CBHSQ State and Substate-Level Data
from the National Survey on Drug Use and Health (NSDUH)
Brief Background and Limitations of NSDUH State and Substate Estimates
The National Survey on Drug Use and Health (NSDUH) is sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA), which is an agency of the U.S. Department of Health and Human Services (HHS), and managed by SAMHSA's Center of Behavioral Health Statistics and Quality (CBHSQ). The NSDUH provides data on the incidence and prevalence of substance use and mental illness as required by Section 505 of the Public Health Service Act (42 USC 290aa4). Conducted on a periodic basis from 1971 through 1988, and annually since 1990, the NSDUH provides current data on substance use and mental health for the U.S. civilian, noninstitutionalized population aged 12 or older.
Prior to 1999, a national design was employed except for the occasional oversampling of specific geographic areas (e.g., rural areas, a few MSAs, etc.). In fact, there were a number of states with very little or no sample prior to 1999. Therefore, state-level estimates are not possible prior to 1999.
A special report was produced based on the combination of 1991-1993 NSDUH data where model-based estimates were generated for a few outcome measures in 26 states and 25 metropolitan areas (see http://www.oas.samhsa.gov/96state/toc.htm). However, the sample was not large enough to produce estimates in every state. The years 1991-1993 were used in part because the survey included additional sample in order to produce annual estimates of substance use in the Chicago, Denver, Los Angeles, Miami, New York, and Washington, D.C. Metropolitan Statistical Areas (MSAs). The six-MSA oversample was discontinued beginning in 1994.
Beginning in 1999, the NSDUH sample expanded significantly so that estimates could be produced for each of the 50 states and the District of Columbia. While state-level estimates are available from 1999 onward, the 1999-2001 estimates will not be comparable to estimates from 2002 and beyond due to significant design changes that took place beginning in 2002. Therefore, data users are encouraged not to perform comparisons over those time periods. Go to http://www.samhsa.gov/data/nsduh/2k4nsduh/2k4results/2k4results.pdf (see Appendix C) or http://www.samhsa.gov/data/nsduh/methods.pdf for additional information.
Since 1999, NSDUH estimates have been based on data from approximately 67,500 individuals per year nationally. That breaks down to 3,600 respondents per year in each of the 8 large states (CA, FL, IL, MI, OH, PA, NY, and TX) and 900 respondents per year in each of the remaining 42 states and the District of Columbia (sometimes collectively called "small states"). This design produces state and substate estimates with adequate precision when combining two or more years of data, especially for the small states. However, the design is inadequate for producing estimates for all 3100+ counties in the U.S. In fact, there are many counties with very little NSDUH data and some with no data even after combining all available years of comparable data (i.e., 2002-2011).
Links to NSDUH State and Substate Estimates
Various state and substate reports can be accessed at the links below. Please check the NSDUH website http://www.samhsa.gov/data/NSDUH.aspx periodically as links to some of the earlier reports will be updated.
These tables contain NSDUH state and substate estimates by age group for a selected set of outcome measures. A list of outcome measures available in these tables is shown here: