Smoking May Harm Mental Health

Research has documented a strong link between smoking and various mental health disorders, including anxiety, depression, schizophrenia and bipolar disorder.  The CDC advises that about 20% of American adults had some form of mental illness in 2009-2011, and the smoking rate for that group was 36%, in contrast to 21% for all others (here).

There is certainly an association, but there are four distinct possibilities with respect to causality:

·       Mental health problems cause people to smoke.

·       Smoking causes mental health problems.

·       Both pathways exist.

·       Neither pathway exists.

My economist colleague Dr. Nantaporn Plurphanswat is the lead author of an innovative analysis that identifies a potential causal pathway for mental illness and smoking; the work appears in the American Journal of Health Behavior (abstract here).  Our co-author is University of Illinois professor Dr. Robert Kaestner.  We used data from people in almost all states participating in the federal Behavioral Risk Factor Surveillance System (BRFSS) from 2000 to 2010.  BRFSS collected information on smoking and asked participants “…for how many days during the past 30 days was your mental health [which includes stress, depression, and problems with emotions] not good?”

Recognizing that traditional approaches cannot identify a causal pathway between smoking and mental illness, Drs. Plurphanswat and Kaestner employed an instrumental variable approach, in which variation in smoking at the state level is strongly associated with cigarette excise taxes, but the excise taxes are completely unrelated to outcomes like mental health.  The IV analysis provides information on whether smoking leads to mental health problems, or vice versa.

Our results indicate that smoking may harm mental health: it is significantly associated with 14 or more days of poor mental health.  Most of the effect due to smoking is from large increases in the number of mentally unhealthy days and not by small increases among many smokers. 

The BRFSS data cannot tell us whether smoking is a form of self-medication practiced by those who suffer from specific mental health disorders.  However, our analysis confirms that smoking may contribute to anxiety, depression and emotional distress.  Thus, policies that reduce smoking may have a positive spill-over effect in improving mental health.