Key Takeaways
Extended Takeaways
- Baseline PHQ-9 item 9 showed a strong dose-response for suicide attempts in the multivariable model: HR=39.58, 95% CI, 14.03–111.64, P<.001 for “nearly every day,” HR= 7.67, 95% CI, 2.32–25.35, P<.001 for “more than half of the days,” and HR =3.17, 95% CI, 1.42–7.07, P=.005 for “several days.”
- For psychiatric hospitalization, broad symptom burden did not add independent predictive value once suicidal ideation was modeled; in multivariable analysis, only nearly every day ideation remained significant (HR = 8.11, 95% CI, 3.10–21.18, P < .001), while depression, anxiety, sleep, and life satisfaction lost significance.
- PHQ-9 item 9 worked better as a rule-out than a rule-in tool in this help-seeking sample, with sensitivity of 71.7%, specificity of 77.8%, PPV of 12.9%, and NPV of 98.3%. Clinicians should avoid using it as a standalone trigger for high-risk decisions and pair it with a fuller suicide risk assessment.
- Sleep disturbance remained independently associated with suicide attempts even after adjustment, with multivariable HR=17.39, 95% CI, 2.05–147.46, P=.009 for severe sleep, HR= 12.10, 95% CI, 1.55–94.24, P=.017 for moderate sleep, and HR =9.96, 95% CI, 1.23–80.43, P=.031 for mild sleep.
- Male sex was associated with higher suicide attempt risk after adjustment (HR=2.08, 95% CI, 1.01–4.26, P=.046) and in IPW models (HR =2.71, 95% CI, 1.27–5.76, P=.010). Because this differs from typical epidemiologic patterns for suicide attempts, it may be most useful as a triage prompt in distressed health care workers rather than a generalizable assumption.
- Attrition was substantial, but the main findings were stable after inverse probability weighting: after IPW, all standardized differences fell below 0.1 (range: 0.001–0.095), and nearly every day ideation and moderate to severe sleep problems remained significant. This supports using baseline risk stratification early, even when longitudinal follow-up is difficult.
