2015-2016
National Survey on Drug Use and Health:
Guide to State Tables and Summary of Small Area Estimation Methodology

Section A: Overview of NSDUH and Model-Based State Estimates

A.1 Introduction

This document provides information on the model-based small area estimates of substance use and mental disorders in states based on data from the combined 2015-2016 National Surveys on Drug Use and Health (NSDUHs). These estimates are available online along with other related information.¹ NSDUH is an annual survey conducted from January through December of the civilian, noninstitutionalized population aged 12 or older and is sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA). The survey collects information from individuals residing in households, noninstitutionalized group quarters (e.g., shelters, rooming houses, dormitories), and civilians living on military bases. In 2015-2016, NSDUH collected data from 136,015 respondents aged 12 or older and was designed to obtain representative samples from the 50 states and the District of Columbia. NSDUH is planned and managed by SAMHSA's Center for Behavioral Health Statistics and Quality (CBHSQ). Data collection and analysis are conducted under contract with RTI International.² A summary of NSDUH's methodology is given in Section A.2. Section A.3 lists all of the tables and files associated with the 2015‑2016 state small area estimates and when and where they can be found. Information is given in Section A.4 on the confidence intervals and margins of error and how to make interpretations with respect to the small area estimates. Section A.5 discusses related substance use measures and warns users about not drawing conclusions by subtracting small area estimates from two different measures. Section A.6 discusses NSDUH questionnaire changes from 2015 and how these changes affect the small area estimates.

The survey-weighted hierarchical Bayes (SWHB) estimation methodology used in the production of state estimates from the 1999 to 2015 surveys also was used in the production of the 2015-2016 state estimates. The SWHB methodology is described in Appendix E of the 2001 state report (Wright, 2003b) and in Folsom, Shah, and Vaish (1999). A general model description is given in Section B.1 of this document. A list of measures for which small area estimates are produced is given in Section B.2. Predictors used in the 2015-2016 small area estimation (SAE) modeling are listed and described in Section B.3. New variable selection was done for all measures, as discussed in Section B.4.

Small area estimates obtained using the SWHB methodology are design consistent (i.e., the small area estimates for states with large sample sizes are close to the robust design-based estimates). The state small area estimates when aggregated using the appropriate population totals result in national small area estimates that are very close to the national design-based estimates. However, to ensure internal consistency, it is desirable to have national small area estimates³ exactly match the national design-based estimates. This process is called benchmarking. The benchmarked state-level estimates are also potentially less biased than the unbenchmarked state-level estimates. Beginning in 2002, exact benchmarking was introduced, as described in Section B.5.⁴ Tables of the estimated numbers of individuals associated with each measure are available online,⁵ and an explanation of how these counts and their respective Bayesian confidence intervals⁶ are calculated can be found in Section B.6. Section B.7 discusses the method to compute aggregate estimates by combining two age groups. The definition and explanation of the formula used in estimating the marijuana incidence rate are given in Section B.8. Note that, unlike the other SAE outcomes discussed in this document, marijuana incidence is calculated as a ratio of two measures.

For all measures except major depressive episode (MDE, i.e., depression), serious mental illness (SMI), any mental illness (AMI), mental health services, and past year serious thoughts of suicide, the age groups for which estimates are provided are 12 to 17, 18 to 25, 26 or older, 18 or older, and 12 or older.⁷

Estimates of underage (aged 12 to 20) alcohol use and binge alcohol use were also produced.⁸ Alcohol consumption is expected to differ significantly across the 18 to 25 age group because of the legalization of alcohol at age 21. Therefore, it was decided that it would be useful to produce small area estimates for individuals aged 12 to 20. A short description of the methodology used to produce underage drinking estimates is provided in Section B.9.

The remainder of Section B covers two topics:

• Section B.10 discusses the criteria used to define substance use disorder and needing but not receiving treatment.
• Section B.11 discusses the production of estimates for MDE (i.e., depression), SMI, AMI, and suicidal thoughts. Note that for MDE, there are no 12 or older estimates published; also, for SMI, AMI, and serious thoughts of suicide, no 12 to 17 estimates are produced because youths are not asked these questions.

In Section C, the 2014, 2015, 2016, pooled 2014-2015, and pooled 2015-2016 survey sample sizes, population estimates, and response rates are included in Tables C.1 to C.14, respectively. Table C.15 lists all of the measures and the years for which small area estimates were produced going back to the 2002 NSDUH, and Table C.16 lists all of the measures by age groups for which small area estimates were produced. In addition, Table C.17 provides a summary of milestones implemented in the SAE production process from 2002 to 2016.

A.2 Summary of NSDUH Methodology

NSDUH is the primary source of statistical information on the use of illicit drugs, alcohol, and tobacco by the U.S. civilian, noninstitutionalized population aged 12 or older. The survey also includes several modules of questions that focus on mental health issues. Conducted by the federal government since 1971, the survey collects data by administering questionnaires to a representative sample of the population through face-to-face interviews at their place of residence.

The survey covers residents of households, noninstitutional group quarters (e.g., shelters, rooming houses, dormitories), and civilians living on military bases. Persons excluded from the survey include homeless people who do not use shelters, military personnel on active duty, and residents of institutional group quarters, such as jails or prisons and long-term hospitals. The 1999 survey marked the first year in which the national sample was interviewed using a computer-assisted interviewing (CAI) method. The survey used a combination of computer-assisted personal interviewing (CAPI) conducted by an interviewer and audio computer-assisted self-interviewing (ACASI). Use of ACASI is designed to provide the respondent with a highly private and confidential means of responding to questions and increases the level of honest reporting of illicit drug use and other sensitive behaviors. For further details on the development of the CAI procedures for the 1999 National Household Survey on Drug Abuse (NHSDA),⁹ see the Office of Applied Studies (OAS, 2001).

The 1999 through 2001 NHSDAs and the 2002 through 2013 NSDUHs employed an independent, multistage area probability sample design for each of the 50 states and the District of Columbia. For this design, eight 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 SAE while at the same time maintaining efficiency for national estimates. The design also oversampled youths and young adults, 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 estimates (note that the target sample size per year in the small states is 960 interviews except for Hawaii where the target sample size is 967 interviews). The fielded sample sizes for each state in 2016 are provided in Table C.5, and the combined 2015-2016 sample sizes are provided in Table C.9.

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, 2015b).

Nationally in 2015-2016, 267,398 addresses were screened, and 136,015 individuals responded (see Table C.9). The screening response rate (SRR) for 2015-2016 combined averaged 78.8 percent, and the interview response rate (IRR) averaged 68.8 percent, for an overall response rate (ORR) of 54.2 percent (Table C.9). The ORRs for 2015-2016 ranged from 40.2 percent in New York to 66.6 percent in New Mexico and Utah. Estimates have been adjusted to reflect the probability of selection, unit nonresponse, poststratification to known census population estimates, item imputation, and other aspects of the estimation process. These procedures are described in detail in the 2014, 2015, and 2016 NSDUHs' methodological resource books (MRBs) (CBHSQ, 2015a, 2016a, in press).

The weighted SRR is defined as the weighted number of successfully screened households (or dwelling units)¹⁰ divided by the weighted number of eligible households, or

,     D

where is the inverse of the unconditional probability of selection for the household (hh) and excludes all adjustments for nonresponse and poststratification.

At the person level, the weighted IRR is defined as the weighted number of respondents divided by the weighted number of selected persons, or

,     D

where is the inverse of the probability of selection for the ith person and includes household-level nonresponse and poststratification adjustments. To be considered a completed interview, a respondent must provide enough data to pass the usable case rule.¹¹

The weighted ORR is defined as the product of the weighted SRR and the weighted IRR or

.     D

A.3 Presentation of Data

In addition to this methodology document for the 2015-2016 state SAE results, the following files are available at https://www.samhsa.gov/data/:

• 2015-2016 NSDUH: Model-Based Prevalence Estimates (50 States and the District of Columbia) (Tables 1 to 30, by Age Group): Tables of percentages and associated 95 percent Bayesian confidence intervals are included for youths aged 12 to 17, young adults aged 18 to 25, adults aged 26 or older, adults 18 or older, and all individuals aged 12 or older. Also included are tables for underage (12 to 20) use of alcohol and underage binge alcohol use. These tables are available in Excel and PDF format. In order to increase the precision of small area estimates and rankings especially for small sample states and to detect year-to-year changes more efficiently, an SAE expert panel¹² recommended producing annual state estimates based on 2 consecutive years of pooled NSDUH data and to base comparisons of estimates (to measure change) on 2-year moving averages.
• 2015-2016 NSDUH National Maps of Prevalence Estimates, by State (Figures 1a to 30d): The color of each state on these U.S. maps indicates how the state ranks, relative to other states for each measure. States could fall into one of five groups according to their ranking by quintiles. Because 51 states were ranked for each measure, the middle quintile was assigned to 11 states, and the remaining quintiles were assigned 10 states each. In some cases, a "quintile" could have more or fewer states than desired because two (or more) states had the same estimate (to two decimal places). When such ties occurred at the "boundary" between two quintiles, all the states with the same estimate were conservatively assigned to the lower quintile. Those states with the highest rates for a given measure are in red, with the exception of the perception of risk measures, for which the lowest perceptions of great risk are in red. Those states with the lowest estimates are in white, with the exception of the perceptions of risk measures, for which the highest perceptions of great risk are in white.
  
  Note that because the past year heroin use for youths aged 12 to 17 was so low and had such an abbreviated range, no U.S. map was included.
• 2015-2016 NSDUH State Estimates Categorized into Five Groups, by Age Group: This Excel table shows the ranges of percentages for each outcome categorized into five groups (used to form the U.S. maps described above) from the lowest to highest estimate for youths aged 12 to 17, young adults aged 18 to 25, adults aged 26 or older, adults aged 18 or older, and all individuals aged 12 or older. Also included are ranges for underage (12 to 20) alcohol use and underage binge alcohol use.
• 2015-2016 NSDUH: Model-Based Estimated Totals (in Thousands) (50 States and the District of Columbia) (Tables 1 to 30): Tables showing estimated numbers (counts in thousands) and confidence intervals are included for youths aged 12 to 17, young adults aged 18 to 25, adults aged 26 or older, adults aged 18 or older, and all individuals aged 12 or older. Also included are tables for underage (12 to 20) alcohol use and underage binge alcohol use. These tables are available in Excel and PDF format.
• 2015-2016 NSDUH State-Specific Tables (Tables 1 to 112): Tables are provided for each individual state and the District of Columbia, as well as for the total United States and the four census regions (i.e., Northeast, Midwest, South, and West). The tables (two per area) show the percentages and the numbers of individuals (counts in thousands).
• NSDUH: Comparison of 2014-2015 and 2015-2016 Population Percentages (50 States and the District of Columbia) (Tables 1 to 15): Tables are presented that show the 2014-2015 (previously published data) and 2015-2016 NSDUH state estimates and an indication of the statistical significance of the difference or change (p value). Estimates are shown for youths aged 12 to 17, young adults aged 18 to 25, adults aged 26 or older, adults aged 18 or older, and all individuals aged 12 or older. Also included are tables for underage (12 to 20) alcohol use. Because annual state-level estimates are based on 2 years of pooled NSDUH data, two consecutive sets of estimates have a 1-year overlap (e.g., 2014-2015 and 2015-2016). If the population totals across the 3 years (e.g., 2014, 2015, and 2016) were the same, then the null hypothesis of no difference between the log odds of the 2014-2015 and 2015-2016 prevalence rates would be equivalent to testing the null hypothesis that the difference between the 2014-2015 and 2015-2016 prevalence rates is zero, which in turn would be equivalent to testing that the difference between the 2014 and 2016 prevalence rates is zero. The methodology used to compare these percentages is provided at the end of the tables.
• NSDUH: Comparison of 2008-2009 and 2015-2016 Population Percentages (50 States and the District of Columbia) (Tables 1 to 13): Tables are presented that show the 2008-2009 (previously published data) and 2015-2016 NSDUH state estimates and an indication of the statistical significance of the difference or change (p value). Estimates are shown for youths aged 12 to 17, young adults aged 18 to 25, adults aged 26 or older, adults aged 18 or older, and all individuals aged 12 or older. Also included are tables for underage (12 to 20) alcohol use. This comparison is done between the most recent estimates (in this case, 2015-2016) and the earliest comparable estimates for all outcomes, including mental health (based on 2008-2009 NSDUH data). The methodology used to compare these percentages is provided at the end of the tables.
• 2015-2016 NSDUH: Other Sources of State-Level Data: This document compares two outcomes (cigarette and alcohol use) from NSDUH with data from the Behavioral Risk Factor Surveillance System (BRFSS).
• 2015-2016 NSDUH: Comparison of Population Percentages from the United States, Census Regions, States, and the District of Columbia: The p values contained in these tables for each outcome and age group can be used to test the null hypothesis of no difference between population percentages for the following types of comparisons: total United States versus census region, total United States versus state, census region versus census region, census region versus state, and state versus state. The methodology used to compute these p values is provided in a document published with these Excel tables.

A.4 Confidence Intervals and Margins of Error

At the top of each of the 30 tables showing state-level model-based estimates¹³ is the design-based national estimate along with a 95 percent design-based confidence interval, all of which are based on the survey design, the survey weights, and the reported data. The state estimates are model-based statistics (using SAE methodology) that have been adjusted (benchmarked) such that the population-weighted mean of the estimates across the 50 states and the District of Columbia equals the design-based national estimate. For more details on this benchmarking, see Section B.5. The region-level estimates are also benchmarked and are obtained by taking the population-weighted mean of the associated state-level benchmarked estimates. Associated with each state and regional estimate is a 95 percent Bayesian confidence interval. These intervals indicate the uncertainty in the estimate due to both sampling variability and model fit. For example, the state with the highest estimate of past month use of marijuana for young adults aged 18 to 25 was Vermont, with an estimate of 38.2 percent and a 95 percent confidence interval that ranged from 33.3 to 43.3 percent (see Table 3 of the state model‑based estimates' tables). Assuming that sampling and modeling conditions held, the Bayes posterior probability was 0.95 that the true percentage of past month marijuana use in Vermont for young adults aged 18 to 25 in 2015-2016 was between 33.3 and 43.3 percent. As noted earlier in a Section A.1 footnote, the term "prediction interval" (PI) was used in the 2004-2005 NSDUH state report (Wright et al., 2007) and prior reports to represent uncertainty in the state and regional estimates. However, that term also is used in other applications to estimate future values of a parameter of interest. That interpretation does not apply to NSDUH state model-based estimates, so PI was replaced with "Bayesian confidence interval."

Margin of error is another term used to describe uncertainty in the estimates. For example, if is a 95 percent symmetric confidence interval for the population proportion (p) and is an estimate of p obtained from the survey data, then the margin of error of is given by or . Because is a symmetric confidence interval, will be the same as . In this case, the probability is 0.95 that the interval ± or ± will contain the true population value (p). This defined margin of error will vary for each estimate and will be affected not only by the sample size (e.g., the larger the sample, the smaller the margin of error), but also by the sample design (e.g., telephone surveys using random digit dialing and surveys employing a stratified multistage cluster design will, more than likely, produce a different margin of error) (Scheuren, 2004).

The confidence intervals shown in NSDUH reports are asymmetric, meaning that the distance between the estimate and the lower confidence limit will not be the same as the distance between the upper confidence limit and the estimate. For example, Utah's past month marijuana use estimate is 12.1 percent for adults aged 18 to 25 years, with a 95 percent confidence interval equal to (9.6 - 15.0) (see Table 3 of the state model-based estimates' tables). Therefore, Utah's estimate is 2.5 (i.e., 12.1 - 9.6) percentage points from the lower 95 percent confidence limit and 2.9 (i.e., 15.0 - 12.1) percentage points from the upper limit. These asymmetric confidence intervals work well for small percentages often found in NSDUH tables and reports while still being appropriate for larger percentages. Some surveys or polls provide only one margin of error for all reported percentages. This single number is usually calculated by setting the sample percentage estimate () equal to 50 percent, which will produce an upper bound or maximum margin of error. Such an approach would not be feasible in NSDUH because the estimates vary from less than 1 percent to over 75 percent; hence, applying a single margin of error to these estimates could significantly overstate or understate the actual precision levels. Therefore, given the differences mentioned above, it is more useful and informative to report the confidence interval for each estimate instead of a margin of error.

When it is indicated that a state has the highest or lowest estimate, it does not imply that the state's estimate is significantly higher or lower than the next highest or lowest state's estimate. Additionally, two significantly different state estimates (at the 5 percent level of significance) may have overlapping 95 percent confidence intervals. For details on a more accurate test to compare state estimates, see the "2015-2016 National Survey on Drug Use and Health: Comparison of Population Percentages from the United States, Census Regions, States, and the District of Columbia" at https://www.samhsa.gov/data/.

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 used only 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 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, such as cigars, pipes, and smokeless tobacco.

A.6 2015 NSDUH Changes and Their Effects on Small Area Estimates

In 2015, a number of changes were made to the NSDUH questionnaire and data collection procedures. These changes were intended to improve the quality of the data that were collected and to address the changing needs of substance use and mental health policy and research.¹⁴ This section briefly summarizes the effect of the redesign on the comparability between the 2015 NSDUH and earlier NSDUHs, specifically related to the SAE outcomes. For a more detailed discussion of the questionnaire redesign and its effect, see Section C of the 2015 NSDUH's methodological summary and definitions report (CBHSQ, 2016b) and a brief report summarizing the implications of the changes for data users (CBHSQ, 2016c).

In the alcohol section of the questionnaire, the threshold for defining binge alcohol use among females was revised from five or more drinks on an occasion to four or more drinks on an occasion to ensure consistency with federal definitions.¹⁵ The threshold for males in 2015 remained at five or more drinks on an occasion. Consequently, a new baseline was established in 2015 for estimates of binge alcohol for the overall population. Small area estimates for past month binge alcohol use using combined 2015 and 2016 data were produced creating a new baseline. Because estimates using combined 2014 and 2015 data were not produced, no comparison between the two sets of years (i.e., 2014-2015 vs. 2015-2016) was done. Note that this change did not affect estimates for alcohol use or alcohol use disorder.

Several changes were made to the various illicit drug modules. Specifically, changes were made to the hallucinogen, inhalant, methamphetamine, and prescription psychotherapeutic modules. For details on these specific changes, see Section C.1 of the 2015 NSDUH methodological summary and definitions report (CBHSQ, 2016b). These changes resulted in the need to revise the baseline using 2015 and 2016 NSDUH data for several small area estimates showing overall illicit drug use (including use disorder and treatment) and pain reliever misuse.¹⁶

Additionally, changes to some of the drug modules might have affected the set of respondents in 2015 who were eligible to be asked questions about treatment for substance use. Hence, SAE outcomes on needing but not receiving treatment (for illicit drugs and alcohol) were potentially affected. Thus, substance use treatment estimates were produced using combined 2015 and 2016 NSDUH data as a new baseline. Because estimates for these treatment outcomes using combined 2014 and 2015 data were not produced, no comparison between the two sets of years was done.

Finally, although questions on the perceptions of risk of harm from using different substances did not change in 2015, data quality checks on preliminary data and the full 2015 data showed deviations from the expected trends for these measures. A survey redesign carries the risk that preceding changes to the questionnaire will affect how respondents answer later questions (e.g., context effects). A context effect may be said to take place when the response to a question is affected by information that is not part of the question itself. For example, the content of a preceding question may affect the interpretation of a subsequent question. Or a respondent may answer a subsequent question in a manner that is consistent with responses to a preceding question if the two questions are closely related to each other. The set of questions preceding the risk and availability module in the 2015 questionnaire had undergone a number of significant changes that could have affected the way in which respondents answered the perceived risk and availability questions. Because of these deviations, the perception of risk estimates were not produced using combined 2014 and 2015 NSDUH data. Estimates were produced using combined 2015 and 2016 NSDUH data, establishing a new baseline.

To summarize, several changes in the 2015 questionnaire had impacts on the comparability of the 2014 and 2015 NSDUH data. It was decided, therefore, that for those measures data across those 2 years could not be pooled, and estimates for those measures could not be produced using 2014 and 2015 NSDUH data. Estimates for these measures are included using the 2015-2016 NSDUH data, establishing a new baseline. Note that because 2014-2015 estimates were not produced for some outcomes, change estimates between 2014-2015 and 2015-2016 were not produced. For a complete list of outcomes for which small area estimates are available using 2014-2015 NSDUH data, refer to Table C.15.

Section B: State Model-Based Estimation Methodology

B.1 General Model Description

The model can be characterized as a complex mixed¹⁷ model (including both fixed and random effects) of the following form:

,     D

where is the probability of engaging in the behavior of interest (e.g., using marijuana in the past month) for person-k belonging to age group-a in grouped state sampling region (SSR)-j of state-i.¹⁸ Let denote a vector of auxiliary (predictor) variables associated with age group-a (12 to 17, 18 to 25, 26 to 34, and 35 or older) and 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 vectors of state-level random effects and grouped SSR-level random effects are assumed to be mutually independent with and where is the total number of individual age groups modeled (generally, ). For hierarchical Bayes (HB) estimation purposes, an improper uniform prior distribution is assumed for , and proper Wishart prior distributions are assumed for and . The HB solution for 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 posterior distribution. The basic process is described in Folsom et al. (1999), Shah, Barnwell, Folsom, and Vaish (2000), and Wright (2003a, 2003b).

Once the required number of MCMC samples (1,250 in all) for the parameters of interest are generated and tested for convergence properties (see Raftery & Lewis, 1992), the small area estimates for each race/ethnicity × gender cell within a block group can be obtained for each age group. These block group-level small area estimates then can be aggregated using the appropriate population count projections for the desired age group(s) to form state-level small area estimates. These state-level small area estimates are benchmarked to the national design-based estimates as described in Section B.5.

B.2 Variables Modeled

The 2016 National Survey on Drug Use and Health (NSDUH) data were pooled with the 2015 NSDUH data, and age group-specific state estimates for 29 binary (0, 1) outcomes listed below were produced. Comparisons between the 2014-2015 and the 2015-2016 state estimates also were produced for measures marked with an asterisk (*).

1. past month use of illicit drugs,
2. past year use of marijuana,*
3. past month use of marijuana,*
4. perceptions of great risk from smoking marijuana once a month,
5. average annual rate of first use of marijuana,*¹⁹
6. past month use of illicit drugs other than marijuana,
7. past year use of cocaine,*
8. perceptions of great risk from using cocaine once a month,
9. past year use of heroin,*
10. perceptions of great risk from trying heroin once or twice,
11. past year misuse of pain relievers,
12. past month use of alcohol,*²⁰
13. past month binge alcohol use,²¹
14. perceptions of great risk from having five or more drinks of an alcoholic beverage once or twice a week,
15. past month use of tobacco products,*
16. past month use of cigarettes,*
17. perceptions of great risk from smoking one or more packs of cigarettes per day,
18. past year illicit drug use disorder,
19. past year pain reliever use disorder,
20. past year alcohol use disorder,*
21. past year substance use disorder,
22. past year needing but not receiving treatment for illicit drug use at a special facility,
23. past year needing but not receiving treatment for alcohol use at a special facility,
24. past year needing but not receiving treatment for substance use at a special facility,
25. serious mental illness (SMI) in the past year,*²²
26. any mental illness (AMI) in the past year,*
27. received mental health services in the past year,*
28. had serious thoughts of suicide in the past year,* and
29. past year major depressive episode (MDE, i.e., depression).*

Some of the outcomes in this list above were not comparable between 2014-2015 and 2015-2016 (as discussed in Section A.6 above). Table C.15 shows all the SAE outcomes and the years they are available; thus, this table can be used to see outcomes for which small area estimates were produced using 2014-2015 and 2015-2016 NSDUH data.

B.3 Predictors Used in Mixed Logistic Regression Models

Local area data used as potential predictor variables in the mixed logistic regression models were obtained from a number of sources, as noted in the following discussion. Variable selection was done using combined 2015 and 2016 data for all outcomes. Fixed-effect predictors for were selected using the method described in Section B.4.

Sources and potential data items used in the 2015-2016 modeling are provided in the following text and lists.

• Claritas. The demographic data package used from Claritas²³ contains data for 2013 with projections to 2018. The population projections are used to update these predictor variables each year. The 2015 and 2016 population projections were used for producing the 2015-2016 state small area estimates.
• U.S. Census Bureau. The 2010 census (demographic and geographic variables) and 2014 food stamp participation estimates were used (https://www.census.gov/data/datasets/time-series/demo/saipe/model-tables.html). The Census Bureau's Small Area Income and Poverty Estimates (SAIPE) program obtains Food Stamp program (now known as the Supplemental Nutrition Assistance Program [SNAP]) participation estimates from the U.S. Department of Agriculture, Food and Nutrition Service. Also, the Census Bureau's 2011-2015 American Community Survey (ACS) 5 year ACS demographic and socioeconomic variables at the tract level and poverty variable at the county level were used (https://www.census.gov/programs-surveys/acs/).
• Federal Bureau of Investigation (FBI). Uniform Crime Report (UCR) arrest totals were obtained from https://www.icpsr.umich.edu/icpsrweb/NACJD/archive.jsp. The most current data used are from 2014 for most counties, with previous years' data substituted in a few cases.
• Bureau of Labor Statistics (BLS). The 2016 county-level unemployment estimates were used (https://www.bls.gov/lau/tables.htm). The BLS uses results from the Current Population Survey (CPS) to provide county-level unemployment estimates. The CPS is a monthly survey of households conducted by the Census Bureau for the BLS.
• Bureau of Economic Analysis (BEA). The 2015 county-level per capita income estimates were used (https://bea.gov/iTable/index.cfm). These county-level per capita income estimates are produced by the Regional Income Division of the BEA.
• National Center for Health Statistics (NCHS). Mortality data using International Classification of Diseases, 10th revision (ICD-10), 2007-2012, were used. The ICD-10 death data are from the NCHS at the Centers for Disease Control and Prevention (CDC).
• SAMHSA, Center for Behavioral Health Statistics and Quality (CBHSQ, formerly the Office of Applied Studies [OAS]). Data were used from the National Survey of Substance Abuse Treatment Services (N-SSATS), formerly known as the Uniform Facility Data Set (UFDS). The 2013 and 2015 data on drug and alcohol treatment estimates were obtained. Maintenance of effort expenditures, block grant awards, cost of services, and total taxable resources data were also used.

The following lists provide the specific independent variables that were potential predictors in the models.

Claritas Data (Description)	Claritas Data (Level)
% Population Aged 0 to 19 in Block Group	Block Group
% Population Aged 20 to 24 in Block Group	Block Group
% Population Aged 25 to 34 in Block Group	Block Group
% Population Aged 35 to 44 in Block Group	Block Group
% Population Aged 45 to 54 in Block Group	Block Group
% Population Aged 55 to 64 in Block Group	Block Group
% Population Aged 65 or Older in Block Group	Block Group
% Non-Hispanic Blacks in Block Group	Block Group
% Hispanics in Block Group	Block Group
% Non-Hispanic Other Races in Block Group	Block Group
% Non-Hispanic Whites in Block Group	Block Group
% Males in Block Group	Block Group
% American Indians, Eskimos, Aleuts in Tract	Tract
% Asians, Pacific Islanders in Tract	Tract
% Population Aged 0 to 19 in Tract	Tract
% Population Aged 20 to 24 in Tract	Tract
% Population Aged 25 to 34 in Tract	Tract
% Population Aged 35 to 44 in Tract	Tract
% Population Aged 45 to 54 in Tract	Tract
% Population Aged 55 to 64 in Tract	Tract
% Population Aged 65 or Older in Tract	Tract
% Non-Hispanic Blacks in Tract	Tract
% Hispanics in Tract	Tract
% Non-Hispanic Other Races in Tract	Tract
% Non-Hispanic Whites in Tract	Tract
% Males in Tract	Tract
% Population Aged 0 to 19 in County	County
% Population Aged 20 to 24 in County	County
% Population Aged 25 to 34 in County	County
% Population Aged 35 to 44 in County	County
% Population Aged 45 to 54 in County	County
% Population Aged 55 to 64 in County	County
% Population Aged 65 or Older in County	County
% Non-Hispanic Blacks in County	County
% Hispanics in County	County
% Non-Hispanic Other Races in County	County
% Non-Hispanic Whites in County	County
% Males in County	County

American Community Survey (ACS) (Description)	ACS Data (Level)
1 This is a new predictor added for the 2015-2016 SAE processing.
% Population Who Dropped Out of High School	Tract
% Housing Units Built in 1940 to 1949	Tract
% Females 16 Years or Older in Labor Force	Tract
% Females Never Married	Tract
% Females Separated, Divorced, Widowed, or Other	Tract
% One-Person Households	Tract
% Males 16 Years or Older in Labor Force	Tract
% Males Never Married	Tract
% Males Separated, Divorced, Widowed, or Other	Tract
% Housing Units Built in 1939 or Earlier	Tract
Average Number of Persons per Room	Tract
% Families below Poverty Level	Tract
% Households with Public Assistance Income	Tract
% Housing Units Rented	Tract
% Population with 9 to 12 Years of School, No High School Diploma	Tract
% Population with 0 to 8 Years of School	Tract
% Population with Associate's Degree	Tract
% Population with Some College and No Degree	Tract
% Population with Bachelor's, Graduate, Professional Degree	Tract
% Housing Units with No Telephone Service Available	Tract
% Households with No Vehicle Available	Tract
% Population with No Health Insurance1	Tract
Median Rents for Rental Units	Tract
Median Value of Owner-Occupied Housing Units	Tract
Median Household Income	Tract
% Families below the Poverty Level	County

Uniform Crime Report (UCR) Data (Description)	UCR Data (Level)
Drug Possession Arrest Rate	County
Drug Sale or Manufacture Arrest Rate	County
Drug Violations' Arrest Rate	County
Marijuana Possession Arrest Rate	County
Marijuana Sale or Manufacture Arrest Rate	County
Opium or Cocaine Possession Arrest Rate	County
Opium or Cocaine Sale or Manufacture Arrest Rate	County
Other Drug Possession Arrest Rate	County
Other Dangerous Non-Narcotics Arrest Rate	County
Serious Crime Arrest Rate	County
Violent Crime Arrest Rate	County
Driving under Influence Arrest Rate	County

Other Categorical Data (Description)	Other Categorical Data (Source)	Other Categorical Data (Level)
= 1 if Hispanic, = 0 Otherwise	National Survey on Drug Use and Health (NSDUH) Sample	Person
= 1 if Non-Hispanic Black, = 0 Otherwise	NSDUH Sample	Person
= 1 if Non-Hispanic Other, = 0 Otherwise	NSDUH Sample	Person
= 1 if Male, = 0 if Female	NSDUH Sample	Person
= 1 if Metropolitan Statistical Area (MSA) with ≥ 1 Million, = 0 Otherwise	2010 Census	County
= 1 if MSA with < 1 Million, = 0 Otherwise	2010 Census	County
= 1 if Non-MSA Urban, = 0 Otherwise	2010 Census	Tract
= 1 if Urban Area, = 0 if Rural Area	2010 Census	Tract
= 1 if No Cubans in Tract, = 0 Otherwise	2010 Census	Tract
= 1 if No Arrests for Dangerous Non-Narcotics, = 0 Otherwise	Uniform Crime Report (UCR)	County
= 1 if No Arrests for Opium or Cocaine Possession = 0 Otherwise	UCR	County
= 1 if No Housing Units Built in 1939 or Earlier, = 0 Otherwise	American Community Survey (ACS)	Tract
= 1 if No Housing Units Built in 1940 to 1949, = 0 Otherwise	ACS	Tract
= 1 if No Households with Public Assistance Income, = 0 Otherwise	ACS	Tract

Miscellaneous Data (Description)	Miscellaneous Data (Source)	Miscellaneous Data (Level)
Alcohol Death Rate, Underlying Cause	National Center for Health Statistics' International Classification of Diseases, 10th revision (NCHS-ICD-10)	County
Cigarette Death Rate, Underlying Cause	NCHS-ICD-10	County
Drug Death Rate, Underlying Cause	NCHS-ICD-10	County
Alcohol Treatment Rate	National Survey of Substance Abuse Treatment Services (N-SSATS) (Formerly Called Uniform Facility Data Set [UFDS])	County
Alcohol and Drug Treatment Rate	N-SSATS (Formerly Called UFDS)	County
Drug Treatment Rate	N-SSATS (Formerly Called UFDS)	County
Unemployment Rate	Bureau of Labor Statistics (BLS)	County
Per Capita Income (in Thousands)	Bureau of Economic Analysis (BEA)	County
Average Suicide Rate (per 10,000)	NCHS-ICD-10	County
Food Stamp Participation Rate	Census Bureau	County
Single State Agency Maintenance of Effort	National Association of State Alcohol and Drug Abuse Directors (NASADAD)	State
Block Grant Awards	Substance Abuse and Mental Health Services Administration (SAMHSA)	State
Cost of Services Factor Index	SAMHSA	State
Total Taxable Resources per Capita Index	U.S. Department of Treasury	State
% Hispanics Who Are Cuban	2010 Census	Tract

B.4 Selection of Independent Variables for the Models

New variable selection was done for all measures listed in Section B.2 using 2015-2016 NSDUH data in a manner consistent with how it was done in prior NSDUHs. To produce small area estimates based on the pooled 2015 and 2016 NSDUH data, the fixed-effect predictors were selected using the following methodology:

1. There were 136,015 respondents in the pooled 2015 and 2016 NSDUH data. Any variable selection performed on such a large dataset would result in an excessive number of predictors in the final model. To avoid this and build parsimonious models, the pooled data were partitioned into modeling and validation samples. For more information on how the data was partitioned, see the 2002-2003 state SAE report (Wright & Sathe, 2005). The modeling sample was first used to get a preliminary list of significant predictors using the variable selection methodology described below. These predictors were further reduced by using SUDAAN^® logistic regression on the validation dataset resulting in parsimonious models (RTI International, 2012).
2. Separate SAS^® stepwise logistic regression models were fit to the modeling sample for all outcomes by four age group domains (12 to 17, 18 to 25, 26 to 34, and 35 or older) (SAS Institute Inc., 2008). The input list to these models included all linear polynomials (constructed from continuous predictor variables) and other categorical or indicator variables given in Section B.3. All predictors that were significant then were input to step 3 of variable selection.
3. Using modeling sample, all significant predictors from step 2 then were input to PROC HPSPLIT to identify significant complex (at most three-way) interaction terms. Proc HPSPLIT is a SAS procedure that uses decision-tree algorithms to build classification systems. The exhaustive chi-squared automatic interaction detector (CHAID) algorithm was used to create the trees. The constraints for making a tree were maximum depth = 3, minimum number of records in child node = 300, and splitting criterion = 3 percent.
4. All the significant variables from step 2 along with their corresponding higher order polynomials (quadratic and cubic), interaction of gender and race, and the significant interactions detected by PROC HPSPLIT in step 3 then were input to SAS stepwise logistic regression models, run on modeling sample. All predictors that remained significant then were input to step 5 of variable selection.
5. All significant variables from step 4 were input to fit SUDAAN logistic regression models on the validation sample, and predictors that remained significant were used in the mixed logistic regression model described in Section B.1. The race and gender predictors were forced in all models.

B.5 Benchmarking the Age Group-Specific Small Area Estimates

The self-calibration built into the survey-weighted hierarchical Bayes (SWHB) solution ensures that the population-weighted average of the state small area estimates will closely match the national design-based estimates. The national design-based estimates in NSDUH are based entirely on survey-weighted data using a direct estimation approach, whereas the state and census region estimates are model-based. Given the self-calibration ensured by the SWHB method, for state reports prior to 2002, the standard Bayes prescription was followed; specifically, the posterior mean was used for the point estimate, and the tail percentiles of the posterior distribution were used for the Bayesian confidence interval limits.

Singh and Folsom (2001) extended Ghosh's (1992) results on constrained Bayes estimation to include exact benchmarking to design-based national estimates. In the simplest version of this constrained Bayes solution where only the design-based mean is imposed as a benchmarking constraint, each of the 2015-2016 state-by-age group small area estimates is adjusted by adding the common factor where is the design-based national estimate and is the population-weighted mean of the state small area estimates for age group-a. The exactly benchmarked state-s and age group-a small area estimates then are given by Experience with such additive adjustments suggests that the resulting exactly benchmarked state small area estimates will always be between 0 percent and 100 percent because the SWHB self-calibration ensures that the adjustment factor is small relative to the size of the state-level small area estimates.

Relative to the Bayes posterior mean, these benchmark-constrained state small area estimates are biased by the common additive adjustment factor. Therefore, the posterior mean squared error (MSE) for each benchmarked state small area estimate has the square of this adjustment factor added to its posterior variance. To achieve the desirable feature of exact benchmarking, this constrained Bayes adjustment factor was implemented for the state-by-age group small area estimates. The associated Bayesian confidence (credible) intervals can be recentered at the benchmarked small area estimates on the logit scale with the symmetric interval end points based on the posterior root mean squared errors (RMSEs). The adjusted 95 percent Bayesian confidence intervals are defined below:

     D

where

     D

     D

and

     D

The associated posterior coverage probabilities for these benchmarked intervals are very close to the prescribed 0.95 value because the state small area estimates have posterior distributions that can be approximated exceptionally well by a Gaussian distribution after the logit transformation.

B.6 Calculation of Estimated Number of Individuals Associated with Each Outcome

Tables 1 to 30 of "2015-2016 NSDUHs: Model-Based Estimated Totals (in Thousands) (50 States and the District of Columbia)" show the estimated numbers of individuals associated with each of the 29 outcomes of interest.²⁴ To calculate these numbers, the benchmarked small area estimates and the associated 95 percent Bayesian confidence intervals are multiplied by the average population across the 2 years (in this case, 2015 and 2016) of the state by the age group of interest.

For example, past month use of alcohol among 18 to 25 year olds in Alabama was 50.76 percent.²⁵ The corresponding Bayesian confidence intervals ranged from 46.91 to 54.60 percent. The population count for 18 to 25 year olds averaged across 2015 and 2016 in Alabama was 522,750 (see Table C.10 in Section C of this methodology document). Hence, the estimated number of 18 to 25 year olds using alcohol in the past month in Alabama was 0.5076 × 522,750, which is 265,348.²⁶ The associated Bayesian confidence intervals ranged from 0.4691 × 522,750 (i.e., 245,222) to 0.5460 × 522,750 (i.e., 285,422). Note that when estimates of the number of individuals are calculated for Tables 1 to 30 in "2015-2016 NSDUHs: Model-Based Estimated Totals (in Thousands) (50 States and the District of Columbia)" (follow the link in footnote 24), the unrounded percentages and population counts are used, then the numbers are reported to the nearest thousand. Hence, the number obtained by multiplying the published estimate with the published population estimate may not exactly match the counts that are published in these tables because of rounding differences.

The only exception to this calculation is the production of the estimated numbers of marijuana initiates. Those estimates cannot be directly calculated as the product of the percentage estimate of first use of marijuana and the population counts available in Section C. That is because the denominator of that percentage estimate is defined as the number of person years at risk for marijuana initiation, which is a combination of individuals who never used marijuana and one half of the individuals who initiated in the past 24 months (see Section B.8 for more details).

B.7 Calculation of Aggregate Age Group Estimates and Limitations

Tables 1 to 30 of "2015-2016 NSDUHs: Model-Based Prevalence Estimates (50 States and the District of Columbia)" show estimates for the following age groups: 12 to 17, 18 to 25, 26 or older, 18 or older, and 12 or older.²⁷ If a user was interested in producing aggregated estimates, such as for those aged 12 to 25, the aggregated estimates could be calculated using prevalence estimates along with the population totals shown in Section C of this document. However, with the information that is provided in the tables, the confidence intervals cannot be calculated. Below is an example of this calculation for a given state.

For example, past month use of alcohol in Alabama among youths 12 to 17 was 8.08 percent, and among young adults 18 to 25 it was 50.76 percent.²⁸ The population counts for 12 to 17 year olds, and 18 to 25 year olds, averaged across 2015 and 2016 in Alabama were 378,330 and 522,750, respectively (see Table C.10 in Section C of this methodology document). Hence, one would calculate the estimate for individuals aged 12 to 25 by first finding the number of users aged 12 to 25, which is 295,917 ([0.0808 × 378,330] + [0.5076 × 522,750]), then dividing that number by the population aged 12 to 25, which results in a rate of 32.84 percent (295,917 / [378,330 + 522,750]).

B.8 Calculation of Average Annual Incidence of Marijuana Use

Incidence rates typically are calculated as the number of new initiates of a substance during a period of time (such as in the past year) divided by an estimate of the number of person-years of exposure (in thousands). The 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 is defined as follows:

     D

where is the number of marijuana initiates in the past 24 months and is the number of persons who never used marijuana.

The incidence rate is expressed as a percentage or rate per 100 person-years of exposure. Note that this estimate uses a 2-year time period to accumulate incidence cases from each annual survey. By assuming further that the distribution of first use for the incidence cases is uniform across the 2-year interval, the total number of person-years of exposure is 1 year on average for the incidence cases plus 2 years for all the "never users" at the end of the time period. This approximation to the person-years of exposure permits one to recast the incidence rate as a function of two population prevalence rates, namely, the fraction of persons who first used marijuana in the past 2 years and the fraction who had never used marijuana. Both of these prevalence estimates were estimated using the SWHB estimation approach.

The count of persons who first used marijuana in the past 2 years is based on a "moving" 2-year period that ranges over 3 calendar years. Subjects were asked when they first used marijuana. If a person indicated first use of marijuana between the day of the interview and 2 years prior, the person was included in the count. Thus, it is possible for a person interviewed in the first part of 2016 to indicate first use as early as the first part of 2014 or as late as the first part of 2016. Similarly, a subject interviewed in the last part of 2016 could indicate first use as early as the last part of 2014 or as late as the last part of 2016. Therefore, in the 2016 survey, the reported period of first use ranged from early 2014 to late 2016 and was "centered" in 2015. For example, about half of the 12 to 17 year olds who reported first use in the past 24 months reported first use in 2015, while a quarter each reported first use in 2014 and 2016. Persons who responded in 2016 that they had never used marijuana were included in the count of "never used." Similarly, reports of first use in the past 24 months from the 2015 survey ranged from early 2013 to late 2015 and were centered in 2014. Half of the 12 to 17 year olds who reported first use in the past 24 months reported first use in 2014, while a quarter each reported first use in 2013 and 2015. Note that only incidence rates for marijuana use are provided here.

B.9 Underage Drinking

To obtain small area estimates for individuals aged 12 to 20 for past month alcohol and binge alcohol use, a separate set of models was fit for these two outcomes for the 12 to 17 age group and the 18 to 20 age group. Model-based estimates for individuals aged 12 to 20 were produced by taking the population-weighted average of the individual age group (12 to 17 and 18 to 20) estimates. Estimates for underage drinking for past month alcohol and binge alcohol use were benchmarked to match national design-based estimates for that age group using the process described in Section B.5.

B.10 Substance Use Disorder / Needing But Not Receiving Treatment

The NSDUH computer-assisted interviewing (CAI) instrumentation includes questions that are designed to measure dependence or abuse of alcohol and illicit drugs (i.e., SUDs). For these substances,²⁹ dependence and abuse questions were based on the criteria in the DSM-IV (APA, 1994).

Specifically, for marijuana, hallucinogens, inhalants, and tranquilizers, a respondent was defined as having dependence if he or she met three or more of the following six dependence criteria:

1. Spent a great deal of time over a period of a month getting, using, or getting over the effects of the substance.
2. Used the substance more often than intended or was unable to keep set limits on the substance use.
3. Needed to use the substance more than before to get desired effects or noticed that the same amount of substance use had less effect than before.
4. Inability to cut down or stop using the substance every time tried or wanted to.
5. Continued to use the substance even though it was causing problems with emotions, nerves, mental health, or physical problems.
6. The substance use reduced or eliminated involvement or participation in important activities.

For alcohol, cocaine, heroin, methamphetamine, pain relievers, sedatives, and prescription stimulants, a seventh withdrawal criterion was added. The seventh withdrawal criterion is defined by a respondent reporting having experienced a certain number of withdrawal symptoms that vary by substance (e.g., having trouble sleeping, cramps, hands tremble). A respondent was defined as having dependence if he or she met three or more of seven dependence criteria for these substances.

For each illicit drug and alcohol, a respondent was defined as having abused that substance if he or she met one or more of the following four abuse criteria and was determined not to be dependent on the respective substance in the past year (i.e., because dependence takes precedence over abuse):

1. Serious problems at home, work, or school caused by the substance, such as neglecting your children, missing work or school, doing a poor job at work or school, or losing a job or dropping out of school.
2. Used the substance regularly and then did something that might have put you in physical danger.
3. Use of the substance caused you to do things that repeatedly got you in trouble with the law.
4. Had problems with family or friends that were probably caused by using the substance and continued to use the substance even though you thought the substance use caused these problems.

For additional details on how respondents were classified as having substance use disorder, see Section B.4.3 in Section B of the 2016 NSDUH methodological summary and definitions report (CBHSQ, 2017).

Additionally, the NSDUH CAI instrument included a series of questions that are designed to measure treatment need for an alcohol or illicit drug use problem and to determine persons needing but not receiving treatment. Respondents were classified as needing substance use treatment in the past year if they met either of the following criteria:

1. presence of an SUD in the past year for alcohol or illicit drugs (i.e., dependence or abuse) (see Section B.4.3 in Section B of CBHSQ, 2017); or
2. receipt of treatment at a specialty facility (i.e., drug and alcohol rehabilitation facility [inpatient or outpatient], hospital [inpatient only], or mental health center) in the past year for the use of alcohol or illicit drugs (or both).

A respondent was classified as needing but not receiving treatment for an alcohol problem if he or she met the criteria for alcohol dependence or abuse in the past year, but did not receive treatment at a specialty facility for an alcohol problem in the past year.

For additional details on how respondents were classified as needing substance use treatment, see Section B.4.4 in Section B of the 2016 NSDUH methodological summary and definitions report (CBHSQ, 2017).

B.11 Mental Health Measures

This section provides a summary of the measurement issues associated with three of the mental health outcome variables—SMI, AMI, and MDE. Additional details can be found in Sections B.4.6 through B.4.8 in Section B of the 2016 NSDUH methodological summary and definitions report (CBHSQ, 2017).

B.11.1 Mental Illness

In the 2000-2001 and 2002-2003 NSDUH state SAE reports (Wright, 2003a, 2003b; Wright & Sathe, 2005), the Kessler-6 (K6) distress scale was used to measure SMI (Kessler et al., 2003). However, SAMHSA discontinued producing state-level SMI estimates beginning with the release of the 2003-2004 state report (Wright & Sathe, 2006) because of concerns about the validity of using only the K6 distress scale without an impairment scale; see Section B.4.4 in Appendix B of the 2004 NSDUH national findings report (OAS, 2005). The use of the K6 distress scale continued in the 2003-2004 and the 2004-2005 state reports (Wright & Sathe, 2006; Wright et al., 2007), not as a measure of SMI, but as a measure of serious psychological distress (SPD) because it was determined that the K6 scale measured only SPD and merely contributed to measuring SMI and AMI (see the details that follow).

In December 2006, a new technical advisory group was convened by SAMHSA's OAS (which later became CBHSQ) and the Center for Mental Health Services (CMHS) to solicit recommendations for data collection strategies to address SAMHSA's legislative requirements. Although it was recognized that the ideal way to estimate SMI in NSDUH would be to administer a clinical diagnostic interview annually to all 45,000 adult respondents, this approach was not feasible because of constraints on the interview time and the need for trained mental health clinicians to conduct the interviews. Therefore, the approach recommended by the technical advisory group and adopted by SAMHSA for NSDUH was to utilize short scales in the NSDUH interview that separately measure psychological distress and functional impairment for use in a statistical model that predicts whether a respondent had mental illness.

In response, SAMHSA's CBHSQ initiated a Mental Health Surveillance Study (MHSS) under its NSDUH contract with RTI International to develop and implement methods to estimate SMI. Based on recommendations from this panel, estimates of SMI were presented based on a revised methodology and, thus, were not comparable with estimates for SMI or SPD shown in NSDUH state reports prior to 2009. However, in 2013, another revision to the methodology for creating SMI estimates was made, and the estimates presented for 2011 and 2012 are based on this revised methodology (and therefore are not comparable with previously published estimates of SMI). Thus, the 2008-2009, 2009-2010, and 2010-2011 SMI estimates were reproduced using the new 2013 methodology.

To develop methods for preparing the estimates of SMI and AMI presented in this and other NSDUH reports and documents, the MHSS was initiated as part of the 2008 NSDUH design and analysis. Because of constraints on the interview time in NSDUH and the need for trained mental health clinicians, it was not possible to administer a full structured diagnostic clinical interview to assess mental illness on approximately 45,000 adult respondents; therefore, the approach adopted by SAMHSA was to utilize short scales separately measuring psychological distress (K6) and functional impairment that could be used in a statistical model to accurately predict whether a respondent had a mental illness. Two impairment scales—the World Health Organization Disability Assessment Schedule (WHODAS) and the Sheehan Disability Scale (SDS)—were included in the 2008 survey for evaluation. The collection of clinical psychiatric interview data was achieved using a subsample of approximately 1,500 adult NSDUH participants in 2008. These participants were recruited for a follow-up clinical interview consisting of a gold-standard diagnostic assessment for mental disorders and functional impairment. In order to determine the optimal scale to measure functional impairment, a split-sample design was incorporated into the full 2008 NSDUH data collection in which half of the adult respondents received the WHODAS and half received the SDS (only the WHODAS scale was used starting in 2009). The 2008 statistical models (subsequently referred to as the "2008 model") using the data from the subsample of respondents collected as part of the MHSS then were developed for each half sample in which the short scales (the K6 in combination with the WHODAS or the K6 in combination with the SDS) were used as predictors in models of mental illness assessed via the clinical interviews. The model parameter estimates then were used to predict SMI in the full 2008 NSDUH sample. SMI probabilities and SMI predicted values (as well as for AMI) were computed for respondents in NSDUH samples from 2008 to 2011 using model parameter estimates from the 2008 model.

In 2010, SAMHSA began preliminary investigations to assess whether improvements to the model were warranted using all of the clinical data that had been collected since 2008. In 2011 and 2012, the clinical sample was augmented to include 1,500 respondents per year, leading to a combined sample of approximately 5,000 clinical interviews for 2008 to 2012. SAMHSA determined that the 2008 model had some important shortcomings that had not been detected in the original model fitting because of the small number of respondents in the 2008 clinical subsample. Specifically, the 2008 model substantially overestimated SMI and AMI among young adults aged 18 to 25 relative to the clinical interview data. In addition, improvements were needed in the weighting procedures for the MHSS sample data to account better for nonresponse and undercoverage. Therefore, SAMHSA decided to modify the model for the 2012 estimates using the combined 2008-2012 clinical data (subsequently referred to as the "2012 model"). To reduce bias and improve prediction, additional mental health-related variables and an age variable were added in the 2012 model. To provide consistent data for trend assessment, state mental illness estimates for 2008-2009, 2009-2010, and 2010-2011 were also recomputed using the new 2012 model. Note that tables or maps showing estimates of AMI and SMI based on these 2012 models include "Revised October 2013" in the source line for estimates using 2008 through 2011 data.

The next few paragraphs describe the instruments and items used to measure the variables employed in the 2012 model. Specifically, the instrument used to measure mental illness in the clinical interviews is described, followed by descriptions of the scales and items in the main NSDUH interviews that were used as predictor variables in the model (e.g., the K6 and WHODAS total scores, age, MDE, and suicidal thoughts).

Clinical Measurement of Mental Illness. Mental illness was measured in the MHSS clinical interviews using an adapted version of the Structured Clinical Interview for the DSM-IV-TR Axis I Disorders, Research Version, Non-patient Edition (SCID-I/NP) (First, Spitzer, Gibbon, & Williams, 2002) and was differentiated by the level of functional impairment based on the Global Assessment of Functioning (GAF) scale (Endicott, Spitzer, Fleiss, & Cohen, 1976). Past year disorders that were assessed through the SCID included mood disorders (e.g., MDE, manic episode), anxiety disorders (e.g., panic disorder, generalized anxiety disorder, posttraumatic stress disorder), eating disorders (e.g., anorexia nervosa), intermittent explosive disorder, and adjustment disorder. In addition, the presence of psychotic symptoms was assessed. SUDs also were assessed, although these disorders were not used to produce estimates of mental illness.

• Respondents were defined as having any mental illness (AMI) if they were determined to have any of the mental disorders assessed in the SCID (not including SUDs), regardless of the level of functional impairment.
• Respondents were defined as having serious mental illness (SMI) if they had any of the mental disorders assessed in the SCID (not including SUDs), and these disorders resulted in substantial impairment in carrying out major life activities, based on GAF scores of 50 or below. The SMI diagnosis was used as the response variable in both the 2008 and 2012 prediction models.

The SCID and the GAF in combination were considered to be the gold standard for measuring mental illness.

Kessler-6 (K6) Distress Scale. The K6 in the main NSDUH consists of two sets of six questions that asked adult respondents how frequently they experienced symptoms of psychological distress during two different time periods: (1) during the past 30 days, and (2) if applicable, the one month in the past year when they were at their worst emotionally. Respondents were asked about the second time period only if they indicated that there was a month in the past 12 months when they felt more depressed, anxious, or emotionally stressed than they felt during the past 30 days.

The six questions comprising the K6 scale for the past month are as follows:

NERVE30 During the past 30 days, how often did you feel nervous?

1 All of the time

2 Most of the time

3 Some of the time

4 A little of the time

5 None of the time

Don't know/Refused

Response categories are the same for the remaining questions shown below.

HOPE30 During the past 30 days, how often did you feel hopeless?

FIDG30 During the past 30 days, how often did you feel restless or fidgety?

NOCHR30 During the past 30 days, how often did you feel so sad or depressed that nothing could cheer you up?

EFFORT30 During the past 30 days, how often did you feel that everything was an effort?

DOWN30 During the past 30 days, how often did you feel down on yourself, no good or worthless?

To create a score, the six items (NERVE30, HOPE30, FIDG30, NOCHR30, EFFORT30, and DOWN30) on the K6 scale were recoded from 0 to 4 so that "all of the time" was coded 4, "most of the time" 3, "some of the time" 2, "a little of the time" 1, and "none of the time" 0, with "don't know" and "refused" also coded as 0. Summing across the transformed responses in these six items resulted in a score with a range from 0 to 24.

If respondents were asked about a month in the past 12 months when they felt more depressed, anxious, or emotionally stressed than they felt during the past 30 days, they were asked comparable K6 items for that particular month in the past 12 months. The scoring procedures for these K6 items for the past 12 months were the same as those described above. The higher of the two K6 total scores for the past 30 days or past 12 months was used both for MHSS analysis purposes and in the adult respondents' final data.

An alternative K6 total score also was created in which K6 scores of less than 8 were recoded as 0 and scores from 8 to 24 were recoded as 1 to 17. The rationale for creating the alternative past year K6 score was that SMI prevalence was typically extremely low for respondents with past year K6 scores of less than 8, and the prevalence rates started increasing only when scores were 8 or greater. The alternative K6 score was used in both the 2008 and 2012 SMI prediction models.

WHODAS. An initial step of the MHSS was to modify the WHODAS for use in a general population survey, including making minor changes to question wording and reducing its length (Novak, 2007). That is, a subset of 8 items was found to capture the information represented in the full 16-item scale with no significant loss of information.

These eight WHODAS items that were included in the main NSDUH interview were assessed on a 0 to 3 scale, with responses of "no difficulty," "don't know," and "refused" coded as 0; "mild difficulty" coded as 1; "moderate difficulty" coded as 2; and "severe difficulty" coded as 3. Some items had an additional category for respondents who did not engage in a particular activity (e.g., they did not leave the house on their own). Respondents who reported that they did not engage in an activity were asked a follow-up question to determine if they did not do so because of emotions, nerves, or mental health. Those who answered "yes" to this follow-up question were subsequently assigned to the "severe difficulty" category; otherwise (i.e., for responses of "no," "don't know," or "refused"), they were assigned to the "no difficulty" category. Summing across these codes for the eight responses resulted in a total score with a range from 0 to 24. More information about scoring of the WHODAS can be found in the 2015 NSDUH public use file codebook (CBHSQ, 2016d).

An alternative WHODAS total score was created in which individual WHODAS item scores of less than 2 were recoded as 0, and item scores of 2 to 3 were recoded as 1. The individual alternative item scores then were summed to yield a total alternative score ranging from 0 to 8. Creation of an alternative version of the WHODAS score was based on the assumption that a dichotomous measure dividing respondents into two groups (i.e., severely impaired vs. less severely impaired) might fit better than a linear continuous measure in models predicting SMI. This alternative WHODAS score was the variable used in both the 2008 and 2012 SMI prediction models.

Suicidal Thoughts, MDE, and Age. In addition to the K6 and WHODAS scales, the 2012 model included the following measures as predictors of SMI: (a) serious thoughts of suicide in the past year; (b) having a past year MDE; and (c) age. The first two variables were added to the model to decrease the error rate in the predictions (i.e., the sum of the false-negative and false-positive rates relative to the clinical interview results). A recoded age variable reduced the biases in estimates for particular age groups, especially 18 to 25 year olds.

Since 2008, all adult respondents in NSDUH have been asked the following question: "At any time in the past 12 months, that is from [DATEFILL] up to and including today, did you seriously think about killing yourself?"³⁰ Definitions for MDE in the lifetime and past year periods are discussed in Section B.11.2 of this document. For respondents aged 18 to 30, an adjusted age was created by subtracting 18 from the respondent's current age, resulting in values ranging from 0 to 12. For a respondent aged 18, for example, the adjusted age was 0 (i.e., 18 minus 18), and for a respondent aged 30, the adjusted age was 12 (i.e., 30 minus 18). For respondents aged 31 or older, the adjusted age was assigned a value of 12.

The 2012 SMI Model. The 2012 SMI prediction model was fit with data from 4,912 WHODAS MHSS respondents from 2008 through 2012. The response variable Y equaled 1 when an SMI diagnosis was positive based on the clinical interview; otherwise, Y was 0. Letting X be a vector of characteristics attached to a NSDUH respondent and letting the probability that this respondent had SMI be , the 2012 SMI prediction model was

     D

where refers to an estimate of the SMI response probability .

These covariates in equation (1) came from the main NSDUH interview data:

• = Alternative Past Year K6 Score: Past year K6 score of less than 8 recoded as 0; past year K6 score of 8 to 24 recoded as 1 to 17.
• = Alternative WHODAS Score: WHODAS item score of less than 2 recoded as 0; WHODAS item score of 2 to 3 recoded as 1, then summed for a score ranging from 0 to 8.
• = Serious Thoughts of Suicide in the Past Year: Coded as 1 if "yes"; coded as 0 otherwise.
• = Past Year MDE: Coded as 1 if criteria for past year MDE were met (see Section B.11.2); coded as 0 otherwise.
• = Adjusted Age: Coded as age minus 18 if aged 18 to 30; coded as 12 otherwise.

As with the 2008 model, a cut point probability was determined, so that if for a particular respondent, then he or she was predicted to be SMI positive; otherwise, he or she was predicted to be SMI negative. The cut point (0.260573529) was chosen so that the weighted number of false positives and false negatives in the MHSS dataset were as close to equal as possible The predicted SMI status for all adult NSDUH respondents was used to compute SMI small area estimates.

A second cut point probability (0.0192519810) was determined so that respondents with an SMI probability greater than or equal to the cut point were predicted to be positive for AMI, and the remainder were predicted to be negative for AMI. The second cut point was chosen so that the weighted numbers of AMI false positives and false negatives were as close to equal as possible.

B.11.2 Major Depressive Episode (Depression)

According to the DSM-IV, a person is defined as having had MDE in his or her lifetime if he or she has had at least five or more of the following nine symptoms nearly every day in the same 2-week period, where at least one of the symptoms is a depressed mood or loss of interest or pleasure in daily activities (APA, 1994): (1) depressed mood most of the day; (2) markedly diminished interest or pleasure in all or almost all activities most of the day; (3) significant weight loss when not sick or dieting, or weight gain when not pregnant or growing, or decrease or increase in appetite; (4) insomnia or hypersomnia; (5) psychomotor agitation or retardation; (6) fatigue or loss of energy; (7) feelings of worthlessness; (8) diminished ability to think or concentrate or indecisiveness; and (9) recurrent thoughts of death or suicidal ideation. Respondents who have had an MDE in their lifetime are asked if, during the past 12 months, they had a period of depression lasting 2 weeks or longer while also having some of the other symptoms mentioned. Those reporting that they have are defined as having had MDE in the past year and then are asked questions from the SDS to measure the level of functional impairment in major life activities reported to be caused by the MDE in the past 12 months (Leon, Olfson, Portera, Farber, & Sheehan, 1997).

Beginning in 2004, modules related to MDE, derived from DSM-IV (APA, 1994) criteria for major depression, were included in the questionnaire. These questions permit prevalence estimates of MDE to be calculated. Separate modules were administered to adults aged 18 or older and youths aged 12 to 17. The adult questions were adapted from the depression section of the National Comorbidity Survey Replication (NCS-R), and the questions for youths were adapted from the depression section of the National Comorbidity Survey Adolescent (NCS-A) (see https://www.hcp.med.harvard.edu/ncs/). To make the modules developmentally appropriate for youths, there are minor wording differences in a few questions between the adult and youth modules. Revisions to the questions in both modules were made primarily to reduce the length and to modify the NCS questions, which are interviewer-administered, to the audio computer-assisted self-interviewing (ACASI) format used in NSDUH. In addition, some revisions, based on cognitive testing, were made to improve comprehension.

Since 2004, the NSDUH questions that determine MDE have remained unchanged for both adults and youths. In the 2008 questionnaire, however, changes were made in other mental health items that precede the MDE questions for adults (K6, suicide, and impairment). Questions also were retained in 2009 for the WHODAS impairment scale, and the questions for the SDS impairment scale were deleted; see Sections B.4.6 and B.4.7 in Section B of the 2016 NSDUH methodological summary and definitions (CBHSQ, 2017) for further details about these questionnaire changes. The questionnaire changes in 2008 appear to have affected the reporting on MDE questions among adults.

Because the WHODAS was selected to be used in the 2009 and subsequent surveys, model-based adjustments were applied to MDE estimates from the SDS half sample in 2008 to remove the context effect differential between the two half samples. Additionally, model-based adjustments were made to the 2005, 2006, and 2007 adult MDE estimates to make them comparable with the 2008 through 2012 MDE estimates (for more information on these adjustments, see CBHSQ, 2012). Thus, the 2008-2009 estimates of MDE were produced using the adjusted 2008 MDE variable along with the unadjusted 2009 MDE variable. Revised estimates for 2005-2006, 2006-2007, and 2007-2008 were produced using the adjusted MDE variable.

In addition, changes to the youth mental health service utilization module questions in 2009 that preceded the questions about adolescent depression could have affected adolescents' responses to the adolescent depression questions and estimates of adolescent MDE. However, these changes in 2009 did not appear to affect the estimates of adolescent MDE. Therefore, data on trends in past year MDE from 2004 to 2012 are available for adolescents aged 12 to 17.

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

Table C.1 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2014

State	Total Selected DUs	Total Eligible DUs	Total Completed Screeners	Weighted DU Screening Response Rate	Total Selected	Total Responded	Population Estimate	Weighted Interview Response Rate	Weighted Overall Response Rate
DU = dwelling unit. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2014.
Total U.S.	185,013	154,533	127,605	81.94%	91,640	67,901	265,122,864	71.20%	58.34%
Northeast	40,667	34,065	26,744	76.59%	18,175	12,999	47,631,944	67.54%	51.73%
Midwest	42,681	35,695	30,189	83.61%	21,523	15,825	56,462,258	71.17%	59.51%
South	61,543	50,983	42,788	84.59%	30,192	22,781	98,843,935	72.44%	61.27%
West	40,122	33,790	27,884	80.21%	21,750	16,296	62,184,728	72.05%	57.79%
Alabama	2,640	2,083	1,730	82.92%	1,272	964	4,042,640	71.97%	59.67%
Alaska	2,985	2,346	1,950	83.13%	1,386	947	580,556	67.80%	56.37%
Arizona	2,514	1,912	1,659	86.87%	1,269	971	5,545,689	74.84%	65.01%
Arkansas	2,674	2,203	1,946	88.05%	1,262	964	2,443,636	72.68%	63.99%
California	10,239	9,203	7,083	76.31%	6,403	4,664	32,201,663	69.82%	53.28%
Colorado	2,607	2,254	1,843	81.83%	1,357	1,008	4,426,092	72.95%	59.70%
Connecticut	2,790	2,484	1,997	80.29%	1,438	980	3,054,946	64.87%	52.08%
Delaware	2,772	2,401	1,855	77.44%	1,264	951	784,117	73.66%	57.05%
District of Columbia	4,330	3,706	2,802	75.60%	1,219	935	564,072	72.83%	55.06%
Florida	10,269	8,222	6,823	82.44%	4,385	3,331	16,916,262	70.33%	57.98%
Georgia	3,693	3,089	2,567	83.01%	2,029	1,549	8,240,647	74.40%	61.76%
Hawaii	2,942	2,469	1,934	77.80%	1,339	968	1,149,245	71.50%	55.63%
Idaho	1,932	1,690	1,477	87.33%	1,267	987	1,326,157	75.54%	65.97%
Illinois	6,904	5,866	4,407	75.00%	3,488	2,397	10,738,476	67.24%	50.43%
Indiana	2,504	2,078	1,782	85.70%	1,294	967	5,460,095	72.26%	61.93%
Iowa	2,496	2,101	1,851	87.94%	1,240	912	2,582,849	71.52%	62.89%
Kansas	2,304	1,990	1,705	85.58%	1,296	982	2,356,686	73.83%	63.19%
Kentucky	2,556	2,080	1,827	87.74%	1,284	946	3,653,138	69.25%	60.76%
Louisiana	2,435	1,987	1,742	87.36%	1,302	992	3,798,948	73.51%	64.22%
Maine	3,342	2,364	2,106	89.08%	1,230	940	1,151,035	75.33%	67.10%
Maryland	2,483	2,251	1,757	77.14%	1,297	971	4,988,662	72.12%	55.63%
Massachusetts	2,948	2,541	2,068	81.37%	1,437	1,000	5,769,623	66.32%	53.97%
Michigan	6,609	5,404	4,498	83.31%	3,269	2,418	8,372,529	70.92%	59.08%
Minnesota	2,375	2,111	1,825	86.44%	1,266	967	4,544,275	75.42%	65.20%
Mississippi	2,199	1,714	1,498	87.30%	1,170	909	2,438,813	76.34%	66.64%
Missouri	2,578	2,116	1,839	86.82%	1,218	934	5,033,932	75.64%	65.67%
Montana	2,829	2,270	2,036	89.64%	1,287	977	857,904	72.51%	65.00%
Nebraska	2,459	2,102	1,842	87.61%	1,268	938	1,536,175	73.47%	64.36%
Nevada	2,421	2,047	1,592	77.33%	1,279	961	2,359,905	72.75%	56.25%
New Hampshire	3,044	2,439	2,055	84.32%	1,288	932	1,144,239	68.75%	57.97%
New Jersey	4,403	3,745	2,951	78.97%	2,167	1,536	7,522,494	69.70%	55.05%
New Mexico	2,313	1,746	1,555	89.09%	1,172	959	1,712,519	80.40%	71.62%
New York	11,063	9,562	6,603	68.76%	4,835	3,284	16,716,169	64.15%	44.11%
North Carolina	4,185	3,443	2,972	86.23%	1,956	1,533	8,216,513	76.58%	66.03%
North Dakota	3,043	2,363	2,136	90.40%	1,240	969	605,994	77.32%	69.89%
Ohio	6,322	5,307	4,531	85.14%	3,337	2,415	9,706,544	69.80%	59.43%
Oklahoma	2,259	1,828	1,609	88.21%	1,284	937	3,156,090	68.47%	60.40%
Oregon	2,529	2,207	1,877	85.36%	1,318	992	3,365,496	72.93%	62.26%
Pennsylvania	7,101	6,028	4,875	80.53%	3,186	2,388	10,828,027	70.81%	57.02%
Rhode Island	2,681	2,251	1,859	82.83%	1,334	991	902,079	72.13%	59.74%
South Carolina	2,843	2,307	1,958	84.71%	1,308	998	4,008,720	75.19%	63.69%
South Dakota	2,163	1,779	1,679	94.39%	1,275	981	691,583	75.06%	70.85%
Tennessee	2,326	1,939	1,676	86.31%	1,204	946	5,459,207	78.68%	67.91%
Texas	7,004	5,857	5,066	86.53%	4,581	3,383	21,690,765	70.38%	60.90%
Utah	1,534	1,344	1,275	94.87%	1,186	972	2,299,458	80.57%	76.44%
Vermont	3,295	2,651	2,230	83.96%	1,260	948	543,332	73.63%	61.82%
Virginia	3,671	3,261	2,678	82.32%	2,020	1,539	6,870,308	73.13%	60.20%
Washington	2,449	2,173	1,705	78.75%	1,241	935	5,879,524	74.01%	58.28%
West Virginia	3,204	2,612	2,282	87.55%	1,355	933	1,571,398	67.70%	59.27%
Wisconsin	2,924	2,478	2,094	84.25%	1,332	945	4,833,121	69.67%	58.70%
Wyoming	2,828	2,129	1,898	89.09%	1,246	955	480,519	74.19%	66.10%

Table C.2 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2014

State	12-17 Total Selected	12-17 Total Responded	12-17 Population Estimate	12-17 Weighted Interview Response Rate	18-25 Total Selected	18-25 Total Responded	18-25 Population Estimate	18-25 Weighted Interview Response Rate	26+ Total Selected	26+ Total Responded	26+ Population Estimate	26+ Weighted Interview Response Rate
NOTE: Computations in this table are based on a respondent's age at screening. Thus, the data in the Total Responded column(s) could differ from data in other NSDUH tables that use the respondent's age recorded during the interview. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2014.
Total U.S.	21,392	17,046	24,874,753	80.03%	21,726	16,570	34,934,625	75.88%	48,522	34,285	205,313,486	69.34%
Northeast	4,205	3,276	4,156,404	77.70%	4,204	3,117	6,150,189	71.74%	9,766	6,606	37,325,350	65.72%
Midwest	4,989	3,919	5,371,702	78.29%	5,143	3,820	7,427,562	73.42%	11,391	8,086	43,662,994	69.94%
South	7,210	5,824	9,410,988	81.01%	7,124	5,622	12,942,634	79.34%	15,858	11,335	76,490,313	70.20%
West	4,988	4,027	5,935,659	81.65%	5,255	4,011	8,414,241	75.77%	11,507	8,258	47,834,829	70.22%
Alabama	282	231	381,574	84.31%	291	236	533,886	80.90%	699	497	3,127,180	69.01%
Alaska	365	253	59,580	67.20%	314	222	83,648	68.72%	707	472	437,329	67.72%
Arizona	270	230	545,127	85.91%	311	244	737,788	78.17%	688	497	4,262,775	72.91%
Arkansas	308	249	236,364	78.53%	257	211	319,018	81.55%	697	504	1,888,254	70.65%
California	1,373	1,115	3,065,381	80.92%	1,531	1,151	4,473,314	74.54%	3,499	2,398	24,662,968	67.62%
Colorado	322	256	411,672	79.70%	409	311	580,685	76.85%	626	441	3,433,735	71.35%
Connecticut	335	256	285,016	78.02%	306	219	384,157	68.85%	797	505	2,385,774	62.71%
Delaware	330	264	68,288	78.60%	302	233	100,409	79.53%	632	454	615,419	72.13%
District of Columbia	273	233	30,727	85.77%	289	235	93,220	81.11%	657	467	440,125	70.19%
Florida	1,060	869	1,392,741	82.44%	1,062	847	1,987,479	79.44%	2,263	1,615	13,536,042	67.74%
Georgia	463	367	841,562	78.40%	543	438	1,112,868	81.03%	1,023	744	6,286,218	72.63%
Hawaii	312	249	96,703	81.76%	298	213	141,189	71.89%	729	506	911,353	70.37%
Idaho	276	233	143,867	84.58%	327	246	174,040	74.71%	664	508	1,008,249	74.52%
Illinois	749	558	1,027,930	74.50%	802	561	1,394,050	71.84%	1,937	1,278	8,316,496	65.66%
Indiana	314	249	540,851	80.33%	301	229	742,327	75.03%	679	489	4,176,917	70.77%
Iowa	268	203	242,540	75.35%	331	256	355,200	78.64%	641	453	1,985,109	69.65%
Kansas	275	213	237,294	78.08%	347	280	327,370	81.11%	674	489	1,792,022	71.94%
Kentucky	319	257	339,725	80.59%	324	243	473,910	75.27%	641	446	2,839,503	66.80%
Louisiana	312	255	367,731	81.26%	353	270	517,271	74.77%	637	467	2,913,946	72.28%
Maine	258	196	93,311	75.75%	278	225	126,789	80.17%	694	519	930,936	74.68%
Maryland	330	262	455,432	79.30%	297	229	628,947	75.83%	670	480	3,904,284	70.56%
Massachusetts	338	268	488,379	78.17%	375	273	786,469	72.66%	724	459	4,494,775	64.05%
Michigan	769	597	793,168	76.39%	730	558	1,116,715	75.04%	1,770	1,263	6,462,646	69.61%
Minnesota	309	252	425,574	81.06%	337	251	571,957	76.87%	620	464	3,546,745	74.56%
Mississippi	262	216	244,895	82.71%	272	231	339,298	85.28%	636	462	1,854,619	73.88%
Missouri	296	239	470,232	82.31%	282	208	657,419	74.23%	640	487	3,906,282	75.09%
Montana	284	222	74,224	79.69%	323	265	111,155	80.21%	680	490	672,526	70.24%
Nebraska	306	242	149,974	79.31%	296	219	210,685	74.17%	666	477	1,175,517	72.54%
Nevada	270	224	221,973	84.05%	318	240	288,475	74.94%	691	497	1,849,457	71.04%
New Hampshire	338	258	99,122	76.99%	294	234	141,805	80.62%	656	440	903,312	65.99%
New Jersey	517	391	699,694	75.24%	533	388	893,781	72.67%	1,117	757	5,929,018	68.64%
New Mexico	308	259	165,894	85.61%	262	220	227,928	84.46%	602	480	1,318,698	78.99%
New York	1,060	817	1,433,846	75.80%	1,077	737	2,238,419	66.42%	2,698	1,730	13,043,905	62.41%
North Carolina	461	380	774,595	82.08%	495	391	1,059,045	80.37%	1,000	762	6,382,874	75.24%
North Dakota	281	228	51,216	81.17%	341	271	102,157	78.81%	618	470	452,621	76.52%
Ohio	764	608	919,721	79.36%	777	550	1,232,774	70.07%	1,796	1,257	7,554,049	68.60%
Oklahoma	265	198	310,671	69.71%	298	235	430,351	77.68%	721	504	2,415,068	66.67%
Oregon	352	284	290,940	82.48%	334	242	413,519	71.42%	632	466	2,661,037	72.14%
Pennsylvania	738	608	937,266	82.54%	760	598	1,374,219	77.83%	1,688	1,182	8,516,542	68.46%
Rhode Island	325	250	75,595	75.22%	288	218	130,594	76.26%	721	523	695,890	70.92%
South Carolina	295	239	363,511	82.24%	304	245	521,002	82.04%	709	514	3,124,207	73.31%
South Dakota	300	251	65,995	83.07%	304	237	93,613	79.14%	671	493	531,976	73.42%
Tennessee	295	238	507,431	80.67%	233	188	703,094	82.76%	676	520	4,248,682	77.82%
Texas	1,137	929	2,342,547	81.93%	1,021	791	3,034,761	78.37%	2,423	1,663	16,313,458	67.20%
Utah	280	242	285,236	87.27%	252	217	374,751	84.88%	654	513	1,639,471	78.58%
Vermont	296	232	44,175	78.65%	293	225	73,958	77.65%	671	491	425,199	72.46%
Virginia	476	391	623,660	83.06%	496	398	897,977	80.79%	1,048	750	5,348,672	70.66%
Washington	272	214	530,698	78.46%	292	224	744,057	76.84%	677	497	4,604,769	73.01%
West Virginia	342	246	129,536	72.19%	287	201	190,099	70.22%	726	486	1,251,764	66.88%
Wisconsin	358	279	447,209	79.03%	295	200	623,296	65.36%	679	466	3,762,616	69.19%
Wyoming	304	246	44,364	79.39%	284	216	63,692	76.18%	658	493	372,464	73.23%

Table C.3 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2015

State	Total Selected DUs	Total Eligible DUs	Total Completed Screeners	Weighted DU Screening Response Rate	Total Selected	Total Responded	Population Estimate	Weighted Interview Response Rate	Weighted Overall Response Rate
DU = dwelling unit. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2015.
Total U.S.	197,962	165,328	132,210	79.69%	94,499	68,073	267,694,489	69.25%	55.19%
Northeast	44,157	37,292	28,065	73.23%	18,988	13,026	47,810,262	65.61%	48.04%
Midwest	46,269	38,853	32,108	81.52%	22,352	15,890	56,662,334	68.39%	55.75%
South	64,177	52,861	43,064	82.87%	30,920	22,768	100,182,409	70.93%	58.78%
West	43,359	36,322	28,973	77.73%	22,239	16,389	63,039,483	70.09%	54.48%
Alabama	2,797	2,185	1,831	83.26%	1,328	953	4,056,416	67.99%	56.61%
Alaska	3,289	2,381	1,892	79.18%	1,373	981	581,652	71.59%	56.68%
Arizona	3,022	2,314	1,949	84.15%	1,363	996	5,645,911	70.73%	59.52%
Arkansas	2,875	2,344	2,005	85.49%	1,343	981	2,457,367	68.96%	58.95%
California	11,282	10,153	7,564	73.80%	6,445	4,671	32,556,837	68.69%	50.69%
Colorado	2,637	2,240	1,795	80.03%	1,328	994	4,526,726	72.42%	57.96%
Connecticut	2,872	2,518	1,936	76.95%	1,411	964	3,058,139	66.21%	50.94%
Delaware	2,701	2,339	1,756	75.03%	1,323	945	795,351	71.21%	53.43%
District of Columbia	5,177	4,341	3,118	71.43%	1,231	924	574,552	74.47%	53.19%
Florida	10,530	8,387	6,793	80.63%	4,665	3,386	17,257,952	70.07%	56.50%
Georgia	4,015	3,307	2,603	78.78%	1,992	1,498	8,359,362	71.79%	56.56%
Hawaii	3,139	2,630	1,959	74.23%	1,389	1,020	1,158,550	70.76%	52.53%
Idaho	2,020	1,813	1,530	84.44%	1,277	949	1,347,084	72.78%	61.46%
Illinois	7,103	6,286	4,639	73.92%	3,592	2,365	10,737,272	63.14%	46.67%
Indiana	2,729	2,292	1,819	79.34%	1,376	973	5,486,199	68.00%	53.95%
Iowa	3,068	2,668	2,265	84.66%	1,357	962	2,597,548	68.53%	58.02%
Kansas	2,640	2,283	1,962	85.92%	1,351	986	2,367,256	71.42%	61.37%
Kentucky	2,469	2,000	1,695	84.66%	1,271	938	3,667,827	72.06%	61.01%
Louisiana	2,618	2,170	1,804	83.66%	1,282	957	3,819,762	73.03%	61.10%
Maine	4,277	3,140	2,643	84.00%	1,400	994	1,151,684	68.79%	57.78%
Maryland	2,308	2,018	1,513	75.20%	1,290	946	5,018,659	69.83%	52.52%
Massachusetts	3,366	2,960	2,131	72.27%	1,591	948	5,822,667	57.99%	41.91%
Michigan	7,166	5,787	4,853	83.66%	3,383	2,441	8,392,983	69.43%	58.08%
Minnesota	2,490	2,149	1,766	82.05%	1,286	951	4,575,592	73.16%	60.02%
Mississippi	2,554	2,060	1,741	84.80%	1,257	921	2,443,849	70.17%	59.51%
Missouri	2,582	2,094	1,846	88.22%	1,342	986	5,057,574	70.25%	61.98%
Montana	3,195	2,528	2,159	85.62%	1,329	977	866,257	69.44%	59.45%
Nebraska	2,510	2,156	1,794	82.82%	1,301	945	1,548,885	71.21%	58.97%
Nevada	2,676	2,287	1,746	76.61%	1,317	997	2,408,267	69.97%	53.60%
New Hampshire	3,324	2,763	2,191	79.00%	1,435	995	1,148,726	68.23%	53.90%
New Jersey	4,076	3,647	2,807	75.90%	2,247	1,517	7,552,211	65.39%	49.63%
New Mexico	2,568	1,853	1,644	88.94%	1,260	959	1,717,549	73.85%	65.68%
New York	12,117	10,496	6,863	64.83%	4,963	3,310	16,779,910	63.60%	41.23%
North Carolina	4,251	3,606	2,990	82.87%	2,125	1,576	8,320,518	69.99%	58.00%
North Dakota	3,425	2,758	2,484	89.86%	1,342	988	618,680	72.44%	65.09%
Ohio	7,032	5,899	4,773	80.86%	3,458	2,428	9,732,558	68.48%	55.38%
Oklahoma	2,857	2,285	1,918	84.37%	1,359	971	3,185,569	67.59%	57.02%
Oregon	2,526	2,195	1,803	82.11%	1,333	962	3,420,080	71.04%	58.33%
Pennsylvania	7,429	6,257	5,054	80.80%	3,232	2,374	10,849,493	71.72%	57.95%
Rhode Island	2,901	2,461	1,915	77.81%	1,354	964	903,886	69.45%	54.04%
South Carolina	2,944	2,436	2,040	83.70%	1,304	987	4,070,523	72.52%	60.70%
South Dakota	2,354	1,968	1,799	91.69%	1,199	904	695,959	74.77%	68.56%
Tennessee	2,670	2,172	1,846	84.96%	1,352	1,004	5,507,975	69.71%	59.22%
Texas	6,227	5,184	4,538	87.56%	4,358	3,308	22,151,524	73.28%	64.16%
Utah	1,506	1,316	1,176	89.31%	1,204	968	2,350,775	77.43%	69.16%
Vermont	3,795	3,050	2,525	82.82%	1,355	960	543,548	68.96%	57.11%
Virginia	3,934	3,410	2,754	80.78%	2,113	1,526	6,928,628	69.71%	56.32%
Washington	2,692	2,423	1,867	76.82%	1,306	944	5,978,195	69.98%	53.76%
West Virginia	3,250	2,617	2,119	80.92%	1,327	947	1,566,577	66.77%	54.03%
Wisconsin	3,170	2,513	2,108	84.08%	1,365	961	4,851,828	68.35%	57.47%
Wyoming	2,807	2,189	1,889	86.02%	1,315	971	481,602	72.26%	62.16%

Table C.4 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2015

State	12-17 Total Selected	12-17 Total Responded	12-17 Population Estimate	12-17 Weighted Interview Response Rate	18-25 Total Selected	18-25 Total Responded	18-25 Population Estimate	18-25 Weighted Interview Response Rate	26+ Total Selected	26+ Total Responded	26+ Population Estimate	26+ Weighted Interview Response Rate
NOTE: Computations in this table are based on a respondent's age at screening. Thus, the data in the Total Responded column(s) could differ from data in other NSDUH tables that use the respondent's age recorded during the interview. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2015.
Total U.S.	21,859	16,955	24,893,417	77.66%	23,211	17,215	34,907,162	74.45%	49,429	33,903	207,893,910	67.36%
Northeast	4,308	3,228	4,124,414	72.98%	4,651	3,233	6,117,578	68.66%	10,029	6,565	37,568,270	64.28%
Midwest	5,296	3,955	5,351,313	73.95%	5,509	4,106	7,415,255	74.10%	11,547	7,829	43,895,766	66.73%
South	7,267	5,767	9,483,323	79.64%	7,496	5,676	12,959,382	76.41%	16,157	11,325	77,739,704	68.96%
West	4,988	4,005	5,934,367	81.09%	5,555	4,200	8,414,946	75.99%	11,696	8,184	48,690,170	67.73%
Alabama	289	229	380,027	78.20%	338	251	527,315	74.78%	701	473	3,149,075	65.56%
Alaska	322	227	58,808	69.67%	331	247	82,845	73.61%	720	507	439,999	71.46%
Arizona	296	239	547,813	80.67%	324	248	745,197	76.07%	743	509	4,352,901	68.60%
Arkansas	323	256	236,353	77.64%	329	245	318,810	74.57%	691	480	1,902,203	66.87%
California	1,411	1,148	3,044,310	80.84%	1,603	1,224	4,441,882	76.89%	3,431	2,299	25,070,645	65.77%
Colorado	320	269	419,211	84.39%	327	241	593,941	73.82%	681	484	3,513,574	70.56%
Connecticut	305	241	281,090	79.35%	347	227	387,506	64.40%	759	496	2,389,542	64.87%
Delaware	302	238	68,905	79.72%	325	221	98,641	67.69%	696	486	627,805	70.81%
District of Columbia	264	210	30,686	80.79%	257	190	94,114	73.72%	710	524	449,752	74.18%
Florida	1,072	844	1,406,795	78.55%	1,159	889	1,981,426	77.16%	2,434	1,653	13,869,730	68.21%
Georgia	524	420	851,391	80.68%	447	358	1,116,369	79.67%	1,021	720	6,391,602	69.17%
Hawaii	286	226	97,117	75.80%	360	275	139,707	76.77%	743	519	921,726	69.35%
Idaho	281	220	145,770	80.39%	346	260	174,661	76.34%	650	469	1,026,653	71.02%
Illinois	887	648	1,018,545	72.96%	809	561	1,382,295	68.56%	1,896	1,156	8,336,432	61.04%
Indiana	316	242	540,488	73.99%	352	256	743,142	73.45%	708	475	4,202,568	66.29%
Iowa	346	253	243,085	73.21%	346	249	358,657	72.25%	665	460	1,995,806	67.26%
Kansas	347	251	237,829	71.04%	296	242	329,951	83.24%	708	493	1,799,476	69.27%
Kentucky	296	232	339,561	77.14%	297	224	471,843	75.59%	678	482	2,856,423	70.90%
Louisiana	311	244	367,609	79.34%	319	233	509,882	73.11%	652	480	2,942,271	72.13%
Maine	382	293	91,980	75.70%	309	217	125,074	69.44%	709	484	934,630	67.99%
Maryland	307	238	453,696	78.67%	326	247	622,611	75.45%	657	461	3,942,353	68.06%
Massachusetts	337	228	487,806	67.52%	375	221	791,046	57.80%	879	499	4,543,815	56.96%
Michigan	798	601	784,266	74.15%	847	653	1,112,424	77.93%	1,738	1,187	6,496,293	67.36%
Minnesota	319	247	426,424	76.74%	304	230	571,849	77.88%	663	474	3,577,318	71.96%
Mississippi	287	231	244,034	81.89%	289	226	335,131	77.47%	681	464	1,864,684	67.41%
Missouri	308	244	470,294	77.78%	384	293	655,956	76.45%	650	449	3,931,325	68.27%
Montana	300	230	74,532	77.20%	302	229	111,838	73.93%	727	518	679,888	67.95%
Nebraska	289	220	152,144	76.73%	338	248	212,640	71.16%	674	477	1,184,101	70.52%
Nevada	324	271	223,603	84.13%	334	254	288,923	75.66%	659	472	1,895,740	67.17%
New Hampshire	322	238	97,633	75.02%	325	235	143,062	74.78%	788	522	908,031	66.49%
New Jersey	527	387	695,324	72.89%	588	411	894,807	69.65%	1,132	719	5,962,081	63.92%
New Mexico	255	215	164,982	84.38%	304	237	226,226	78.86%	701	507	1,326,341	71.89%
New York	1,065	766	1,421,217	69.93%	1,302	909	2,218,443	67.76%	2,596	1,635	13,140,250	62.15%
North Carolina	539	438	780,506	82.17%	515	397	1,065,839	77.39%	1,071	741	6,474,173	67.38%
North Dakota	318	231	52,164	71.69%	328	259	104,459	77.80%	696	498	462,057	71.27%
Ohio	803	589	914,823	72.84%	827	599	1,225,255	73.19%	1,828	1,240	7,592,481	67.22%
Oklahoma	349	260	313,866	75.40%	289	215	431,841	71.97%	721	496	2,439,862	65.76%
Oregon	281	214	291,606	77.27%	335	244	415,899	72.61%	717	504	2,712,575	70.12%
Pennsylvania	742	574	931,284	77.42%	794	596	1,354,815	76.16%	1,696	1,204	8,563,393	70.38%
Rhode Island	286	228	74,717	79.60%	332	235	128,339	71.08%	736	501	700,830	68.02%
South Carolina	344	282	366,745	82.77%	274	219	519,107	79.59%	686	486	3,184,672	70.29%
South Dakota	300	230	65,584	77.20%	297	233	93,003	77.41%	602	441	537,373	73.96%
Tennessee	295	230	508,351	77.48%	414	318	703,173	74.53%	643	456	4,296,451	67.99%
Texas	959	780	2,380,293	80.39%	1,085	849	3,080,905	78.32%	2,314	1,679	16,690,326	71.33%
Utah	299	262	292,037	88.19%	308	250	383,514	81.11%	597	456	1,675,224	74.73%
Vermont	342	273	43,364	79.72%	279	182	74,485	66.68%	734	505	425,699	68.21%
Virginia	490	392	625,315	79.95%	504	357	895,251	70.76%	1,119	777	5,408,062	68.32%
Washington	285	227	530,641	79.31%	350	250	747,302	71.32%	671	467	4,700,252	68.75%
West Virginia	316	243	129,191	78.60%	329	237	187,125	73.58%	682	467	1,250,260	64.34%
Wisconsin	265	199	445,668	72.18%	381	283	625,624	72.36%	719	479	3,780,537	67.14%
Wyoming	328	257	43,939	77.94%	331	241	63,010	74.06%	656	473	374,652	71.28%

Table C.5 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2016

State	Total Selected DUs	Total Eligible DUs	Total Completed Screeners	Weighted DU Screening Response Rate	Total Selected	Total Responded	Population Estimate	Weighted Interview Response Rate	Weighted Overall Response Rate
DU = dwelling unit. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2016.
Total U.S.	205,589	173,149	135,188	77.88%	95,607	67,942	269,430,135	68.44%	53.30%
Northeast	45,388	38,488	28,275	71.60%	18,782	12,711	47,797,488	64.63%	46.28%
Midwest	46,850	39,972	32,231	79.66%	22,649	16,023	56,744,903	68.00%	54.17%
South	67,261	56,067	44,353	80.59%	31,462	22,833	101,241,206	70.62%	56.91%
West	46,090	38,622	30,329	76.52%	22,714	16,375	63,646,539	68.21%	52.20%
Alabama	2,996	2,478	2,026	82.04%	1,392	983	4,064,691	66.70%	54.72%
Alaska	3,272	2,386	1,901	79.52%	1,325	960	585,025	69.03%	54.90%
Arizona	2,921	2,203	1,835	83.43%	1,313	982	5,742,769	74.79%	62.39%
Arkansas	3,036	2,503	2,041	81.73%	1,381	992	2,468,292	69.49%	56.80%
California	12,192	11,070	7,993	72.01%	6,720	4,619	32,689,876	65.40%	47.10%
Colorado	2,570	2,163	1,757	80.69%	1,324	920	4,612,005	67.04%	54.10%
Connecticut	2,980	2,559	1,931	75.41%	1,392	937	3,052,524	65.01%	49.03%
Delaware	2,953	2,459	1,880	76.98%	1,330	928	802,361	67.70%	52.12%
District of Columbia	5,940	5,119	3,401	65.20%	1,260	967	580,859	74.11%	48.32%
Florida	11,282	9,267	7,135	77.11%	4,794	3,435	17,554,248	68.22%	52.60%
Georgia	3,619	3,139	2,443	77.88%	1,998	1,508	8,462,591	71.10%	55.37%
Hawaii	3,949	3,329	2,478	73.74%	1,458	1,004	1,157,906	66.33%	48.91%
Idaho	2,653	2,151	1,842	85.77%	1,429	1,088	1,373,371	74.13%	63.59%
Illinois	7,222	6,310	4,501	71.35%	3,789	2,467	10,702,668	61.81%	44.10%
Indiana	2,560	2,149	1,665	77.38%	1,286	933	5,503,158	69.65%	53.90%
Iowa	2,893	2,461	2,076	84.27%	1,414	1,028	2,607,021	71.71%	60.43%
Kansas	2,522	2,204	1,848	83.82%	1,363	996	2,369,503	71.16%	59.64%
Kentucky	3,162	2,586	2,104	81.27%	1,445	953	3,684,220	62.76%	51.00%
Louisiana	2,946	2,381	1,934	81.24%	1,328	959	3,831,309	70.61%	57.37%
Maine	3,941	3,022	2,473	82.01%	1,394	992	1,154,268	71.53%	58.66%
Maryland	2,418	2,120	1,550	72.57%	1,317	990	5,027,075	73.23%	53.14%
Massachusetts	3,700	3,252	2,365	72.42%	1,596	988	5,849,205	61.77%	44.73%
Michigan	7,090	5,893	4,809	81.40%	3,311	2,420	8,406,442	70.59%	57.46%
Minnesota	2,596	2,278	1,855	81.33%	1,375	962	4,605,050	68.58%	55.78%
Mississippi	2,382	1,949	1,617	83.00%	1,283	934	2,447,209	71.09%	59.00%
Missouri	2,612	2,247	1,926	85.56%	1,334	938	5,069,324	66.20%	56.65%
Montana	3,217	2,602	2,247	86.51%	1,433	1,018	874,320	71.23%	61.62%
Nebraska	2,696	2,350	1,881	80.01%	1,364	964	1,557,938	68.95%	55.16%
Nevada	2,379	2,095	1,526	72.71%	1,268	966	2,448,780	72.48%	52.70%
New Hampshire	3,244	2,763	2,148	77.51%	1,355	936	1,153,236	67.19%	52.08%
New Jersey	4,370	3,866	2,791	71.09%	2,149	1,433	7,550,513	63.19%	44.92%
New Mexico	2,907	2,023	1,720	84.86%	1,215	980	1,719,897	79.43%	67.41%
New York	12,398	10,716	6,932	63.92%	4,934	3,232	16,748,367	61.44%	39.27%
North Carolina	4,122	3,470	2,832	81.56%	2,089	1,508	8,419,860	71.49%	58.31%
North Dakota	3,511	2,882	2,521	87.70%	1,344	960	617,001	69.08%	60.58%
Ohio	6,804	5,933	4,700	79.21%	3,363	2,377	9,738,448	67.60%	53.55%
Oklahoma	2,654	2,198	1,794	81.39%	1,374	965	3,198,970	68.24%	55.54%
Oregon	3,160	2,765	2,224	80.46%	1,391	1,004	3,478,192	71.05%	57.17%
Pennsylvania	7,825	6,665	5,277	79.17%	3,308	2,360	10,840,710	70.48%	55.80%
Rhode Island	3,072	2,653	2,043	77.12%	1,356	937	905,791	67.37%	51.96%
South Carolina	2,832	2,251	1,849	81.99%	1,326	970	4,133,914	72.46%	59.41%
South Dakota	2,813	2,338	2,037	86.96%	1,338	960	701,645	70.92%	61.67%
Tennessee	3,034	2,416	2,002	82.87%	1,373	993	5,556,863	70.57%	58.48%
Texas	6,793	5,725	4,877	84.53%	4,255	3,293	22,490,422	74.68%	63.13%
Utah	1,483	1,331	1,138	85.78%	1,215	936	2,403,330	74.82%	64.18%
Vermont	3,858	2,992	2,315	77.15%	1,298	896	542,875	71.09%	54.85%
Virginia	3,920	3,376	2,743	81.20%	2,077	1,493	6,961,461	68.86%	55.91%
Washington	2,779	2,421	1,911	78.99%	1,362	934	6,080,095	66.41%	52.45%
West Virginia	3,172	2,630	2,125	80.79%	1,440	962	1,556,861	63.87%	51.60%
Wisconsin	3,531	2,927	2,412	82.32%	1,368	1,018	4,866,705	73.22%	60.27%
Wyoming	2,608	2,083	1,757	84.46%	1,261	964	480,973	75.14%	63.46%

Table C.6 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2016

State	12-17 Total Selected	12-17 Total Responded	12-17 Population Estimate	12-17 Weighted Interview Response Rate	18-25 Total Selected	18-25 Total Responded	18-25 Population Estimate	18-25 Weighted Interview Response Rate	26+ Total Selected	26+ Total Responded	26+ Population Estimate	26+ Weighted Interview Response Rate
NOTE: Computations in this table are based on a respondent's age at screening. Thus, the data in the Total Responded column(s) could differ from data in other NSDUH tables that use the respondent's age recorded during the interview. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2016.
Total U.S.	22,323	17,109	24,896,527	76.95%	22,836	16,573	34,570,728	72.66%	50,448	34,260	209,962,880	66.74%
Northeast	4,417	3,193	4,097,263	70.97%	4,459	3,059	6,052,258	67.96%	9,906	6,459	37,647,967	63.41%
Midwest	5,355	4,105	5,326,597	76.54%	5,444	3,896	7,367,324	71.62%	11,850	8,022	44,050,981	66.38%
South	7,219	5,625	9,530,368	78.55%	7,519	5,623	12,828,550	75.69%	16,724	11,585	78,882,288	68.85%
West	5,332	4,186	5,942,298	78.87%	5,414	3,995	8,322,597	72.31%	11,968	8,194	49,381,644	66.19%
Alabama	304	234	376,632	79.57%	313	243	518,185	76.56%	775	506	3,169,874	63.84%
Alaska	317	236	59,359	75.48%	362	276	77,379	76.69%	646	448	448,287	66.89%
Arizona	316	234	549,195	75.37%	317	237	747,345	74.16%	680	511	4,446,229	74.82%
Arkansas	307	235	236,955	78.47%	347	260	317,177	73.69%	727	497	1,914,160	67.68%
California	1,509	1,187	3,034,119	79.22%	1,517	1,092	4,358,028	71.70%	3,694	2,340	25,297,729	62.56%
Colorado	307	243	423,725	78.45%	303	212	599,128	68.58%	714	465	3,589,152	65.26%
Connecticut	303	224	278,000	75.81%	366	251	388,847	68.19%	723	462	2,385,677	63.36%
Delaware	288	217	69,423	77.17%	344	245	95,867	71.38%	698	466	637,071	66.16%
District of Columbia	292	240	30,940	82.15%	327	251	93,288	76.72%	641	476	456,632	72.98%
Florida	1,107	859	1,404,808	77.61%	1,031	793	1,961,863	76.96%	2,656	1,783	14,187,577	66.26%
Georgia	461	370	859,100	78.55%	432	352	1,107,792	80.49%	1,105	786	6,495,700	68.62%
Hawaii	388	282	96,028	71.79%	326	243	131,256	73.17%	744	479	930,622	64.71%
Idaho	334	270	147,812	79.99%	376	286	175,630	74.50%	719	532	1,049,928	73.19%
Illinois	884	641	1,012,090	72.69%	918	614	1,363,215	66.25%	1,987	1,212	8,327,363	59.80%
Indiana	283	222	538,647	78.86%	317	241	743,072	76.19%	686	470	4,221,440	67.20%
Iowa	349	272	243,421	78.47%	343	243	359,699	71.52%	722	513	2,003,901	70.90%
Kansas	337	258	237,465	75.77%	306	223	325,008	73.30%	720	515	1,807,031	70.19%
Kentucky	345	250	340,245	71.68%	359	233	470,276	65.18%	741	470	2,873,699	61.30%
Louisiana	325	249	367,320	75.79%	307	221	496,651	72.36%	696	489	2,967,339	69.64%
Maine	314	227	90,994	72.99%	312	225	124,447	73.55%	768	540	938,827	71.13%
Maryland	264	209	453,651	79.62%	309	231	612,960	74.02%	744	550	3,960,463	72.40%
Massachusetts	367	228	486,692	62.45%	347	212	793,386	62.16%	882	548	4,569,126	61.63%
Michigan	762	610	774,747	80.16%	800	598	1,104,650	75.06%	1,749	1,212	6,527,045	68.74%
Minnesota	314	239	428,949	76.11%	335	223	574,038	64.59%	726	500	3,602,063	68.38%
Mississippi	305	235	244,408	76.88%	305	235	326,958	78.37%	673	464	1,875,843	69.05%
Missouri	282	216	468,693	76.84%	309	232	649,195	75.02%	743	490	3,951,436	63.73%
Montana	333	258	74,323	76.38%	371	267	110,690	71.37%	729	493	689,307	70.64%
Nebraska	313	241	153,264	77.62%	350	236	213,572	67.37%	701	487	1,191,102	68.11%
Nevada	291	249	224,692	84.52%	296	230	285,894	77.28%	681	487	1,938,194	70.39%
New Hampshire	321	236	95,915	74.44%	298	203	142,331	68.39%	736	497	914,990	66.22%
New Jersey	483	369	693,040	76.68%	487	333	889,421	67.89%	1,179	731	5,968,052	60.92%
New Mexico	315	269	165,841	87.57%	273	220	221,098	82.25%	627	491	1,332,957	77.94%
New York	1,228	862	1,411,235	66.91%	1,142	779	2,176,812	66.82%	2,564	1,591	13,160,320	60.00%
North Carolina	463	350	787,252	75.62%	486	353	1,042,023	73.36%	1,140	805	6,590,585	70.67%
North Dakota	361	277	52,057	77.94%	326	236	99,863	70.30%	657	447	465,081	67.70%
Ohio	771	581	905,155	73.88%	809	582	1,215,046	72.06%	1,783	1,214	7,618,247	66.19%
Oklahoma	341	264	315,530	77.50%	347	237	425,978	67.58%	686	464	2,457,462	67.17%
Oregon	331	244	291,562	72.28%	310	215	420,001	70.39%	750	545	2,766,628	71.02%
Pennsylvania	814	614	925,024	74.86%	803	571	1,334,425	72.14%	1,691	1,175	8,581,261	69.74%
Rhode Island	295	224	73,856	76.68%	348	237	127,610	69.94%	713	476	704,325	65.94%
South Carolina	288	228	368,554	77.77%	324	240	511,293	75.12%	714	502	3,254,067	71.45%
South Dakota	332	255	66,650	76.73%	311	227	92,952	73.75%	695	478	542,043	69.60%
Tennessee	315	235	508,796	74.37%	315	230	698,244	73.51%	743	528	4,349,823	69.66%
Texas	1,001	826	2,410,422	82.34%	1,060	847	3,086,091	79.55%	2,194	1,620	16,993,908	72.64%
Utah	286	240	297,786	81.97%	266	206	390,726	79.39%	663	490	1,714,818	72.56%
Vermont	292	209	42,507	72.18%	356	248	74,978	72.38%	650	439	425,389	70.74%
Virginia	492	391	628,350	79.49%	539	394	880,842	72.90%	1,046	708	5,452,270	66.92%
Washington	324	253	533,613	79.36%	338	232	744,179	68.26%	700	449	4,802,304	64.65%
West Virginia	321	233	127,982	74.10%	374	258	183,063	66.48%	745	471	1,245,817	62.43%
Wisconsin	367	293	445,459	80.36%	320	241	627,016	74.85%	681	484	3,794,230	72.12%
Wyoming	281	221	44,244	76.40%	359	279	61,241	76.61%	621	464	375,489	74.74%

Table C.7 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2014 and 2015

State	Total Selected DUs	Total Eligible DUs	Total Completed Screeners	Weighted DU Screening Response Rate	Total Selected	Total Responded	Population Estimate	Weighted Interview Response Rate	Weighted Overall Response Rate
DU = dwelling unit. NOTE: To compute the pooled 2014-2015 weighted response rates, two samples were combined, and the individual year weights were used for the pooled sample. Thus, the response rates presented here are weighted across 2 years of data rather than being a simple average of the 2014 and 2015 individual response rates. The 2014-2015 population estimate is the average of the 2014 and the 2015 population. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2014 and 2015.
Total U.S.	382,975	319,861	259,815	80.81%	186,139	135,974	266,408,677	70.22%	56.75%
Northeast	84,824	71,357	54,809	74.92%	37,163	26,025	47,721,103	66.57%	49.88%
Midwest	88,950	74,548	62,297	82.56%	43,875	31,715	56,562,296	69.79%	57.62%
South	125,720	103,844	85,852	83.72%	61,112	45,549	99,513,172	71.68%	60.01%
West	83,481	70,112	56,857	78.97%	43,989	32,685	62,612,105	71.07%	56.12%
Alabama	5,437	4,268	3,561	83.09%	2,600	1,917	4,049,528	70.00%	58.17%
Alaska	6,274	4,727	3,842	81.18%	2,759	1,928	581,104	69.70%	56.59%
Arizona	5,536	4,226	3,608	85.51%	2,632	1,967	5,595,800	72.80%	62.25%
Arkansas	5,549	4,547	3,951	86.78%	2,605	1,945	2,450,501	70.86%	61.50%
California	21,521	19,356	14,647	75.06%	12,848	9,335	32,379,250	69.25%	51.98%
Colorado	5,244	4,494	3,638	80.95%	2,685	2,002	4,476,409	72.69%	58.84%
Connecticut	5,662	5,002	3,933	78.65%	2,849	1,944	3,056,542	65.54%	51.54%
Delaware	5,473	4,740	3,611	76.24%	2,587	1,896	789,734	72.42%	55.21%
District of Columbia	9,507	8,047	5,920	73.51%	2,450	1,859	569,312	73.67%	54.16%
Florida	20,799	16,609	13,616	81.52%	9,050	6,717	17,087,107	70.19%	57.22%
Georgia	7,708	6,396	5,170	80.87%	4,021	3,047	8,300,005	73.09%	59.11%
Hawaii	6,081	5,099	3,893	76.04%	2,728	1,988	1,153,898	71.14%	54.09%
Idaho	3,952	3,503	3,007	85.87%	2,544	1,936	1,336,620	74.19%	63.71%
Illinois	14,007	12,152	9,046	74.45%	7,080	4,762	10,737,874	65.21%	48.55%
Indiana	5,233	4,370	3,601	82.54%	2,670	1,940	5,473,147	70.16%	57.91%
Iowa	5,564	4,769	4,116	86.27%	2,597	1,874	2,590,199	69.99%	60.38%
Kansas	4,944	4,273	3,667	85.75%	2,647	1,968	2,361,971	72.63%	62.28%
Kentucky	5,025	4,080	3,522	86.20%	2,555	1,884	3,660,483	70.68%	60.92%
Louisiana	5,053	4,157	3,546	85.63%	2,584	1,949	3,809,355	73.28%	62.75%
Maine	7,619	5,504	4,749	86.51%	2,630	1,934	1,151,359	72.09%	62.37%
Maryland	4,791	4,269	3,270	76.18%	2,587	1,917	5,003,661	70.91%	54.02%
Massachusetts	6,314	5,501	4,199	76.88%	3,028	1,948	5,796,145	62.17%	47.79%
Michigan	13,775	11,191	9,351	83.49%	6,652	4,859	8,382,756	70.19%	58.60%
Minnesota	4,865	4,260	3,591	84.26%	2,552	1,918	4,559,933	74.31%	62.62%
Mississippi	4,753	3,774	3,239	86.02%	2,427	1,830	2,441,331	73.26%	63.02%
Missouri	5,160	4,210	3,685	87.52%	2,560	1,920	5,045,753	72.95%	63.85%
Montana	6,024	4,798	4,195	87.66%	2,616	1,954	862,081	70.88%	62.14%
Nebraska	4,969	4,258	3,636	85.24%	2,569	1,883	1,542,530	72.31%	61.64%
Nevada	5,097	4,334	3,338	76.96%	2,596	1,958	2,384,086	71.41%	54.96%
New Hampshire	6,368	5,202	4,246	81.65%	2,723	1,927	1,146,483	68.49%	55.92%
New Jersey	8,479	7,392	5,758	77.41%	4,414	3,053	7,537,352	67.53%	52.28%
New Mexico	4,881	3,599	3,199	89.01%	2,432	1,918	1,715,034	76.99%	68.53%
New York	23,180	20,058	13,466	66.83%	9,798	6,594	16,748,040	63.87%	42.69%
North Carolina	8,436	7,049	5,962	84.53%	4,081	3,109	8,268,515	73.18%	61.86%
North Dakota	6,468	5,121	4,620	90.12%	2,582	1,957	612,337	74.86%	67.46%
Ohio	13,354	11,206	9,304	82.99%	6,795	4,843	9,719,551	69.14%	57.38%
Oklahoma	5,116	4,113	3,527	86.27%	2,643	1,908	3,170,829	68.03%	58.69%
Oregon	5,055	4,402	3,680	83.70%	2,651	1,954	3,392,788	71.97%	60.24%
Pennsylvania	14,530	12,285	9,929	80.66%	6,418	4,762	10,838,760	71.26%	57.48%
Rhode Island	5,582	4,712	3,774	80.29%	2,688	1,955	902,983	70.75%	56.80%
South Carolina	5,787	4,743	3,998	84.19%	2,612	1,985	4,039,622	73.83%	62.15%
South Dakota	4,517	3,747	3,478	93.04%	2,474	1,885	693,771	74.92%	69.70%
Tennessee	4,996	4,111	3,522	85.64%	2,556	1,950	5,483,591	74.16%	63.51%
Texas	13,231	11,041	9,604	87.06%	8,939	6,691	21,921,145	71.84%	62.54%
Utah	3,040	2,660	2,451	92.12%	2,390	1,940	2,325,116	79.00%	72.78%
Vermont	7,090	5,701	4,755	83.39%	2,615	1,908	543,440	71.33%	59.48%
Virginia	7,605	6,671	5,432	81.58%	4,133	3,065	6,899,468	71.42%	58.27%
Washington	5,141	4,596	3,572	77.76%	2,547	1,879	5,928,859	71.97%	55.97%
West Virginia	6,454	5,229	4,401	84.30%	2,682	1,880	1,568,988	67.25%	56.70%
Wisconsin	6,094	4,991	4,202	84.17%	2,697	1,906	4,842,475	69.01%	58.08%
Wyoming	5,635	4,318	3,787	87.57%	2,561	1,926	481,060	73.23%	64.13%

Table C.8 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2014 and 2015

State	12-17 Total Selected	12-17 Total Responded	12-17 Population Estimate	12-17 Weighted Interview Response Rate	18-25 Total Selected	18-25 Total Responded	18-25 Population Estimate	18-25 Weighted Interview Response Rate	26+ Total Selected	26+ Total Responded	26+ Population Estimate	26+ Weighted Interview Response Rate
NOTE: Computations in this table are based on a respondent's age at screening. Thus, the data in the Total Responded column(s) could differ from data in other NSDUH tables that use the respondent's age recorded during the interview. NOTE: To compute the pooled 2014-2015 weighted response rates, two samples were combined, and the individual year weights were used for the pooled sample. Thus, the response rates presented here are weighted across 2 years of data rather than being a simple average of the 2014 and 2015 individual response rates. The 2014-2015 population estimate is the average of the 2014 and the 2015 population. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2014 and 2015.
Total U.S.	43,251	34,001	24,884,085	78.85%	44,937	33,785	34,920,893	75.16%	97,951	68,188	206,603,698	68.35%
Northeast	8,513	6,504	4,140,409	75.34%	8,855	6,350	6,133,884	70.19%	19,795	13,171	37,446,810	65.00%
Midwest	10,285	7,874	5,361,507	76.13%	10,652	7,926	7,421,409	73.76%	22,938	15,915	43,779,380	68.35%
South	14,477	11,591	9,447,156	80.32%	14,620	11,298	12,951,008	77.86%	32,015	22,660	77,115,008	69.57%
West	9,976	8,032	5,935,013	81.37%	10,810	8,211	8,414,593	75.88%	23,203	16,442	48,262,499	68.97%
Alabama	571	460	380,801	81.22%	629	487	530,600	77.87%	1,400	970	3,138,127	67.31%
Alaska	687	480	59,194	68.45%	645	469	83,247	71.20%	1,427	979	438,664	69.59%
Arizona	566	469	546,470	83.32%	635	492	741,492	77.11%	1,431	1,006	4,307,838	70.77%
Arkansas	631	505	236,359	78.09%	586	456	318,914	77.91%	1,388	984	1,895,228	68.83%
California	2,784	2,263	3,054,845	80.88%	3,134	2,375	4,457,598	75.71%	6,930	4,697	24,866,806	66.69%
Colorado	642	525	415,441	82.10%	736	552	587,313	75.37%	1,307	925	3,473,655	70.95%
Connecticut	640	497	283,053	78.71%	653	446	385,831	66.63%	1,556	1,001	2,387,658	63.79%
Delaware	632	502	68,597	79.16%	627	454	99,525	73.56%	1,328	940	621,612	71.46%
District of Columbia	537	443	30,707	83.27%	546	425	93,667	77.48%	1,367	991	444,939	72.25%
Florida	2,132	1,713	1,399,768	80.50%	2,221	1,736	1,984,453	78.28%	4,697	3,268	13,702,886	67.98%
Georgia	987	787	846,476	79.53%	990	796	1,114,618	80.35%	2,044	1,464	6,338,910	70.90%
Hawaii	598	475	96,910	78.81%	658	488	140,448	74.29%	1,472	1,025	916,539	69.86%
Idaho	557	453	144,818	82.39%	673	506	174,351	75.52%	1,314	977	1,017,451	72.83%
Illinois	1,636	1,206	1,023,238	73.72%	1,611	1,122	1,388,172	70.17%	3,833	2,434	8,326,464	63.39%
Indiana	630	491	540,670	77.25%	653	485	742,735	74.23%	1,387	964	4,189,743	68.56%
Iowa	614	456	242,812	74.29%	677	505	356,929	75.39%	1,306	913	1,990,458	68.43%
Kansas	622	464	237,562	74.56%	643	522	328,661	82.18%	1,382	982	1,795,749	70.61%
Kentucky	615	489	339,643	78.88%	621	467	472,877	75.43%	1,319	928	2,847,963	68.89%
Louisiana	623	499	367,670	80.31%	672	503	513,576	73.93%	1,289	947	2,928,109	72.21%
Maine	640	489	92,645	75.72%	587	442	125,931	74.65%	1,403	1,003	932,783	71.40%
Maryland	637	500	454,564	79.00%	623	476	625,779	75.64%	1,327	941	3,923,318	69.22%
Massachusetts	675	496	488,093	72.67%	750	494	788,758	65.15%	1,603	958	4,519,295	60.54%
Michigan	1,567	1,198	788,717	75.26%	1,577	1,211	1,114,570	76.49%	3,508	2,450	6,479,469	68.51%
Minnesota	628	499	425,999	78.88%	641	481	571,903	77.37%	1,283	938	3,562,031	73.29%
Mississippi	549	447	244,465	82.33%	561	457	337,215	81.26%	1,317	926	1,859,652	70.64%
Missouri	604	483	470,263	80.03%	666	501	656,687	75.35%	1,290	936	3,918,803	71.70%
Montana	584	452	74,378	78.48%	625	494	111,496	77.00%	1,407	1,008	676,207	69.01%
Nebraska	595	462	151,059	78.03%	634	467	211,662	72.60%	1,340	954	1,179,809	71.51%
Nevada	594	495	222,788	84.09%	652	494	288,699	75.29%	1,350	969	1,872,598	69.19%
New Hampshire	660	496	98,378	76.04%	619	469	142,433	77.73%	1,444	962	905,671	66.24%
New Jersey	1,044	778	697,509	74.07%	1,121	799	894,294	71.13%	2,249	1,476	5,945,550	66.26%
New Mexico	563	474	165,438	85.01%	566	457	227,077	81.69%	1,303	987	1,322,519	75.24%
New York	2,125	1,583	1,427,531	72.87%	2,379	1,646	2,228,431	67.09%	5,294	3,365	13,092,078	62.28%
North Carolina	1,000	818	777,550	82.13%	1,010	788	1,062,442	78.88%	2,071	1,503	6,428,523	71.17%
North Dakota	599	459	51,690	76.55%	669	530	103,308	78.29%	1,314	968	457,339	73.87%
Ohio	1,567	1,197	917,272	76.13%	1,604	1,149	1,229,015	71.61%	3,624	2,497	7,573,265	67.91%
Oklahoma	614	458	312,268	72.62%	587	450	431,096	74.87%	1,442	1,000	2,427,465	66.22%
Oregon	633	498	291,273	79.81%	669	486	414,709	72.00%	1,349	970	2,686,806	71.11%
Pennsylvania	1,480	1,182	934,275	80.00%	1,554	1,194	1,364,517	77.00%	3,384	2,386	8,539,968	69.41%
Rhode Island	611	478	75,156	77.41%	620	453	129,467	73.65%	1,457	1,024	698,360	69.42%
South Carolina	639	521	365,128	82.51%	578	464	520,055	80.79%	1,395	1,000	3,154,439	71.77%
South Dakota	600	481	65,789	80.12%	601	470	93,308	78.30%	1,273	934	534,674	73.67%
Tennessee	590	468	507,891	79.10%	647	506	703,134	78.42%	1,319	976	4,272,566	72.89%
Texas	2,096	1,709	2,361,420	81.17%	2,106	1,640	3,057,833	78.34%	4,737	3,342	16,501,892	69.29%
Utah	579	504	288,637	87.75%	560	467	379,132	82.96%	1,251	969	1,657,347	76.68%
Vermont	638	505	43,770	79.19%	572	407	74,221	72.16%	1,405	996	425,449	70.38%
Virginia	966	783	624,487	81.50%	1,000	755	896,614	75.80%	2,167	1,527	5,378,367	69.49%
Washington	557	441	530,669	78.87%	642	474	745,679	74.11%	1,348	964	4,652,511	70.84%
West Virginia	658	489	129,363	75.39%	616	438	188,612	71.91%	1,408	953	1,251,012	65.67%
Wisconsin	623	478	446,438	75.68%	676	483	624,460	68.97%	1,398	945	3,771,576	68.16%
Wyoming	632	503	44,151	78.66%	615	457	63,351	75.14%	1,314	966	373,558	72.26%

Table C.9 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2015 and 2016

State	Total Selected DUs	Total Eligible DUs	Total Completed Screeners	Weighted DU Screening Response Rate	Total Selected	Total Responded	Population Estimate	Weighted Interview Response Rate	Weighted Overall Response Rate
DU = dwelling unit. NOTE: To compute the pooled 2015-2016 weighted response rates, two samples were combined, and the individual year weights were used for the pooled sample. Thus, the response rates presented here are weighted across 2 years of data rather than being a simple average of the 2015 and 2016 individual response rates. The 2015-2016 population estimate is the average of the 2015 and the 2016 population. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2015 and 2016.
Total U.S.	403,551	338,477	267,398	78.78%	190,106	136,015	268,562,312	68.84%	54.24%
Northeast	89,545	75,780	56,340	72.42%	37,770	25,737	47,803,875	65.12%	47.15%
Midwest	93,119	78,825	64,339	80.59%	45,001	31,913	56,703,618	68.20%	54.96%
South	131,438	108,928	87,417	81.72%	62,382	45,601	100,711,808	70.78%	57.84%
West	89,449	74,944	59,302	77.13%	44,953	32,764	63,343,011	69.15%	53.34%
Alabama	5,793	4,663	3,857	82.65%	2,720	1,936	4,060,554	67.33%	55.65%
Alaska	6,561	4,767	3,793	79.36%	2,698	1,941	583,339	70.31%	55.80%
Arizona	5,943	4,517	3,784	83.81%	2,676	1,978	5,694,340	72.79%	61.01%
Arkansas	5,911	4,847	4,046	83.63%	2,724	1,973	2,462,829	69.23%	57.90%
California	23,474	21,223	15,557	72.90%	13,165	9,290	32,623,357	67.06%	48.89%
Colorado	5,207	4,403	3,552	80.36%	2,652	1,914	4,569,366	69.67%	55.99%
Connecticut	5,852	5,077	3,867	76.19%	2,803	1,901	3,055,331	65.59%	49.98%
Delaware	5,654	4,798	3,636	76.00%	2,653	1,873	798,856	69.45%	52.78%
District of Columbia	11,117	9,460	6,519	68.24%	2,491	1,891	577,705	74.29%	50.70%
Florida	21,812	17,654	13,928	78.86%	9,459	6,821	17,406,100	69.13%	54.51%
Georgia	7,634	6,446	5,046	78.32%	3,990	3,006	8,410,977	71.43%	55.94%
Hawaii	7,088	5,959	4,437	73.99%	2,847	2,024	1,158,228	68.58%	50.74%
Idaho	4,673	3,964	3,372	85.11%	2,706	2,037	1,360,227	73.45%	62.52%
Illinois	14,325	12,596	9,140	72.68%	7,381	4,832	10,719,970	62.47%	45.41%
Indiana	5,289	4,441	3,484	78.35%	2,662	1,906	5,494,678	68.81%	53.91%
Iowa	5,961	5,129	4,341	84.47%	2,771	1,990	2,602,285	70.11%	59.22%
Kansas	5,162	4,487	3,810	84.85%	2,714	1,982	2,368,380	71.29%	60.49%
Kentucky	5,631	4,586	3,799	82.96%	2,716	1,891	3,676,023	67.37%	55.89%
Louisiana	5,564	4,551	3,738	82.47%	2,610	1,916	3,825,536	71.78%	59.20%
Maine	8,218	6,162	5,116	82.99%	2,794	1,986	1,152,976	70.16%	58.23%
Maryland	4,726	4,138	3,063	73.86%	2,607	1,936	5,022,867	71.51%	52.81%
Massachusetts	7,066	6,212	4,496	72.35%	3,187	1,936	5,835,936	59.87%	43.32%
Michigan	14,256	11,680	9,662	82.52%	6,694	4,861	8,399,712	70.01%	57.78%
Minnesota	5,086	4,427	3,621	81.68%	2,661	1,913	4,590,321	70.84%	57.87%
Mississippi	4,936	4,009	3,358	83.89%	2,540	1,855	2,445,529	70.62%	59.25%
Missouri	5,194	4,341	3,772	86.85%	2,676	1,924	5,063,449	68.16%	59.19%
Montana	6,412	5,130	4,406	86.07%	2,762	1,995	870,289	70.32%	60.52%
Nebraska	5,206	4,506	3,675	81.41%	2,665	1,909	1,553,412	70.08%	57.05%
Nevada	5,055	4,382	3,272	74.68%	2,585	1,963	2,428,523	71.30%	53.24%
New Hampshire	6,568	5,526	4,339	78.24%	2,790	1,931	1,150,981	67.72%	52.98%
New Jersey	8,446	7,513	5,598	73.51%	4,396	2,950	7,551,362	64.31%	47.27%
New Mexico	5,475	3,876	3,364	86.94%	2,475	1,939	1,718,723	76.58%	66.58%
New York	24,515	21,212	13,795	64.38%	9,897	6,542	16,764,138	62.50%	40.24%
North Carolina	8,373	7,076	5,822	82.22%	4,214	3,084	8,370,189	70.72%	58.15%
North Dakota	6,936	5,640	5,005	88.76%	2,686	1,948	617,841	70.78%	62.83%
Ohio	13,836	11,832	9,473	80.02%	6,821	4,805	9,735,503	68.04%	54.44%
Oklahoma	5,511	4,483	3,712	82.82%	2,733	1,936	3,192,269	67.92%	56.25%
Oregon	5,686	4,960	4,027	81.27%	2,724	1,966	3,449,136	71.04%	57.74%
Pennsylvania	15,254	12,922	10,331	79.98%	6,540	4,734	10,845,101	71.10%	56.86%
Rhode Island	5,973	5,114	3,958	77.46%	2,710	1,901	904,838	68.42%	53.00%
South Carolina	5,776	4,687	3,889	82.87%	2,630	1,957	4,102,218	72.49%	60.07%
South Dakota	5,167	4,306	3,836	89.29%	2,537	1,864	698,802	72.83%	65.03%
Tennessee	5,704	4,588	3,848	83.93%	2,725	1,997	5,532,419	70.14%	58.87%
Texas	13,020	10,909	9,415	86.01%	8,613	6,601	22,320,973	73.98%	63.63%
Utah	2,989	2,647	2,314	87.51%	2,419	1,904	2,377,053	76.12%	66.62%
Vermont	7,653	6,042	4,840	79.94%	2,653	1,856	543,211	70.02%	55.97%
Virginia	7,854	6,786	5,497	80.99%	4,190	3,019	6,945,044	69.29%	56.12%
Washington	5,471	4,844	3,778	77.87%	2,668	1,878	6,029,145	68.19%	53.10%
West Virginia	6,422	5,247	4,244	80.85%	2,767	1,909	1,561,719	65.27%	52.78%
Wisconsin	6,701	5,440	4,520	83.20%	2,733	1,979	4,859,267	70.85%	58.95%
Wyoming	5,415	4,272	3,646	85.26%	2,576	1,935	481,287	73.68%	62.82%

Table C.10 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2015 and 2016

State	12-17 Total Selected	12-17 Total Responded	12-17 Population Estimate	12-17 Weighted Interview Response Rate	18-25 Total Selected	18-25 Total Responded	18-25 Population Estimate	18-25 Weighted Interview Response Rate	26+ Total Selected	26+ Total Responded	26+ Population Estimate	26+ Weighted Interview Response Rate
NOTE: Computations in this table are based on a respondent's age at screening. Thus, the data in the Total Responded column(s) could differ from data in other NSDUH tables that use the respondent's age recorded during the interview. NOTE: To compute the pooled 2015-2016 weighted response rates, two samples were combined, and the individual year weights were used for the pooled sample. Thus, the response rates presented here are weighted across 2 years of data rather than being a simple average of the 2015 and 2016 individual response rates. The 2015-2016 population estimate is the average of the 2015 and the 2016 population. Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2015 and 2016.
Total U.S.	44,182	34,064	24,894,972	77.31%	46,047	33,788	34,738,945	73.57%	99,877	68,163	208,928,395	67.05%
Northeast	8,725	6,421	4,110,839	71.98%	9,110	6,292	6,084,918	68.31%	19,935	13,024	37,608,119	63.84%
Midwest	10,651	8,060	5,338,955	75.24%	10,953	8,002	7,391,290	72.87%	23,397	15,851	43,973,374	66.56%
South	14,486	11,392	9,506,846	79.09%	15,015	11,299	12,893,966	76.05%	32,881	22,910	78,310,996	68.91%
West	10,320	8,191	5,938,333	79.97%	10,969	8,195	8,368,771	74.15%	23,664	16,378	49,035,907	66.96%
Alabama	593	463	378,330	78.87%	651	494	522,750	75.65%	1,476	979	3,159,474	64.66%
Alaska	639	463	59,083	72.56%	693	523	80,112	75.10%	1,366	955	444,143	69.16%
Arizona	612	473	548,504	78.00%	641	485	746,271	75.11%	1,423	1,020	4,399,565	71.77%
Arkansas	630	491	236,654	78.06%	676	505	317,993	74.15%	1,418	977	1,908,182	67.28%
California	2,920	2,335	3,039,214	80.03%	3,120	2,316	4,399,955	74.31%	7,125	4,639	25,184,187	64.18%
Colorado	627	512	421,468	81.37%	630	453	596,535	71.18%	1,395	949	3,551,363	67.84%
Connecticut	608	465	279,545	77.63%	713	478	388,177	66.27%	1,482	958	2,387,609	64.09%
Delaware	590	455	69,164	78.47%	669	466	97,254	69.48%	1,394	952	632,438	68.45%
District of Columbia	556	450	30,813	81.46%	584	441	93,701	75.27%	1,351	1,000	453,192	73.60%
Florida	2,179	1,703	1,405,801	78.08%	2,190	1,682	1,971,645	77.06%	5,090	3,436	14,028,654	67.21%
Georgia	985	790	855,245	79.58%	879	710	1,112,080	80.07%	2,126	1,506	6,443,651	68.89%
Hawaii	674	508	96,573	73.86%	686	518	135,482	74.95%	1,487	998	926,174	67.07%
Idaho	615	490	146,791	80.19%	722	546	175,145	75.43%	1,369	1,001	1,038,291	72.10%
Illinois	1,771	1,289	1,015,317	72.82%	1,727	1,175	1,372,755	67.41%	3,883	2,368	8,331,898	60.42%
Indiana	599	464	539,568	76.49%	669	497	743,107	74.81%	1,394	945	4,212,004	66.73%
Iowa	695	525	243,253	75.83%	689	492	359,178	71.88%	1,387	973	1,999,854	69.06%
Kansas	684	509	237,647	73.41%	602	465	327,480	78.33%	1,428	1,008	1,803,253	69.73%
Kentucky	641	482	339,903	74.44%	656	457	471,060	70.15%	1,419	952	2,865,061	66.07%
Louisiana	636	493	367,464	77.59%	626	454	503,266	72.73%	1,348	969	2,954,805	70.84%
Maine	696	520	91,487	74.37%	621	442	124,761	71.44%	1,477	1,024	936,729	69.58%
Maryland	571	447	453,674	79.16%	635	478	617,785	74.75%	1,401	1,011	3,951,408	70.20%
Massachusetts	704	456	487,249	65.06%	722	433	792,216	59.97%	1,761	1,047	4,556,470	59.29%
Michigan	1,560	1,211	779,507	77.10%	1,647	1,251	1,108,537	76.52%	3,487	2,399	6,511,669	68.05%
Minnesota	633	486	427,687	76.43%	639	453	572,944	71.18%	1,389	974	3,589,691	70.14%
Mississippi	592	466	244,221	79.24%	594	461	331,045	77.89%	1,354	928	1,870,263	68.22%
Missouri	590	460	469,494	77.31%	693	525	652,575	75.74%	1,393	939	3,941,380	65.89%
Montana	633	488	74,427	76.77%	673	496	111,264	72.71%	1,456	1,011	684,598	69.27%
Nebraska	602	461	152,704	77.18%	688	484	213,106	69.37%	1,375	964	1,187,602	69.31%
Nevada	615	520	224,147	84.33%	630	484	287,409	76.48%	1,340	959	1,916,967	68.90%
New Hampshire	643	474	96,774	74.72%	623	438	142,696	71.64%	1,524	1,019	911,511	66.35%
New Jersey	1,010	756	694,182	74.78%	1,075	744	892,114	68.78%	2,311	1,450	5,965,066	62.45%
New Mexico	570	484	165,412	86.00%	577	457	223,662	80.57%	1,328	998	1,329,649	74.82%
New York	2,293	1,628	1,416,226	68.42%	2,444	1,688	2,197,628	67.30%	5,160	3,226	13,150,285	61.05%
North Carolina	1,002	788	783,879	78.87%	1,001	750	1,053,931	75.40%	2,211	1,546	6,532,379	68.98%
North Dakota	679	508	52,111	74.85%	654	495	102,161	74.02%	1,353	945	463,569	69.53%
Ohio	1,574	1,170	909,989	73.36%	1,636	1,181	1,220,151	72.62%	3,611	2,454	7,605,364	66.70%
Oklahoma	690	524	314,698	76.44%	636	452	428,910	69.76%	1,407	960	2,448,662	66.47%
Oregon	612	458	291,584	74.83%	645	459	417,950	71.50%	1,467	1,049	2,739,602	70.57%
Pennsylvania	1,556	1,188	928,154	76.13%	1,597	1,167	1,344,620	74.15%	3,387	2,379	8,572,327	70.06%
Rhode Island	581	452	74,286	78.17%	680	472	127,975	70.53%	1,449	977	702,577	66.98%
South Carolina	632	510	367,649	80.27%	598	459	515,200	77.38%	1,400	988	3,219,369	70.86%
South Dakota	632	485	66,117	76.96%	608	460	92,977	75.58%	1,297	919	539,708	71.77%
Tennessee	610	465	508,573	75.94%	729	548	700,709	74.02%	1,386	984	4,323,137	68.83%
Texas	1,960	1,606	2,395,358	81.38%	2,145	1,696	3,083,498	78.94%	4,508	3,299	16,842,117	71.98%
Utah	585	502	294,912	85.09%	574	456	387,120	80.24%	1,260	946	1,695,021	73.64%
Vermont	634	482	42,936	76.00%	635	430	74,732	69.62%	1,384	944	425,544	69.47%
Virginia	982	783	626,833	79.72%	1,043	751	888,046	71.83%	2,165	1,485	5,430,166	67.62%
Washington	609	480	532,127	79.33%	688	482	745,740	69.76%	1,371	916	4,751,278	66.71%
West Virginia	637	476	128,586	76.36%	703	495	185,094	70.04%	1,427	938	1,248,039	63.35%
Wisconsin	632	492	445,564	76.30%	701	524	626,320	73.57%	1,400	963	3,787,384	69.73%
Wyoming	609	478	44,091	77.18%	690	520	62,125	75.33%	1,277	937	375,070	72.99%