Surveying Cellular Telephone Users
In the past few years, all large population health surveys that depend on telephone interviews, including the BRFSS, have had to adjust to the rapid rise in the proportion of U.S. households that have a cellular telephone but no landline telephone. In 2004 an expert panel of survey methodologists recommended that BRFSS address the growing effects of cellular telephone-only households. These recommendations included adding cellular telephones in the BRFSS sample and developing improved weighting and adjustment methodologies. Since 2004, CDC has been planning and testing these changes, which were fully implemented with the May 2012 release of the 2011 BRFSS dataset. The BRFSS changes are discussed in detail in the June 8, 2012, MMWR Policy Note titled, "Methodologic Changes in the Behavioral Risk Factor Surveillance System in 2011 and Potential Effects on Prevalence Estimates."
Geographic Stratification within Cellular Samples
Geographic stratification allows states to sample smaller geographic areas within the states. Landline telephone sampling allows for geographic sampling within each state, but cellular telephone numbers are not tied to geographic areas, which complicates geographic sampling. In 2013, the BRFSS is piloting cellular geographic stratification using cellular sampling frames derived from rate centers. A rate center separates the local call boundaries set by service providers for billing purposes; as such, subsets of 1000-series blocks of cellular numbers assigned to specific wireless service providers can be mapped to geographic areas of interest for targeting purposes. This method allows for a better geographic distribution of the cellular sample within the state.
Opt-in Internet Panel Surveys
Opt-in Internet surveys are completed by individuals who have been recruited to fill out questionnaires online. These methods are very flexible, allowing national data collection in a matter of weeks. They are also much more cost effective than random digit dialed surveys and have the potential for following the same groups of people over time. Given these potential advantages, CDC’s Division of Behavioral Surveillance is conducting a pilot project to assess the feasibility and accuracy of opt-in Internet panel surveys for behavioral health surveillance. In collaboration with four state departments of health (Georgia, Illinois, New York, and Texas), opt-in Internet panels are being used to assess a range of health-related measures including healthcare access and use, health behaviors, and health outcomes. Three Internet panel firms that use sampling methods with different recruitment strategies, sample selection, and sample matching to the U.S. adult population were used to collect national-, state-, and metropolitan statistical area (MSA)-level data. At the national level, 3,500 Internet panel surveys will be completed; at the state- and MSA-level, 11,000 completed surveys, respectively (Georgia, Atlanta; Illinois, Chicago; New York, New York City; and Texas, Houston). A Web-based questionnaire was developed to establish standards with other federal surveys, such as the Behavioral Risk Factor Surveillance System (BRFSS), Selected Metropolitan/Micropolitan Area Risk Trends (SMART) BRFSS, the National Health Interview Survey (NHIS), the National Survey on Drug Use and Health (NSDUH), the National Health and Nutrition Examination Survey (NHANES), and Consumer Survey of Attitudes Toward the Privacy and Security Aspects of Electronic Health Records and Health Information Exchanges. Data collection is anticipated to be completed by the end of March 2013. Comparative analyses will then be conducted to assess the advantages and disadvantages of different Internet sampling methodologies across a range of parameters including cost, geography, timeliness, usability, and ease of use for technology transfer to states and local communities. Recommendations for future efforts in behavioral health surveillance will be based on these results.