The temporal relationship between smoking and suicidality is not yet clear. In order to clarify this relationship, we examined prospectively bi-directional associations between smoking and suicidality and their temporal ordering of onset.

A representative community sample of 2548 young adults aged 14-26 years at baseline was folllowed up over a period of 4 years. Smoking (occasional and regular), nicotine dependence, suicidal ideation and suicide attempts were assessed using the standardized Munich-Composite International Diagnostic Interview (M-CIDI).

Suicide ideation and suicide attempts were strongly associated with occasional and regular smoking and nicotine dependence at baseline (Odds ratios [OR] range from 1.4 to 16.4). In the prospective analyses, prior occasional, regular smoking and nicotine dependence increased the risk for new onset of suicide ideation (OR range from 1.5 to 2.7) and prior regular smoking and nicotine dependence increased also the risk for onset of suicide attempt(s) (OR range between 3.1 and 4.5). Pre-existing suicidality could not be shown to be associated with subsequent smoking or nicotine dependence. Associations remained stable when participants who fulfilled DSM-IV-criteria for major depression were excluded.

The presence of associations between prior smoking and subsequent suicidality, in concert with the lack of associations between prior suicidality and subsequent smoking suggests the existence of a specific, causal pathway from smoking to suicidality.