Modeling smoking and depression comorbidity in the U.S
Jamie Tam 1  
,  
Gemma Taylor 2
,  
Kara Zivin 1
,  
 
 
More details
Hide details
1
University of Michigan, Health Management and Policy, United States of America
2
University of Bristol, United Kingdom
3
University of Michigan, Epidemiology, United States of America
Publish date: 2018-03-01
 
Tob. Induc. Dis. 2018;16(Suppl 1):A369
KEYWORDS:
WCTOH
 
TOPICS:
Download abstract book (PDF)

ABSTRACT:
Background:
Evidence shows that the relationship between depression and smoking may be bi-directional, with smoking increasing the risk for subsequent depression, and vice versa—translating into significantly higher smoking rates among those with depression compared to the general population. Computational models can be used to assess trends in smoking but have not been utilized to examine populations with behavioral health comorbidities. We develop the first joint model of smoking and depression, projecting future smoking patterns under a 'status quo' scenario.

Methods:
We use a compartmental model to simulate individuals transitioning across smoking and depressive states from birth to age 99 or death. The model is calibrated to reproduce smoking and depression data from the 2005-2014 U.S. National Survey on Drug Use and Health. Current smokers report smoking within the past year and smoked at least 100 cigarettes in their lifetime while those with depression report a major depressive episode in the past year. Parameters for the effects of smoking and depression on each other were drawn from existing literature.



[Smoking and depression model diagram]



Results:

From 2015 to 2050, smoking prevalence among women with depression is projected to decline by 43.5% (27.6% to 15.6%), compared to 46.9% (17.9% to 9.5%) for women who report never having had a depressive episode. Among men with depression, smoking prevalence is expected to decline by 32.0% (34.4% to 23.4%), compared to 39.0% (24.1% to 14.7%) for men who report never being depressed. The likelihood of smoking among those with depression is projected to increase.

Conclusions:

This model can be used to examine trends and identify optimal interventions to address smoking disparities by depressive status. Smoking prevalence among people with depression is expected to decline, though not as rapidly as for those without depression. In the absence of population-level strategies, smoking and tobacco-related diseases will continue to further disproportionately burden people with depression.

eISSN:1617-9625