Using Small-Area Estimation to Calculate the Prevalence of Smoking by Subcounty Geographic Areas in King County, Washington, Behavioral Risk Factor Surveillance System, 2009–2013
ORIGINAL RESEARCH — Volume 13 — April 19, 2016
PEER REVIEWED
Three maps illustrate 3 methods for estimating smoking prevalence rates by census tract. Map A is based on the direct estimation method. Because sample sizes for many of the census tracts are too small (n <50), prevalence rates estimated by this method are not reliable. Map B shows smoothed estimates derived from our small area estimation model (hierarchical Bayesian model) for respondents with complete data on geocoded census tracts. Respondents with missing data on census tracts were excluded from this analysis. Map B combines the smoothed and the multiple imputation methods to present estimates generated using both the small area estimation model and multiple imputation to include all respondents. Map A shows high smoking rates scattered throughout different regions of the county. Maps B and C show high smoking rates concentrated in the south Seattle and South County areas.
Figure 1. Current smoking prevalence by census tract among King County adults. Maps illustrate 3 methods for estimating smoking prevalence rates by census tract. Map A is based on the direct estimation method. Because sample sizes for many of the census tracts are too small (n <50), prevalence rates estimated by this method are unreliable. Map B shows smoothed estimates derived from our small area estimation model (hierarchical Bayesian model) for respondents with complete information on geocoded census tracts. Respondents with missing data on census tracts were excluded from this analysis. Map C combines the smoothed and the multiple imputation methods to present estimates generated by using both the small area estimation model and multiple imputation to include all respondents. Data are from the King County sample of the Behavioral Risk Factor Surveillance System for 2009 through 2013 combined.
The map shows that among the 48 health reporting areas in King County, higher smoking rates are concentrated in some of the South Seattle and the South County areas. The smoking rates are generally lower in the East County and North County areas.
Figure 2. Model-based current smoking prevalence (percentage) among King County adults by King County health reporting areas. The map shows smoothed smoking prevalence rates. Estimates were generated by using a spatial hierarchical Bayesian model. Data are from the King County sample of the Behavioral Risk Factor Surveillance System for 2009 through 2013 combined.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.