Long-term predictors of suicide after concussion*
| Predictor | RR (95% CI) | |
|---|---|---|
| Univariable analysis† | Multivariable analysis‡ | |
| Weekend concussion | 1.36 (1.14–1.64) | 1.27 (1.06–1.53) |
| Age, per yr older | 1.00 (1.00–1.00) | NA |
| Sex, male | 2.28 (1.92–2.70) | 2.47 (2.08–2.94) |
| Income, low | 1.68 (1.33–2.13) | 1.30 (1.02–1.65) |
| Home, rural | 0.97 (0.78–1.20) | NA |
| Enrolment, recent§ | 0.96 (0.95–0.98) | 0.96 (0.94–0.97) |
| Imaging¶ | 1.38 (1.15–1.66) | 1.31 (1.09–1.57) |
| Fracture | 1.09 (0.59–2.04) | NA |
| Schizophrenia | 10.78 (7.89–14.73) | 2.38 (1.68–3.37) |
| Depression | 4.32 (3.49–5.35) | 1.65 (1.30–2.11) |
| Bipolar disorder | 7.03 (5.31–9.32) | 1.96 (1.43–2.68) |
| Substance abuse | 8.43 (7.04–10.10) | 3.60 (2.94–4.41) |
| Anxiety disorder | 3.98 (3.42–4.64) | 3.04 (2.57–3.60) |
| Prior hospital admission | 2.28 (1.53–3.40) | 1.40 (0.89–2.21) |
| Prior suicide attempt | 38.83 (23.94–62.99) | 5.65 (3.26–9.81) |
Note: CI = confidence interval, NA = not applicable, RR = relative risk.
↵* Results from proportional hazards analysis, with weekdays defined as referrant.
↵† Basic comparison with no adjustments for other baseline differences.
↵‡ Adjusted comparison accounting for other baseline differences in demographic characteristics, psychiatric diagnoses and history of suicide attempts.
↵§ “Recent” denotes years 2002–2012 (years 1992–2001 denoted as “remote”).
↵¶ Includes skull radiography, computed tomography and magnetic resonance imaging.