Latent class analysis of depressive symptoms and associations with suicidal thoughts, plans, and attempts among a large national sample.

Background: Depression is strongly associated with risk for suicidal behaviors. However, depression is a highly heterogeneous condition (i.e. there are more than 200 combinations of DSM-5-TR depressive symptoms to correspond to a depression diagnosis). Limited research to date has taken an empirical approach to see how people cluster together based on their classification of depressive symptoms and whether people in certain classes are more likely to report suicide outcomes than other classes. This analysis leverages the National Survey on Drug Use and Health and examines classes of depressive symptoms to explore differences in suicide-related outcomes by class among adults endorsing depressive symptoms (n = 41 969).

Methods: We used latent class analysis (LCA) to identify classes of individuals' DSM-5 depressive symptoms presentation and then explored differences in suicide-related outcomes (i.e. suicide plans, suicide attempts) by the resulting classes.

Results: A four-class model was determined to optimize the fit criteria. Class 3 (high depressive symptoms) had significantly greater rates of suicide-related outcomes, followed by class 1 (high depressed mood and moderate worthlessness), with classes 4 and 2 having significantly lower rates of suicide-related outcomes.

Conclusions: The use of LCA provided valuable findings on the importance of leveraging both a multi-faceted assessment of depressive symptoms to identify cases where a high number of depressive symptoms are endorsed, and review of the specific symptoms endorsed. Worthlessness, in particular, may be of particular value to focus on within the context of suicide prevention.