Consequences of declining survey response rates for smoking prevalence estimates

Background: Response rates have been declining in statewide tobacco surveys. This study investigated whether there was associated evidence of increasing bias in smoking prevalence estimates. Methods: Demographic characteristics of respondents to tobacco surveys in Massachusetts and California were compared to population data in the early 1990s, when response rates were high, and in more recent years, when response rates were lower. State estimates of smoking prevalence at three times were compared with estimates from the Current Population Survey Tobacco Use Supplement (CPS-TUS), conducted by the U.S. Census Bureau. Results: Under- and over-representation of population subgroups has not changed as response rates have declined. Smoking prevalence estimates from state surveys remain relatively close to the state-specific CPS-TUS estimates. Conclusions: There is no evidence that declining response rates have resulted in less accurate or biased estimates of smoking behavior. (C) 2004 American Journal of Preventive Medicine.