Substate Small Area Estimates Using Data from the 2014-2016 National Surveys on Drug Use and Health (NSDUHs)

• Activity Number: 596 - Advances in Small Area Estimation
• Type: Contributed
• Date/Time: Wednesday, August 1, 2018 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Neeraja Sathe* and Matthew Williams and Kathy Spagnola and Akhil Vaish
• Companies: RTI International and SAMHSA/CBHSQ and RTI International and RTI International
• Keywords: : small area estimation; National Survey on Drug Use and Health (NSDUH); substate estimates; substance use; mental health; hierarchical Bayes

Abstract

The National Survey on Drug Use and Health (NSDUH) is the primary source of information on substance use and mental disorders for the U.S. civilian, noninstitutionalized population aged 12 years old or older. State and local agencies concerned with the health and welfare of the population within their jurisdiction face difficulties when designing health promotion and prevention activities because of the lack of local area-specific data. Typically, they want information on substance use or mental health indicators for states or preferably even substates (e.g., counties, groups of counties, health service areas, etc.) so that policymakers can use these estimates to decide how treatment and prevention money will be spent in their state. Substate estimates may also help to get a better idea about how existing treatment and prevention programs are working within the states. This paper presents substate estimates for a few outcomes using data from the 2014, 2015, and 2016 NSDUHs. Also included in this paper is a brief description of the small area estimation method and models used, and examples of how substate estimates for certain areas in a state compare with overall state estimates.