Abstract

Background Depression severity is a well-established risk factor for suicidal ideation, but the extent to which sociodemographic and employment-related factors contribute independently remains unclear. Aims Complete data from doctors (N = 4055) presenting to National Health Service Practitioner Health (NHS-PH) in 2022–2023 were used to test the hypothesis that depression severity is the largest determinant of suicide ideation risk (defined by Patient Health Questionnaire 9 (PHQ-9) question 9 score) among doctors. Method Using PHQ-8 score (PHQ-9, excluding the item on suicide ideation) as a proxy for depression severity, the case–control discriminatory capacity of receiver operating characteristic curves (AUCs) were evaluated for (a) a univariable model studying modified PHQ-9 alone as the predictor of severe suicide ideation; and (b) a multivariable model integrating modified PHQ-9 and multiple sociodemographic and employment factors as the predictor of severe suicide ideation. Models were compared both descriptively and through a likelihood ratio test. Results The univariable model using depression severity alone as the predictor of severe suicide ideation yielded an AUC of 0.921. The addition of sociodemographic and employment factors improved the fit significantly (likelihood ratio test with (χ2(14) = 50.26, P < 0.001), amended AUC 0.930). Having both a disability and a relationship status of ‘no partner’ was significantly independently associated with suicide ideation in the multivariable model. Conclusions In this national cohort of doctors, depression severity was strongly associated with suicidal ideation. However, disability and lack of a partner were also independently linked to increased risk, suggesting that suicidal ideation is not solely driven by symptom severity. Social and functional factors may help identify higher-risk individuals and inform targeted support.