2012-2014 National Surveys on Drug Use and Health:
Guide to Substate Tables and Summary
of Small Area Estimation Methodology

Section A: Overview

A.1. Introduction

This document provides a guide on the development and presentation of model-based small area estimates of the prevalence of substance use and mental disorders in substate regions based on data from the combined 2012-2014 National Surveys on Drug Use and Health (NSDUHs). The estimates along with this document and other related information are available at http://www.samhsa.gov/data/.

An annual survey of the civilian, noninstitutionalized population aged 12 or older, NSDUH is sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA). It collects information from individuals residing in households, noninstitutionalized group quarters (e.g., shelters, rooming houses, dormitories), and civilians living on military bases. In 2012-2014, NSDUH collected data from 204,048 respondents aged 12 or older and was designed to obtain representative samples from the 50 states and the District of Columbia. The survey is planned and managed by SAMHSA's Center for Behavioral Health Statistics and Quality (CBHSQ). Data collection and analysis were conducted under contract with RTI International.1

The 1999 through 2001 National Household Surveys on Drug Abuse (NHSDAs)2 and the 2002 through 2013 NSDUHs employed a 50-state design with an independent, multistage area probability sample for each of the 50 states and the District of Columbia. For the 50-state design, 8 states were designated as large sample states (California, Florida, Illinois, Michigan, New York, Ohio, Pennsylvania, and Texas) with target sample sizes of 3,600 per year. For the remaining 42 states and the District of Columbia, the target sample size was 900 per year. This approach ensured that there was sufficient sample in every state to support small area estimation (SAE) production at the state and substate level, while at the same time maintaining efficiency for national estimates. The design also oversampled youths aged 12 to 17 and young adults aged 18 to 25 so that each state's sample was approximately equally distributed among three major age groups: 12 to 17 years, 18 to 25 years, and 26 years or older.

A coordinated design was developed for the 2014 through 2017 NSDUHs. Similar to the 1999 through 2013 surveys, the coordinated 4-year design is state-based with an independent, multistage area probability sample within each state and the District of Columbia. This design designates 12 states as large sample states. These 12 states have the following target sample sizes per year: 4,560 interviews in California; 3,300 interviews in Florida, New York, and Texas; 2,400 interviews in Illinois, Michigan, Ohio, and Pennsylvania; and 1,500 interviews in Georgia, New Jersey, North Carolina, and Virginia. Making the sample sizes more proportional to the state population sizes improves the precision of national NSDUH estimates. This change also allows for a more cost-efficient sample allocation to the largest states while slightly increasing the sample sizes in smaller states to improve the precision of state and substate estimates (note that the target sample size per year in the small states is 960 interviews with the exception of Hawaii where the target sample size is 967 interviews).

Starting in 2014, the allocation of the NSDUH sample is 25 percent for adolescents aged 12 to 17, 25 percent for adults aged 18 to 25, and 50 percent for adults aged 26 or older. The sample of adults aged 26 or older is further divided into three subgroups: aged 26 to 34 (15 percent), aged 35 to 49 (20 percent), and aged 50 or older (15 percent). For more information on the 2014 through the 2017 NSDUH sample design and for differences between the 2013 and 2014 surveys, refer to the 2014 NSDUH sample design report (CBHSQ, 2015).

This marks the seventh time3 that estimates for substate regions (also referred to as planning regions or substate areas) in all 50 states and the District of Columbia have been presented by SAMHSA. Here, substate-level small area estimates are provided for 25 binary (0, 1) substance use or mental health measures using combined data from the 2012-2014 NSDUHs for individuals aged 12 or older (or adults 18 or older for the four mental disorders, and individuals aged 12 to 20 for underage alcohol and binge alcohol use). For a list of outcomes for which substate-level estimates are available, refer to Table A1 at the end of this section. These substate estimates are available at http://www.samhsa.gov/data/. The list of products (e.g., tables, maps, substate region definitions) related to the 2012-2014 substate estimates is provided in Section A.2.

Estimates for 384 substate regions were generated using the 2012-2014 NSDUH data. These substate regions were defined by government officials from each of the 50 states and the District of Columbia and were typically based on the substance abuse treatment planning regions specified by the states in their applications for the Substance Abuse Prevention and Treatment Block Grant (SABG) administered by SAMHSA. The SABG program provides financial and technical assistance to the 50 states, the District of Columbia, and other jurisdictions to support substance abuse prevention and treatment programs and to promote public health. States use NSDUH substate estimates for a variety of purposes, including strategic planning and program development, production of epidemiological profiles for briefing state legislatures and informing the public, allocating funds to substate areas based on the need for services, and other uses. More information on defining these regions is available in Section A.3.

Section A.4 discusses the methodological changes that were introduced in the 2002 NSDUH. An unanticipated result of these changes was that the prevalence rates for 2002 were in general substantially higher than those for 2001. As a result, the 1999-2001 substate estimates are not comparable with the other substate estimates. Section A.5 discussed related substance use measures and warns users about not drawing conclusions by subtracting small area estimates from two different measures.

Section B provides information on the SAE methodology used to produce substate estimates. Section C includes the population estimates and the combined 2012, 2013, and 2014 NSDUH sample sizes and response rates for each substate region. Users may find the population estimates helpful in calculating the prevalence estimate for any combination of substate regions or to determine the number of people using a particular substance in a substate region. For example, the number of individuals aged 12 or older who used marijuana in the past month in Alabama's Region 1 (53,712 individuals) can be obtained by multiplying the prevalence rate (4.71 percent) from Table 3 in the "2012-2014 NSDUH Substate Regions: Excel Tables" (see http://www.samhsa.gov/data/) and the population estimate from Table C1 (1,140,380) in this document. Section D lists the references cited in this document, and Section E provides a list of contributors to the production of the 2012-2014 substate estimates.

A.2. Presentation of Findings

The following products associated with the 2012-2014 substate estimates are available at http://www.samhsa.gov/data/:

Note that other products may be added to the 2012-2014 NSDUH substate homepage at a later date.

A.3. Formation and Ranking of Substate Regions

The substate regions for each state were developed in a series of communications during the fall of 2015 between SAMHSA staff and state officials responsible for the SABG application. There is extensive variation in the size and use of substate regions across the states. In some states, the substate regions are used more for administrative purposes rather than for planning purposes. The goal of the project was to provide substate-level estimates showing the geographic distribution of substance use prevalence for regions that states would find useful for planning and reporting purposes. The final substate region boundaries were based on the state's recommendations, assuming that the NSDUH sample sizes were large enough to provide estimates with adequate precision. Most states defined regions in terms of counties or groups of counties. A few states defined the regions in terms of census tracts. Several states also requested estimates for aggregate substate regions, along with the estimates for their substate regions. An aggregate substate region is made up of two or more substate regions. These substate region definitions are available in a document titled "2012-2014 NSDUH Substate Region Definitions" (see http://www.samhsa.gov/data/ as listed in Section A.2). Some of the states (specifically, New Hampshire, New York, Texas, and Washington) wanted the maps to be produced only for the aggregate regions. For example, New York has 15 substate regions, and those 15 regions were combined to create 4 aggregate regions that are used in the maps. Hence, for each measure, maps were produced for 362 regions and not for 384 regions.

These 362 substate regions used in the maps were ranked from lowest to highest for each measure and were divided into seven categories (based on estimates rounded to 2 decimals) designed to represent distributions that are somewhat symmetric, like a normal distribution. Colors were assigned to all substate regions such that the third having the lowest prevalence are in blue (121 substate regions), the middle third are in white (120 substate regions), and the third with the highest prevalence are in red (121 substate regions). The only exceptions were the three perception-of-risk outcomes shown in Figure 4 (marijuana), Figure 11 (alcohol), and Figure 16 (cigarettes) of the national maps, which have the highest estimates represented in blue and the lowest estimates represented in red to reflect the inverse relationship between substance use and the perception of risk of using that substance. To further distinguish among the substate regions that display relatively higher prevalence, the "highest" third in red has been subdivided into (a) dark red for the 16 substate regions with the highest estimates, (b) medium red for the 33 substate regions with the next highest estimates, and (c) light red for the 72 substate regions in the third highest group. The "lowest" third is categorized in a similar way using three distinct shades of blue. In some cases, a group (or category) could have more or fewer substate regions because two (or more) substate regions have the same estimate (to two decimal places). When such ties occurred at the "boundary" between two groups, all substate regions with the same estimate (to two decimal places) were assigned to the lower group. The upper and lower limits of each of the seven categories shown in the map legend collectively define a continuum and are not necessarily the actual values of a particular substate region. For example, in Figure 9 (national map showing alcohol use in the past month among persons aged 12 or older) (see the "2012-2014 NSDUH National Maps of Prevalence Estimates, by Substate Region" at http://www.samhsa.gov/data/), the values on the boundary in the lowest category (group 1) correspond to Utah County in Utah (21.37 percent) and Catchment Area 7 in Arkansas (36.38 percent) and are displayed in the legend. In the next to lowest category, Catchment Area 4 in Arkansas (37.39 percent) and Communicare and River Valley in Kentucky (43.54 percent) are the regions with the lowest and highest values; however, in the continuum of the legend, the lower limit for group 2, was assigned a value of 36.39 percent because the upper limit of group 1 is 36.38 percent. These national maps are available at http://www.samhsa.gov/data/ as listed in Section A.2.

The 2012-2014 substate estimates and corresponding Bayesian CIs are available in the "2012-2014 NSDUH Substate Region Estimates: Excel Tables" (as mentioned in Section A.2, see http://www.samhsa.gov/data/). These tables also contain a sort order number and a map-group indicator (= 1 for the nation, = 2 for census regions, = 3 for states, = 4 if a region is part of the 362 mapping regions, and = 5 for all other substate/aggregate regions not included on the maps).

In addition to the substate region estimates, comparable estimates are provided for the 50 states and the District of Columbia and also for the four census regions. Design-based estimates and corresponding CIs for the nation are also included. Because these estimates are based on 3 consecutive years of data, they are not directly comparable with the annually published state, census region, or national estimates that are based on only 2 consecutive years of NSDUH data. The U.S. Census Bureau defines the census regions as follows:

Northeast Region - Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont.
Midwest Region - Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, and Wisconsin.
South Region - Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia.
West Region - Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming.

A.4. Comparability with Past Estimates

For the 2002 NSDUH, a number of methodological changes were introduced, including a $30 incentive for participating in the survey, additional training for interviewers to encourage adherence to survey protocols, a change in the survey name, and a shift to the 2000 decennial census (from the 1990 census) as a basis for population counts used in sample selection, weighting, and estimation. Additional information on these methodological changes is available in a report by the Office of Applied Studies (OAS, 2005).4 An unanticipated result of these changes was that the prevalence rates for 2002 were in general substantially higher than the 2001 prevalence rates. The jump in the prevalence rates between 2001 and 2002 was more than the usual year-to-year change. Because of the changes in the survey that took place in 2002, substate estimates for 1999-2001 are not comparable with the other substate estimates. It is not possible to separate the effect of the methodological changes from the true trends in substance use.

However, estimates from 2002-2004, 2004-2006, 2006-2008, 2008-2010, 2010-2012, and 2012-2014 are comparable for outcome measures that were defined in a similar manner and for substate regions defined consistently across these time periods. Table A1 at the end of this section lists the outcome measures for which substate estimates were produced using 2012-2014 NSDUH data and shows the outcome measures that remained comparable over time (indicated with an "X" in the table) since 2002-2004.

It is useful to note that the 2002-2004 to 2008-2010 substate estimates were produced using predictors from the 2000 census; also, the survey weights used population projections from the 2000 census. The 2010-2012 and 2012-2014 estimates, on the other hand, were produced using 2010 census data. Hence, when reviewing changes between 2002-2004 and 2012-2014, it is important to note that they may be confounded with changes due to differences in the source of the predictors and the population projections (referred to as "census effects"). The impact of such census effects on national and state estimates is discussed in Section B.4.3 in Appendix B of the 2011 NSDUH national findings report (CBHSQ, 2012) and the "2011-2012 NSDUH: Impact of Using 2010 Census Data on 2010-2011 Small Area Estimates" at http://www.samhsa.gov/data/.

During regular data collection and processing checks for the 2011 NSDUH, data errors were identified. These errors were falsified cases submitted by field interviewers that affected the data for Pennsylvania (2006 to 2010) and Maryland (2008 and 2009). Cases with erroneous data were removed from the data files, and the remaining cases were reweighted to provide representative estimates (for more details on this data error, refer to Section A.7 of the "2011-2012 NSDUH Guide to State Tables and Summary of Small Area Estimation Methodology" at http://www.samhsa.gov/data/). The 2008-2010 and the 2010-2012 substate estimates exclude data based on falsified cases.

A.5. Related Substance Use Measures

Small area estimates are produced for a number of related drug measures, such as marijuana use and illicit drug use. It might appear that one could draw conclusions by subtracting one from the other (e.g., subtracting the percentage who used illicit drugs other than marijuana in the past month from the percentage who used illicit drugs in the past month to find the percentage who only used marijuana in the past month). Because related measures have been estimated with different models (i.e., separate models by age group and outcome), subtracting one measure from another related measure at the substate region, state, or census region level can give misleading results, perhaps even a "negative" estimate, and should be avoided. However, these comparisons can be made at the national level because these estimates are design-based estimates. For example, at the national level, subtracting cigarette use estimates from tobacco use estimates will give the estimate of individuals who did not use cigarettes, but used other forms of tobacco.

Table A1. – Outcomes, by Survey Year, for Which Substate Small Area Estimates Are Available
Measure 2002-2004 2004-2006 2006-2008 2008-2010 2010-2012 2012-2014
X = available; -- = not available.
NOTE: The measures included in the 1999-2001 substate small area estimation (SAE) report are not included here. Because of the changes in the survey that took place in 2002, the 1999-2001 estimates are not comparable with the 2002-2004 or subsequent year estimates. Estimates using the combined 2002-2004, 2004-2006, and 2006-2008 data can be found at http://www.samhsa.gov/data/. Estimates using the combined 2008-2010, 2010-2012, and 2012-2014 data can also be found at http://www.samhsa.gov/data/.
1 The marijuana incidence definition used here employs a simpler form of the at-risk population based on the model-based methodology. This model-based average annual incidence rate for first use of marijuana is defined as follows: Average annual marijuana initiation rate (%) = 100 × {[X1 ÷ (0.5 × X1 + X2)] ÷ 2}, where X1 is the number of marijuana initiates in the past 24 months and X2 is the number of individuals who never used marijuana (with the at-risk population defined as 0.5 × X1 + X2). Both X1 and X2 are based on binary measures that correspond to questions with a "yes" or "no" response option. For details on calculating the average annual rate of first use of marijuana from NSDUH data, see Section A.8 in Appendix A of the 2009-2010 state estimates report (Hughes, Muhuri, Sathe, & Spagnola, 2012).
2 Dependence or abuse is based on definitions found in the 4th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). For more details on these measures, see Section B.4.2 in Appendix B of the 2012 NSDUH national findings report (Center for Behavioral Health Statistics and Quality [CBHSQ], 2013b).
3 Because of questionnaire changes, SPD estimates in 2002-2004 are not comparable with the 2004-2006 SPD estimates. For details, see Section B.7 of the report on Substate Estimates from the 2004-2006 National Surveys on Drug Use and Health (Office of Applied Studies [OAS], 2008). Additional questionnaire changes were made in 2008 that affected past year SPD trends. However, revised past year SPD measures were created for 2005 through 2007 that are comparable with the 2008 through 2014 past year SPD measure. Substate small area estimates for 2006-2008, 2008-2010, 2010-2012, and 2012-2014 were not created for this measure.
4 Questions used to determine MDE were added in 2004. The 2004-2006 MDE estimates are not comparable with the 2006-2008 and subsequent year estimates. For details on MDE, see Sections B.4.2 and B.4.4 in Appendix B of the 2012 NSDUH mental health findings report (CBHSQ, 2013a).
5 The mental illness measures are based on a predictive model and are not direct measures of diagnostic status (i.e., a small subsample of the respondents received a clinical follow-up, and these responses along with questionnaire data were used to create predictive models for all of the respondents). For details about these measures and the predictive models, see Section B.4.3 in Appendix B of the 2012 NSDUH mental health findings report (CBHSQ, 2013a).
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Surveys on Drug Use and Health, 2002-2014.
Illicit Drug Use in the Past Month X X X X X X
Marijuana Use in the Past Year X X X X X X
Marijuana Use in the Past Month X X X X X X
Perceptions of Great Risk from Smoking Marijuana Once a Month X X X X X X
First Use of Marijuana1 X X X X X X
Illicit Drug Use Other Than Marijuana in the Past Month X X X X X X
Cocaine Use in the Past Year X X X X X X
Nonmedical Use of Pain Relievers in the Past Year X X X X X X
Alcohol Use in the Past Month X X X X X X
Underage Past Month Use of Alcohol (among Individuals Aged 12 to 20) X X X X X X
Binge Alcohol Use in the Past Month X X X X X X
Underage Past Month Binge Alcohol Use (among Individuals Aged 12 to 20) X X X X X X
Perceptions of Great Risk from Having Five or More Drinks of an Alcoholic Beverage Once or Twice a Week X X X X X X
Tobacco Product Use in the Past Month X X X X X X
Cigarette Use in the Past Month X X X X X X
Perceptions of Great Risk from Smoking One or More Packs of Cigarettes per Day X X X X X X
Alcohol Dependence or Abuse in the Past Year2 X X X X X X
Alcohol Dependence in the Past Year2 X X X X X X
Illicit Drug Dependence or Abuse in the Past Year2 X X X X X X
Illicit Drug Dependence in the Past Year2 X X X X X X
Dependence or Abuse of Illicit Drugs or Alcohol in the Past Year2 X X X X X X
Needing But Not Receiving Treatment for Illicit Drug Use in the Past Year X X X X X X
Needing But Not Receiving Treatment for Alcohol Use in the Past Year X X X X X X
Serious Psychological Distress (SPD) in the Past Year3 X X -- -- -- --
Had at Least One Major Depressive Episode (MDE) in the Past Year4 -- X X X X X
Serious Mental Illness (SMI) in the Past Year5 -- -- -- X X X
Any Mental Illness (AMI) in the Past Year5 -- -- -- X X X
Had Serious Thoughts of Suicide in the Past Year -- -- -- X X X
Table A2. – NSDUH Substate Region Counts and Overlap, by State and Estimation Period
State 2002-2004 2010-2012 2012-2014
Number of
Substate
Regions1
Number of
Map
Regions2
Number of
Substate
Regions1
Number of
Map
Regions2
Number of
Substate
Regions1
Number of
Map
Regions2
Number of
Substate
Regions
Overlapping
with
2002-20043
Number of
Substate
Regions
Overlapping
with
2010-20123
1 Number of regions only include the main substate regions and not the aggregate regions.
2 More information on the map regions can be found in Section A.3.
3 The names of some of the substate regions have changed across the time periods. However, as long as the two regions are made of the same counties or tracts, they are included in the count of overlapping regions.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Surveys on Drug Use and Health, 2002-2014.
Total U.S. 357 340 384 363 384 362 264 356
Alabama 4 4 4 4 4 4 4 4
Alaska 4 4 4 4 4 4 1 4
Arizona 4 4 4 4 4 4 4 4
Arkansas 7 7 8 8 8 8 3 8
California 15 15 27 27 26 26 3 25
Colorado 5 5 5 5 5 5 5 5
Connecticut 5 5 5 5 5 5 5 5
Delaware 4 4 4 4 4 4 4 4
District of Columbia 8 8 8 8 8 8 8 8
Florida 14 14 18 18 18 18 8 16
Georgia 5 5 6 6 6 6 0 6
Hawaii 3 3 4 4 4 4 2 4
Idaho 7 7 7 7 7 7 5 5
Illinois 5 5 5 5 5 5 5 5
Indiana 8 8 8 8 8 8 8 8
Iowa 6 6 6 6 6 6 6 6
Kansas 6 6 6 6 6 6 6 6
Kentucky 6 6 6 6 6 6 6 6
Louisiana 7 7 6 6 6 6 5 6
Maine 7 7 8 8 8 8 3 8
Maryland 7 7 9 9 9 9 4 9
Massachusetts 6 6 6 6 6 6 6 6
Michigan 16 16 15 15 10 10 3 3
Minnesota 6 6 6 6 6 6 6 6
Mississippi 7 7 7 7 7 7 7 7
Missouri 7 7 7 7 7 7 7 7
Montana 5 5 5 5 5 5 5 5
Nebraska 5 5 6 6 6 6 4 6
Nevada 3 3 4 4 4 4 2 4
New Hampshire 5 3 5 3 5 3 5 5
New Jersey 4 4 4 4 4 4 4 4
New Mexico 5 5 5 5 5 5 5 5
New York 15 4 15 4 15 4 15 15
North Carolina 4 4 12 11 14 14 0 10
North Dakota 5 5 5 5 7 7 3 3
Ohio 21 21 21 21 21 21 21 21
Oklahoma 7 7 7 7 7 7 7 7
Oregon 5 5 6 6 6 6 4 6
Pennsylvania 13 13 13 13 13 13 9 13
Rhode Island 4 4 4 4 4 4 4 4
South Carolina 4 4 4 4 4 4 2 2
South Dakota 7 7 5 5 5 5 0 5
Tennessee 7 7 7 7 7 7 5 7
Texas 15 11 15 11 15 11 15 15
Utah 6 6 6 6 6 6 6 6
Vermont 4 4 4 4 4 4 4 4
Virginia 5 5 5 5 5 5 3 5
Washington 6 6 6 3 8 3 2 2
West Virginia 8 8 6 6 6 6 2 6
Wisconsin 6 6 6 6 6 6 4 6
Wyoming 9 9 9 9 9 9 9 9

Section B: Substate Region Estimation Methodology

The survey-weighted hierarchical Bayes (SWHB) methodology used in the production of state estimates from the 1999 to 2014 National Surveys on Drug Use and Health (NSDUHs) also was used in the production of the 2012-2014 substate estimates. The SWHB methodology is described by Folsom, Shah, and Vaish (1999). A general model description is given in Section B.1. A brief discussion of the precision of the estimates and interpretation of the Bayesian confidence intervals (CIs) is given in Section B.2. The goals of the small area estimation (SAE) modeling, the general model description, and the implementation of SAE modeling remain the same and are described in Appendix E of the 2001 state report (Wright, 2003).

Small area estimates obtained using the SWHB methodology are design consistent (i.e., for states or substate areas with large sample sizes, the small area estimates are close to the corresponding robust design-based estimates). The substate small area estimates when aggregated by using the appropriate population totals result in national small area estimates that are very close to the national design-based estimates. However, for many reasons, including internal consistency, it is desirable to have the national small area estimates exactly match the national design-based estimates. Beginning in 2002, exact benchmarking was introduced (see Appendix A, Section A.4, in Wright & Sathe, 2005). The 2012-2014 substate small area estimates have been benchmarked to the national design-based estimates.

B.1. General SAE Model Description

The model described here to produce the 2012-2014 substate small area estimates is the same logistic mixed5 hierarchical Bayes (HB) model that was used to produce the 2010-2012 substate small area estimates (see "2010-2012 National Survey on Drug Use and Health: Overview and Summary of Substate Region Estimation Methodology" at http://www.samhsa.gov/data/). The following model was used:

The model is given by the following equation: log of pi sub a, i, j, k divided by 1 minus pi sub a, i, j, k is equal to the sum of three terms. The first term is given by x transpose sub a, i, j, k times beta sub a. The second term is eta sub a, i. And the third term is nu sub a, i, j.,

where pi sub a, i, j, k is the probability of engaging in the behavior of interest (e.g., using marijuana in the past month, or cocaine use in past year) for person-k belonging to age group-a in substate region-j of state-i.

Let x sub a, i, j, k denote a p sub a times 1 vector of the auxiliary (predictor) variables associated with age group-a (12 to 17, 18 to 25, 26 to 34, and 35 or older) and Beta sub a denote the associated vector of the regression parameters. The age group-specific vectors of the auxiliary variables are defined for every block group in the nation and also include person-level demographic variables, such as race/ethnicity and gender. The auxiliary variables include block group, census tract, county, and state-level data. These predictor variables include demographic and socioeconomic data from the American Community Survey (ACS), population projections obtained from Nielsen Claritas, food stamp participation rates from the U.S. Census Bureau, Uniform Crime Report (UCR) arrest totals from the Federal Bureau of Investigation (FBI), per capita income from the Bureau of Economic Analysis (BEA), unemployment rates from the Bureau of Labor Statistics (BLS), mortality rates from the National Center for Health Statistics (NCHS), treatment rates from the National Survey of Substance Abuse Treatment Services (N-SSATS), and Block Grant awards, cost of services, and total taxable resources from the Substance Abuse and Mental Health Services Administration (SAMHSA). For a complete list of predictors, refer to Section B of the "2013-2014 NSDUH Guide to State Tables and Summary of Small Area Estimation Methodology" at http://www.samhsa.gov/data/.

The vectors An eta sub i is a transposed vector of values eta sub 1, i and so on until eta sub capital A, i. and A nu sub i, j is a vector of transposed values nu sub 1, i, j and so on until nu sub capital A, i, j., defined as state- and substate-level random effects, respectively, are assumed to be mutually independent with An eta sub i is normally distributed with mean 0 and variance denoted by matrix capital D sub eta. and A nu sub i, j is normally distributed with mean 0 and variance denoted by matrix captal D sub nu. where A is the total number of individual age groups modeled (generally, Capital A equals 4.). For HB estimation purposes, an improper uniform prior distribution6 is assumed for Beta sub a, and proper Wishart prior distributions are assumed for inverse of capital D sub eta. and inverse of capital D sub nu.. The HB solution for pi sub a, i, j, k involves a series of complex Markov Chain Monte Carlo (MCMC) steps to generate values of the desired fixed and random effects from the underlying joint distribution. The basic process is described in Folsom et al. (1999), Shah, Barnwell, Folsom, and Vaish (2000), and Wright (2003). Once the required number of MCMC samples for the parameters of interest are generated and tested for convergence properties (see Raftery & Lewis, 1992), the small area estimates for each age group × race/ethnicity × gender cell within a block group can be obtained. These block group-level small area estimates then can be aggregated using the appropriate population projections to form substate- and state-level small area estimates for the desired age group(s). These small area estimates then are benchmarked to the national design-based estimates (for details on exact benchmarking, see Section B.6 of the "2011-2012 NSDUH Guide to State Tables and Summary of Small Area Estimation Methodology" at http://www.samhsa.gov/data/).

Because the objective here was to produce small area estimates for substate regions, it was decided to ratio adjust the person-level sampling weights to match population projections available from Claritas at the substate × age group × gender level. These adjusted sampling weights are used in the estimation because they reflect the demography of the substate regions better than the unadjusted weights. This ratio adjustment was done at the substate region (384 regions) × age group (12 to 17, 18 to 25, 26 to 34, and 35 or older) × gender (male and female) level collectively over 3 years (2012, 2013, and 2014) of data.

The SAE methodology used here tends to borrow strength from both the national model and the state-level random effects. Estimates for substate regions with smaller sample sizes tend to be shrunk more toward the corresponding state and national prevalence estimates than substate regions with larger sample sizes. This methodology tends to cluster the small sample substate estimates around their state prevalence estimates. Thus, relatively high estimates for substate regions with small sample sizes tend to shrink toward the state prevalence estimates, while relatively low estimates tend to increase toward the state prevalence estimates. On the other hand, for substate regions with large sample sizes, the methodology produces estimates that are close to the weighted average of the sample data in that substate region. In addition, these estimates are design consistent so that, as the sample size for a substate region increases, the estimate approaches the corresponding design-based estimate.

B.2. Precision, Comparison, and Validation of the Estimates

The primary purpose of producing substate estimates is to give policy officials and data users a better perspective on the range of prevalence estimates within and across states. Because the data were collected in a consistent manner by field interviewers who adhered to the same procedures and administered the same questions across all states and substate regions, the results are comparable within and across the 50 states and the District of Columbia.

The 95 percent Bayesian CI associated with each estimate provides a measure of the accuracy of the estimate. It defines the range within which the true value can be expected to fall 95 percent of the time. For example, the estimated prevalence of past month use of marijuana in Region 1 in Alabama is 4.7 percent, and the 95 percent CI ranges from 3.5 to 6.3 percent.7 Therefore, the probability is 0.95 that the true value is within that range. The CI indicates the uncertainty due to both sampling variability and model bias. The key assumption underlying the validity of the CIs is that the state- and substate-level error (or bias correction) terms in the models behave like random effects with zero means and common variance components.

When comparing prevalence rates for two substate regions, it is tempting and often convenient to look at their 95 percent Bayesian CIs to decide whether the difference in the two substate regions' prevalence rates is significant. If the two Bayesian CIs overlap, one would conclude that the difference is not statistically significant. If the two Bayesian CIs do not overlap, it implies that the two substate regions prevalence rates are significantly different from each other. However, as discussed in Schenker and Gentleman (2001), the method of overlapping Bayesian CIs is more conservative (i.e., it rejects the null hypothesis of no difference less often) than the standard method based on z statistics when the null hypothesis is true. For details on the z statistics method to test the differences between the estimates of two substate regions, refer to Section B.12 (which describes the test for two state estimates) of the "2011-2012 NSDUH Guide to State Tables and Summary of Small Area Estimation Methodology" at http://www.samhsa.gov/data/. The same test can be extended to test the differences between two substate regions' prevalence rates. Note that the referenced test assumes zero correlation between the two estimates. Two substate regions in one state (or across states) are likely correlated, but the test can still be used knowing that the resulting test will be more conservative than if the correlation was accounted for. This is still a better alternative than using the overlapping CI test. Even if Bayesian CIs for two substate regions overlap, the two prevalence rates may be declared significantly different by the test based on z statistics. Hence, the method of overlapping Bayesian CIs is not recommended to test the equivalence of two substate prevalence rates. A detailed description of the method of overlapping CIs and its comparison with the standard methods for testing of a hypothesis is given in Schenker and Gentleman (2001) and Payton, Greenstone, and Schenker (2003).

A comparison of the standard errors (SEs) among substate regions with small (n ≤ 500), medium (500 < n ≤ 1,000), and large (n > 1,000) sample sizes for certain measures shows that the small area estimates behave in predictable ways. Regardless of whether the substate region is from a state with a larger annual sample size or one of the other states, the sizes of the CIs are very similar and are primarily a function of the sample size of the substate region and the prevalence estimate of the measure. Substate regions with large sample sizes had the smallest SEs. For past month use of alcohol, where the national prevalence for all individuals aged 12 or older was 52.3 percent (for 2012-2014), the average relative standard error (RSE)8 was about 5.1 percent, and the RSE for substate regions with a large sample size was about 3.2 percent. For substate regions with a medium sample size, the average RSE was 4.3 percent; for small sample sizes, the average RSE was 5.7 percent. For past month use of marijuana (with a national prevalence of 7.7 percent), the average RSE was 9.6 percent for substate regions with large samples. For medium sample sizes, the average RSE was 12.6 percent, and for small samples, the RSE was 15.2 percent, whereas the overall national average RSE was 13.9 percent. Substance use measures with lower prevalence rates, such as past year use of cocaine (1.7 percent nationally), displayed larger average RSEs. For substate regions with large sample sizes, the average RSE was 19.3 percent. For those with medium sample sizes, the average RSE was 23.0 percent, and for those with small sample sizes, the average RSE was 26.0 percent. The overall national RSE for past year use of cocaine was 24.5 percent.

Section C: Sample Sizes, Response Rates, and Population Estimates

160201
Table C1. – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by Substate Region, for Individuals Aged 12 or Older: 2012, 2013, and 2014 NSDUHs
State/Substate Region Total
Selected DUs
Total Eligible
DUs
Total
Completed
Screeners
Weighted DU
Screening
Response
Rate
(Percentage)
Total
Selected
Total
Responded
Population
Estimate
Weighted
Interview
Response
Rate
(Percentage)
Weighted
Overall
Response
Rate
(Percentage)
DU = dwelling unit; SPA = service planning area.
NOTE: For substate region definitions, see the "2012-2014 National Survey on Drug Use and Health Substate Region Definitions" at http://www.samhsa.gov/data/.
NOTE: To compute the pooled 2012-2014 weighted response rates, the three samples were combined, and the individual-year weights were used for the pooled sample. Thus, the response rates presented here are weighted across 3 years of data rather than being a simple average of the 2012, 2013, and 2014 individual response rates.
NOTE: The total responded column represents the combined sample size from the 2012, 2013, and 2014 NSDUHs.
NOTE: The population estimate is the simple average of the 2012, 2013, and 2014 population counts for persons aged 12 or older. Because of rounding, the sum of the substate region population counts within a state may not exactly match the state population count listed in the table.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2012, 2013, and 2014.
Total United States 626,362 523,186 441,803 83.96% 268,038 204,048 262,720,104 71.97% 60.43%
Northeast 139,742 118,083 94,399 78.34% 54,810 40,433 47,408,955 68.62% 53.76%
Midwest 162,920 136,982 117,579 85.62% 70,864 53,789 56,241,587 72.32% 61.92%
South 197,620 162,556 139,570 86.12% 83,229 64,449 97,650,147 73.32% 63.14%
West 126,080 105,565 90,255 83.26% 59,135 45,377 61,419,415 72.10% 60.03%
Alabama 8,762 6,977 6,012 85.73% 3,573 2,765 4,046,816 71.86% 61.61%
Region 1 2,332 1,869 1,578 83.63% 856 634 1,140,380 68.91% 57.63%
Region 2 2,649 2,149 1,790 82.80% 1,113 833 1,294,847 69.88% 57.87%
Region 3 2,041 1,578 1,433 90.82% 874 716 719,075 76.62% 69.58%
Region 4 1,740 1,381 1,211 87.53% 730 582 892,515 73.59% 64.41%
Alaska 8,586 6,562 5,636 85.94% 3,584 2,639 581,260 72.00% 61.87%
Anchorage 3,220 2,793 2,392 85.72% 1,515 1,112 239,664 72.27% 61.95%
Northern 1,837 1,279 1,101 86.14% 775 563 128,698 69.11% 59.53%
South Central 2,627 1,812 1,520 84.10% 934 678 152,325 70.82% 59.56%
Southeast 902 678 623 91.56% 360 286 60,572 79.85% 73.11%
Arizona 8,298 6,379 5,578 87.44% 3,578 2,775 5,450,630 73.81% 64.54%
Maricopa 4,626 3,867 3,407 88.11% 2,245 1,708 3,244,780 71.15% 62.69%
Pima 1,358 1,017 866 85.94% 531 422 844,533 80.72% 69.37%
Rural North 1,065 750 655 86.56% 343 255 621,580 70.50% 61.02%
Rural South 1,249 745 650 86.83% 459 390 739,737 80.64% 70.03%
Arkansas 8,171 6,684 6,020 89.88% 3,667 2,785 2,436,479 71.92% 64.64%
Catchment Area 1 1,234 1,062 931 87.52% 665 495 394,933 70.19% 61.43%
Catchment Area 2 1,047 786 701 89.50% 381 292 296,809 76.56% 68.52%
Catchment Area 3 1,072 902 834 92.17% 513 398 325,708 72.47% 66.80%
Catchment Area 4 785 654 585 89.38% 367 279 215,114 69.78% 62.37%
Catchment Area 5 1,140 891 806 90.23% 533 393 363,227 70.04% 63.20%
Catchment Area 6 680 548 517 94.30% 284 243 174,036 85.38% 80.51%
Catchment Area 7 682 579 552 95.21% 316 249 189,808 75.38% 71.77%
Catchment Area 8 1,531 1,262 1,094 86.41% 608 436 476,844 64.66% 55.87%
California 29,722 26,482 21,146 79.61% 16,046 12,001 31,788,545 70.15% 55.85%
Region 1R 785 659 574 87.15% 387 304 813,670 75.19% 65.53%
Region 2R 1,101 753 634 83.94% 432 334 860,841 72.71% 61.03%
Region 3R (Sacramento) 1,067 979 800 81.88% 528 421 1,202,211 78.44% 64.23%
Region 4R 1,461 1,297 1,097 84.62% 690 536 1,108,749 78.65% 66.55%
Region 5R (San Francisco) 731 636 420 65.81% 293 197 746,157 57.61% 37.91%
Region 6 (Santa Clara) 1,314 1,243 984 79.34% 678 503 1,533,142 67.17% 53.29%
Region 7R (Contra Costa) 840 777 648 83.29% 453 341 906,155 72.27% 60.20%
Region 8R (Alameda) 1,326 1,194 958 80.38% 656 489 1,309,079 70.28% 56.49%
Region 9R (San Mateo) 665 612 506 82.64% 315 230 622,271 63.95% 52.85%
Region 10 908 822 693 84.33% 569 400 1,061,304 63.80% 53.81%
Region 11 (Los Angeles) 7,544 6,956 5,507 79.27% 4,410 3,188 8,347,839 67.45% 53.47%
LA SPA 1 and 5 779 717 546 76.40% 375 271 895,830 68.25% 52.14%
LA SPA 2 1,767 1,643 1,211 73.77% 955 680 1,834,499 65.93% 48.64%
LA SPA 3 1,198 1,119 889 79.95% 756 523 1,490,785 66.40% 53.09%
LA SPA 4 1,008 892 670 74.88% 471 326 977,073 65.62% 49.14%
LA SPA 6 612 545 478 88.12% 431 324 800,750 67.75% 59.70%
LA SPA 7 879 837 723 86.30% 692 522 1,063,490 69.09% 59.62%
LA SPA 8 1,301 1,203 990 82.07% 730 542 1,285,413 69.68% 57.19%
Region 12R 467 411 341 83.07% 258 194 702,429 68.88% 57.21%
Regions 13 and 19R 1,746 1,417 1,115 74.88% 1,019 796 2,006,173 74.09% 55.48%
Region 14 (Orange) 2,395 2,201 1,638 73.62% 1,289 931 2,609,808 66.61% 49.04%
Region 15R (Fresno) 634 587 476 81.13% 417 328 763,896 77.00% 62.47%
Region 16R (San Diego) 2,594 2,313 1,753 75.92% 1,237 936 2,682,797 70.80% 53.75%
Region 17R 1,185 1,077 883 81.84% 750 583 1,182,387 75.38% 61.69%
Region 18R (San Bernardino) 1,353 1,177 971 82.79% 877 686 1,694,198 75.16% 62.23%
Region 20R 786 708 601 85.05% 459 362 773,194 69.58% 59.18%
Region 21R 820 663 547 82.26% 329 242 862,248 67.52% 55.54%
Colorado 8,468 7,269 6,060 83.34% 3,718 2,820 4,358,205 73.03% 60.86%
Region 1 943 815 680 83.04% 433 322 575,705 75.63% 62.81%
Regions 2 and 7 4,454 4,002 3,221 80.57% 2,046 1,556 2,441,220 73.05% 58.86%
Region 3 1,438 1,205 1,026 84.20% 626 489 644,112 72.41% 60.97%
Region 4 671 475 440 93.15% 231 187 232,934 77.59% 72.27%
Regions 5 and 6 962 772 693 89.52% 382 266 464,234 68.02% 60.90%
Connecticut 8,634 7,710 6,398 82.78% 3,897 2,837 3,044,939 69.23% 57.30%
Eastern 1,227 1,046 896 85.48% 543 423 372,631 74.22% 63.44%
North Central 2,475 2,264 1,881 82.87% 1,098 795 852,501 68.21% 56.52%
Northwestern 1,427 1,257 1,049 83.33% 680 489 526,119 66.62% 55.52%
South Central 2,053 1,856 1,578 84.96% 908 667 713,456 72.93% 61.96%
Southwest 1,452 1,287 994 76.75% 668 463 580,232 64.91% 49.81%
Delaware 8,661 7,178 5,936 82.68% 3,487 2,706 777,549 75.35% 62.30%
Kent 1,541 1,353 1,145 84.34% 724 561 139,696 73.87% 62.30%
New Castle (excluding Wilmington City) 4,205 3,805 3,108 81.62% 1,853 1,423 400,243 75.14% 61.33%
Sussex 2,392 1,586 1,313 82.96% 692 539 174,945 74.07% 61.45%
Wilmington City 523 434 370 85.64% 218 183 62,665 87.56% 74.98%
District of Columbia 14,851 12,364 9,829 78.97% 3,486 2,804 554,678 76.32% 60.27%
Ward 1 1,863 1,624 1,326 82.09% 436 343 69,539 70.83% 58.15%
Ward 2 1,676 1,257 893 69.90% 306 242 73,677 80.00% 55.92%
Ward 3 2,056 1,642 1,334 80.33% 448 352 72,326 78.50% 63.05%
Ward 4 1,871 1,641 1,286 78.48% 460 352 68,750 69.80% 54.78%
Ward 5 2,130 1,812 1,425 77.39% 522 414 69,786 75.01% 58.06%
Ward 6 2,180 1,853 1,530 82.42% 439 349 73,115 74.97% 61.79%
Ward 7 1,200 1,018 810 78.90% 316 279 64,161 87.09% 68.72%
Ward 8 1,875 1,517 1,225 80.99% 559 473 63,324 80.38% 65.10%
Florida 37,211 29,333 24,515 82.79% 13,756 10,524 16,632,820 70.85% 58.66%
Broward (Circuit 17) 2,988 2,420 1,962 80.56% 1,123 872 1,536,230 73.94% 59.56%
Central I 4,368 3,768 3,302 87.19% 2,080 1,579 2,090,644 69.63% 60.71%
Circuit 9 2,492 2,164 1,927 88.76% 1,323 1,046 1,238,291 74.56% 66.18%
Circuit 18 1,876 1,604 1,375 85.35% 757 533 852,353 63.27% 54.00%
Central II 11,058 8,108 6,720 82.25% 3,504 2,674 4,657,305 70.97% 58.37%
Circuit 6 2,823 2,163 1,744 79.37% 883 657 1,229,871 68.82% 54.62%
Circuit 10 1,230 1,020 889 87.26% 530 419 630,086 76.04% 66.35%
Circuit 12 1,969 1,431 1,209 84.45% 500 360 668,199 68.97% 58.24%
Circuit 13 (Hillsborough) 2,356 1,973 1,594 80.68% 964 762 1,085,807 73.95% 59.66%
Circuit 20 2,680 1,521 1,284 82.77% 627 476 1,043,341 68.77% 56.93%
Northeast 7,078 5,718 5,019 87.65% 2,599 1,956 3,121,138 69.54% 60.95%
Circuit 4 1,989 1,682 1,408 83.16% 768 579 963,449 70.58% 58.69%
Circuit 5 2,072 1,601 1,420 88.43% 682 528 934,745 71.43% 63.16%
Circuit 7 1,894 1,558 1,386 89.07% 672 490 762,973 64.19% 57.17%
Circuit 8 plus Columbia, Dixie, Hamilton,
   Lafayette, and Suwannee
1,123 877 805 92.03% 477 359 459,971 74.26% 68.34%
Northwest 3,003 2,357 1,797 76.28% 994 778 1,234,666 74.17% 56.58%
Circuit 1 1,319 1,147 746 66.54% 432 318 601,463 66.92% 44.53%
Circuit 2 plus Madison and Taylor 937 753 658 88.21% 379 317 377,593 81.17% 71.60%
Circuit 14 747 457 393 80.57% 183 143 255,610 75.39% 60.74%
South (Circuits 11 and 16) 4,425 3,725 3,045 78.64% 1,962 1,554 2,285,489 74.04% 58.23%
Southeast 4,291 3,237 2,670 81.72% 1,494 1,111 1,707,347 66.67% 54.48%
Circuit 15 (Palm Beach) 3,046 2,259 1,809 79.21% 1,080 789 1,172,607 64.67% 51.23%
Circuit 19 1,245 978 861 88.06% 414 322 534,740 72.15% 63.54%
Georgia 8,718 7,349 6,199 84.56% 4,266 3,286 8,151,672 73.51% 62.16%
Region 1 2,041 1,759 1,500 85.14% 1,013 755 2,100,001 73.11% 62.25%
Region 2 1,364 1,130 980 87.15% 636 494 1,052,144 74.39% 64.83%
Region 3 2,569 2,237 1,840 82.76% 1,377 1,039 2,431,810 71.52% 59.19%
Region 4 725 592 496 83.46% 314 254 505,270 74.40% 62.09%
Region 5 879 674 556 82.76% 367 296 923,299 74.83% 61.93%
Region 6 1,140 957 827 86.66% 559 448 1,139,148 76.75% 66.51%
Hawaii 9,448 8,091 6,408 78.67% 3,864 2,830 1,145,942 69.09% 54.35%
Hawaii Island 1,544 1,252 981 78.19% 488 366 154,193 70.27% 54.95%
Honolulu 6,222 5,470 4,257 77.21% 2,584 1,864 806,141 67.94% 52.46%
Kauai 677 525 473 90.11% 296 231 56,012 76.88% 69.28%
Maui 1,005 844 697 82.93% 496 369 129,596 70.21% 58.22%
Idaho 6,620 5,649 5,161 91.10% 3,566 2,815 1,310,838 76.44% 69.64%
Region 1 987 831 742 88.71% 438 333 184,241 76.63% 67.97%
Region 2 466 389 362 92.52% 221 166 92,647 71.29% 65.96%
Region 3 1,051 912 842 92.06% 620 486 207,040 71.83% 66.13%
Region 4 1,728 1,586 1,401 88.37% 940 753 374,859 76.51% 67.61%
Region 5 891 749 693 92.01% 491 385 151,475 77.37% 71.18%
Region 6 455 390 366 93.64% 271 210 98,121 78.77% 73.75%
Region 7 1,042 792 755 95.22% 585 482 202,455 81.08% 77.21%
Illinois 30,056 26,209 19,997 76.05% 13,294 9,572 10,710,970 68.05% 51.75%
Region I (Cook) 12,222 10,706 7,185 67.12% 4,935 3,480 4,357,973 67.49% 45.29%
Region II 8,626 7,815 6,321 80.05% 4,541 3,266 3,378,920 67.75% 54.24%
Region III 3,602 3,055 2,558 83.97% 1,612 1,199 1,212,886 69.93% 58.72%
Region IV 2,660 2,231 1,898 85.35% 1,034 742 757,300 68.73% 58.67%
Region V 2,946 2,402 2,035 83.96% 1,172 885 1,003,891 68.63% 57.62%
Indiana 7,987 6,701 5,885 87.79% 3,630 2,772 5,432,794 72.26% 63.44%
Central 1,882 1,627 1,377 84.36% 854 621 1,429,110 67.43% 56.88%
East 793 683 622 91.03% 373 273 459,906 71.27% 64.88%
North Central 1,316 1,070 956 89.32% 607 483 767,055 78.67% 70.27%
Northeast 727 612 533 86.69% 340 262 535,186 74.18% 64.30%
Northwest 917 799 686 85.81% 442 342 618,458 73.37% 62.96%
Southeast 843 717 645 90.17% 364 282 584,474 69.24% 62.44%
Southwest 624 542 490 90.86% 283 219 426,617 70.88% 64.40%
West 885 651 576 88.63% 367 290 611,988 74.71% 66.22%
Iowa 7,725 6,618 5,993 90.47% 3,541 2,712 2,573,570 72.51% 65.60%
Central 1,412 1,231 1,086 88.70% 666 489 463,423 69.62% 61.75%
North Central 911 743 676 90.81% 380 299 286,618 76.42% 69.40%
Northeast 1,825 1,620 1,476 90.96% 855 654 626,017 72.66% 66.09%
Northwest 1,061 903 833 92.20% 473 378 393,835 73.92% 68.16%
Southeast 1,590 1,365 1,228 89.82% 746 571 545,834 70.29% 63.14%
Southwest 926 756 694 91.47% 421 321 257,843 75.86% 69.39%
Kansas 7,510 6,379 5,626 88.04% 3,570 2,781 2,349,071 74.93% 65.97%
Kansas City Metro 2,320 2,074 1,808 87.05% 1,223 938 790,271 75.19% 65.46%
Northeast 1,379 1,198 972 80.66% 628 491 452,635 72.60% 58.56%
South Central 1,043 873 811 92.78% 523 417 291,751 78.84% 73.14%
Southeast 670 445 407 91.13% 222 173 156,861 70.58% 64.32%
West 861 714 649 90.90% 345 259 255,068 69.56% 63.23%
Wichita (Sedgwick) 1,237 1,075 979 91.16% 629 503 402,484 78.30% 71.37%
Kentucky 8,493 7,043 6,370 90.30% 3,628 2,777 3,634,802 72.08% 65.08%
Adanta, Cumberland River, and Lifeskills 1,408 1,121 1,002 89.73% 577 446 610,288 76.26% 68.43%
Bluegrass, Comprehend, and North Key 2,600 2,216 2,016 90.65% 1,126 844 1,064,608 70.69% 64.08%
Communicare and River Valley 1,053 886 816 91.91% 455 336 403,536 67.82% 62.34%
Four Rivers and Pennyroyal 719 569 521 91.28% 287 222 343,249 74.25% 67.77%
Kentucky River, Mountain, and Pathways 897 691 622 89.45% 358 279 408,053 70.21% 62.80%
Seven Counties 1,816 1,560 1,393 89.29% 825 650 805,068 73.75% 65.85%
Louisiana 8,053 6,451 5,815 89.79% 3,562 2,796 3,772,866 74.77% 67.13%
Regions 1 and 10 1,531 1,213 1,084 88.30% 655 493 727,409 71.53% 63.16%
Region 1 724 497 453 90.84% 262 201 370,733 78.77% 71.55%
Region 10 (Jefferson) 807 716 631 87.10% 393 292 356,676 67.67% 58.94%
Regions 2 and 9 2,152 1,776 1,605 90.25% 1,039 830 1,001,912 77.91% 70.32%
Region 3 650 563 511 90.72% 327 250 328,066 75.06% 68.10%
Regions 4, 5, and 6 2,073 1,616 1,438 88.68% 858 654 972,236 71.14% 63.08%
Regions 7 and 8 1,647 1,283 1,177 91.80% 683 569 743,243 77.93% 71.54%
Maine 10,832 7,930 7,135 89.88% 3,489 2,804 1,151,999 77.55% 69.71%
Aroostook/Downeast 1,598 923 869 94.11% 390 323 137,788 82.13% 77.30%
Aroostook 721 523 490 93.48% 230 205 61,785 88.67% 82.89%
Downeast 877 400 379 94.85% 160 118 76,003 74.01% 70.19%
Central 1,525 1,148 1,041 90.59% 460 375 149,949 80.46% 72.89%
Cumberland 2,110 1,836 1,576 85.83% 824 634 247,034 74.11% 63.61%
Midcoast 1,075 730 662 90.62% 277 235 128,140 84.23% 76.32%
Penquis 1,513 1,067 985 92.20% 479 401 148,493 80.43% 74.16%
Western 1,517 1,120 1,007 89.80% 530 447 168,151 80.07% 71.91%
York 1,494 1,106 995 89.60% 529 389 172,445 67.00% 60.03%
Maryland 7,922 6,989 5,478 78.16% 3,554 2,770 4,947,177 74.97% 58.59%
Anne Arundel 853 743 573 76.53% 361 282 464,152 74.53% 57.04%
Baltimore City 1,037 865 682 78.96% 434 367 521,363 84.61% 66.81%
Baltimore County 990 899 664 73.25% 423 311 693,237 74.43% 54.52%
Montgomery 1,219 1,133 860 74.73% 549 425 838,186 74.66% 55.79%
North Central 489 452 349 77.23% 236 186 392,997 77.02% 59.48%
Northeast 644 554 468 84.51% 331 251 415,119 66.42% 56.13%
Prince George's 1,037 940 715 76.27% 492 359 729,740 69.37% 52.91%
South 901 754 615 81.95% 365 304 473,748 77.25% 63.31%
West 752 649 552 85.37% 363 285 418,634 76.17% 65.02%
Massachusetts 9,019 7,886 6,465 81.81% 3,930 2,852 5,714,249 69.08% 56.51%
Boston 1,040 942 764 80.83% 531 375 710,161 68.43% 55.31%
Central 1,139 1,028 852 82.59% 486 330 736,188 64.21% 53.03%
Metrowest 1,766 1,646 1,312 79.54% 737 529 1,323,939 68.31% 54.34%
Northeast 1,767 1,646 1,347 81.62% 836 628 1,121,285 69.80% 56.97%
Southeast 1,979 1,510 1,270 84.10% 784 541 1,097,752 67.13% 56.45%
Western 1,328 1,114 920 82.34% 556 449 724,925 77.93% 64.17%
Michigan 30,130 24,549 20,634 83.90% 12,591 9,709 8,345,968 73.13% 61.36%
Region 1 1,332 891 795 89.45% 396 334 271,878 80.74% 72.22%
Region 2 1,839 1,102 942 85.31% 460 345 438,239 76.74% 65.46%
Region 3 3,398 2,948 2,539 85.53% 1,660 1,326 1,021,185 75.96% 64.96%
Region 4 2,509 2,099 1,884 89.13% 1,141 900 705,626 75.35% 67.16%
Region 5 5,074 3,945 3,420 86.39% 1,996 1,625 1,403,699 79.16% 68.39%
Region 6 2,450 2,072 1,790 85.83% 1,182 919 671,218 73.65% 63.22%
Region 7 5,034 4,021 3,160 78.41% 2,029 1,539 1,480,129 69.16% 54.23%
Region 8 3,431 3,091 2,517 81.56% 1,513 1,106 1,039,135 70.06% 57.14%
Region 9 2,489 2,287 1,782 78.43% 1,164 824 718,525 66.54% 52.18%
Region 10 2,574 2,093 1,805 86.37% 1,050 791 596,334 70.16% 60.60%
Minnesota 7,453 6,543 5,856 89.65% 3,484 2,775 4,518,658 77.91% 69.85%
Regions 1 and 2 1,041 751 667 88.92% 308 225 453,480 69.66% 61.94%
Regions 3 and 4 1,164 1,004 908 90.34% 508 422 771,938 84.30% 76.16%
Regions 5 and 6 1,295 1,128 1,049 93.08% 638 489 845,719 72.04% 67.06%
Region 7A (Hennepin) 1,495 1,382 1,195 86.48% 735 589 1,002,294 80.76% 69.84%
Region 7B (Ramsey) 639 579 519 89.58% 335 284 439,425 81.56% 73.06%
Region 7C 1,819 1,699 1,518 89.89% 960 766 1,005,802 77.94% 70.07%
Mississippi 7,193 5,820 5,278 90.52% 3,358 2,728 2,430,687 78.04% 70.64%
Region 1 1,506 1,246 1,133 90.79% 741 625 551,657 82.59% 74.99%
Region 2 928 724 651 89.26% 416 341 300,761 75.13% 67.07%
Region 3 1,233 974 905 92.38% 522 437 341,670 78.89% 72.88%
Region 4 1,170 979 843 86.18% 553 423 447,201 74.10% 63.87%
Region 5 384 273 258 94.23% 163 121 148,199 74.01% 69.74%
Region 6 923 763 728 95.30% 484 405 251,105 79.93% 76.18%
Region 7 1,049 861 760 88.28% 479 376 390,092 76.94% 67.92%
Missouri 8,601 7,111 6,357 89.24% 3,550 2,766 5,009,763 74.40% 66.40%
Central 1,224 1,031 939 91.03% 542 432 680,338 75.32% 68.56%
Eastern 2,930 2,581 2,320 89.70% 1,252 961 1,742,375 71.94% 64.54%
Eastern (St. Louis City and County) 1,884 1,623 1,463 89.85% 748 572 1,099,907 71.48% 64.23%
Eastern (excluding St. Louis) 1,046 958 857 89.47% 504 389 642,468 72.61% 64.96%
Northwest 1,921 1,578 1,443 91.29% 815 647 1,211,706 77.74% 70.97%
Northwest (Jackson) 1,046 871 798 91.46% 447 359 554,310 78.70% 71.98%
Northwest (excluding Jackson) 875 707 645 91.09% 368 288 657,395 76.66% 69.83%
Southeast 1,139 825 740 89.99% 453 354 595,780 75.46% 67.91%
Southwest 1,387 1,096 915 82.99% 488 372 779,564 73.80% 61.25%
Montana 9,115 7,309 6,702 91.57% 3,573 2,763 854,275 74.73% 68.44%
Region 1 711 495 474 95.72% 238 196 67,224 85.72% 82.05%
Region 2 1,441 1,102 1,019 92.11% 518 430 120,568 82.51% 75.99%
Region 3 2,141 1,765 1,587 89.71% 833 647 176,644 75.19% 67.45%
Region 4 2,022 1,622 1,476 90.98% 840 627 226,881 72.63% 66.08%
Region 5 2,800 2,325 2,146 92.21% 1,144 863 262,959 69.80% 64.36%
Nebraska 8,067 6,777 6,139 90.37% 3,584 2,788 1,533,852 73.60% 66.51%
Regions 1 and 2 856 643 596 92.36% 358 268 156,367 68.28% 63.07%
Region 1 528 375 345 91.68% 222 170 73,199 69.69% 63.89%
Region 2 328 268 251 93.35% 136 98 83,168 65.53% 61.17%
Region 3 1,258 1,017 945 92.76% 516 431 188,369 78.66% 72.97%
Region 4 775 616 579 93.94% 313 240 170,429 72.62% 68.22%
Region 5 2,041 1,761 1,604 90.96% 936 735 378,877 72.52% 65.96%
Region 6 3,137 2,740 2,415 87.80% 1,461 1,114 639,810 74.10% 65.06%
Nevada 7,528 6,211 5,317 84.69% 3,550 2,796 2,316,939 74.33% 62.95%
Clark – Region 1 5,197 4,268 3,631 84.05% 2,391 1,897 1,667,917 74.56% 62.66%
Region 3 1,046 827 744 89.49% 436 322 285,362 73.04% 65.36%
Capital District 374 325 289 87.56% 169 124 135,596 71.30% 62.42%
Rural/Frontier 672 502 455 90.72% 267 198 149,766 74.35% 67.45%
Washoe – Region 2 1,285 1,116 942 83.69% 723 577 363,660 74.45% 62.30%
New Hampshire 9,522 7,865 6,744 85.70% 3,790 2,835 1,141,956 72.55% 62.17%
Central 2,864 2,397 2,091 87.18% 1,192 936 324,468 74.51% 64.96%
Central 1 1,318 1,138 1,009 88.56% 622 486 160,191 72.47% 64.18%
Central 2 1,546 1,259 1,082 85.91% 570 450 164,278 76.61% 65.82%
Northern 1,613 974 851 87.08% 433 330 148,840 76.58% 66.69%
Southern 5,045 4,494 3,802 84.62% 2,165 1,569 668,648 70.81% 59.92%
Southern 1 (Rockingham) 2,077 1,880 1,613 85.75% 902 623 257,005 68.28% 58.55%
Southern 2 2,968 2,614 2,189 83.81% 1,263 946 411,642 72.67% 60.91%
New Jersey 10,189 8,746 7,167 82.65% 4,560 3,347 7,480,144 70.73% 58.46%
Central 2,417 2,019 1,656 82.65% 1,030 755 1,718,758 69.87% 57.74%
Metropolitan 2,628 2,283 1,821 80.65% 1,147 839 1,797,944 68.89% 55.56%
Northern 3,085 2,690 2,222 82.99% 1,435 1,041 2,403,370 71.91% 59.67%
Southern 2,059 1,754 1,468 84.58% 948 712 1,560,072 72.32% 61.17%
New Mexico 7,952 6,052 5,482 90.44% 3,441 2,760 1,718,212 76.10% 68.82%
Region 1 1,461 1,178 1,081 91.13% 696 576 356,636 79.56% 72.50%
Region 2 1,322 926 821 88.69% 449 358 248,167 72.89% 64.64%
Region 3 (Bernalillo) 2,349 1,932 1,711 88.50% 1,061 846 559,454 76.53% 67.73%
Region 4 1,096 896 815 91.42% 562 435 214,977 73.31% 67.02%
Region 5 1,724 1,120 1,054 93.60% 673 545 338,978 76.14% 71.26%
New York 40,767 35,101 24,961 70.60% 15,350 10,601 16,622,553 64.06% 45.23%
Region A 17,099 15,079 9,526 62.85% 6,219 4,014 7,075,349 59.55% 37.43%
Region 1 2,717 2,441 1,775 72.86% 1,264 940 1,152,598 72.11% 52.54%
Region 2 6,020 5,355 3,378 63.02% 2,198 1,348 2,539,400 56.31% 35.49%
Region 3 4,222 3,606 2,092 57.05% 1,139 761 1,441,473 64.88% 37.02%
Region 4 4,140 3,677 2,281 61.97% 1,618 965 1,941,877 52.80% 32.72%
Region B 9,019 8,127 5,652 68.38% 3,732 2,475 4,301,696 60.93% 41.67%
Region 5 4,871 4,422 3,187 69.91% 2,218 1,443 2,422,547 60.26% 42.13%
Region 6 2,422 2,234 1,402 63.12% 880 570 1,155,102 59.61% 37.63%
Region 7 1,726 1,471 1,063 71.90% 634 462 724,048 65.70% 47.24%
Region C 10,334 8,887 7,217 80.93% 4,081 3,114 3,921,179 72.80% 58.91%
Region 8 2,127 1,824 1,454 79.28% 842 603 858,772 71.52% 56.70%
Region 9 2,408 2,084 1,717 82.26% 963 751 814,993 73.74% 60.66%
Region 10 841 688 556 80.18% 313 252 378,512 75.69% 60.69%
Region 11 2,384 2,054 1,667 80.67% 966 738 898,353 73.36% 59.18%
Region 12 2,574 2,237 1,823 81.49% 997 770 970,550 71.76% 58.48%
Region D 4,315 3,008 2,566 84.71% 1,318 998 1,324,329 71.82% 60.84%
Region 13 1,856 1,131 964 84.64% 461 349 408,192 72.42% 61.29%
Region 14 1,024 688 581 84.56% 327 248 456,356 70.23% 59.39%
Region 15 1,435 1,189 1,021 84.86% 530 401 459,781 72.26% 61.32%
North Carolina 9,905 8,071 7,052 87.44% 4,176 3,330 8,112,661 76.00% 66.45%
Alliance Behavioral Healthcare 1 803 675 612 90.47% 392 334 636,566 84.01% 76.00%
Alliance Behavioral Healthcare 2 894 781 701 89.55% 388 313 782,617 75.45% 67.57%
Cardinal Innovations Healthcare Solutions 1 829 707 578 82.76% 328 246 619,013 72.63% 60.11%
Cardinal Innovations Healthcare Solutions 2 772 671 617 91.80% 361 295 557,384 80.56% 73.95%
Cardinal Innovations Healthcare Solutions 3 753 657 530 80.49% 361 259 789,444 70.64% 56.86%
CenterPoint Human Services 559 487 410 84.26% 274 237 450,760 83.11% 70.03%
Eastpointe 844 707 641 90.76% 379 299 679,883 70.37% 63.86%
Partners Behavioral Health Management 987 828 725 87.81% 468 370 757,835 75.17% 66.01%
Sandhills Center 1 428 321 272 82.86% 153 124 467,131 73.34% 60.77%
Sandhills Center 2 595 493 420 85.36% 235 191 419,943 83.68% 71.42%
Smoky Mountain Center 1 530 345 308 89.18% 185 141 457,963 68.83% 61.38%
Smoky Mountain Center 2 713 543 473 87.35% 191 153 451,426 78.59% 68.65%
Trillium Health Resources 1 544 396 348 89.35% 232 191 516,013 80.51% 71.93%
Trillium Health Resources 2 654 460 417 89.98% 229 177 526,682 64.99% 58.47%
North Dakota 10,051 7,763 7,159 92.06% 3,653 2,809 594,462 73.12% 67.32%
Badlands and West Central 2,569 1,951 1,860 95.33% 884 702 159,784 76.21% 72.65%
Lake Region 875 662 598 90.29% 273 209 33,321 72.96% 65.88%
North Central 1,336 910 847 92.62% 426 329 82,028 75.32% 69.75%
Northeast 1,332 1,092 987 90.60% 570 437 76,284 69.58% 63.04%
Northwest 568 372 359 96.40% 209 139 30,735 64.55% 62.22%
South Central 944 704 650 92.46% 263 207 48,377 75.32% 69.64%
Southeast 2,427 2,072 1,858 89.04% 1,028 786 163,934 72.37% 64.44%
Ohio 29,584 25,253 22,004 86.76% 12,898 9,670 9,674,385 71.18% 61.75%
Boards 2, 46, 55, and 68 1,050 907 764 84.07% 481 343 428,541 66.48% 55.89%
Boards 3, 52, and 85 966 875 768 87.45% 410 304 318,153 72.48% 63.38%
Boards 4 and 78 791 669 611 91.30% 285 222 262,041 77.42% 70.68%
Boards 5 and 60 750 636 601 94.04% 400 324 286,254 79.15% 74.43%
Boards 7, 15, 41, 79, and 84 1,342 1,122 1,019 88.59% 528 405 391,238 73.11% 64.77%
Boards 8, 13, and 83 1,221 1,043 883 84.89% 571 406 415,759 67.34% 57.17%
Board 9 (Butler) 804 692 586 84.66% 389 272 307,246 64.96% 54.99%
Board 12 743 622 533 85.98% 329 233 291,381 63.08% 54.24%
Boards 18 and 47 4,269 3,544 3,017 84.37% 1,640 1,299 1,316,932 75.55% 63.73%
Boards 20, 32, 54, and 69 970 877 841 95.65% 459 354 286,134 74.38% 71.14%
Boards 21, 39, 51, 70, and 80 1,352 1,213 1,087 89.92% 709 527 463,444 71.12% 63.95%
Boards 22, 74, and 87 999 769 708 91.52% 401 312 326,476 78.47% 71.82%
Boards 23 and 45 966 799 723 90.27% 464 360 314,595 72.62% 65.56%
Board 25 (Franklin) 2,945 2,559 2,101 81.65% 1,310 957 986,745 70.98% 57.95%
Boards 27, 71, and 73 1,257 1,069 964 89.18% 592 421 413,378 63.56% 56.68%
Boards 28, 43, and 67 1,223 1,026 929 90.53% 613 481 413,943 75.41% 68.27%
Board 31 (Hamilton) 2,319 1,944 1,576 81.06% 897 691 662,797 73.53% 59.60%
Board 48 (Lucas) 1,095 984 860 87.60% 507 362 363,646 65.64% 57.50%
Boards 50 and 76 1,779 1,520 1,390 91.48% 758 549 518,114 66.33% 60.68%
Board 57 (Montgomery) 1,480 1,266 1,045 82.24% 589 445 452,661 71.36% 58.69%
Board 77 (Summit) 1,263 1,117 998 88.67% 566 403 454,908 66.82% 59.24%
Oklahoma 8,049 6,536 5,882 89.95% 3,723 2,795 3,128,664 69.88% 62.86%
Central 838 744 663 88.90% 442 312 394,726 62.74% 55.77%
East Central 895 761 714 94.11% 441 333 359,859 74.73% 70.32%
Northeast 1,075 859 799 92.89% 466 356 399,926 70.47% 65.46%
Northwest and Southwest 1,229 888 803 90.43% 462 362 447,891 71.91% 65.02%
Oklahoma County 1,532 1,310 1,140 87.13% 761 557 600,549 66.34% 57.81%
Southeast 1,304 1,009 934 92.23% 591 450 426,132 71.32% 65.78%
Tulsa County 1,176 965 829 85.94% 560 425 499,582 73.20% 62.91%
Oregon 7,846 6,915 6,049 87.50% 3,576 2,776 3,337,930 75.43% 66.00%
Region 1 (Multnomah) 1,603 1,461 1,263 86.78% 700 528 656,607 74.00% 64.22%
Region 2 1,554 1,444 1,264 87.65% 863 667 785,496 73.42% 64.35%
Region 3 2,461 2,157 1,889 87.25% 1,095 865 1,040,424 78.60% 68.58%
Region 4 1,253 1,055 901 85.50% 470 370 475,661 77.22% 66.03%
Region 5 (Central) 424 361 330 91.34% 214 173 174,095 76.62% 69.98%
Region 6 (Eastern) 551 437 402 92.22% 234 173 205,647 66.95% 61.74%
Pennsylvania 32,300 27,774 22,541 80.87% 12,651 9,631 10,808,980 71.52% 57.84%
Region 1 (Allegheny) 3,364 2,963 2,474 83.47% 1,308 905 1,057,711 65.55% 54.71%
Regions 3, 8, 9, and 51 1,924 1,682 1,442 86.07% 770 574 598,769 69.67% 59.97%
Regions 4, 11, 37, and 49 2,337 1,799 1,483 82.93% 881 661 766,595 70.20% 58.21%
Regions 5, 18, 23, 24, and 46 2,068 1,793 1,563 86.33% 831 671 632,134 74.27% 64.12%
Regions 6, 12, 16, 31, 35, 45, and 47 1,758 1,492 1,328 87.85% 765 610 607,968 75.65% 66.46%
Regions 7, 13, 20, and 33 6,109 5,529 4,056 72.03% 2,312 1,737 2,105,420 70.17% 50.55%
Regions 10, 15, 27, 32, 43,and 44 1,294 1,085 984 90.54% 505 409 440,206 78.50% 71.07%
Regions 17 and 21 917 797 717 89.53% 351 288 311,928 76.52% 68.51%
Regions 19, 26, 28, and 42 3,506 3,107 2,721 87.50% 1,597 1,238 1,227,294 73.51% 64.32%
Regions 22, 38, 40, 41, and 48 2,026 1,622 1,403 86.21% 742 546 710,814 69.90% 60.26%
Regions 29 and 34 1,506 1,250 1,104 89.19% 572 416 552,687 66.13% 58.98%
Regions 30 and 50 1,606 1,358 1,131 83.50% 627 457 520,153 72.25% 60.33%
Region 36 (Philadelphia) 3,885 3,297 2,135 65.37% 1,390 1,119 1,277,300 75.58% 49.41%
Rhode Island 8,270 6,956 6,021 86.51% 3,632 2,818 898,242 73.97% 64.00%
Bristol and Newport 1,004 792 700 88.33% 363 281 114,467 73.14% 64.60%
Kent 1,217 1,087 944 87.06% 533 393 142,012 68.12% 59.31%
Providence 4,825 4,090 3,505 85.58% 2,242 1,759 532,030 75.85% 64.91%
Washington 1,224 987 872 88.17% 494 385 109,733 73.57% 64.86%
South Carolina 9,440 7,736 6,640 85.63% 3,613 2,844 3,956,936 75.58% 64.72%
Region 1 2,901 2,409 2,047 84.64% 1,079 821 1,243,056 72.37% 61.26%
Region 2 2,111 1,827 1,564 85.53% 866 696 970,647 77.23% 66.05%
Region 3 1,941 1,460 1,261 86.16% 678 515 687,638 74.18% 63.91%
Region 4 2,487 2,040 1,768 86.50% 990 812 1,055,596 78.74% 68.11%
South Dakota 7,527 6,146 5,769 93.90% 3,494 2,748 687,598 75.99% 71.35%
Region 1 1,965 1,618 1,504 93.06% 892 705 169,081 77.41% 72.03%
Region 2 702 573 538 93.98% 357 284 63,097 73.87% 69.42%
Region 3 1,879 1,436 1,347 93.78% 778 615 162,926 78.39% 73.51%
Region 4 1,007 831 793 95.48% 464 345 97,103 71.30% 68.08%
Region 5 1,974 1,688 1,587 94.00% 1,003 799 195,392 75.75% 71.20%
Tennessee 7,825 6,465 5,757 88.84% 3,430 2,767 5,424,854 77.56% 68.90%
Region 1 719 609 548 89.72% 304 234 439,416 75.15% 67.42%
Region 2 1,509 1,210 1,085 89.48% 617 493 1,015,153 79.60% 71.23%
Region 3 1,161 982 894 91.09% 549 461 812,846 79.57% 72.47%
Region 4 (Davidson) 850 697 606 86.84% 359 275 538,805 72.47% 62.93%
Region 5 1,644 1,434 1,268 88.17% 807 642 1,307,310 78.29% 69.02%
Region 6 883 649 613 94.22% 331 283 533,058 78.84% 74.28%
Region 7 (Shelby) 1,059 884 743 83.43% 463 379 778,267 75.80% 63.24%
Texas 25,375 21,395 18,731 87.41% 13,936 10,612 21,255,571 71.92% 62.86%
Region 1 1,006 866 775 89.06% 553 404 700,556 68.20% 60.74%
Region 2 530 398 373 93.51% 231 198 458,176 82.76% 77.39%
Region 3 6,321 5,621 5,151 91.49% 3,879 3,092 5,709,770 75.33% 68.92%
Region 3a 3,925 3,534 3,202 90.39% 2,382 1,867 3,636,876 73.39% 66.34%
Region 3bc 2,396 2,087 1,949 93.45% 1,497 1,225 2,072,894 78.66% 73.50%
Region 4 1,290 998 931 92.96% 614 487 937,270 73.94% 68.74%
Region 5 976 754 669 88.92% 395 287 646,406 67.17% 59.73%
Region 6 6,093 5,239 4,221 80.49% 3,267 2,213 5,140,988 63.27% 50.92%
Region 6a 5,355 4,669 3,750 80.19% 2,934 1,990 4,598,455 63.83% 51.18%
Region 6bc 738 570 471 83.00% 333 223 542,533 58.63% 48.66%
Region 7 3,330 2,635 2,382 90.10% 1,696 1,327 2,564,103 75.48% 68.01%
Region 7a 1,977 1,623 1,452 89.39% 990 754 1,622,678 73.13% 65.37%
Region 7bcd 1,353 1,012 930 91.31% 706 573 941,425 79.17% 72.29%
Region 8 2,675 2,291 1,855 81.17% 1,372 1,073 2,220,314 74.61% 60.56%
Region 9 620 501 469 93.67% 342 233 479,518 65.10% 60.98%
Region 10 734 670 621 92.22% 514 416 691,723 78.84% 72.71%
Region 11 1,800 1,422 1,284 90.26% 1,073 882 1,706,747 78.29% 70.67%
Region 11abd 1,184 904 824 91.04% 627 500 1,082,350 74.22% 67.57%
Region 11c (Hidalgo) 616 518 460 88.90% 446 382 624,398 84.43% 75.05%
Utah 5,359 4,673 4,427 94.88% 3,435 2,828 2,267,830 79.70% 75.62%
Bear River, Northeastern, Summit, Tooele, and Wasatch 690 598 561 93.69% 432 365 272,267 82.94% 77.70%
Central, Four Corners, San Juan, and Southwest 685 511 478 93.15% 326 250 275,710 72.42% 67.46%
Davis County 556 512 484 94.49% 392 340 245,606 86.09% 81.35%
Salt Lake County 2,148 1,897 1,805 95.15% 1,377 1,114 863,824 78.82% 75.00%
Utah County 893 810 763 95.11% 642 528 412,454 78.46% 74.63%
Weber, Morgan 387 345 336 97.67% 266 231 197,968 82.26% 80.34%
Vermont 10,209 8,115 6,967 85.74% 3,511 2,708 545,894 74.75% 64.09%
Champlain Valley 4,144 3,424 2,957 86.32% 1,649 1,246 218,396 72.09% 62.23%
Rural Northeast 2,130 1,738 1,389 79.77% 698 538 128,688 75.86% 60.52%
Rural Southeast 2,225 1,684 1,474 87.36% 695 539 113,325 75.60% 66.04%
Rural Southwest 1,710 1,269 1,147 90.36% 469 385 85,486 79.64% 71.96%
Virginia 9,039 7,967 6,777 85.31% 4,263 3,335 6,803,172 75.33% 64.27%
Region 1 1,305 1,161 1,042 89.78% 679 579 1,057,936 83.35% 74.84%
Region 2 2,324 2,169 1,801 83.86% 1,216 921 1,910,006 72.23% 60.57%
Region 3 1,781 1,487 1,260 84.80% 722 570 1,155,467 77.20% 65.46%
Region 4 1,521 1,324 1,124 84.20% 676 512 1,160,224 71.50% 60.21%
Region 5 2,108 1,826 1,550 85.37% 970 753 1,519,538 74.50% 63.60%
Washington 7,747 6,714 5,720 85.15% 3,634 2,763 5,809,672 72.43% 61.67%
Region 1 1,559 1,300 1,153 88.61% 729 571 1,263,639 73.45% 65.08%
Greater Columbia and North Central 895 758 664 87.56% 436 345 748,582 74.78% 65.47%
Spokane 664 542 489 90.09% 293 226 515,057 71.71% 64.60%
Region 2 3,787 3,336 2,799 83.88% 1,741 1,291 2,670,225 71.47% 59.95%
King 2,278 2,090 1,699 81.35% 1,084 792 1,700,584 71.52% 58.19%
North Sound 1,509 1,246 1,100 88.20% 657 499 969,640 71.38% 62.96%
Region 3 2,401 2,078 1,768 85.03% 1,164 901 1,875,809 73.28% 62.31%
Peninsula 407 337 301 88.58% 161 119 311,642 68.62% 60.78%
Pierce 959 819 691 84.49% 490 399 683,820 78.75% 66.54%
SW WA and Timberlands 740 681 574 84.27% 373 271 606,660 66.52% 56.06%
Thurston-Mason 295 241 202 84.08% 140 112 273,686 81.26% 68.32%
West Virginia 9,952 8,198 7,279 88.75% 3,751 2,825 1,582,742 72.65% 64.48%
Region I 634 560 496 88.82% 246 170 126,303 60.42% 53.67%
Region II 1,158 996 858 86.10% 480 372 223,427 74.41% 64.06%
Region III 808 699 615 87.86% 280 201 145,864 68.65% 60.32%
Region IV 2,229 1,852 1,645 88.47% 888 685 347,725 75.54% 66.83%
Region V 3,056 2,550 2,274 89.17% 1,119 820 450,354 71.22% 63.50%
Region VI 2,067 1,541 1,391 90.47% 738 577 289,068 76.10% 68.84%
Wisconsin 8,229 6,933 6,160 88.80% 3,575 2,687 4,810,496 72.98% 64.81%
Milwaukee 1,323 1,169 1,017 86.95% 641 468 777,283 69.49% 60.43%
Northeastern 1,789 1,495 1,326 88.60% 765 575 1,042,686 75.46% 66.86%
Northern 972 702 604 85.96% 303 228 416,255 75.10% 64.55%
Southeastern 1,511 1,358 1,202 88.51% 715 511 977,794 70.71% 62.58%
Southern 1,475 1,272 1,134 89.13% 660 515 938,374 74.09% 66.04%
Western 1,159 937 877 93.66% 491 390 658,104 73.85% 69.17%
Wyoming 9,391 7,259 6,569 90.38% 3,570 2,811 479,134 76.79% 69.40%
Judicial District 1 (Laramie) 1,579 1,361 1,208 88.71% 609 462 79,182 74.04% 65.68%
Judicial District 2 1,010 683 628 91.86% 368 308 46,192 78.18% 71.82%
Judicial District 3 1,460 1,028 981 95.33% 583 478 67,390 81.87% 78.05%
Judicial District 4 701 536 469 87.38% 250 196 32,066 70.54% 61.64%
Judicial District 5 685 545 475 86.62% 230 189 45,709 83.67% 72.47%
Judicial District 6 935 776 709 91.48% 420 315 49,732 72.47% 66.30%
Judicial District 7 (Natrona) 892 786 700 89.11% 410 325 64,704 75.61% 67.38%
Judicial District 8 915 733 675 92.06% 332 253 33,335 75.41% 69.43%
Judicial District 9 1,214 811 724 88.99% 368 285 60,825 79.07% 70.37%
160201
Table C2. – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by Substate Region, for Individuals Aged 12 to 20 and Adults Aged 18 or Older: 2012, 2013, and 2014 NSDUHs
State/Substate Region 12-20
Total
Selected
12-20
Total
Responded
12-20
Population
Estimate
12-20
Weighted
Interview
Response Rate
(Percentage)
18+
Total
Selected
18+
Total
Responded
18+
Population
Estimate
18+
Weighted
Interview
Response Rate
(Percentage)
DU = dwelling unit; SPA = service planning area.
NOTE: For substate region definitions, see the "2012-2014 National Survey on Drug Use and Health Substate Region Definitions" at http://www.samhsa.gov/data.
NOTE: To compute the pooled 2012-2014 weighted response rates, the three samples were combined, and the individual-year weights were used for the pooled sample. Thus, the response rates presented here are weighted across 3 years of data rather than being a simple average of the 2012, 2013, and 2014 individual response rates.
NOTE: The total responded column represents the combined sample size from the 2012, 2013, and 2014 NSDUHs.
NOTE: The population estimate is the simple average of the 2012, 2013, and 2014 population counts for individuals aged 12 to 20 and adults aged 18 or older. Because of rounding, the sum of the substate region population counts within a state may not exactly match the state population count listed in the table.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2012, 2013, and 2014.
Total United States 104,160 84,746 38,119,368 81.31% 191,869 141,978 237,815,123 70.96%
Northeast 21,218 16,934 6,512,384 79.01% 39,392 28,175 43,215,107 67.62%
Midwest 27,903 22,483 8,167,441 80.00% 50,412 37,251 50,845,917 71.44%
South 32,337 26,621 14,080,628 82.14% 59,359 44,748 88,292,583 72.35%
West 22,702 18,708 9,358,915 82.79% 42,706 31,804 55,461,515 70.92%
Alabama 1,341 1,111 574,793 82.57% 2,568 1,934 3,663,979 70.76%
Region 1 305 232 155,640 76.92% 619 455 1,033,386 68.16%
Region 2 410 336 184,610 81.25% 815 589 1,173,788 68.72%
Region 3 376 322 110,715 85.66% 590 475 650,749 75.61%
Region 4 250 221 123,828 87.82% 544 415 806,056 72.14%
Alaska 1,398 1,065 92,936 75.88% 2,551 1,877 521,224 71.86%
Anchorage 573 440 37,968 77.30% 1,099 806 215,669 72.15%
Northern 318 234 22,276 70.90% 551 403 114,716 69.26%
South Central 360 272 24,285 76.64% 646 466 135,963 70.27%
Southeast 147 119 8,407 79.96% 255 202 54,877 79.93%
Arizona 1,404 1,160 809,966 82.48% 2,546 1,910 4,908,587 72.64%
Maricopa 882 730 493,252 82.99% 1,615 1,177 2,908,477 69.63%
Pima 193 161 118,807 81.79% 398 308 770,148 80.19%
Rural North 149 110 88,418 71.88% 222 167 565,487 70.49%
Rural South 180 159 109,490 89.62% 311 258 664,475 79.33%
Arkansas 1,412 1,124 352,319 79.08% 2,638 1,969 2,200,019 71.24%
Catchment Area 1 255 207 64,364 82.28% 507 370 352,318 69.24%
Catchment Area 2 122 93 37,333 76.90% 287 222 271,614 76.74%
Catchment Area 3 217 173 48,480 78.10% 341 261 293,037 71.82%
Catchment Area 4 144 124 32,026 87.86% 255 184 192,768 68.08%
Catchment Area 5 190 149 53,652 77.47% 410 297 330,119 69.42%
Catchment Area 6 95 82 24,711 83.95% 214 181 157,928 85.09%
Catchment Area 7 137 110 27,022 80.27% 208 159 171,384 74.30%
Catchment Area 8 252 186 64,730 71.44% 416 295 430,851 63.94%
California 6,027 4,983 4,980,281 82.53% 11,774 8,464 28,688,457 68.80%
Region 1R 157 127 115,230 78.56% 282 221 744,769 75.14%
Region 2R 180 147 135,097 83.50% 297 225 777,854 71.36%
Region 3R (Sacramento) 185 152 185,633 78.64% 399 312 1,083,281 78.30%
Region 4R 256 205 153,980 79.86% 482 372 1,009,809 78.74%
Region 5R (San Francisco) 74 63 58,844 84.11% 264 175 713,281 56.95%
Region 6 (Santa Clara) 212 174 215,073 81.84% 522 372 1,393,058 65.72%
Region 7R (Contra Costa) 158 130 136,022 80.22% 338 246 815,822 71.35%
Region 8R (Alameda) 196 163 180,548 85.35% 515 371 1,198,818 68.83%
Region 9R (San Mateo) 100 85 77,266 85.87% 249 174 569,928 62.47%
Region 10 227 169 177,020 73.32% 408 279 957,493 62.73%
Region 11 (Los Angeles) 1,603 1,309 1,289,339 80.56% 3,279 2,257 7,553,191 66.03%
LA SPA 1 and 5 125 107 129,907 87.02% 282 192 822,243 66.59%
LA SPA 2 372 306 265,950 81.15% 695 466 1,665,774 64.25%
LA SPA 3 279 207 229,849 72.66% 540 359 1,348,424 65.39%
LA SPA 4 129 105 116,414 78.89% 383 252 907,093 64.42%
LA SPA 6 170 145 163,130 84.32% 311 221 702,729 65.87%
LA SPA 7 274 231 190,582 82.80% 504 362 944,015 67.23%
LA SPA 8 254 208 193,505 81.13% 564 405 1,162,914 68.76%
Region 12R 113 93 120,358 85.30% 172 127 625,135 67.03%
Regions 13 and 19R 492 419 365,434 85.96% 729 552 1,774,072 72.66%
Region 14 (Orange) 469 392 402,287 83.63% 953 644 2,356,009 64.55%
Region 15R (Fresno) 164 140 143,653 85.50% 319 241 673,696 75.47%
Region 16R (San Diego) 460 388 402,281 83.89% 909 660 2,442,309 69.38%
Region 17R 339 278 221,581 82.13% 471 352 1,040,107 73.95%
Region 18R (San Bernardino) 349 300 318,118 87.31% 618 462 1,491,700 73.55%
Region 20R 190 165 142,522 87.40% 311 236 682,632 67.59%
Region 21R 103 84 139,997 83.43% 257 186 785,491 66.82%
Colorado 1,408 1,153 610,116 82.60% 2,698 1,986 3,952,890 72.10%
Region 1 157 115 88,152 72.15% 328 245 521,941 75.87%
Regions 2 and 7 792 658 333,695 84.17% 1,469 1,073 2,214,402 71.80%
Region 3 237 200 94,650 85.31% 452 346 581,345 71.42%
Region 4 95 85 33,809 88.45% 161 123 211,284 76.57%
Regions 5 and 6 127 95 59,809 76.17% 288 199 423,918 67.57%
Connecticut 1,503 1,201 438,656 80.47% 2,810 1,977 2,757,464 68.14%
Eastern 204 175 58,926 87.02% 422 322 341,160 73.42%
North Central 427 337 119,098 79.65% 757 528 772,148 66.77%
Northwestern 271 221 74,616 83.58% 468 312 473,950 64.57%
South Central 346 276 99,984 81.38% 678 489 650,631 72.33%
Southwest 255 192 86,032 72.84% 485 326 519,575 64.25%
Delaware 1,397 1,144 105,479 81.75% 2,447 1,854 709,231 74.81%
Kent 302 242 20,743 78.80% 494 377 126,396 73.44%
New Castle (excluding Wilmington City) 744 601 57,534 80.61% 1,313 984 363,974 74.65%
Sussex 257 219 18,843 86.47% 488 369 161,718 73.28%
Wilmington City 94 82 8,359 87.61% 152 124 57,143 87.25%
District of Columbia 1,292 1,133 57,432 87.75% 2,477 1,915 523,865 75.59%
Ward 1 116 104 6,673 93.51% 354 271 67,110 70.34%
Ward 2 103 89 5,935 86.58% 280 218 72,780 79.77%
Ward 3 127 103 5,426 77.84% 343 265 69,420 78.28%
Ward 4 176 151 6,651 85.99% 303 217 64,027 68.67%
Ward 5 206 183 7,746 88.48% 361 269 65,600 73.92%
Ward 6 111 99 4,319 86.03% 347 264 70,224 74.19%
Ward 7 157 143 9,603 92.48% 185 161 58,182 86.80%
Ward 8 296 261 11,079 88.72% 304 250 56,522 79.54%
Florida 5,299 4,363 2,092,834 82.19% 9,870 7,306 15,244,962 69.75%
Broward (Circuit 17) 435 360 195,904 83.52% 782 588 1,399,241 72.85%
Central I 830 681 294,041 81.88% 1,475 1,086 1,899,753 68.44%
Circuit 9 535 450 184,932 84.11% 943 728 1,121,864 73.56%
Circuit 18 295 231 109,108 77.78% 532 358 777,889 62.03%
Central II 1,335 1,102 553,444 81.70% 2,495 1,829 4,280,251 69.89%
Circuit 6 345 281 134,224 80.69% 615 432 1,136,991 67.19%
Circuit 10 233 189 83,377 82.11% 353 276 573,539 75.36%
Circuit 12 163 128 68,987 77.57% 376 261 620,143 68.39%
Circuit 13 (Hillsborough) 370 313 154,192 83.72% 709 540 983,526 72.83%
Circuit 20 224 191 112,665 82.59% 442 320 966,053 67.88%
Northeast 1,014 816 390,397 79.63% 1,869 1,354 2,866,045 68.39%
Circuit 4 287 236 132,579 82.49% 558 401 873,523 69.18%
Circuit 5 280 231 97,652 80.98% 469 350 866,597 70.51%
Circuit 7 252 206 93,805 81.57% 492 341 700,237 62.66%
Circuit 8 plus Columbia, Dixie, Hamilton,
   Lafayette, and Suwannee
195 143 66,361 70.71% 350 262 425,688 74.02%
Northwest 378 308 170,611 83.53% 737 570 1,132,870 73.32%
Circuit 1 176 138 82,028 78.59% 312 219 549,058 65.07%
Circuit 2 plus Madison and Taylor 137 120 57,444 91.52% 289 242 349,303 80.66%
Circuit 14 65 50 31,138 78.33% 136 109 234,509 75.66%
South (Circuits 11 and 16) 766 658 288,055 86.82% 1,435 1,109 2,097,732 73.05%
Southeast 541 438 200,383 80.64% 1,077 770 1,569,071 65.44%
Circuit 15 (Palm Beach) 395 318 137,395 79.77% 782 546 1,078,349 63.33%
Circuit 19 146 120 62,988 83.37% 295 224 490,722 71.22%
Georgia 1,611 1,308 1,268,845 81.47% 3,101 2,341 7,316,745 72.65%
Region 1 393 313 325,139 80.53% 726 530 1,874,488 72.38%
Region 2 231 186 167,580 80.31% 466 356 949,652 73.90%
Region 3 526 427 364,507 81.77% 1,002 734 2,185,513 70.37%
Region 4 116 93 81,766 76.07% 224 177 455,203 73.31%
Region 5 117 102 144,719 88.56% 288 230 835,534 73.94%
Region 6 228 187 185,133 82.78% 395 314 1,016,355 76.10%
Hawaii 1,406 1,121 147,189 80.19% 2,807 1,991 1,048,507 68.14%
Hawaii Island 188 153 20,051 81.64% 350 254 140,541 69.12%
Honolulu 931 732 103,868 79.43% 1,883 1,315 738,904 66.95%
Kauai 122 102 6,939 85.54% 199 151 51,085 76.26%
Maui 165 134 16,332 79.67% 375 271 117,976 69.57%
Idaho 1,401 1,168 209,051 83.60% 2,564 1,957 1,168,987 75.25%
Region 1 162 130 25,696 80.96% 325 240 166,188 75.85%
Region 2 80 62 13,683 76.86% 170 127 85,384 70.85%
Region 3 266 223 36,161 84.93% 436 327 182,126 69.83%
Region 4 332 285 55,911 86.21% 701 544 335,312 75.30%
Region 5 229 188 24,714 79.59% 321 240 134,420 76.26%
Region 6 100 81 16,208 83.18% 198 148 87,108 77.69%
Region 7 232 199 36,678 85.97% 413 331 178,449 80.15%
Illinois 5,100 3,985 1,562,197 78.10% 9,568 6,635 9,671,148 66.93%
Region I (Cook) 1,805 1,399 599,958 77.24% 3,668 2,464 3,959,568 66.25%
Region II 1,873 1,460 537,681 78.15% 3,122 2,160 3,002,794 66.51%
Region III 611 484 182,831 79.36% 1,172 855 1,105,771 69.26%
Region IV 377 284 104,606 75.93% 742 525 688,792 68.21%
Region V 434 358 137,120 82.00% 864 631 914,223 67.37%
Indiana 1,390 1,112 805,744 79.10% 2,620 1,960 4,891,833 71.42%
Central 301 225 207,789 73.09% 638 460 1,279,161 66.74%
East 126 98 68,097 78.33% 287 207 418,235 70.87%
North Central 234 194 114,701 80.03% 431 334 688,476 78.08%
Northeast 139 111 82,735 80.48% 235 175 476,675 72.79%
Northwest 170 138 91,531 79.91% 306 229 554,496 72.50%
Southeast 155 129 81,532 83.57% 256 193 526,227 67.58%
Southwest 107 92 60,030 86.45% 205 153 386,336 69.58%
West 158 125 99,331 77.93% 262 209 562,228 74.51%
Iowa 1,364 1,105 376,485 80.21% 2,543 1,908 2,331,516 71.81%
Central 262 212 66,819 77.48% 462 324 418,051 68.71%
North Central 158 128 45,576 82.47% 273 213 261,694 75.67%
Northeast 303 240 91,562 77.50% 637 481 567,910 72.28%
Northwest 198 174 57,333 90.37% 330 250 355,324 71.75%
Southeast 307 243 78,823 75.60% 519 393 496,432 69.86%
Southwest 136 108 36,372 82.96% 322 247 232,104 75.77%
Kansas 1,398 1,154 360,624 82.42% 2,538 1,929 2,111,849 74.10%
Kansas City Metro 500 397 119,069 78.88% 850 640 709,882 74.70%
Northeast 207 173 70,186 83.32% 491 375 409,624 71.64%
South Central 225 188 44,976 83.87% 362 282 262,360 78.20%
Southeast 89 74 23,858 84.57% 151 114 142,077 68.79%
West 121 106 40,187 87.34% 252 179 228,432 68.00%
Wichita (Sedgwick) 256 216 62,348 84.51% 432 339 359,474 77.74%
Kentucky 1,441 1,176 517,498 81.23% 2,559 1,902 3,294,920 71.07%
Adanta, Cumberland River, and Lifeskills 226 179 88,471 78.29% 413 318 553,262 76.05%
Bluegrass, Comprehend, and North Key 430 347 156,418 80.25% 809 590 964,590 69.71%
Communicare and River Valley 176 143 58,902 81.62% 314 220 363,284 66.22%
Four Rivers and Pennyroyal 117 96 47,520 82.32% 214 160 312,746 73.35%
Kentucky River, Mountain, and Pathways 133 117 54,925 87.56% 262 195 371,879 68.69%
Seven Counties 359 294 111,261 81.61% 547 419 729,160 72.69%
Louisiana 1,395 1,173 574,987 83.43% 2,550 1,952 3,405,070 73.90%
Regions 1 and 10 214 179 98,913 80.00% 496 359 665,793 70.69%
Region 1 89 76 52,326 82.54% 196 144 340,233 78.22%
Region 10 (Jefferson) 125 103 46,587 78.30% 300 215 325,560 66.79%
Regions 2 and 9 435 365 157,519 83.25% 750 592 901,628 77.53%
Region 3 148 117 51,559 80.46% 207 157 293,967 74.40%
Regions 4, 5, and 6 323 266 150,915 81.41% 609 449 872,660 70.01%
Regions 7 and 8 275 246 116,082 90.65% 488 395 671,023 76.68%
Maine 1,392 1,162 145,270 82.81% 2,482 1,975 1,057,570 77.20%
Aroostook/Downeast 157 130 16,252 84.77% 271 227 127,118 82.34%
Aroostook 91 81 7,625 88.11% 160 145 56,743 89.36%
Downeast 66 49 8,627 80.05% 111 82 70,375 73.67%
Central 182 152 18,909 84.33% 324 262 137,358 80.21%
Cumberland 320 260 31,579 77.65% 580 435 226,494 73.59%
Midcoast 90 83 14,526 91.76% 212 177 117,872 84.02%
Penquis 190 160 20,442 82.33% 333 282 137,197 80.62%
Western 228 194 22,053 84.55% 367 307 153,929 79.36%
York 225 183 21,508 82.93% 395 285 157,603 66.15%
Maryland 1,377 1,118 664,412 80.35% 2,519 1,924 4,490,598 74.22%
Anne Arundel 155 124 60,580 78.95% 236 183 422,356 73.96%
Baltimore City 132 120 64,746 87.46% 334 273 481,314 83.76%
Baltimore County 161 117 90,244 72.14% 310 228 634,067 74.59%
Montgomery 212 174 102,652 83.38% 374 284 758,355 73.85%
North Central 100 87 57,300 86.17% 164 124 350,403 76.30%
Northeast 132 108 58,062 81.34% 224 165 374,855 64.94%
Prince George's 169 124 102,304 72.69% 383 275 662,200 68.46%
South 153 137 70,306 90.34% 249 201 427,589 76.14%
West 163 127 58,218 78.00% 245 191 379,457 75.60%
Massachusetts 1,508 1,200 782,267 79.36% 2,842 1,990 5,223,941 68.20%
Boston 150 112 95,009 74.21% 443 310 668,386 68.09%
Central 195 146 107,491 75.25% 333 214 665,374 62.93%
Metrowest 263 208 174,380 80.31% 520 357 1,207,726 67.16%
Northeast 336 281 152,156 83.69% 580 417 1,018,681 68.37%
Southeast 305 233 142,882 75.15% 576 383 1,001,184 66.64%
Western 259 220 110,348 84.39% 390 309 662,591 77.54%
Michigan 5,061 4,078 1,223,572 79.85% 8,889 6,740 7,544,403 72.47%
Region 1 153 138 35,390 89.62% 280 231 250,499 80.28%
Region 2 188 132 52,904 66.84% 321 248 400,863 77.66%
Region 3 663 548 160,133 82.54% 1,138 893 915,546 75.19%
Region 4 450 379 103,771 83.79% 789 607 638,370 74.75%
Region 5 747 624 216,428 83.87% 1,486 1,208 1,276,129 78.91%
Region 6 467 386 104,461 82.05% 861 660 608,005 72.99%
Region 7 842 669 224,967 79.39% 1,463 1,089 1,329,957 68.09%
Region 8 640 494 141,735 76.61% 999 708 937,373 69.11%
Region 9 476 358 96,285 74.42% 827 571 651,472 65.69%
Region 10 435 350 87,497 77.61% 725 525 536,189 69.09%
Minnesota 1,350 1,145 634,874 84.68% 2,477 1,912 4,093,708 77.10%
Regions 1 and 2 118 88 60,648 78.20% 227 168 414,929 69.41%
Regions 3 and 4 201 175 112,158 87.54% 361 294 697,366 83.96%
Regions 5 and 6 238 198 121,304 81.94% 454 336 767,548 70.78%
Region 7A (Hennepin) 249 212 127,780 86.51% 547 425 917,410 80.13%
Region 7B (Ramsey) 139 127 63,220 90.76% 242 197 401,071 80.28%
Region 7C 405 345 149,764 83.22% 646 492 895,383 77.03%
Mississippi 1,367 1,165 374,676 85.43% 2,335 1,862 2,184,218 77.33%
Region 1 305 266 88,001 88.31% 503 419 494,304 82.04%
Region 2 168 146 45,448 87.87% 302 241 270,315 73.80%
Region 3 205 184 54,742 88.24% 377 306 307,535 77.93%
Region 4 227 180 68,432 78.59% 378 288 400,607 74.10%
Region 5 58 43 21,163 78.12% 118 89 133,781 74.21%
Region 6 198 174 38,159 88.82% 340 280 226,620 79.18%
Region 7 206 172 58,732 83.71% 317 239 351,055 75.69%
Missouri 1,358 1,123 703,170 82.32% 2,529 1,913 4,537,760 73.44%
Central 203 173 102,483 84.98% 411 316 619,647 74.18%
Eastern 464 382 240,646 81.56% 903 668 1,574,889 70.75%
Eastern (St. Louis City and County) 284 230 147,370 81.47% 539 401 998,801 70.46%
Eastern (excluding St. Louis) 180 152 93,276 81.71% 364 267 576,087 71.17%
Northwest 301 253 169,079 83.03% 578 449 1,095,401 77.25%
Northwest (Jackson) 167 146 74,140 85.69% 316 247 502,397 78.09%
Northwest (excluding Jackson) 134 107 94,939 79.74% 262 202 593,004 76.30%
Southeast 199 162 82,010 81.38% 300 228 540,963 74.57%
Southwest 191 153 108,952 81.42% 337 252 706,860 72.93%
Montana 1,457 1,176 117,946 80.98% 2,507 1,911 780,158 74.14%
Region 1 117 97 9,385 81.52% 148 121 60,830 86.39%
Region 2 188 163 17,648 87.77% 366 300 109,284 82.19%
Region 3 334 274 23,978 81.74% 574 435 160,458 74.43%
Region 4 344 257 31,306 76.01% 601 452 208,544 72.45%
Region 5 474 385 35,628 81.24% 818 603 241,042 68.73%
Nebraska 1,419 1,180 229,797 82.63% 2,566 1,947 1,385,174 72.62%
Regions 1 and 2 166 137 22,699 83.02% 240 172 141,371 66.45%
Region 1 94 80 10,592 85.53% 144 104 66,422 67.69%
Region 2 72 57 12,107 80.12% 96 68 74,949 64.00%
Region 3 201 179 28,388 88.43% 363 292 169,893 77.33%
Region 4 131 108 25,973 78.06% 206 154 153,104 72.36%
Region 5 372 311 56,452 83.62% 690 539 345,998 71.82%
Region 6 549 445 96,284 81.06% 1,067 790 574,808 73.07%
Nevada 1,346 1,154 339,747 86.15% 2,592 1,972 2,095,504 73.08%
Clark – Region 1 888 775 246,185 87.66% 1,761 1,344 1,506,486 73.12%
Region 3 173 133 39,964 78.00% 313 226 258,414 72.61%
Capital District 68 52 17,924 77.02% 122 87 123,581 70.96%
Rural/Frontier 105 81 22,040 78.50% 191 139 134,833 73.90%
Washoe – Region 2 285 246 53,598 86.31% 518 402 330,603 73.27%
New Hampshire 1,597 1,251 167,083 79.22% 2,654 1,968 1,041,444 72.17%
Central 513 429 48,867 85.03% 868 674 297,358 73.92%
Central 1 313 268 25,768 87.29% 444 344 147,376 71.70%
Central 2 200 161 23,098 80.81% 424 330 149,982 76.14%
Northern 205 159 21,158 78.85% 307 237 137,687 76.90%
Southern 879 663 97,059 75.56% 1,479 1,057 606,399 70.40%
Southern 1 (Rockingham) 362 261 35,718 72.83% 617 419 232,337 67.83%
Southern 2 517 402 61,341 77.42% 862 638 374,062 72.30%
New Jersey 1,702 1,337 1,044,003 79.28% 3,314 2,372 6,776,162 69.85%
Central 411 319 241,494 77.54% 730 523 1,552,030 69.02%
Metropolitan 391 308 261,669 80.32% 854 609 1,626,795 67.82%
Northern 504 402 322,437 81.53% 1,080 759 2,184,675 71.05%
Southern 396 308 218,403 77.15% 650 481 1,412,662 71.66%
New Mexico 1,338 1,139 251,927 85.28% 2,461 1,914 1,549,877 74.91%
Region 1 268 239 56,486 90.22% 499 401 318,051 78.39%
Region 2 155 138 31,035 90.07% 325 246 226,769 71.30%
Region 3 (Bernalillo) 397 332 77,475 83.10% 785 607 508,046 75.34%
Region 4 264 210 34,860 80.35% 366 279 191,653 72.06%
Region 5 254 220 52,071 86.25% 486 381 305,358 75.17%
New York 5,799 4,437 2,249,511 75.89% 11,041 7,288 15,173,526 62.86%
Region A 2,177 1,598 878,530 73.03% 4,696 2,901 6,518,835 58.42%
Region 1 530 432 186,007 82.00% 877 620 1,036,410 70.93%
Region 2 788 533 337,616 66.19% 1,665 993 2,317,580 55.50%
Region 3 285 225 126,518 77.32% 960 620 1,370,195 64.42%
Region 4 574 408 228,389 74.40% 1,194 668 1,794,650 50.86%
Region B 1,530 1,138 624,703 73.98% 2,548 1,581 3,871,194 59.31%
Region 5 912 667 344,235 72.64% 1,521 927 2,183,053 58.78%
Region 6 360 258 169,363 70.31% 598 362 1,035,505 58.20%
Region 7 258 213 111,106 84.65% 429 292 652,636 63.26%
Region C 1,618 1,316 559,345 81.56% 2,849 2,111 3,575,427 71.88%
Region 8 354 257 118,195 73.37% 560 399 784,327 71.47%
Region 9 409 345 121,183 84.90% 654 491 740,418 72.51%
Region 10 120 100 58,625 79.41% 241 189 348,455 74.85%
Region 11 365 293 130,480 80.24% 679 508 816,246 72.68%
Region 12 370 321 130,863 87.56% 715 524 885,981 70.16%
Region D 474 385 186,933 80.50% 948 695 1,208,070 70.85%
Region 13 168 136 57,291 81.90% 328 238 373,284 71.31%
Region 14 107 88 64,012 86.12% 243 181 416,796 69.19%
Region 15 199 161 65,630 76.18% 377 276 417,990 71.47%
North Carolina 1,538 1,285 1,120,046 83.86% 3,051 2,386 7,344,723 75.11%
Alliance Behavioral Healthcare 1 135 118 90,634 86.30% 299 251 575,093 83.80%
Alliance Behavioral Healthcare 2 164 144 114,862 87.53% 246 188 700,520 73.94%
Cardinal Innovations Healthcare Solutions 1 132 99 91,249 73.35% 225 166 551,825 72.09%
Cardinal Innovations Healthcare Solutions 2 111 96 78,842 87.16% 279 224 506,508 79.97%
Cardinal Innovations Healthcare Solutions 3 111 87 107,636 77.59% 281 196 712,067 69.92%
CenterPoint Human Services 102 90 61,945 88.69% 193 166 407,724 82.63%
Eastpointe 157 133 95,280 84.58% 268 204 612,590 68.79%
Partners Behavioral Health Management 190 160 104,105 86.68% 330 257 683,738 73.80%
Sandhills Center 1 56 50 64,933 91.77% 109 84 420,365 71.26%
Sandhills Center 2 85 71 60,589 82.70% 180 141 380,568 82.80%
Smoky Mountain Center 1 81 65 59,264 83.19% 139 108 421,307 68.20%
Smoky Mountain Center 2 61 46 52,322 73.31% 149 123 414,641 79.20%
Trillium Health Resources 1 87 71 70,193 85.11% 169 141 472,296 80.32%
Trillium Health Resources 2 66 55 68,193 83.34% 184 137 485,480 62.83%
North Dakota 1,385 1,131 87,446 81.65% 2,633 1,975 544,336 72.37%
Badlands and West Central 313 268 21,759 83.98% 632 483 145,880 75.15%
Lake Region 119 95 5,244 81.94% 182 135 29,650 71.75%
North Central 177 133 11,978 72.97% 272 213 75,117 75.53%
Northeast 223 189 12,583 85.75% 438 326 70,654 68.29%
Northwest 83 62 4,133 71.18% 144 89 27,982 63.48%
South Central 87 74 6,300 84.92% 191 146 44,398 74.76%
Southeast 383 310 25,450 81.60% 774 583 150,655 71.93%
Ohio 5,290 4,192 1,409,063 78.94% 8,964 6,545 8,750,593 70.31%
Boards 2, 46, 55, and 68 200 159 64,265 75.34% 315 210 385,490 64.93%
Boards 3, 52, and 85 182 142 50,146 78.99% 271 197 283,559 71.75%
Boards 4 and 78 111 90 35,066 80.08% 197 150 237,734 76.95%
Boards 5 and 60 199 167 46,604 83.34% 301 242 260,144 78.60%
Boards 7, 15, 41, 79, and 84 212 168 50,931 78.67% 374 284 356,843 72.61%
Boards 8, 13, and 83 250 189 62,641 75.91% 370 252 370,805 66.00%
Board 9 (Butler) 159 116 50,152 70.23% 268 181 276,181 64.19%
Board 12 136 102 44,196 74.45% 230 159 264,936 62.06%
Boards 18 and 47 657 564 184,762 86.06% 1,137 865 1,193,450 74.49%
Boards 20, 32, 54, and 69 188 153 42,380 82.62% 314 235 256,829 73.47%
Boards 21, 39, 51, 70, and 80 327 245 68,568 73.11% 440 329 414,969 70.99%
Boards 22, 74, and 87 176 137 48,899 77.41% 293 232 296,234 78.97%
Boards 23 and 45 202 162 49,176 79.67% 296 225 281,631 71.71%
Board 25 (Franklin) 495 386 143,523 78.76% 952 677 897,304 70.25%
Boards 27, 71, and 73 252 191 57,958 75.58% 422 291 373,892 62.16%
Boards 28, 43, and 67 271 222 61,584 81.27% 431 334 374,770 74.95%
Board 31 (Hamilton) 357 292 95,290 80.68% 625 465 601,566 72.64%
Board 48 (Lucas) 216 168 54,209 78.41% 337 228 329,559 63.84%
Boards 50 and 76 294 227 71,614 76.80% 544 383 470,658 65.25%
Board 57 (Montgomery) 205 162 63,420 81.35% 436 324 411,534 70.60%
Board 77 (Summit) 201 150 63,682 74.00% 411 282 412,503 65.74%
Oklahoma 1,480 1,159 476,983 76.92% 2,650 1,948 2,820,561 69.13%
Central 172 134 63,066 76.40% 326 221 355,862 61.35%
East Central 200 149 56,086 73.23% 295 222 322,240 74.73%
Northeast 183 146 63,827 81.24% 345 264 360,940 69.93%
Northwest and Southwest 214 171 68,431 75.51% 304 233 404,489 71.37%
Oklahoma County 278 215 86,764 76.20% 543 392 543,373 65.54%
Southeast 237 192 63,929 80.11% 408 295 384,299 69.98%
Tulsa County 196 152 74,879 75.88% 429 321 449,359 72.78%
Oregon 1,377 1,132 457,538 82.57% 2,592 1,959 3,046,250 74.69%
Region 1 (Multnomah) 235 187 76,731 77.71% 530 388 607,794 73.39%
Region 2 355 291 112,092 82.10% 597 447 708,028 72.44%
Region 3 429 348 155,869 82.18% 813 635 949,188 78.30%
Region 4 180 153 59,634 85.48% 337 257 436,964 76.47%
Region 5 (Central) 82 74 23,165 90.72% 159 123 158,143 75.43%
Region 6 (Eastern) 96 79 30,047 86.00% 156 109 186,134 64.71%
Pennsylvania 4,954 4,072 1,480,904 82.02% 9,105 6,708 9,861,971 70.47%
Region 1 (Allegheny) 466 358 128,851 77.45% 961 628 975,588 64.17%
Regions 3, 8, 9, and 51 294 241 81,459 82.38% 561 402 549,711 68.70%
Regions 4, 11, 37, and 49 371 304 109,140 80.87% 612 442 695,119 69.01%
Regions 5, 18, 23, 24, and 46 322 278 85,000 84.68% 592 460 577,717 72.97%
Regions 6, 12, 16, 31, 35, 45, and 47 306 249 91,779 82.17% 565 452 561,746 75.39%
Regions 7, 13, 20, and 33 871 734 294,951 85.28% 1,668 1,193 1,902,261 68.68%
Regions 10, 15, 27, 32, 43,and 44 201 164 56,235 80.37% 349 282 402,461 78.55%
Regions 17 and 21 123 106 46,366 81.87% 263 213 283,624 76.53%
Regions 19, 26, 28, and 42 700 571 172,159 81.63% 1,104 841 1,110,681 72.71%
Regions 22, 38, 40, 41, and 48 299 229 86,166 76.13% 518 376 652,989 69.43%
Regions 29 and 34 227 186 78,637 80.99% 401 273 501,438 64.64%
Regions 30 and 50 254 196 66,633 79.55% 461 329 477,864 71.64%
Region 36 (Philadelphia) 520 456 183,530 84.48% 1,050 817 1,170,774 74.43%
Rhode Island 1,402 1,168 132,087 83.53% 2,615 1,980 822,015 73.23%
Bristol and Newport 158 131 16,503 86.85% 263 204 104,975 72.45%
Kent 190 155 17,532 80.69% 401 284 130,120 67.24%
Providence 846 713 79,554 83.56% 1,628 1,241 486,412 75.13%
Washington 208 169 18,498 83.03% 323 251 100,508 73.00%
South Carolina 1,401 1,148 541,337 82.26% 2,541 1,969 3,596,083 74.98%
Region 1 415 335 173,813 81.27% 761 566 1,127,070 71.55%
Region 2 362 311 139,965 86.82% 587 457 878,504 76.26%
Region 3 263 199 88,126 74.22% 470 357 627,339 74.23%
Region 4 361 303 139,433 84.25% 723 589 963,170 78.27%
South Dakota 1,383 1,162 105,160 83.71% 2,519 1,928 621,956 75.21%
Region 1 336 276 24,999 80.15% 633 490 153,104 77.05%
Region 2 160 143 9,467 89.98% 234 174 56,396 71.89%
Region 3 324 273 26,074 84.29% 574 445 148,447 77.93%
Region 4 173 136 15,361 80.13% 339 249 88,007 70.84%
Region 5 390 334 29,259 85.06% 739 570 176,003 74.62%
Tennessee 1,305 1,101 759,270 83.92% 2,465 1,951 4,918,833 76.93%
Region 1 109 91 54,945 81.37% 225 169 403,218 74.69%
Region 2 238 196 134,754 83.78% 441 349 926,189 79.44%
Region 3 217 190 109,776 87.04% 387 317 741,941 78.64%
Region 4 (Davidson) 115 88 68,238 75.41% 280 215 497,055 72.43%
Region 5 318 259 195,168 80.51% 561 442 1,172,253 78.10%
Region 6 120 110 75,704 93.04% 243 204 482,718 77.73%
Region 7 (Shelby) 188 167 120,684 88.08% 328 255 695,458 73.76%
Texas 5,491 4,504 3,444,138 81.60% 9,923 7,298 18,944,344 70.65%
Region 1 187 151 115,574 79.02% 417 296 628,963 67.31%
Region 2 76 68 69,557 90.81% 177 150 415,696 82.12%
Region 3 1,529 1,309 906,472 85.68% 2,746 2,114 5,078,952 73.94%
Region 3a 906 774 577,395 85.42% 1,703 1,282 3,233,114 71.80%
Region 3bc 623 535 329,078 86.12% 1,043 832 1,845,838 77.62%
Region 4 223 190 136,677 85.44% 449 345 845,637 72.62%
Region 5 167 129 95,227 75.95% 259 179 584,572 65.84%
Region 6 1,229 907 821,986 72.93% 2,371 1,543 4,573,572 61.91%
Region 6a 1,095 809 738,683 72.97% 2,137 1,391 4,088,986 62.49%
Region 6bc 134 98 83,303 72.56% 234 152 484,586 57.08%
Region 7 637 536 404,643 83.98% 1,275 975 2,316,920 74.65%
Region 7a 341 286 245,257 82.22% 747 552 1,465,297 72.26%
Region 7bcd 296 250 159,386 85.98% 528 423 851,623 78.44%
Region 8 588 502 360,371 85.81% 922 685 1,980,282 72.96%
Region 9 134 96 76,366 72.70% 241 160 430,196 63.92%
Region 10 229 196 129,511 83.46% 346 273 604,974 78.27%
Region 11 492 420 327,753 86.29% 720 578 1,484,582 76.82%
Region 11abd 288 241 198,702 84.29% 421 324 948,499 72.33%
Region 11c (Hidalgo) 204 179 129,051 89.06% 299 254 536,082 83.69%
Utah 1,321 1,147 420,016 87.10% 2,465 1,981 1,986,263 78.53%
Bear River, Northeastern, Summit, Tooele, and Wasatch 184 167 52,068 90.81% 285 230 236,606 81.21%
Central, Four Corners, San Juan, and Southwest 129 106 52,115 86.37% 227 169 241,440 70.68%
Davis County 167 142 47,471 82.24% 275 240 211,915 86.41%
Salt Lake County 490 417 144,536 84.39% 1,011 795 765,196 77.78%
Utah County 263 233 89,148 89.91% 465 375 356,816 77.01%
Weber, Morgan 88 82 34,678 94.15% 202 172 174,291 80.96%
Vermont 1,361 1,106 72,604 81.47% 2,529 1,917 501,015 74.27%
Champlain Valley 664 541 32,271 82.13% 1,218 899 200,149 71.31%
Rural Northeast 247 197 16,517 80.17% 503 384 117,880 75.62%
Rural Southeast 267 210 13,034 78.80% 481 373 104,234 75.54%
Rural Southwest 183 158 10,782 84.97% 327 261 78,752 79.03%
Virginia 1,643 1,379 944,663 84.75% 3,020 2,291 6,181,981 74.38%
Region 1 269 238 157,157 89.39% 487 406 958,791 82.57%
Region 2 479 393 253,841 84.32% 837 611 1,723,482 70.87%
Region 3 256 212 152,331 82.96% 526 408 1,063,921 76.80%
Region 4 261 221 163,956 85.41% 469 338 1,053,897 69.92%
Region 5 378 315 217,378 82.87% 701 528 1,381,891 73.77%
Washington 1,404 1,145 855,057 82.00% 2,641 1,951 5,279,538 71.53%
Region 1 329 277 214,765 85.58% 499 376 1,135,716 72.12%
Greater Columbia and North Central 202 170 135,663 84.39% 295 228 668,527 73.84%
Spokane 127 107 79,102 87.46% 204 148 467,190 69.90%
Region 2 605 477 363,834 78.88% 1,297 942 2,444,573 70.85%
King 367 283 220,868 76.96% 809 581 1,564,010 71.16%
North Sound 238 194 142,966 81.79% 488 361 880,563 70.35%
Region 3 470 391 276,458 83.41% 845 633 1,699,249 72.20%
Peninsula 60 53 41,543 87.33% 121 81 285,757 66.15%
Pierce 190 164 104,520 85.42% 362 289 618,236 78.09%
SW WA and Timberlands 155 122 92,120 80.17% 271 188 546,023 65.06%
Thurston-Mason 65 52 38,275 81.30% 91 75 249,234 82.22%
West Virginia 1,547 1,230 210,916 79.43% 2,645 1,946 1,452,450 72.02%
Region I 101 77 16,330 76.44% 164 108 116,314 58.81%
Region II 206 170 32,591 80.76% 309 227 202,057 73.27%
Region III 113 83 18,740 69.56% 188 129 133,231 67.80%
Region IV 402 324 49,781 81.44% 678 524 322,096 75.50%
Region V 426 329 58,067 78.04% 799 573 412,851 70.61%
Region VI 299 247 35,407 82.01% 507 385 265,900 75.45%
Wisconsin 1,405 1,116 669,309 78.16% 2,566 1,859 4,361,642 72.01%
Milwaukee 228 182 114,736 79.18% 481 336 702,647 68.09%
Northeastern 305 234 140,203 75.19% 545 402 946,149 75.24%
Northern 128 94 53,634 70.95% 209 157 379,798 75.25%
Southeastern 290 223 140,634 76.64% 491 330 878,892 69.45%
Southern 255 212 126,323 80.92% 479 359 855,445 72.78%
Western 199 171 93,779 85.23% 361 275 598,710 72.26%
Wyoming 1,415 1,165 67,146 81.47% 2,508 1,931 435,273 76.27%
Judicial District 1 (Laramie) 238 193 10,846 81.85% 424 315 72,026 73.63%
Judicial District 2 140 123 7,579 82.44% 281 230 43,225 77.22%
Judicial District 3 239 200 10,453 84.10% 396 317 59,981 81.36%
Judicial District 4 101 82 4,110 83.21% 191 151 29,252 69.91%
Judicial District 5 91 80 6,284 90.34% 158 127 41,458 83.21%
Judicial District 6 175 134 7,253 73.19% 282 205 44,585 72.16%
Judicial District 7 (Natrona) 165 136 8,859 79.17% 281 213 58,836 74.47%
Judicial District 8 139 114 4,519 79.52% 226 168 30,325 75.15%
Judicial District 9 127 103 7,243 82.72% 269 205 55,583 78.85%

Section D: References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (DSM-IV) (4th ed.). Washington, DC: Author.

Box, G. E. P., & Tiao, G. C. (1992). Bayesian inference in statistical analysis (Wiley Classics Library). Hoboken, NJ: John Wiley & Sons.

Center for Behavioral Health Statistics and Quality. (2012). Results from the 2011 National Survey on Drug Use and Health: Summary of national findings (HHS Publication No. SMA 12-4713, NSDUH Series H-44). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Center for Behavioral Health Statistics and Quality. (2013a). Results from the 2012 National Survey on Drug Use and Health: Mental health findings (HHS Publication No. SMA 13-4805, NSDUH Series H-47). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Center for Behavioral Health Statistics and Quality. (2013b). Results from the 2012 National Survey on Drug Use and Health: Summary of national findings (HHS Publication No. SMA 13-4795, NSDUH Series H-46). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Center for Behavioral Health Statistics and Quality. (2015). 2014 National Survey on Drug Use and Health: Methodological resource book (Section 2, Sample design report). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Folsom, R. E., Shah, B., & Vaish, A. (1999). Substance abuse in states: A methodological report on model based estimates from the 1994-1996 National Household Surveys on Drug Abuse. In Proceedings of the 1999 Joint Statistical Meetings, American Statistical Association, Survey Research Methods Section, Baltimore, MD (pp. 371-375). Alexandria, VA: American Statistical Association.

Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2003). Bayesian data analysis (2nd ed., Chapman & Hall/CRC Texts in Statistical Science). Boca Raton, FL: Chapman and Hall/CRC.

Hughes, A., Muhuri, P., Sathe, N., & Spagnola, K. (2012). State estimates of substance use and mental disorders from the 2009-2010 National Surveys on Drug Use and Health (HHS Publication No. SMA 12-4703, NSDUH Series H-43). Rockville, MD: Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality.

Office of Applied Studies. (2005, September). Appendix C: Research on the impact of changes in NSDUH methods. In Results from the 2004 National Survey on Drug Use and Health: National findings (HHS Publication No. SMA 05-4062, NSDUH Series H-28, pp. 145-154). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Office of Applied Studies. (2008, June). Substate estimates from the 2004-2006 National Surveys on Drug Use and Health. Rockville, MD: Substance Abuse and Mental Health Services Administration.

Payton, M. E., Greenstone, M. H., & Schenker, N. (2003). Overlapping confidence intervals or standard error intervals: What do they mean in terms of statistical significance? Journal of Insect Science, 3, 34.

Raftery, A. L., & Lewis, S. (1992). How many iterations in the Gibbs sampler? In J. M. Bernardo, J. O. Berger, A. P. Dawid, & A. F. M. Smith (Eds.), Bayesian statistics 4 (pp. 763-774). London, England: Oxford University Press.

Rao, J. N. K. (2003). Small area estimation (Wiley Series in Survey Methodology). Hoboken, NJ: John Wiley & Sons.

Schenker, N., & Gentleman, J. F. (2001). On judging the significance of differences by examining the overlap between confidence intervals. American Statistician, 55(3), 182-186.

Shah, B. V., Barnwell, B. G., Folsom, R., & Vaish, A. (2000). Design consistent small area estimates using Gibbs algorithm for logistic models. In Proceedings of the 2000 Joint Statistical Meetings, American Statistical Association, Survey Research Methods Section, Indianapolis, IN (pp. 105-111). Alexandria, VA: American Statistical Association.

Wright, D. (2003, July). State estimates of substance use from the 2001 National Household Survey on Drug Abuse: Volume II. Individual state tables and technical appendices (HHS Publication No. SMA 03-3826, NHSDA Series H-20). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies.

Wright, D., & Sathe, N. (2005, January). State estimates of substance use from the 2002-2003 National Surveys on Drug Use and Health (HHS Publication No. SMA 05-3989, NSDUH Series H-26). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies.

Section E: List of Contributors

This National Survey on Drug Use and Health (NSDUH) document was prepared by the Center for Behavioral Health Statistics and Quality (CBHSQ), Substance Abuse and Mental Health Services Administration (SAMHSA), U.S. Department of Health and Human Services (HHS), and by RTI International (a registered trademark and a trade name of Research Triangle Institute), Research Triangle Park, North Carolina. Work by RTI was performed under Contract No. HHSS283201300001C.

At RTI, Neeraja S. Sathe and Kathryn Spagnola were responsible for the writing of the document, and Ralph E. Folsom and Akhil K. Vaish were responsible for the overall methodology and estimation for the model-based Bayes estimates and confidence intervals. At SAMHSA, Arthur Hughes reviewed the document and provided substantive revisions.

The following staff were responsible for generating the estimates: Akhil K. Vaish, Neeraja S. Sathe, and Kathryn Spagnola. The following staff provided other support: Brenda K. Porter and Nicole Mack. Ms. Spagnola also provided oversight for production of the document. Richard S. Straw edited it; Debbie F. Bond formatted its text and tables; and Teresa F. Bass, Kimberly Cone, Danny Occoquan, and Pamela Tuck prepared the web versions. Justine L. Allpress prepared the maps used in the associated files.



End Notes

1 RTI International is a registered trademark and a trade name of Research Triangle Institute, Research Triangle Park, North Carolina.

2 In 2002, the survey's name changed from the National Household Survey on Drug Abuse (NHSDA) to the National Survey on Drug Use and Health (NSDUH).

3 Prior to this effort, substate small area estimates using the combined 1999-2001, 2002-2004, 2004-2006, 2006-2008, 2008-2010, and 2010-2012 data have been produced by SAMHSA. These estimates can be found at http://www.samhsa.gov/data/.

4 OAS is the former name of SAMHSA's CBHSQ.

5 The use of mixed models (fixed and random effects) allows additional error components (random effects) to be included. These account for differences between states and within-state variations that are not taken into account by the predictor variables (fixed effects) alone. It is also difficult (if not impossible) to produce valid mean squared errors (MSEs) for small area estimates based solely on a fixed-effect national regression model (i.e., synthetic estimation) (Rao, 2003, p. 52). The mixed models produce estimates that are approximately represented by a weighted combination of the direct estimate from the substate data and a regression estimate from the national model. The regression coefficients of the national model are estimated using data from all of the substate areas (i.e., borrowing strength), and the regression estimate for a particular substate area is obtained by applying the national model to the substate-specific predictor data. The regression estimate for the substate area is then combined with the direct estimate from the substate data in a weighted combination where the weights are obtained by minimizing the MSE (variance + squared bias) of the small area estimate.

6 Improper noninformative priors were used for the fixed regression coefficients. Because NSDUH's sample size is very large (approximately 200,000 observations across 3 years), it can be said with certainty that the posterior distribution for the fixed regression coefficients will be proper. For the W1 and W2 matrices, noninformative proper inverse Wishart prior distributions were used so that proper posterior distributions are guaranteed. Noninformative priors are used so that the data speak for themselves and inferences are unaffected by information external to the current data. For further information on Bayesian inference and analysis, see Box and Tiao (1992) and Gelman, Carlin, Stern, and Rubin (2003).

7 See Table 3 in the "2012-2014 NSDUH Substate Region Estimates: Excel Tables" at http://www.samhsa.gov/data/.

8 The RSE of an estimate is the posterior SE divided by the estimate itself. Note that the RSEs have been calculated based on the unbenchmarked small area estimates.

 

Go to Top of Page