Smoking and hospital costs during pregnancy and the first year of life.

Smoking during pregnancy has been cited as a contributor to low birth weight and increased costs. However, much of the literature has used separate data sets for smoking status, birth weight and costs. This dissertation uses smoking and birth weight data from birth certificates liked to cost data from a health maintenance organization. Cost data include only hospital costs and are for the time period of nine months prior to delivery until six months post delivery for the mother and for the first year of life for the baby. The study population numbered 3833 mother-baby pairs and exhibited the expected significant relationships between smoking and low birth weight. Smokers were twice as likely to have low birth weight babies and four times as likely to have moderately low birth weight babies, but no more likely to have very low birth weight babies. Smokers were associated with premature delivery at a similar rate among lower birth weight babies as non-smokers and a higher rate of intrauterine growth retardation (low birth weight for a given age). Unadjusted mean costs were $439 higher for smokers as compared to non-smokers, a difference that was not statistically significant (p = .67). Regression analysis adjusting for age, race, drinking, payment status, pre-delivery care, smoking, birth weight and smoking-birth weight interactions revealed no birth weight independent effects of smoking, but significant birth weight effects and smoking-birth weight interaction effects. Low birth weight was associated with higher costs and smoking was associated with lower costs within the very low birth weight and low birth weight groups. Adjusted costs were higher for smokers based on a higher likelihood of being in birth weight groups ($583), which were partially off-set by lower costs for smokers within birth weight groups ($195), yielding an adjusted excess cost of $388 associated with smokers. Given the cost of smoking cessation programs ($200--677), efforts to reduce smoking among pregnant women are not likely to reduce short-run hospital costs. Policy makers will find it difficult to use smoking cessation programs as a solution to the problem of low birth weight/high cost pregnancies. However, adoption of smoking cessation programs aimed at pregnant women may be based on long-term costs and other factors outside of the scope of this dissertation.Dr.P.H.Health and Environmental SciencesPublic healthUniversity of Michigan, Horace H. Rackham School of Graduate Studies, School of Public Healthhttp://deepblue.lib.umich.edu/bitstream/2027.42/124725/2/3155719.pd