Suicidal ideation risk among LGB Spanish university students: The role of childhood and adolescence adversities and mental disorders.

BACKGROUND: Childhood/adolescence adversities and mental disorders are higher among LGB youths. AIMS: To evaluate the role of childhood maltreatment, bullying, and mental disorders on the association between sexual orientation and suicidal ideation (SI); and the role of mental disorders on the association between sexual orientation discrimination and SI. METHODS: Baseline and 12-month follow-up online surveys of Spanish first-year university students (18-24-year-olds). Multivariable logistic regression models assessed the effects of childhood/adolescence adversities and mental disorders in the relationship between sexual orientation, discrimination and SI. RESULTS: A total of 1224 students were included (16.4 % LGBs). Risk factors of lifetime SI were sexual orientation (OR 2.4), any bullying (OR 2.4), any childhood maltreatment (OR 4.0), and any mental disorders (OR 3.8). Final model Area Under the Curve (AUC) 0.78. Among homosexual and bisexual students, discrimination showed increased risk of 12-month SI (OR 2.2), but this effect was no longer statistically significant when any 12-month mental disorder was added (OR 7.8). Final model AUC 0.72. LIMITATIONS: Sample of interest was relatively small. But it was similar to comparable studies and statistical adjustments have been performed. Assessment of mental disorders and SI was not based on clinical assessment. However, validated scales showing good diagnostic agreement with clinical judgement were used. CONCLUSIONS: Childhood/adolescence adversities and mental disorders interact in the association between sexual orientation and SI. Mental disorders may mediate the association between sexual orientation discrimination and SI. Further research using larger samples and causal modelling approach assessing the mediators of SI risk among LGBs is needed.

Partial likelihood estimation of IRT models with censored lifetime data: an application to mental disorders in the ESEMeD surveys.

Developmental studies of mental disorders based on epidemiological data are often based on cross-sectional retrospective surveys. Under such designs, observations are right-censored, causing underestimation of lifetime prevalences and correlations, and inducing bias in latent trait models on the observations. In this paper we propose a Partial Likelihood (PL) method to estimate unbiased IRT models of lifetime predisposition to develop a certain outcome. A two-step estimation procedure corrects the IRT likelihood of outcome appearance with a function depending on (a) projected outcome frequencies at the end of the risk period, and (b) outcome censoring status at the time of the observation. Simulation results showed that the PL method yielded good recovery of true frequencies and intercepts. Slopes were best estimated when events were sufficiently correlated. When PL is applied to lifetime mental health disorders (assessed in the ESEMeD project surveys), estimated univariate prevalences were, on average, 1.4 times above raw estimates, and 2.06 higher in the case of bivariate prevalences.