Cannabis addiction was a stable predictor of suicidal thoughts, planning, and attempts across 12 years of national survey data

Addictions to alcohol, pain relievers, marijuana, and cocaine were stable and reliable predictors of all three STB outcomes (ideation, planning, attempt) across 12 years. Cocaine was the exception, showing instability for suicide attempt prediction. The selected four-substance model had greater predictive accuracy than demographic factors alone or the remaining seven substances. Alcohol addiction alone had comparable predictive accuracy to all other ten addictions combined.

Analysis of National Survey on Drug Use and Health (NSDUH) data from 2008-2020 examining eleven substance addictions as predictors of suicidal thoughts and behaviors. Stability of predictions tested across multiple survey years.

Identifying which substance addictions most reliably predict suicidality helps prioritize clinical screening. Cannabis addiction being among the four most stable predictors elevates its importance in suicide risk assessment, alongside the better-recognized risks of alcohol and opioids.

The stability of marijuana addiction as a suicidality predictor across 12 years of data, spanning periods of changing legalization and social norms, suggests this is a robust clinical relationship that should inform suicide risk assessment protocols.

Cross-sectional survey data collected annually; cannot determine temporal ordering. Self-reported substance use and suicidality. "Addiction" defined by survey criteria, not clinical diagnosis. Association does not prove causation.