Findings can help state and local public health departments tailor tobacco prevention and control interventions.

Large variations were found in cigarette smoking patterns and in current and former smoking status between states and within each state in the country, according to a study published in Cancer Epidemiology, Biomarkers & Prevention, a journal of the American Association for Cancer Research.

Counties with the highest prevalence of current cigarette smokers were not always from the states with the highest prevalence of current cigarette smokers.

"Cigarette smoking is the leading preventable cause of disease, disability, and death in the United States. Each year, nearly half a million American adults die prematurely from cigarette smoking or exposure to secondhand smoke," said Zahava Berkowitz, MSPH, MSc, statistician in the Epidemiology and Applied Research Branch of the Division of Cancer Prevention and Control at the National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention (CDC).

In the United States, the estimated cost of smoking-related medical care and loss of productivity is more than $300 billion a year, Berkowitz noted. Her team examined county-level estimates in the United States for different categories of cigarette smoking, including people who smoke every day, people who smoke occasionally, former smokers, and never smokers.

The researchers found that the highest prevalence of current cigarette smokers was in parts of the Midwest and the South, with the highest percentages in West Virginia (24.81 percent) and Kentucky (24.56 percent); other states with high prevalence of current smokers were Alabama, Tennessee, Missouri, Arkansas, Oklahoma, Indiana, Louisiana, Ohio, and Mississippi.

Of note, counties with the highest prevalence of current cigarette smokers were in Arkansas, South Dakota, North Dakota, and Wisconsin, even though Arkansas was the only one of these states with a relatively high prevalence of current smokers.

The highest prevalence of former cigarette smokers was in Maine, Vermont, New Hampshire, Connecticut, Rhode Island, Massachusetts, and Florida.

Utah had the lowest prevalence of current cigarette smokers and former smokers. Counties with the lowest prevalence of current cigarette smokers were in Utah and California.

"We were struck by the great variation in county-level cigarette smoking patterns across the United States," Berkowitz noted. "We have known about the state-by-state variation for some time, but it is important to hone in on county-level variation to have a more fine-grained picture of the geographic variability. County-level estimates can help identify areas where targeted tobacco prevention and control interventions may be warranted.

"This is the first study to estimate the percentages of people who are former smokers within counties. This information is important because some former smokers may benefit from lung cancer screening," she added.

Berkowitz and Xingyou Zhang, who developed the analytical method to conduct this study, and colleagues used data from the Behavioral Risk Factor Surveillance System (BRFSS), a large nationwide survey of 475,687 people to generate county-level estimates and predict cigarette smoking patterns for all counties in the United States.

In order to validate the multilevel small area estimation approach used in the study, the authors compared their model-based state-level estimated percentages for each cigarette smoking category with the BRFSS state estimates and found a high correlation.

"This analysis highlights the value of having nationwide surveys such as BRFSS to provide data that can also be used at the local level, including county-level patterns in cigarette smoking. These data are extremely useful to monitor public health progress in reducing tobacco use at the local level," Berkowitz said.

A limitation of the study is that these are model-based predictions of cigarette smoking prevalence, meaning they are estimates and not actual measurements, Berkowitz said.

All authors were full-time federal government employees. The authors declare no conflicts of interest.

Article: Multilevel Small-Area Estimation of Multiple Cigarette Smoking Status Categories Using the 2012 Behavioral Risk Factor Surveillance System, Zahava Berkowitz, Xingyou Zhang, Thomas B. Richards, Lucy Peipins, S. Jane Henley and James Holt, Cancer Epidemiology, Biomarkers & Prevention, doi: 10.1158/1055-9965.EPI-16-0244, published 2 October 2016.