This study has a number of significant strengths, including its longitudinal design, the large clinical sample recruited from two sites (university and private psychiatric hospitals), and the use of survival analyses that made use of time-varying covariates to take into account changes in the magnitude and direction of predictor variables over time. The examination of parental mental health histories and, specifically, the mental health histories of mothers and fathers is also a unique strength of this study; to our knowledge, this has not been investigated previously in a large clinical sample of suicidal adolescents. Brent et al. (1993) examined family history of psychiatric condition (suicidal behavior, substance abuse, affective disorder) as a predictor of suicide attempts. However, they had a much smaller sample size (13 adolescents had made a suicide attempt prior to follow-up) and examined only univariate associations.
