This study aimed to explore prediction of diagnosis of mental illness in suicide attempters in Singapore, using available variables. Three years of medical records of 462 suicide attempters were analyzed. As hypothesized, the full model with available risk and protective factors and features of the attempt was significant in predicting cases with and without diagnosis of mental illness and accurately classi- fied 79% of suicide attempters with diagnosis of mental illness. The results offer preliminary evidence to support the current usage of brief assessment at the emergency department. Together with the mental status examination that was conducted routinely, the psychosocial assessment using a brief semistructured interview was supported by the results. The strongest predictor of diagnosis of mental illness among suicide attempters was mental illness or suicide in

family, followed by unemployment, willingness to seek help, habitual poor coping, alcohol or drug abuse, and lastly, lack of positive future planning. The findings were consistent with previous literature associating suicide risk with mental illness or suicide in the family [37], unemployment, and alcohol or drug abuse [3]. Poor coping and lack of positive future planning could be associated with poor problem solving and reduced social support, in suicide attempters with mental illness [48].

The risk factors for mental illness in suicide attempters included psychosocial variables which often co-occurred in the psychiatric population [32], e.g., depressed individuals often suffered from impairment of planning and problem solving [48].These variables might reflect the adverse impact of untreated debilitating mental illness which in turn have impact on employment and coping and also exacerbate the suicide risk. Patterns of co-occurring risk factors had been observed in suicides, e.g.,mental illnesswith substance abuse, financial problems, relationship problems, recent crisis, and medical problems [3]; these multiple factors challenged the individuals’ coping capacity and further increased the risk for repeated suicide attempts [29]. It is desirable to have multilevel modelling analysis to identify different layers of factors contributing to diagnosis of mental illness in suicide attempters, which take into account the social cultural environment, and other contextual factors, that might have impact on suicide attempters, e.g., socioeconomic status [49], and whether the attempters had received treatment and/or are currently receiving treatment for mental illness.