | 1.14 (0.69–1.91) | 1.04 (0.97–1.11) | 2.15 (1.34–3.44) | 1.06 (1.01–1.11) |
| Model 3** | 1.25 (1.06–1.47) | 1.06 (1.01–1.12) | 1.24 (1.00–1.54) | 1.10 (0.98–1.24) | 0.94 (0.58–1.52) | 1.03 (0.97–1.10) | 1.65 (1.02–2.69) | 1.05 (0.99–1.12) |
Source: Scottish Longitudinal Study.
Note: Bold text indicates significant associations (p<0.05). Models were fitted with logistic regression applying cluster robust estimation at local authority level; estimates are expressed in OR with 95% CI. Average crime exposure is log10-transformed, change in crime exposure is standardised. Models included average and change variables at the same time.
†Main residential location changed between 2004/2006 and 2007/2009.
‡Main residential location changed between 2007/2009 and 2010/2012.
§Model 1: Controlled for sex, age and age-squared.
¶Model 2: Model 1+2001 baseline covariates (ethnicity, education, social class, employment, marital status, living status, long-term illness) and psychiatric inpatient service use in 2001/2003.
**Model 3: Model 2+2001–2011 change indicators (gained higher level of education; separated, divorced or widowed; started to live alone; became unemployed or left labour force) and social grade in 2011.