Research objective: Much of the work on estimating health care costs attributable to smoking has failed to capture the effects and related costs of smoking during pregnancy. The goal of this study is to use data on smoking behavior, birth outcomes and resource utilization to estimate neonatal costs attributable to maternal smoking during pregnancy.
Study design: We use 1995 data from the Center for Disease Control's (CDC) Pregnancy Risk Assessment Monitoring System (PRAMS) database. The PRAMS collects representative samples of births from 13 states (Alabama, Alaska, California, Florida, Georgia, Indiana, Maine, Michigan, New York (excluding New York City), Oklahoma, South Carolina, Washington, and West Virginia), and the District of Columbia. The 1995 PRAMS sample is approximately 25 000. Multivariate analysis is used to estimate the relationship of smoking to probability of admission to an NICU and, separately, the length of stay for those admitted or not admitted to an NICU. Neonatal costs are predicted for infants 'as is' and 'as if' their mother did not smoke. The difference between these constitutes smoking attributable neonatal costs; this divided by total neonatal costs constitutes the smoking attributable fraction (SAF). We use data from the MarketScantrade mark database of the MedStattrade mark Corporation to attach average dollar amounts to NICU and non-NICU nursery nights and data from the 1997 birth certificates to extrapolate the SAFs and attributable expenses to all states.
Principal findings: The analysis showed that maternal smoking increased the relative risk of admission to an NICU by almost 20%. For infants admitted to the NICU, maternal smoking increased length of stay while for non- NICU infants it appeared to lower it. Over all births, however, smoking increased infant length of stay by 1.1%. NICU infants cost $2496 per night while in the NICU and $1796 while in a regular nursery compared to only $748 for non-NICU infants. The combination of the increased NICU use, longer stays and higher costs result in a positive smoking attributable fraction (SAF) for neonatal costs. The SAF across all states is 2.2%. Across the states, the SAF varied from a low of 1.3% in Texas to a high of 4.6% in Indiana.
Conclusions: These results further confirm the adverse effects of smoking. Among mothers who smoke, smoking adds over $700 in neonatal costs. The smoking attributable neonatal costs in the US represent almost $367 million in 1996 dollars; these costs vary from less than a million in smaller states to over $35 million in California. These costs are highly preventable since the adverse effects of maternal smoking occur in the short-run and can be avoided by even a temporary cessation of maternal smoking. These cost estimates can be used by managed care plans, state and local public health officials and others to evaluate alternative smoking cessation programs.
Copyright 2002 John Wiley & Sons, Ltd.