There is uncertainty about factors associated with involuntary in-patient psychiatric care. Understanding these factors would help in reducing coercion in psychiatry.

To explore variables associated with involuntary care in the largest database of involuntary admissions published.

We identified 166 102 public mental health hospital admissions over 5 years in New South Wales, Australia. Demographic, clinical and episode-of-care variables were examined in an exploratory, multivariable logistic regression.

A total of 54% of eligible admissions included involuntary care. The strongest associations with involuntary care were referral from the legal system (odds ratio 4.98, 95% CI 4.61–5.38), and psychosis (odds ratio 4.48, 95% CI 4.31–4.64) or organic mental disorder (odds ratio 4.40, 95% CI 3.85–5.03). There were moderately strong associations between involuntary treatment and substance use disorder (odds ratio 2.68, 95% CI 2.56–2.81) or affective disorder (odds ratio 2.06, 95% CI 1.99–2.14); comorbid cannabis and amphetamine use disorders (odds ratio 1.65, 95% CI 1.57–1.74); unmarried status (odds ratio 1.62, 95% CI 1.49–1.76) and being born in Asia (odds ratio 1.42, 95% CI 1.35–1.50), Africa or the Middle East (odds ratio 1.32, 95% CI 1.24–1.40). Involuntary care was less likely for people aged >75 years (odds ratio 0.68, 95% CI 0.62–0.74), with comorbid personality disorder (odds ratio 0.90, 95% CI 0.87–0.94) or with private health insurance (odds ratio 0.89, 95% CI 0.86–0.93).

This research strengthens the evidence linking diagnostic, socioeconomic and cultural factors to involuntary treatment. Targeted interventions are needed to reduce involuntary admissions in disadvantaged groups.