Skip to main content
. 2020 Oct 2;42(1):103–116. doi: 10.1093/epirev/mxaa006

Table 2.

Risk Estimates for Violence in Drug Use Disorders According to Sample or Study Characteristics, Multiple Countries, 1990–2019

Source of Heterogeneity No. of Studies No. in Population OR 95% CI
Sex
 Male 4 43,976 3.9 1.7, 8.9
 Female 4 800 2.2 1.8, 2.7
 Mixed 11 546,635 5.4 4.1, 7.0
Study location
 United States 14 542,393 4.2 2.9, 6.1
 Other high-income counties 3 48,888 7.1 4.1, 12.2
US study locations
 National 8 533,339 4.4 2.9, 6.1
 Region-based 6 9,054 3.9 2.0, 7.6
Measures of outcome
 Self-reported outcome 12 546,877 4.6 3.0, 7.2
 Others’ report/official records 3 43,962 3.2 1.3, 7.8
 Combined measures 3 572 4.4 1.3, 14.5
Temporality in cohort studies
 Drug prior to violence 4 85,105 3.8 1.6, 9.1
 Others 6 26,764 2.6 1.6, 4.3
Study design
 Cohort study 10 111,869 2.7 2.1, 3.5
 Case-control study 8 473,850 6.6 5.1, 8.6
Years of follow-up
 <9.5 years 5 67485 1.9 1.5, 2.4
 ≥9.5 years 5 44384 4.5 2.2, 9.1
sample size
 <500 8 1503 2.6 2.0, 3.4
 ≥500 8 166285 5.1 2.9, 9.2
Violent outcome
 Intimate partner violence 4 31,621 1.7 1.4, 2.1
 Intimate partner violence with general controls 2 31,325 1.8 0.8, 4.2
 General violence 14 559,790 5.7 3.8, 8.6
Clinical criteria
 DSM-III (Revision) 5 14,901 5.7 2.5, 13.0
 DSM-IV (Text Revision) 10 491,106 3.4 2.0, 6.0

Abbreviations: CI, confidence interval; DSM, Diagnostic and Statistical Manual of Mental Disorders; OR, odds ratio.