2013-2014
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 2013-2014 National Surveys on Drug Use and Health (NSDUHs). These estimates are available online along with other related information.1 An annual survey of the civilian, noninstitutionalized population aged 12 or older, NSDUH is sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA), and it collects information from individuals residing in households, noninstitutionalized group quarters (e.g., shelters, rooming houses, dormitories), and civilians living on military bases. In 2013-2014, NSDUH collected data from 135,739 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 The survey is conducted annually from January through December. 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 2013-2014 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.

The survey-weighted hierarchical Bayes (SWHB) estimation methodology used in the production of state estimates from the 1999 to 2013 surveys also was used in the production of the 2013-2014 state estimates. The SWHB methodology is described in Appendix E of the 2001 state report (Wright, 2003b) and in Folsom, Shah, and Vaish (1999). The goals and implementation of small area estimation (SAE) modeling remain the same and are described in Appendix E of the 2001 state report (Wright, 2003b). 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 2013-2014 SAE modeling are listed and described in Section B.3.

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. 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.4.4 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.5. Section B.6 discusses the method to compare the estimates of a particular measure between two states. For all measures except major depressive episode (MDE, i.e., depression), serious mental illness (SMI), any mental illness (AMI), 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. 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.

In Section C, the 2012, 2013, 2014, pooled 2012-2013, and pooled 2013-2014 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 2014.

A.2 Summary of NSDUH Methodology

This section provides a brief overview of the NSDUH methodology, specifically the sample design. For additional details on NSDUH's methodology, see Section A.2 of the 2011-2012 state SAE methodology document.8

The 1999 through 2001 National Household Surveys on Drug Abuse (NHSDAs)9 and the 2002 through 2013 NSDUHs employed a 50-state design with an independent, multistage area probability sample for each of the 50 states and the District of Columbia. For the 50-state design, 8 states were designated as large sample states (California, Florida, Illinois, Michigan, New York, Ohio, Pennsylvania, and Texas) with target sample sizes of 3,600 per year. For the remaining 42 states and the District of Columbia, the target sample size was 900 per year. This approach ensured that there was sufficient sample in every state to support 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 with the exception of Hawaii where the target sample size is 967 interviews). The fielded sample sizes for each state in 2014 are provided in Table C.5, and the combined 2013-2014 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 2013-2014, 287,930 addresses were screened, and 135,739 individuals responded within the screened addresses (see Table C.9). The screening response rate (SRR) for 2013-2014 combined averaged 82.9 percent, and the interview response rate (IRR) averaged 71.4 percent, for an overall response rate (ORR) of 59.2 percent (Table C.9). The ORRs for 2013-2014 ranged from 44.7 percent in New York to 74.0 percent in 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 2012, 2013, and 2014 NSDUHs' methodological resource books (MRBs) (CBHSQ, 2014, 2015a, in press).

A.3 Presentation of Data

In addition to this methodology document for the 2013-2014 state SAE results, the following files are available at http://www.samhsa.gov/data/:

A.4 Confidence Intervals and Margins of Error

At the top of each of the 26 state model-based estimate tables11 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.4. 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 Colorado, with an estimate of 31.2 percent and a 95 percent confidence interval that ranged from 27.7 to 35.1 percent (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 Colorado for young adults aged 18 to 25 in 2013-2014 was between 27.7 and 35.1 percent. As noted earlier in a Section A.1 footnote, the term "prediction interval" (PI) was used in the 2004-2005 NSDUH state report 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 lower interval l comma and upper interval u is a 95 percent symmetric confidence interval for the population proportion (p) and p hat is an estimate of p obtained from the survey data, then the margin of error of p hat is given by u minus p hat or p hat minus l. Because lower interval l comma and upper interval u is a symmetric confidence interval, u minus p hat will be the same as p hat minus l. In this case, the probability is 0.95 that the interval ±u minus p hat or ±p hat minus l will contain the true population value (p). The margin of error defined above 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 of 11.6 percent for adults aged 18 to 25 years with a 95 percent confidence interval equal to (9.3, 14.3) (see Table 3 of the state model-based estimates' tables). Therefore Utah's estimate is 2.3 (i.e., 11.6 − 9.3) percentage points from the lower 95 percent confidence limit and 2.7 (i.e., 14.3 − 11.6) 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 (p hat) 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 Section B.6.

A.5 Related Substance Use Measures

Small area estimates are produced for a number of related drug measures, such as marijuana use and illicit drug use. It might appear that one could draw conclusions by subtracting one from the other (e.g., subtracting the percentage who used illicit drugs other than marijuana in the past month from the percentage who used illicit drugs in the past month to find the percentage who only used marijuana in the past month). Because related measures have been estimated with different models (i.e., separate models by age group and outcome), subtracting one measure from another related measure at the 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.

Section B: State Model-Based Estimation Methodology

B.1 General Model Description

The model can be characterized as a complex mixed12 model (including both fixed and random effects) of the following form:

Equation 1,     D

where pi sub a, i, j, k is the probability of engaging in the behavior of interest (e.g., using marijuana in the past month) for person-k belonging to age group-a in grouped state sampling region (SSR)-j of state-i.13 Let x sub a, i, j, k denote a p sub a times 1 vector of auxiliary (predictor) variables associated with age group-a (12 to 17, 18 to 25, 26 to 34, and 35 or older) and beta sub a denote the associated vector of the regression parameters. The age group-specific vectors of the auxiliary variables are defined for every block group in the nation and also include person-level demographic variables, such as race/ethnicity and gender. The vectors of state-level random effects An eta sub i is a transposed vector of values eta sub 1, i and so on until eta sub capital A, i. and grouped SSR-level random effects A nu sub i, j is a vector of transposed values nu sub 1, i, j and so on until nu sub capital A, i, j. are assumed to be mutually independent with An eta sub i is normally distributed with mean 0 and variance denoted by matrix capital D sub eta. and A nu sub i, j is normally distributed with mean 0 and variance denoted by matrix capital D sub nu., where capital A is the total number of individual age groups modeled (generally, Capital A equals 4.). For hierarchical Bayes (HB) estimation purposes, an improper uniform prior distribution is assumed for beta sub a, and proper Wishart prior distributions are assumed for inverse of capital D sub eta and inverse of capital D sub nu. The HB solution for pi sub a, i, j, k involves a series of complex Markov Chain Monte Carlo (MCMC) steps to generate values of the desired fixed and random effects from the underlying joint 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.4.

B.2 Variables Modeled

The 2014 NSDUH data were pooled with the 2013 NSDUH data, and age group-specific state estimates for 25 binary (0, 1) measures were produced for the following outcomes:

  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,14
  6. past month use of illicit drugs other than marijuana,
  7. past year use of cocaine,
  8. past year nonmedical use of pain relievers,
  9. past month use of alcohol,
  10. past month binge alcohol use,
  11. perceptions of great risk from having five or more drinks of an alcoholic beverage once or twice a week,
  12. past month use of tobacco products,
  13. past month use of cigarettes,
  14. perceptions of great risk from smoking one or more packs of cigarettes per day,
  15. past year alcohol dependence or abuse,
  16. past year alcohol dependence,
  17. past year illicit drug dependence or abuse,
  18. past year illicit drug dependence,
  19. past year dependence or abuse of illicit drugs or alcohol,
  20. needing but not receiving treatment for illicit drug use in the past year,
  21. needing but not receiving treatment for alcohol use in the past year,
  22. serious mental illness (SMI) in the past year,
  23. any mental illness (AMI) in the past year,
  24. serious thoughts of suicide in the past year, and
  25. past year major depressive episode (MDE, i.e., depression).

Estimates of underage (aged 12 to 20) alcohol use and binge alcohol use were also produced. Comparisons between the 2012-2013 and the 2013-2014 state estimates were produced for all of these measures as well. For details on how measures such as AMI, SMI, MDE, illicit drugs, dependence or abuse, and average annual rate of first use of marijuana are defined, see "2011-2012 National Surveys on Drug Use and Health: Guide to State Tables and Summary of Small Area Estimation Methodology" at http://www.samhsa.gov/data/.

Illicit drug use includes the use of methamphetamines. NSDUH questions on methamphetamine use are asked in the stimulant module of the core section of the questionnaire in the context of questions about the nonmedical use of prescription stimulants. Beginning in 2005, new questions were added to the special drug module in the noncore section of the NSDUH questionnaire to capture information from respondents who may have used methamphetamines but did not recognize it as a prescription drug and therefore did not report use in the core stimulant module. However, the illicit drug estimates (including dependence, abuse, and treatment of illicit drugs) shown in the 2013-2014 small area estimation (SAE) documents include data from the original methamphetamine questions, but they do not include the new methamphetamine items added in 2005 and 2006 (i.e., the responses to the noncore questions). For more information on these new methamphetamine questions, see the findings from the methamphetamine analysis section of the 2005 NSDUH Methodological Resource Book (Center for Behavioral Health Statistics and Quality [CBHSQ], 2007).

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 several sources, including Nielsen Claritas, the U.S. Census Bureau, the Federal Bureau of Investigation (FBI) (Uniform Crime Reports [UCRs]), the Bureau of Labor Statistics (BLS), the Bureau of Economic Analysis (BEA), the Substance Abuse and Mental Health Services Administration (SAMHSA) (National Survey of Substance Abuse Treatment Services [N-SSATS]), and the National Center for Health Statistics (NCHS) (mortality data). Note that the predictors used to produce the 2013-2014 state small area estimates are the same as the predictors used to produce the 2012-2013 state small area estimates (however, values of the data were updated when possible). That is, no new variable selection was done for 2013-2014.

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

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

Nielsen Claritas Data (Description) Nielsen 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

2010 Census Data (Description) 2010 Census Data (Level)
% Hispanics Who Are Cuban Tract

American Community Survey (ACS) (Description) ACS Data (Level)
% 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
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 greater than or equal to 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 (NCHS)
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

B.4 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 solution, 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 2013-2014 state-by-age group small area estimates is adjusted by adding the common factor Delta sub a is defined as the national design-based estimate, capital D sub a, minus the national model-based small area estimate, capital P sub a., where capital D sub a is the design-based national estimate and capital P sub a is the population-weighted mean of the state small area estimates capital P sub s and a for age group-a. The exactly benchmarked state-s and age group-a small area estimates then are given by The benchmarked state-s and age group-a small area estimate, Theta sub s and a, is defined as the sum of capital P sub s and a and Delta sub a.. 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 Lower sub s and a is the lower bound of the 95 percent Bayesian confidence interval of Theta sub s and a; upper sub s and a is the upper bound of the 95 confidence interval of Theta sub s and a. are defined below:

Equation 2,     D

where

Equation 3,     D

Equation 4,     D     and

Equation 5.     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.

B.5 Calculation of Estimated Number of Individuals Associated with Each Outcome

Tables 1 to 26 of "2013-2014 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 25 outcomes of interest.15 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, 2013 and 2014) 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 51.36 percent.16 The corresponding Bayesian confidence intervals ranged from 47.49 to 55.21 percent. The population count for 18 to 25 year olds averaged across 2013 and 2014 in Alabama was 535,409 (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 0.5136 × 535,409, which is 274,986.17 The associated Bayesian confidence intervals ranged from 0.4749 × 535,409 (i.e., 254,266) to 0.5521 × 535,409 (i.e., 295,599). Note that when estimates of the number of individuals are calculated for Tables 1 to 26 in "2013-2014 NSDUHs: Model-Based Estimated Totals (in Thousands) (50 States and the District of Columbia)" (follow the link in footnote 17), 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.

B.6 Comparison of Two 2013-2014 Small Area Estimates

This section describes a method for determining whether differences between two 2013-2014 State population percentages are statistically significant. This procedure can be used for any two State population percentages representing the same age group (e.g., young adults aged 18 to 25) and time period (e.g., 2013-2014).

Let pi 1 sub a and pi 2 sub a denote the 2013-2014 age group-a specific prevalence rates for two different states, state 1 and state 2, respectively. The null hypothesis of no difference, that is, Pi 1 sub a is equal to pi 2 sub a., is equivalent to the log-odds ratio equal to zero, that is, Log-odds ratio lor sub a is equal to zero., where lor sub a is defined as The log-odds ratio, lor sub a, is defined as the natural logarithm of the ratio of two quantities. The numerator of the ratio is pi 2 sub a divided by 1 minus pi 2 sub a. The denominator of the ratio is pi 1 sub a divided by 1 minus pi 1 sub a., where ln denotes the natural logarithm. An estimate of lor sub a is given by The estimate of the log-odds ratio, lor hat sub a, is defined as the natural logarithm of the ratio of two quantities. The numerator of the ratio is p 2 sub a divided by 1 minus p 2 sub a. The denominator of the ratio is p 1 sub a divided by 1 minus p 1 sub a., where p 1 sub a and p 2 sub a are the 2013-2014 state estimates given in the "2013-2014 NSDUH: Model-Based Prevalence Estimates (50 States and the District of Columbia) (Tables 1 to 26, by Age Group)" (follow the link in footnote 16). To compute the variance of the estimate of the log-odds ratio, lor hat sub a, that is, variance v of the estimate of the log-odds ratio, lor hat sub a, let Theta 1 hat is defined as the ratio of p 1 sub a and 1 minus p 1 sub a. and Theta 2 hat is defined as the ratio of p 2 sub a and 1 minus p 2 sub a., then Variance v of the estimate of the log-odds ratio, lor hat sub a, is a function of three quantities: q1, q2, and q3. It is expressed as the sum of q1 and q2 minus q3. Quantity q1 is the variance v of the natural logarithm of Theta 1 hat, quantity q2 is the variance v of the natural logarithm of Theta 2 hat, and quantity q3 is 2 times the covariance between the natural logarithm of Theta 1 hat and the natural logarithm of Theta 2 hat., where the covariance between the natural logarithm of Theta 1 hat and the natural logarithm of Theta 2 hat denotes the covariance between the natural logarithm of Theta 1 hat and the natural logarithm of Theta 2 hat. This covariance is defined in terms of the associated correlation as follows:

Equation 6.     D

The quantities variance v of the natural logarithm of Theta 1 hat and variance v of the natural logarithm of Theta 2 hat can be obtained by using the 95 percent Bayesian confidence intervals given in the "2013-2014 NSDUH: Model-Based Prevalence Estimates (50 States and the District of Columbia) (Tables 1 to 26, by Age Group)" (follow the link in footnote 16). For this purpose, let lower sub 1 and upper sub 1 and lower sub 2 and upper sub 2 denote the 95 percent Bayesian confidence intervals for the two states, state 1 and state 2, respectively. Then

Equation 7,     D

where Capital U sub i is the natural logarithm of upper sub i divided by 1 minus upper sub i, and capital L sub i is the natural logarithm of lower sub i divided by 1 minus lower sub i..

For all practical purposes, the correlation between the natural logarithm of Theta 1 hat and the natural logarithm of Theta 2 hat is assumed to be negligible; hence, variance v of the estimate of the log-odds ratio, lor hat sub a can be approximated by the sum of the variance v of the natural logarithm of Theta 1 hat and the variance v of the natural logarithm of Theta 2 hat. The correlation is assumed to be negligible because each State was a stratum in the first level of stratification; therefore, each State sample is selected independently. However, the correlation between the two State estimates is theoretically nonzero because State estimates share common fixed-effect parameters in the SAE models. Hence, the test statistic quantity z (defined below) might result in a different conclusion in a few cases when the correlation between the State estimates is incorporated in calculating variance v of the estimate of the log-odds ratio, lor hat sub a. To calculate the p value for testing the null hypothesis of no difference (Log-odds ratio, lor sub a, is equal to zero.), it is assumed that the posterior distribution of log-odds ratio, lor sub a is normal with Mean is equal to the estimate of the log-odds ratio, lor hat sub a. and Variance is equal to the variance v of the estimate of the log-odds ratio, lor hat sub a.. With the null value of Log-odds ratio, lor sub a, is equal to zero., the Bayes p value or significance level for the null hypothesis of no difference is The p value is equal to 2 times the probability of realizing a standard normal variate greater than or equal to the absolute value of a quantity z., where capital Z is a standard normal random variate, Quantity z is the estimate of the log-odds ratio, lor hat sub a, divided by the square root of the sum of the variance v of the natural logarithm of Theta 1 hat and the variance v of the natural logarithm of Theta 2 hat., and absolute value of quantity z denotes the absolute value of quantity z. This Bayesian significance level (or p value) for the null value of Log-odds ratio lor, say log-odds ratio lor sub zero, is defined following Rubin (1987) as the posterior probability for the collection of the Log-odds ratio lor values that are less likely or have smaller posterior density d of the log-odds ratio lor than the null (no change) value log-odds ratio lor sub zero. That is, The p value of log-odds ratio lor sub zero is equal to the probability of d of the log-odds ratio lor when it is less than or equal to d of the log-odds ratio lor sub zero.. With the posterior distribution of Log-odds ratio lor approximately normal, the p value of log-odds ratio lor sub zero is given by the above expression.

Hence, to test whether differences between two 2013-2014 state estimates are statistically significant, the test statistic quantity z and the associated p value can be used. If p less than or equal to 0.05, then the two state estimates can be considered different at the 5 percent level of significance. Because age group estimates within a state are correlated, the method described here cannot be used to test whether differences between two age group estimates within a state are statistically significant.

When comparing estimates for two states, it is tempting and often convenient to look at their 95 percent Bayesian confidence intervals to decide whether the difference in the state estimates is significant. If the two Bayesian confidence intervals overlap, one would conclude that the difference is not statistically significant. If the two Bayesian confidence intervals do not overlap, it implies that the state estimates are significantly different from each other. However, the type-I error for the overlapping 95 percent Bayesian confidence intervals test may be as low as 0.6 percent (assuming that the two state estimates are uncorrelated and have the same variances) as compared with the 5 percent type-I error of the test based on the quantity z statistics defined above (Payton, Greenstone, & Schenker, 2003).

As discussed in Schenker and Gentleman (2001), the method of overlapping Bayesian confidence intervals is more conservative (i.e., it rejects the null hypothesis of no difference less often) than the standard method based on quantity z statistics when the null hypothesis is true. Even if Bayesian confidence intervals for two states overlap, the two estimates may be declared significantly different by the test based on quantity z statistics. Hence, the method of overlapping Bayesian confidence intervals is not recommended to test the difference of two state estimates. A detailed description of the method of overlapping confidence intervals and its comparison with the standard methods for testing of a hypothesis is given in Schenker and Gentleman (2001) and Payton et al. (2003).

Example. The percentages for past month alcohol use among 12 to 17 year olds in New Jersey and Oklahoma are shown in the following exhibit and also in Table 9 of the "2013-2014 NSDUH: Model-Based Prevalence Estimates (50 States and the District of Columbia)" at http://www.samhsa.gov/data/. Looking at the two 95 percent Bayesian confidence intervals, it would appear that the Oklahoma and New Jersey percentages for past month alcohol use are not statistically different at the 5 percent level of significance because the two Bayesian confidence intervals overlap:

State Point Estimate (%) 95% Bayesian Confidence Interval (%)
New Jersey 14.31 (12.14, 16.78)
Oklahoma 10.89   (9.00, 13.13)

However, in the following example, the test based on the quantity z statistic described earlier concludes that they are significantly different at the 5 percent level of significance.

Let p 1 sub a equal 0.1431, lower sub 1 equal 0.1214, upper sub 1 equal 0.1678, p 2 sub a equal 0.1089, lower sub 2 equal 0.0900, upper sub 2 equal 0.1313. Then,

Equation 8,     D

Equation 9,     D

Equation 10,     D

Equation 11,     D

Equation 12,     D   and

Equation 13.     D


Because the computed absolute value of quantity z is greater than or equal to 1.96 (the critical value of the quantity z statistic), then at the 5 percent level of significance, the hypothesis of no difference (Oklahoma prevalence rate = New Jersey prevalence rate) is rejected. Thus, the two state prevalence rates are statistically different. The Bayes p value or significance level for the null hypothesis of no difference is p value is equal to 2 times probability that the z statistic is greater than or equal to the absolute value of negative 2.1546, which is 0.0312.

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: 2012
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, 2012.
Total U.S. 214,274 178,586 153,873 86.07% 87,656 68,309 260,057,325 73.04% 62.87%
Northeast 47,763 40,410 32,868 79.93% 18,301 13,773 47,174,958 69.59% 55.62%
Midwest 58,534 49,381 43,010 87.61% 24,499 19,142 55,924,697 74.27% 65.06%
South 66,141 54,110 47,494 88.15% 26,279 20,886 96,373,144 74.22% 65.42%
West 41,836 34,685 30,501 86.04% 18,577 14,508 60,584,526 72.75% 62.59%
Alabama 3,012 2,372 2,141 90.30% 1,145 901 4,005,432 74.57% 67.34%
Alaska 2,424 1,869 1,642 87.82% 1,076 829 577,147 73.34% 64.40%
Arizona 2,771 2,143 1,928 90.16% 1,139 922 5,362,657 77.11% 69.52%
Arkansas 2,776 2,292 2,090 90.92% 1,212 913 2,422,926 69.77% 63.43%
California 9,489 8,314 6,852 82.37% 4,779 3,608 31,424,054 70.20% 57.82%
Colorado 3,071 2,579 2,201 85.23% 1,188 927 4,260,412 74.95% 63.88%
Connecticut 2,855 2,535 2,107 82.76% 1,261 964 3,034,241 72.36% 59.88%
Delaware 2,847 2,292 2,008 87.57% 1,110 893 765,733 79.90% 69.97%
District of Columbia 5,055 4,104 3,327 80.90% 1,125 962 544,627 80.64% 65.24%
Florida 12,768 10,055 8,516 84.67% 4,579 3,544 16,382,543 70.57% 59.75%
Georgia 2,365 2,042 1,796 87.94% 1,144 885 8,040,955 73.07% 64.26%
Hawaii 3,212 2,761 2,239 80.80% 1,285 938 1,130,820 68.98% 55.73%
Idaho 2,300 1,939 1,821 93.92% 1,136 921 1,288,271 78.38% 73.61%
Illinois 11,385 9,964 7,678 77.04% 4,871 3,672 10,680,769 70.95% 54.66%
Indiana 2,491 2,110 1,921 91.01% 1,171 911 5,391,372 72.95% 66.39%
Iowa 2,529 2,199 2,022 91.72% 1,137 900 2,550,660 74.74% 68.55%
Kansas 2,598 2,198 1,977 89.98% 1,109 912 2,336,047 77.88% 70.07%
Kentucky 2,852 2,407 2,202 91.46% 1,184 927 3,607,428 73.49% 67.21%
Louisiana 2,741 2,143 1,977 92.28% 1,100 901 3,745,460 77.61% 71.63%
Maine 3,866 2,858 2,585 90.56% 1,134 938 1,145,565 79.20% 71.72%
Maryland 2,680 2,308 1,802 78.13% 1,074 874 4,905,827 75.90% 59.30%
Massachusetts 3,064 2,653 2,208 83.22% 1,253 955 5,661,530 71.52% 59.52%
Michigan 11,441 9,207 7,826 85.05% 4,606 3,655 8,319,227 75.75% 64.43%
Minnesota 2,483 2,160 1,975 91.57% 1,092 902 4,470,679 81.16% 74.32%
Mississippi 2,553 2,087 1,951 93.50% 1,100 901 2,419,811 78.58% 73.48%
Missouri 2,879 2,409 2,188 90.88% 1,149 915 4,985,565 74.36% 67.58%
Montana 3,295 2,610 2,415 92.62% 1,109 876 842,009 77.46% 71.74%
Nebraska 2,556 2,175 2,018 92.74% 1,170 940 1,511,302 73.14% 67.83%
Nevada 2,354 1,879 1,721 91.75% 1,134 903 2,278,656 75.62% 69.38%
New Hampshire 2,990 2,507 2,191 87.40% 1,259 950 1,133,661 73.08% 63.87%
New Jersey 2,622 2,227 1,935 86.87% 1,155 898 7,440,994 73.64% 63.97%
New Mexico 2,771 2,052 1,889 92.22% 1,101 879 1,702,667 74.17% 68.39%
New York 14,547 12,547 9,115 71.89% 5,267 3,680 16,532,006 64.38% 46.28%
North Carolina 2,848 2,246 1,990 88.48% 1,117 917 8,007,328 75.46% 66.77%
North Dakota 3,374 2,633 2,461 93.42% 1,156 895 577,526 73.47% 68.64%
Ohio 11,722 10,122 9,023 89.14% 4,827 3,687 9,638,652 72.73% 64.84%
Oklahoma 2,960 2,382 2,173 91.22% 1,189 908 3,099,247 72.38% 66.03%
Oregon 2,547 2,250 2,019 89.57% 1,165 923 3,293,097 76.48% 68.51%
Pennsylvania 11,907 10,256 8,453 82.09% 4,705 3,580 10,790,033 70.67% 58.02%
Rhode Island 2,620 2,190 1,957 89.37% 1,131 923 895,345 77.76% 69.50%
South Carolina 3,306 2,666 2,374 88.97% 1,171 938 3,900,041 75.13% 66.85%
South Dakota 2,636 2,163 2,031 93.92% 1,113 878 676,283 76.12% 71.49%
Tennessee 2,532 2,095 1,929 91.91% 1,105 927 5,363,074 81.06% 74.50%
Texas 9,048 7,651 6,792 88.52% 4,612 3,625 20,852,844 73.36% 64.94%
Utah 1,793 1,558 1,474 94.67% 1,099 926 2,214,352 83.26% 78.83%
Vermont 3,292 2,637 2,317 87.81% 1,136 885 541,583 73.81% 64.82%
Virginia 2,576 2,293 2,027 88.47% 1,095 894 6,735,698 76.50% 67.68%
Washington 2,700 2,306 2,078 90.10% 1,218 928 5,736,136 71.82% 64.71%
West Virginia 3,222 2,675 2,399 89.39% 1,217 976 1,574,171 74.07% 66.21%
Wisconsin 2,440 2,041 1,890 92.37% 1,098 875 4,786,617 75.55% 69.79%
Wyoming 3,109 2,425 2,222 91.72% 1,148 928 474,248 77.48% 71.07%
Table C.2 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2012
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, 2012.
Total U.S. 27,147 22,492 24,933,051 82.84% 28,639 22,762 34,589,953 79.26% 31,870 23,055 200,534,321 70.76%
Northeast 5,513 4,421 4,237,419 79.81% 6,114 4,720 6,153,492 76.54% 6,674 4,632 36,784,047 67.26%
Midwest 7,733 6,399 5,416,148 83.34% 7,891 6,270 7,361,823 79.64% 8,875 6,473 43,146,726 72.22%
South 8,292 6,973 9,305,299 83.52% 8,583 7,012 12,758,779 81.70% 9,404 6,901 74,309,066 71.75%
West 5,609 4,699 5,974,186 83.44% 6,051 4,760 8,315,859 77.22% 6,917 5,049 46,294,482 70.61%
Alabama 342 278 384,244 80.41% 383 312 536,932 80.90% 420 311 3,084,257 72.65%
Alaska 304 233 60,308 76.07% 348 286 81,619 82.25% 424 310 435,220 71.44%
Arizona 366 312 539,163 85.61% 371 293 713,584 74.97% 402 317 4,109,911 76.39%
Arkansas 394 312 236,048 78.13% 404 310 317,735 75.45% 414 291 1,869,143 67.71%
California 1,409 1,159 3,139,169 81.82% 1,584 1,216 4,452,711 76.51% 1,786 1,233 23,832,173 67.51%
Colorado 376 319 399,087 86.13% 390 301 560,123 78.11% 422 307 3,301,202 73.13%
Connecticut 361 288 289,862 79.74% 426 339 373,279 80.56% 474 337 2,371,100 70.39%
Delaware 376 307 68,973 82.59% 305 246 102,090 83.85% 429 340 594,670 79.02%
District of Columbia 362 329 31,338 91.77% 398 344 95,556 87.06% 365 289 417,734 78.39%
Florida 1,419 1,193 1,383,312 83.48% 1,535 1,222 1,970,724 79.16% 1,625 1,129 13,028,506 67.81%
Georgia 344 287 828,383 81.72% 360 284 1,096,583 79.58% 440 314 6,115,989 70.82%
Hawaii 377 284 96,933 75.93% 382 308 140,267 80.83% 526 346 893,621 66.50%
Idaho 389 345 139,664 88.85% 334 262 173,325 80.12% 413 314 975,282 76.28%
Illinois 1,517 1,234 1,051,880 81.95% 1,562 1,190 1,393,334 76.45% 1,792 1,248 8,235,555 68.62%
Indiana 330 271 540,535 82.24% 408 328 731,531 80.64% 433 312 4,119,306 70.63%
Iowa 373 314 241,376 82.15% 362 287 347,524 79.41% 402 299 1,961,760 72.90%
Kansas 388 343 236,447 88.15% 318 265 322,233 84.49% 403 304 1,777,368 75.30%
Kentucky 384 318 339,442 81.85% 380 302 461,441 80.21% 420 307 2,806,546 71.39%
Louisiana 330 292 367,661 88.75% 364 303 523,034 82.65% 406 306 2,854,766 75.23%
Maine 359 305 95,666 85.30% 387 325 129,416 84.13% 388 308 920,484 77.79%
Maryland 330 282 458,368 85.48% 363 306 631,975 83.31% 381 286 3,815,483 73.39%
Massachusetts 380 309 493,395 81.19% 408 312 772,360 77.20% 465 334 4,395,776 69.50%
Michigan 1,445 1,178 809,401 81.72% 1,508 1,231 1,101,787 81.78% 1,653 1,246 6,408,038 73.97%
Minnesota 363 324 424,357 89.54% 339 272 571,203 79.91% 390 306 3,475,119 80.32%
Mississippi 384 313 248,208 80.62% 338 297 336,270 88.22% 378 291 1,835,332 76.36%
Missouri 367 312 474,059 85.89% 356 290 654,819 82.34% 426 313 3,856,687 71.53%
Montana 388 316 73,775 81.81% 350 279 107,843 78.48% 371 281 660,391 76.71%
Nebraska 322 278 147,378 86.79% 433 365 205,771 84.84% 415 297 1,158,152 69.50%
Nevada 333 290 220,899 86.58% 368 289 284,532 79.10% 433 324 1,773,226 73.75%
New Hampshire 405 305 102,103 75.51% 417 324 139,482 78.95% 437 321 892,076 71.84%
New Jersey 349 291 708,659 83.09% 378 292 881,583 78.25% 428 315 5,850,752 71.73%
New Mexico 332 290 168,839 87.22% 369 303 226,708 81.39% 400 286 1,307,120 71.17%
New York 1,564 1,193 1,466,519 75.84% 1,778 1,266 2,246,785 71.75% 1,925 1,221 12,818,701 61.76%
North Carolina 354 298 760,601 83.53% 382 337 1,033,454 87.89% 381 282 6,213,274 72.36%
North Dakota 371 309 48,912 83.61% 339 268 93,645 79.86% 446 318 434,970 70.99%
Ohio 1,628 1,297 926,791 79.72% 1,475 1,148 1,232,694 77.78% 1,724 1,242 7,479,167 71.02%
Oklahoma 385 303 305,458 78.05% 383 297 424,952 76.87% 421 308 2,368,838 70.82%
Oregon 311 270 292,395 87.03% 407 318 409,756 79.10% 447 335 2,590,946 75.05%
Pennsylvania 1,425 1,169 958,552 82.15% 1,536 1,218 1,404,841 79.74% 1,744 1,193 8,426,641 67.91%
Rhode Island 320 276 77,245 86.40% 391 329 132,691 84.47% 420 318 685,409 75.55%
South Carolina 385 317 358,471 81.59% 349 295 515,765 84.67% 437 326 3,025,806 72.71%
South Dakota 316 265 64,543 84.11% 371 300 91,525 82.83% 426 313 520,215 74.13%
Tennessee 299 261 505,108 85.96% 419 352 688,253 83.32% 387 314 4,169,713 80.11%
Texas 1,472 1,246 2,279,511 84.37% 1,471 1,183 2,943,283 80.38% 1,669 1,196 15,630,050 70.45%
Utah 319 287 272,004 90.49% 384 310 363,798 81.78% 396 329 1,578,549 82.34%
Vermont 350 285 45,420 80.52% 393 315 73,055 80.92% 393 285 423,108 71.93%
Virginia 373 322 619,042 85.05% 316 270 891,542 85.19% 406 302 5,225,114 73.95%
Washington 368 301 528,812 81.58% 406 310 737,911 75.50% 444 317 4,469,414 70.24%
West Virginia 359 315 131,131 87.64% 433 352 189,192 81.40% 425 309 1,253,848 71.61%
Wisconsin 313 274 450,470 86.72% 420 326 615,758 77.80% 365 275 3,720,389 73.85%
Wyoming 337 293 43,140 85.79% 358 285 63,681 78.58% 453 350 367,427 76.36%
Table C.3 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2013
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, 2013.
Total U.S. 227,075 190,067 160,325 83.93% 88,742 67,838 262,391,455 71.69% 60.18%
Northeast 51,312 43,608 34,787 78.54% 18,334 13,661 47,388,235 68.75% 54.00%
Midwest 61,705 51,906 44,380 85.68% 24,842 18,822 56,214,652 71.54% 61.30%
South 69,936 57,463 49,288 85.67% 26,758 20,782 97,513,014 73.32% 62.81%
West 44,122 37,090 31,870 83.74% 18,808 14,573 61,275,553 71.48% 59.86%
Alabama 3,110 2,522 2,141 84.04% 1,156 900 4,025,044 69.26% 58.21%
Alaska 3,177 2,347 2,044 87.05% 1,122 863 577,309 74.91% 65.21%
Arizona 3,013 2,324 1,991 85.43% 1,170 882 5,443,545 69.25% 59.16%
Arkansas 2,721 2,189 1,984 90.66% 1,193 908 2,435,182 73.21% 66.38%
California 9,994 8,965 7,211 80.33% 4,864 3,729 31,739,919 70.45% 56.60%
Colorado 2,790 2,436 2,016 82.93% 1,173 885 4,339,337 71.19% 59.04%
Connecticut 2,989 2,691 2,294 85.25% 1,198 893 3,045,630 70.24% 59.88%
Delaware 3,042 2,485 2,073 83.64% 1,113 862 774,640 72.21% 60.40%
District of Columbia 5,466 4,554 3,700 80.83% 1,142 907 555,335 75.40% 60.95%
Florida 14,174 11,056 9,176 81.41% 4,792 3,649 16,599,656 71.63% 58.31%
Georgia 2,660 2,218 1,836 82.63% 1,093 852 8,133,541 73.03% 60.34%
Hawaii 3,294 2,861 2,235 77.45% 1,240 924 1,135,919 66.79% 51.73%
Idaho 2,388 2,020 1,863 92.19% 1,163 907 1,305,833 75.66% 69.75%
Illinois 11,767 10,379 7,912 76.19% 4,935 3,503 10,713,667 65.98% 50.27%
Indiana 2,992 2,513 2,182 86.71% 1,165 894 5,430,975 71.51% 62.00%
Iowa 2,700 2,318 2,120 91.46% 1,164 900 2,566,989 71.34% 65.25%
Kansas 2,608 2,191 1,944 88.60% 1,165 887 2,344,171 73.15% 64.81%
Kentucky 3,085 2,556 2,341 91.53% 1,160 904 3,633,237 73.51% 67.28%
Louisiana 2,877 2,321 2,096 90.32% 1,160 903 3,774,189 73.28% 66.19%
Maine 3,624 2,708 2,444 90.02% 1,125 926 1,147,984 78.25% 70.44%
Maryland 2,759 2,430 1,919 79.18% 1,183 925 4,947,041 76.85% 60.85%
Massachusetts 3,007 2,692 2,189 80.96% 1,240 897 5,711,595 69.49% 56.26%
Michigan 12,080 9,938 8,310 83.39% 4,716 3,636 8,346,148 72.79% 60.70%
Minnesota 2,595 2,272 2,056 90.74% 1,126 906 4,509,704 77.38% 70.21%
Mississippi 2,441 2,019 1,829 90.55% 1,088 918 2,428,802 79.27% 71.77%
Missouri 3,144 2,586 2,330 89.93% 1,183 917 5,009,791 73.20% 65.83%
Montana 2,991 2,429 2,251 92.54% 1,177 910 850,469 74.42% 68.87%
Nebraska 3,052 2,500 2,279 91.03% 1,146 910 1,524,399 74.27% 67.61%
Nevada 2,753 2,285 2,004 87.68% 1,137 932 2,312,257 74.64% 65.44%
New Hampshire 3,488 2,919 2,498 85.43% 1,243 953 1,137,904 76.03% 64.95%
New Jersey 3,164 2,774 2,281 82.31% 1,238 913 7,476,944 68.88% 56.70%
New Mexico 2,868 2,254 2,038 90.20% 1,168 922 1,707,564 73.84% 66.60%
New York 15,157 12,992 9,243 71.27% 5,248 3,637 16,619,482 63.66% 45.36%
North Carolina 2,872 2,382 2,090 87.63% 1,103 880 8,114,142 75.94% 66.55%
North Dakota 3,634 2,767 2,562 92.58% 1,257 945 593,987 68.81% 63.71%
Ohio 11,540 9,824 8,450 85.92% 4,734 3,568 9,677,958 71.01% 61.01%
Oklahoma 2,830 2,326 2,100 90.39% 1,250 950 3,130,656 68.89% 62.27%
Oregon 2,770 2,458 2,153 87.44% 1,093 861 3,327,918 76.84% 67.19%
Pennsylvania 13,292 11,490 9,213 80.00% 4,760 3,663 10,808,879 73.13% 58.50%
Rhode Island 2,969 2,515 2,205 87.59% 1,167 904 897,301 71.97% 63.04%
South Carolina 3,291 2,763 2,308 83.36% 1,134 908 3,952,463 76.40% 63.69%
South Dakota 2,728 2,204 2,059 93.35% 1,106 889 685,112 76.78% 71.68%
Tennessee 2,967 2,431 2,152 88.53% 1,121 894 5,407,982 73.11% 64.72%
Texas 9,323 7,887 6,873 87.12% 4,743 3,604 21,223,105 72.07% 62.79%
Utah 2,032 1,771 1,678 95.05% 1,150 930 2,258,561 75.09% 71.37%
Vermont 3,622 2,827 2,420 85.51% 1,115 875 542,516 76.92% 65.78%
Virginia 2,792 2,413 2,072 85.14% 1,148 902 6,803,508 76.51% 65.15%
Washington 2,598 2,235 1,937 86.55% 1,175 900 5,797,644 71.56% 61.93%
West Virginia 3,526 2,911 2,598 89.32% 1,179 916 1,574,493 76.28% 68.13%
Wisconsin 2,865 2,414 2,176 90.41% 1,145 867 4,811,751 73.66% 66.60%
Wyoming 3,454 2,705 2,449 90.40% 1,176 928 479,279 78.69% 71.14%
Table C.4 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2013
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, 2013.
Total U.S. 27,630 22,532 24,892,618 81.95% 28,921 22,458 34,785,501 77.34% 32,191 22,848 202,713,336 69.45%
Northeast 5,700 4,561 4,187,318 79.38% 5,915 4,465 6,149,025 74.20% 6,719 4,635 37,051,892 66.60%
Midwest 7,730 6,220 5,398,028 80.27% 8,236 6,328 7,406,554 76.24% 8,876 6,274 43,410,071 69.65%
South 8,368 6,904 9,356,405 82.51% 8,566 6,762 12,857,518 78.55% 9,824 7,116 75,299,092 71.29%
West 5,832 4,847 5,950,868 84.38% 6,204 4,903 8,372,403 78.74% 6,772 4,823 46,952,282 68.53%
Alabama 381 322 382,694 82.54% 377 304 536,933 78.79% 398 274 3,105,417 66.03%
Alaska 364 276 60,220 76.37% 380 301 83,264 77.91% 378 286 433,826 74.16%
Arizona 396 323 541,841 81.38% 385 293 727,937 76.31% 389 266 4,173,767 66.25%
Arkansas 327 255 236,968 78.23% 454 350 319,725 76.45% 412 303 1,878,489 72.01%
California 1,490 1,263 3,095,715 85.24% 1,571 1,236 4,464,898 78.73% 1,803 1,230 24,179,306 66.97%
Colorado 322 259 405,187 80.90% 399 304 570,429 75.38% 452 322 3,363,721 69.41%
Connecticut 391 316 287,546 82.74% 351 271 378,789 78.01% 456 306 2,379,294 67.41%
Delaware 334 281 67,694 82.04% 396 309 102,069 78.44% 383 272 604,877 70.04%
District of Columbia 374 327 30,375 88.49% 304 237 93,799 80.28% 464 343 431,161 73.41%
Florida 1,407 1,156 1,387,520 82.81% 1,513 1,184 1,973,936 77.89% 1,872 1,309 13,238,200 69.64%
Georgia 358 291 834,836 82.28% 384 306 1,103,523 79.41% 351 255 6,195,182 70.39%
Hawaii 368 306 97,238 81.23% 417 321 140,183 75.08% 455 297 898,498 64.16%
Idaho 337 280 142,022 84.51% 429 341 172,682 82.06% 397 286 991,129 73.13%
Illinois 1,460 1,145 1,039,658 79.14% 1,661 1,201 1,395,665 71.65% 1,814 1,157 8,278,344 63.39%
Indiana 366 292 541,496 78.05% 365 288 738,003 77.25% 434 314 4,151,475 69.66%
Iowa 357 287 242,247 79.14% 395 315 350,483 80.07% 412 298 1,974,259 68.83%
Kansas 369 296 237,924 80.42% 386 295 324,627 77.64% 410 296 1,781,619 71.39%
Kentucky 366 300 340,478 82.34% 365 296 468,033 81.37% 429 308 2,824,726 71.05%
Louisiana 370 297 367,993 78.65% 340 276 520,801 79.72% 450 330 2,885,395 71.59%
Maine 390 328 94,311 82.76% 361 306 127,972 84.65% 374 292 925,702 76.97%
Maryland 375 302 455,935 81.11% 389 306 630,762 76.22% 419 317 3,860,344 76.45%
Massachusetts 370 285 489,152 76.58% 427 311 777,767 73.11% 443 301 4,444,677 68.04%
Michigan 1,488 1,194 802,126 80.07% 1,550 1,220 1,112,833 78.07% 1,678 1,222 6,431,190 70.93%
Minnesota 335 287 424,921 87.36% 391 307 571,675 76.12% 400 312 3,513,108 76.46%
Mississippi 377 337 246,305 88.95% 328 287 338,137 87.14% 383 294 1,844,359 76.42%
Missouri 358 302 471,719 82.66% 381 292 655,369 76.22% 444 323 3,882,703 71.61%
Montana 394 314 74,018 79.63% 397 309 110,155 77.44% 386 287 666,296 73.30%
Nebraska 390 321 148,681 80.79% 371 309 208,331 82.84% 385 280 1,167,387 71.59%
Nevada 355 310 221,435 88.57% 351 314 286,394 87.34% 431 308 1,804,427 70.98%
New Hampshire 393 304 100,312 76.63% 414 319 140,525 77.94% 436 330 897,067 75.64%
New Jersey 380 293 703,594 78.88% 404 313 887,966 77.36% 454 307 5,885,384 66.32%
New Mexico 340 297 167,385 87.52% 378 297 229,365 77.50% 450 328 1,310,813 71.52%
New York 1,685 1,303 1,446,714 77.33% 1,649 1,136 2,239,850 68.87% 1,914 1,198 12,932,918 61.18%
North Carolina 310 266 768,619 87.00% 368 290 1,050,264 77.57% 425 324 6,295,258 74.28%
North Dakota 368 297 50,250 78.97% 402 315 99,046 78.91% 487 333 444,691 65.58%
Ohio 1,542 1,220 924,863 78.72% 1,525 1,173 1,238,671 78.36% 1,667 1,175 7,514,424 68.82%
Oklahoma 423 346 308,182 82.96% 412 319 428,032 77.07% 415 285 2,394,443 65.70%
Oregon 321 263 291,705 80.87% 361 289 413,732 79.98% 411 309 2,622,480 75.89%
Pennsylvania 1,383 1,146 945,209 82.78% 1,575 1,220 1,391,012 77.81% 1,802 1,297 8,472,657 71.23%
Rhode Island 372 312 75,840 84.51% 360 289 131,461 79.12% 435 303 690,001 69.39%
South Carolina 392 319 360,578 80.86% 345 285 522,722 82.89% 397 304 3,069,164 74.75%
South Dakota 359 304 65,259 84.23% 361 286 93,194 78.68% 386 299 526,659 75.61%
Tennessee 371 317 505,527 85.19% 359 292 697,396 81.65% 391 285 4,205,059 70.31%
Texas 1,404 1,139 2,311,623 80.63% 1,588 1,219 2,985,606 76.39% 1,751 1,246 15,925,876 70.06%
Utah 371 318 279,317 86.38% 419 340 370,856 81.41% 360 272 1,608,388 71.37%
Vermont 336 274 44,641 81.36% 374 300 73,683 80.65% 405 301 424,193 75.81%
Virginia 394 331 620,869 85.27% 322 247 895,156 79.29% 432 324 5,287,483 74.84%
Washington 353 297 530,892 85.62% 365 289 738,379 78.95% 457 314 4,528,373 68.85%
West Virginia 405 318 130,210 78.65% 322 255 190,624 79.31% 452 343 1,253,658 75.55%
Wisconsin 338 275 448,884 80.11% 448 327 618,657 71.94% 359 265 3,744,210 73.13%
Wyoming 421 341 43,892 80.89% 352 269 64,129 78.38% 403 318 371,258 78.50%
Table C.5 – 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 67,901 265,122,865 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 4,426,093 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% 1,334 991 902,080 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% 1,246 955 480,520 74.19% 66.10%
Table C.6 – 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 34,934,626 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% 272 231 339,299 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.7 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2012 and 2013
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 2012-2013 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 2012 and 2013 individual response rates. The 2012-2013 population estimate is the average of the 2012 and the 2013 population.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2012 and 2013.
Total U.S. 441,349 368,653 314,198 85.00% 176,398 136,147 261,224,390 72.37% 61.51%
Northeast 99,075 84,018 67,655 79.23% 36,635 27,434 47,281,597 69.17% 54.80%
Midwest 120,239 101,287 87,390 86.65% 49,341 37,964 56,069,675 72.91% 63.17%
South 136,077 111,573 96,782 86.90% 53,037 41,668 96,943,079 73.77% 64.11%
West 85,958 71,775 62,371 84.88% 37,385 29,081 60,930,039 72.12% 61.22%
Alabama 6,122 4,894 4,282 87.14% 2,301 1,801 4,015,238 71.81% 62.58%
Alaska 5,601 4,216 3,686 87.42% 2,198 1,692 577,228 74.10% 64.78%
Arizona 5,784 4,467 3,919 87.78% 2,309 1,804 5,403,101 73.26% 64.31%
Arkansas 5,497 4,481 4,074 90.79% 2,405 1,821 2,429,054 71.52% 64.93%
California 19,483 17,279 14,063 81.35% 9,643 7,337 31,581,986 70.32% 57.21%
Colorado 5,861 5,015 4,217 84.13% 2,361 1,812 4,299,874 73.07% 61.47%
Connecticut 5,844 5,226 4,401 84.03% 2,459 1,857 3,039,935 71.34% 59.95%
Delaware 5,889 4,777 4,081 85.61% 2,223 1,755 770,186 76.17% 65.21%
District of Columbia 10,521 8,658 7,027 80.87% 2,267 1,869 549,981 78.06% 63.13%
Florida 26,942 21,111 17,692 82.98% 9,371 7,193 16,491,099 71.12% 59.02%
Georgia 5,025 4,260 3,632 85.39% 2,237 1,737 8,087,248 73.06% 62.38%
Hawaii 6,506 5,622 4,474 79.11% 2,525 1,862 1,133,370 67.89% 53.71%
Idaho 4,688 3,959 3,684 93.03% 2,299 1,828 1,297,052 76.96% 71.59%
Illinois 23,152 20,343 15,590 76.61% 9,806 7,175 10,697,218 68.46% 52.45%
Indiana 5,483 4,623 4,103 88.87% 2,336 1,805 5,411,173 72.26% 64.22%
Iowa 5,229 4,517 4,142 91.59% 2,301 1,800 2,558,825 73.00% 66.86%
Kansas 5,206 4,389 3,921 89.29% 2,274 1,799 2,340,109 75.50% 67.41%
Kentucky 5,937 4,963 4,543 91.49% 2,344 1,831 3,620,332 73.50% 67.25%
Louisiana 5,618 4,464 4,073 91.33% 2,260 1,804 3,759,825 75.38% 68.85%
Maine 7,490 5,566 5,029 90.29% 2,259 1,864 1,146,775 78.71% 71.07%
Maryland 5,439 4,738 3,721 78.66% 2,257 1,799 4,926,434 76.39% 60.09%
Massachusetts 6,071 5,345 4,397 82.04% 2,493 1,852 5,686,563 70.51% 57.85%
Michigan 23,521 19,145 16,136 84.20% 9,322 7,291 8,332,687 74.28% 62.54%
Minnesota 5,078 4,432 4,031 91.15% 2,218 1,808 4,490,191 79.20% 72.19%
Mississippi 4,994 4,106 3,780 92.05% 2,188 1,819 2,424,306 78.93% 72.65%
Missouri 6,023 4,995 4,518 90.41% 2,332 1,832 4,997,678 73.77% 66.69%
Montana 6,286 5,039 4,666 92.58% 2,286 1,786 846,239 75.85% 70.22%
Nebraska 5,608 4,675 4,297 91.86% 2,316 1,850 1,517,851 73.67% 67.68%
Nevada 5,107 4,164 3,725 89.51% 2,271 1,835 2,295,456 75.13% 67.25%
New Hampshire 6,478 5,426 4,689 86.38% 2,502 1,903 1,135,783 74.56% 64.40%
New Jersey 5,786 5,001 4,216 84.47% 2,393 1,811 7,458,969 71.25% 60.18%
New Mexico 5,639 4,306 3,927 91.19% 2,269 1,801 1,705,115 74.00% 67.48%
New York 29,704 25,539 18,358 71.58% 10,515 7,317 16,575,744 64.02% 45.83%
North Carolina 5,720 4,628 4,080 88.05% 2,220 1,797 8,060,735 75.70% 66.66%
North Dakota 7,008 5,400 5,023 92.99% 2,413 1,840 585,756 71.05% 66.07%
Ohio 23,262 19,946 17,473 87.57% 9,561 7,255 9,658,305 71.87% 62.94%
Oklahoma 5,790 4,708 4,273 90.80% 2,439 1,858 3,114,952 70.60% 64.10%
Oregon 5,317 4,708 4,172 88.54% 2,258 1,784 3,310,508 76.65% 67.87%
Pennsylvania 25,199 21,746 17,666 81.03% 9,465 7,243 10,799,456 71.88% 58.25%
Rhode Island 5,589 4,705 4,162 88.48% 2,298 1,827 896,323 74.83% 66.20%
South Carolina 6,597 5,429 4,682 86.09% 2,305 1,846 3,926,252 75.78% 65.23%
South Dakota 5,364 4,367 4,090 93.65% 2,219 1,767 680,698 76.45% 71.60%
Tennessee 5,499 4,526 4,081 90.21% 2,226 1,821 5,385,528 77.01% 69.47%
Texas 18,371 15,538 13,665 87.82% 9,355 7,229 21,037,974 72.71% 63.86%
Utah 3,825 3,329 3,152 94.88% 2,249 1,856 2,236,456 79.22% 75.17%
Vermont 6,914 5,464 4,737 86.66% 2,251 1,760 542,049 75.33% 65.28%
Virginia 5,368 4,706 4,099 86.82% 2,243 1,796 6,769,603 76.51% 66.43%
Washington 5,298 4,541 4,015 88.37% 2,393 1,828 5,766,890 71.69% 63.35%
West Virginia 6,748 5,586 4,997 89.36% 2,396 1,892 1,574,332 75.15% 67.15%
Wisconsin 5,305 4,455 4,066 91.41% 2,243 1,742 4,799,184 74.64% 68.23%
Wyoming 6,563 5,130 4,671 91.06% 2,324 1,856 476,764 78.09% 71.10%
Table C.8 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2012 and 2013
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 2012-2013 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 2012 and 2013 individual response rates. The 2012-2013 population estimate is the average of the 2012 and the 2013 population.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2012 and 2013.
Total U.S. 54,777 45,024 24,912,835 82.39% 57,560 45,220 34,687,727 78.30% 64,061 45,903 201,623,828 70.10%
Northeast 11,213 8,982 4,212,368 79.60% 12,029 9,185 6,151,259 75.38% 13,393 9,267 36,917,970 66.93%
Midwest 15,463 12,619 5,407,088 81.81% 16,127 12,598 7,384,188 77.93% 17,751 12,747 43,278,398 70.94%
South 16,660 13,877 9,330,852 83.02% 17,149 13,774 12,808,148 80.11% 19,228 14,017 74,804,079 71.51%
West 11,441 9,546 5,962,527 83.91% 12,255 9,663 8,344,131 77.98% 13,689 9,872 46,623,382 69.58%
Alabama 723 600 383,469 81.47% 760 616 536,933 79.81% 818 585 3,094,837 69.18%
Alaska 668 509 60,264 76.22% 728 587 82,442 80.12% 802 596 434,523 72.75%
Arizona 762 635 540,502 83.49% 756 586 720,761 75.65% 791 583 4,141,839 71.47%
Arkansas 721 567 236,508 78.18% 858 660 318,730 75.95% 826 594 1,873,816 69.89%
California 2,899 2,422 3,117,442 83.51% 3,155 2,452 4,458,805 77.62% 3,589 2,463 24,005,740 67.24%
Colorado 698 578 402,137 83.52% 789 605 565,276 76.77% 874 629 3,332,461 71.26%
Connecticut 752 604 288,704 81.26% 777 610 376,034 79.25% 930 643 2,375,197 68.98%
Delaware 710 588 68,333 82.31% 701 555 102,080 81.10% 812 612 599,773 74.72%
District of Columbia 736 656 30,856 90.16% 702 581 94,677 83.79% 829 632 424,447 75.93%
Florida 2,826 2,349 1,385,416 83.15% 3,048 2,406 1,972,330 78.53% 3,497 2,438 13,133,353 68.77%
Georgia 702 578 831,609 82.00% 744 590 1,100,053 79.50% 791 569 6,155,586 70.62%
Hawaii 745 590 97,086 78.59% 799 629 140,225 78.05% 981 643 896,059 65.33%
Idaho 726 625 140,843 86.67% 763 603 173,004 81.11% 810 600 983,206 74.62%
Illinois 2,977 2,379 1,045,769 80.56% 3,223 2,391 1,394,500 74.06% 3,606 2,405 8,256,950 66.00%
Indiana 696 563 541,016 80.14% 773 616 734,767 78.96% 867 626 4,135,390 70.17%
Iowa 730 601 241,811 80.67% 757 602 349,004 79.74% 814 597 1,968,010 70.80%
Kansas 757 639 237,185 84.32% 704 560 323,430 81.02% 813 600 1,779,494 73.33%
Kentucky 750 618 339,960 82.10% 745 598 464,737 80.80% 849 615 2,815,636 71.22%
Louisiana 700 589 367,827 83.68% 704 579 521,917 81.23% 856 636 2,870,080 73.33%
Maine 749 633 94,988 84.06% 748 631 128,694 84.38% 762 600 923,093 77.36%
Maryland 705 584 457,152 83.28% 752 612 631,368 79.76% 800 603 3,837,914 74.99%
Massachusetts 750 594 491,273 78.88% 835 623 775,063 75.12% 908 635 4,420,226 68.78%
Michigan 2,933 2,372 805,764 80.89% 3,058 2,451 1,107,310 79.91% 3,331 2,468 6,419,614 72.46%
Minnesota 698 611 424,639 88.46% 730 579 571,439 78.02% 790 618 3,494,114 78.30%
Mississippi 761 650 247,257 84.79% 666 584 337,204 87.67% 761 585 1,839,846 76.40%
Missouri 725 614 472,889 84.31% 737 582 655,094 79.29% 870 636 3,869,695 71.57%
Montana 782 630 73,896 80.70% 747 588 108,999 77.95% 757 568 663,344 74.89%
Nebraska 712 599 148,030 83.78% 804 674 207,051 83.81% 800 577 1,162,770 70.47%
Nevada 688 600 221,167 87.57% 719 603 285,463 83.25% 864 632 1,788,826 72.37%
New Hampshire 798 609 101,207 76.06% 831 643 140,004 78.44% 873 651 894,572 73.74%
New Jersey 729 584 706,126 81.00% 782 605 884,775 77.80% 882 622 5,868,068 69.01%
New Mexico 672 587 168,112 87.37% 747 600 228,037 79.38% 850 614 1,308,966 71.35%
New York 3,249 2,496 1,456,617 76.58% 3,427 2,402 2,243,318 70.34% 3,839 2,419 12,875,810 61.47%
North Carolina 664 564 764,610 85.24% 750 627 1,041,859 82.67% 806 606 6,254,266 73.32%
North Dakota 739 606 49,581 81.26% 741 583 96,345 79.36% 933 651 439,830 68.18%
Ohio 3,170 2,517 925,827 79.23% 3,000 2,321 1,235,683 78.07% 3,391 2,417 7,496,795 69.92%
Oklahoma 808 649 306,820 80.51% 795 616 426,492 76.97% 836 593 2,381,640 68.19%
Oregon 632 533 292,050 83.94% 768 607 411,744 79.54% 858 644 2,606,713 75.45%
Pennsylvania 2,808 2,315 951,880 82.46% 3,111 2,438 1,397,926 78.78% 3,546 2,490 8,449,649 69.54%
Rhode Island 692 588 76,543 85.46% 751 618 132,076 81.81% 855 621 687,705 72.40%
South Carolina 777 636 359,524 81.22% 694 580 519,243 83.75% 834 630 3,047,485 73.75%
South Dakota 675 569 64,901 84.17% 732 586 92,359 80.75% 812 612 523,437 74.88%
Tennessee 670 578 505,317 85.57% 778 644 692,825 82.46% 778 599 4,187,386 75.12%
Texas 2,876 2,385 2,295,567 82.49% 3,059 2,402 2,964,444 78.37% 3,420 2,442 15,777,963 70.25%
Utah 690 605 275,660 88.43% 803 650 367,327 81.59% 756 601 1,593,468 76.95%
Vermont 686 559 45,030 80.94% 767 615 73,369 80.79% 798 586 423,650 73.81%
Virginia 767 653 619,956 85.16% 638 517 893,349 82.17% 838 626 5,256,298 74.38%
Washington 721 598 529,852 83.62% 771 599 738,145 77.21% 901 631 4,498,893 69.56%
West Virginia 764 633 130,671 83.20% 755 607 189,908 80.33% 877 652 1,253,753 73.53%
Wisconsin 651 549 449,677 83.42% 868 653 617,207 74.83% 724 540 3,732,300 73.51%
Wyoming 758 634 43,516 83.37% 710 554 63,905 78.47% 856 668 369,343 77.41%
Table C.9 – Sample Sizes, Weighted Screening and Interview Response Rates, and Population Estimates, by State, for Individuals Aged 12 or Older: 2013 and 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.
NOTE: To compute the pooled 2013-2014 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 2013 and 2014 individual response rates. The 2013-2014 population estimate is the average of the 2013 and the 2014 population.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2013 and 2014.
Total U.S. 412,088 344,600 287,930 82.92% 180,382 135,739 263,757,160 71.44% 59.24%
Northeast 91,979 77,673 61,531 77.56% 36,509 26,660 47,510,090 68.14% 52.85%
Midwest 104,386 87,601 74,569 84.64% 46,365 34,647 56,338,455 71.36% 60.39%
South 131,479 108,446 92,076 85.12% 56,950 43,563 98,178,474 72.88% 62.04%
West 84,244 70,880 59,754 81.92% 40,558 30,869 61,730,140 71.77% 58.80%
Alabama 5,750 4,605 3,871 83.48% 2,428 1,864 4,033,842 70.62% 58.95%
Alaska 6,162 4,693 3,994 85.07% 2,508 1,810 578,933 71.30% 60.65%
Arizona 5,527 4,236 3,650 86.20% 2,439 1,853 5,494,617 72.15% 62.20%
Arkansas 5,395 4,392 3,930 89.35% 2,455 1,872 2,439,409 72.94% 65.17%
California 20,233 18,168 14,294 78.27% 11,267 8,393 31,970,791 70.13% 54.89%
Colorado 5,397 4,690 3,859 82.35% 2,530 1,893 4,382,715 72.05% 59.33%
Connecticut 5,779 5,175 4,291 82.79% 2,636 1,873 3,050,288 67.55% 55.92%
Delaware 5,814 4,886 3,928 80.36% 2,377 1,813 779,378 72.94% 58.62%
District of Columbia 9,796 8,260 6,502 78.17% 2,361 1,842 559,703 74.11% 57.93%
Florida 24,443 19,278 15,999 81.93% 9,177 6,980 16,757,959 70.99% 58.16%
Georgia 6,353 5,307 4,403 82.83% 3,122 2,401 8,187,094 73.74% 61.08%
Hawaii 6,236 5,330 4,169 77.63% 2,579 1,892 1,142,582 69.15% 53.68%
Idaho 4,320 3,710 3,340 89.78% 2,430 1,894 1,315,995 75.60% 67.87%
Illinois 18,671 16,245 12,319 75.59% 8,423 5,900 10,726,071 66.62% 50.35%
Indiana 5,496 4,591 3,964 86.19% 2,459 1,861 5,445,535 71.89% 61.97%
Iowa 5,196 4,419 3,971 89.78% 2,404 1,812 2,574,919 71.43% 64.13%
Kansas 4,912 4,181 3,649 87.08% 2,461 1,869 2,350,428 73.49% 64.00%
Kentucky 5,641 4,636 4,168 89.66% 2,444 1,850 3,643,187 71.36% 63.98%
Louisiana 5,312 4,308 3,838 88.65% 2,462 1,895 3,786,568 73.39% 65.06%
Maine 6,966 5,072 4,550 89.55% 2,355 1,866 1,149,510 76.78% 68.75%
Maryland 5,242 4,681 3,676 78.18% 2,480 1,896 4,967,852 74.52% 58.26%
Massachusetts 5,955 5,233 4,257 81.16% 2,677 1,897 5,740,609 67.87% 55.08%
Michigan 18,689 15,342 12,808 83.35% 7,985 6,054 8,359,339 71.83% 59.87%
Minnesota 4,970 4,383 3,881 88.66% 2,392 1,873 4,526,990 76.40% 67.74%
Mississippi 4,640 3,733 3,327 88.96% 2,258 1,827 2,433,807 77.78% 69.19%
Missouri 5,722 4,702 4,169 88.41% 2,401 1,851 5,021,862 74.42% 65.79%
Montana 5,820 4,699 4,287 91.06% 2,464 1,887 854,187 73.49% 66.92%
Nebraska 5,511 4,602 4,121 89.28% 2,414 1,848 1,530,287 73.85% 65.94%
Nevada 5,174 4,332 3,596 82.06% 2,416 1,893 2,336,081 73.68% 60.46%
New Hampshire 6,532 5,358 4,553 84.88% 2,531 1,885 1,141,071 72.29% 61.36%
New Jersey 7,567 6,519 5,232 80.70% 3,405 2,449 7,499,719 69.30% 55.92%
New Mexico 5,181 4,000 3,593 89.62% 2,340 1,881 1,710,041 77.04% 69.04%
New York 26,220 22,554 15,846 69.96% 10,083 6,921 16,667,826 63.90% 44.71%
North Carolina 7,057 5,825 5,062 86.93% 3,059 2,413 8,165,327 76.26% 66.30%
North Dakota 6,677 5,130 4,698 91.45% 2,497 1,914 599,990 72.95% 66.71%
Ohio 17,862 15,131 12,981 85.53% 8,071 5,983 9,692,251 70.40% 60.21%
Oklahoma 5,089 4,154 3,709 89.32% 2,534 1,887 3,143,373 68.68% 61.35%
Oregon 5,299 4,665 4,030 86.40% 2,411 1,853 3,346,707 74.87% 64.69%
Pennsylvania 20,393 17,518 14,088 80.26% 7,946 6,051 10,818,453 71.95% 57.75%
Rhode Island 5,650 4,766 4,064 85.14% 2,501 1,895 899,690 72.04% 61.34%
South Carolina 6,134 5,070 4,266 84.02% 2,442 1,906 3,980,592 75.79% 63.68%
South Dakota 4,891 3,983 3,738 93.89% 2,381 1,870 688,348 75.92% 71.28%
Tennessee 5,293 4,370 3,828 87.37% 2,325 1,840 5,433,594 75.87% 66.29%
Texas 16,327 13,744 11,939 86.84% 9,324 6,987 21,456,935 71.22% 61.84%
Utah 3,566 3,115 2,953 94.96% 2,336 1,902 2,279,009 77.97% 74.04%
Vermont 6,917 5,478 4,650 84.72% 2,375 1,823 542,924 75.24% 63.74%
Virginia 6,463 5,674 4,750 83.72% 3,168 2,441 6,836,908 74.75% 62.58%
Washington 5,047 4,408 3,642 82.59% 2,416 1,835 5,838,584 72.75% 60.09%
West Virginia 6,730 5,523 4,880 88.41% 2,534 1,849 1,572,945 71.93% 63.59%
Wisconsin 5,789 4,892 4,270 87.10% 2,477 1,812 4,822,436 71.62% 62.38%
Wyoming 6,282 4,834 4,347 89.73% 2,422 1,883 479,899 76.43% 68.58%
Table C.10 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates, by State and Three Age Groups: 2013 and 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.
NOTE: To compute the pooled 2013-2014 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 2013 and 2014 individual response rates. The 2013-2014 population estimate is the average of the 2013 and the 2014 population.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2013 and 2014.
Total U.S. 49,022 39,578 24,883,686 80.99% 50,647 39,028 34,860,063 76.61% 80,713 57,133 204,013,411 69.39%
Northeast 9,905 7,837 4,171,861 78.55% 10,119 7,582 6,149,607 72.98% 16,485 11,241 37,188,621 66.16%
Midwest 12,719 10,139 5,384,865 79.29% 13,379 10,148 7,417,058 74.82% 20,267 14,360 43,536,532 69.79%
South 15,578 12,728 9,383,697 81.75% 15,690 12,384 12,900,076 78.95% 25,682 18,451 75,894,702 70.74%
West 10,820 8,874 5,943,263 83.02% 11,459 8,914 8,393,322 77.24% 18,279 13,081 47,393,555 69.39%
Alabama 663 553 382,134 83.41% 668 540 535,409 79.84% 1,097 771 3,116,298 67.52%
Alaska 729 529 59,900 71.89% 694 523 83,456 73.24% 1,085 758 435,577 70.87%
Arizona 666 553 543,484 83.67% 696 537 732,863 77.21% 1,077 763 4,218,271 69.76%
Arkansas 635 504 236,666 78.38% 711 561 319,372 78.89% 1,109 807 1,883,371 71.31%
California 2,863 2,378 3,080,548 83.11% 3,102 2,387 4,469,106 76.62% 5,302 3,628 24,421,137 67.30%
Colorado 644 515 408,429 80.29% 808 615 575,557 76.15% 1,078 763 3,398,728 70.34%
Connecticut 726 572 286,281 80.43% 657 490 381,473 73.50% 1,253 811 2,382,534 65.04%
Delaware 664 545 67,991 80.28% 698 542 101,239 78.97% 1,015 726 610,148 71.09%
District of Columbia 647 560 30,551 87.12% 593 472 93,509 80.69% 1,121 810 435,643 71.78%
Florida 2,467 2,025 1,390,131 82.62% 2,575 2,031 1,980,707 78.66% 4,135 2,924 13,387,121 68.71%
Georgia 821 658 838,199 80.33% 927 744 1,108,195 80.23% 1,374 999 6,240,700 71.56%
Hawaii 680 555 96,971 81.50% 715 534 140,686 73.43% 1,184 803 904,925 67.25%
Idaho 613 513 142,945 84.55% 756 587 173,361 78.38% 1,061 794 999,689 73.87%
Illinois 2,209 1,703 1,033,794 76.86% 2,463 1,762 1,394,857 71.74% 3,751 2,435 8,297,420 64.55%
Indiana 680 541 541,174 79.19% 666 517 740,165 76.12% 1,113 803 4,164,196 70.23%
Iowa 625 490 242,393 77.21% 726 571 352,842 79.35% 1,053 751 1,979,684 69.23%
Kansas 644 509 237,609 79.26% 733 575 325,999 79.43% 1,084 785 1,786,820 71.67%
Kentucky 685 557 340,101 81.45% 689 539 470,972 78.33% 1,070 754 2,832,115 68.90%
Louisiana 682 552 367,862 79.96% 693 546 519,036 77.25% 1,087 797 2,899,670 71.93%
Maine 648 524 93,811 79.34% 639 531 127,380 82.41% 1,068 811 928,319 75.81%
Maryland 705 564 455,684 80.19% 686 535 629,854 76.02% 1,089 797 3,882,314 73.58%
Massachusetts 708 553 488,765 77.37% 802 584 782,118 72.88% 1,167 760 4,469,726 65.98%
Michigan 2,257 1,791 797,647 78.27% 2,280 1,778 1,114,774 76.55% 3,448 2,485 6,446,918 70.25%
Minnesota 644 539 425,247 84.20% 728 558 571,816 76.50% 1,020 776 3,529,926 75.52%
Mississippi 639 553 245,600 85.82% 600 518 338,718 86.22% 1,019 756 1,849,489 75.13%
Missouri 654 541 470,976 82.48% 663 500 656,394 75.21% 1,084 810 3,894,492 73.34%
Montana 678 536 74,121 79.66% 720 574 110,655 78.84% 1,066 777 669,411 71.82%
Nebraska 696 563 149,327 80.04% 667 528 209,508 78.45% 1,051 757 1,171,452 72.08%
Nevada 625 534 221,704 86.30% 669 554 287,435 80.89% 1,122 805 1,826,942 71.01%
New Hampshire 731 562 99,717 76.82% 708 553 141,165 79.27% 1,092 770 900,190 70.65%
New Jersey 897 684 701,644 77.06% 937 701 890,874 75.11% 1,571 1,064 5,907,201 67.49%
New Mexico 648 556 166,639 86.54% 640 517 228,647 80.87% 1,052 808 1,314,755 75.15%
New York 2,745 2,120 1,440,280 76.57% 2,726 1,873 2,239,134 67.62% 4,612 2,928 12,988,411 61.79%
North Carolina 771 646 771,607 84.54% 863 681 1,054,654 78.99% 1,425 1,086 6,339,066 74.77%
North Dakota 649 525 50,733 80.11% 743 586 100,601 78.86% 1,105 803 448,656 70.85%
Ohio 2,306 1,828 922,292 79.04% 2,302 1,723 1,235,723 74.20% 3,463 2,432 7,534,236 68.71%
Oklahoma 688 544 309,426 76.31% 710 554 429,192 77.37% 1,136 789 2,404,755 66.18%
Oregon 673 547 291,323 81.68% 695 531 413,626 75.68% 1,043 775 2,641,759 74.00%
Pennsylvania 2,121 1,754 941,238 82.66% 2,335 1,818 1,382,616 77.82% 3,490 2,479 8,494,600 69.81%
Rhode Island 697 562 75,717 79.86% 648 507 131,028 77.69% 1,156 826 692,945 70.11%
South Carolina 687 558 362,044 81.55% 649 530 521,862 82.47% 1,106 818 3,096,686 74.02%
South Dakota 659 555 65,627 83.65% 665 523 93,403 78.92% 1,057 792 529,318 74.53%
Tennessee 666 555 506,479 82.84% 592 480 700,245 82.17% 1,067 805 4,226,870 74.05%
Texas 2,541 2,068 2,327,085 81.29% 2,609 2,010 3,010,183 77.38% 4,174 2,909 16,119,667 68.63%
Utah 651 560 282,277 86.82% 671 557 372,803 83.11% 1,014 785 1,623,929 75.24%
Vermont 632 506 44,408 80.00% 667 525 73,820 79.15% 1,076 792 424,696 74.09%
Virginia 870 722 622,264 84.16% 818 645 896,567 80.03% 1,480 1,074 5,318,077 72.63%
Washington 625 511 530,795 82.05% 657 513 741,218 77.88% 1,134 811 4,566,571 70.86%
West Virginia 747 564 129,873 75.44% 609 456 190,362 74.84% 1,178 829 1,252,711 71.12%
Wisconsin 696 554 448,046 79.56% 743 527 620,976 68.68% 1,038 731 3,753,413 71.10%
Wyoming 725 587 44,128 80.13% 636 485 63,910 77.28% 1,061 811 371,861 75.86%
Table C.11 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates among Individuals Aged 12 to 20, by State: 2012, 2013, and 2014
State 2012
Total
Selected
2012
Total
Responded
2012
Population
Estimate
2012
Weighted
Interview
Response
Rate
2013
Total
Selected
2013
Total
Responded
2013
Population
Estimate
2013
Weighted
Interview
Response
Rate
2014
Total
Selected
2014
Total
Responded
2014
Population
Estimate
2014
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, 2012, 2013, and 2014.
Total U.S. 37,391 30,912 38,205,953 82.59% 37,820 30,801 38,086,579 81.70% 28,949 23,033 37,981,012 79.64%
Northeast 7,735 6,239 6,646,927 80.21% 7,770 6,238 6,379,509 79.42% 5,713 4,457 6,502,814 77.38%
Midwest 10,454 8,616 8,152,530 82.67% 10,686 8,592 8,217,933 80.04% 6,763 5,275 8,114,553 77.28%
South 11,385 9,547 14,063,463 83.57% 11,306 9,274 14,070,964 81.83% 9,646 7,800 14,076,323 81.03%
West 7,817 6,510 9,343,033 82.70% 8,058 6,697 9,418,173 84.56% 6,827 5,501 9,287,322 81.13%
Alabama 469 384 584,363 81.07% 497 421 570,714 82.97% 375 306 564,703 83.74%
Alaska 441 352 95,819 80.24% 490 383 91,357 77.84% 467 330 91,021 69.24%
Arizona 503 424 816,941 83.45% 526 428 816,730 81.20% 375 308 796,228 82.79%
Arkansas 550 439 370,165 79.62% 457 357 334,342 77.85% 405 328 352,450 79.72%
California 2,016 1,646 5,018,845 81.44% 2,070 1,767 5,008,517 85.96% 1,941 1,570 4,913,481 80.22%
Colorado 501 421 594,406 85.04% 450 367 609,754 82.09% 457 365 626,186 80.80%
Connecticut 520 427 455,720 82.40% 534 431 421,506 81.80% 449 343 438,741 77.16%
Delaware 493 407 107,644 84.15% 460 379 99,907 80.87% 444 358 108,885 80.32%
District of Columbia 498 451 64,190 91.18% 452 387 54,486 84.22% 342 295 52,520 87.27%
Florida 1,980 1,649 2,109,563 82.68% 1,929 1,574 2,127,386 81.54% 1,390 1,140 2,041,554 82.35%
Georgia 478 397 1,309,366 82.78% 502 405 1,278,777 81.65% 631 506 1,218,390 79.90%
Hawaii 500 388 145,487 78.38% 508 416 146,388 80.45% 398 317 146,275 81.78%
Idaho 515 441 206,195 85.69% 483 398 202,212 84.41% 403 329 217,741 80.74%
Illinois 2,036 1,637 1,553,772 80.89% 2,048 1,582 1,571,014 77.50% 1,016 766 1,561,804 75.84%
Indiana 480 393 813,060 81.75% 490 392 794,141 77.86% 420 327 810,033 77.67%
Iowa 485 404 353,403 82.15% 484 396 365,893 81.48% 395 305 406,568 77.47%
Kansas 508 443 380,034 86.86% 499 404 360,191 81.57% 391 307 341,647 78.63%
Kentucky 511 422 505,420 82.23% 491 400 507,396 81.31% 439 354 536,524 80.24%
Louisiana 451 395 552,954 87.18% 487 399 574,885 80.70% 457 379 597,123 82.46%
Maine 504 433 145,895 86.56% 523 448 146,805 85.44% 365 281 140,376 76.29%
Maryland 438 372 655,351 84.43% 505 403 653,828 79.02% 434 343 684,058 77.90%
Massachusetts 520 420 763,162 80.74% 499 385 723,842 76.61% 489 395 859,796 80.58%
Michigan 1,992 1,638 1,251,079 82.84% 2,054 1,654 1,239,358 80.23% 1,015 786 1,180,278 76.23%
Minnesota 471 411 629,891 86.19% 456 393 626,747 86.71% 423 341 647,983 81.36%
Mississippi 517 426 376,196 82.30% 493 437 363,901 88.44% 357 302 379,058 85.68%
Missouri 486 407 700,548 84.33% 493 412 714,528 81.35% 379 304 694,435 81.24%
Montana 522 431 123,289 83.41% 550 440 120,530 79.55% 385 305 109,111 80.01%
Nebraska 475 413 228,674 87.51% 539 452 240,691 82.96% 405 315 217,731 77.42%
Nevada 474 403 339,091 85.10% 486 431 343,860 89.80% 386 320 336,291 83.66%
New Hampshire 599 472 181,715 80.39% 556 444 173,109 80.69% 442 335 143,093 76.34%
New Jersey 475 389 1,041,104 81.91% 506 400 1,028,297 80.63% 721 548 1,062,607 75.32%
New Mexico 459 396 247,385 86.18% 477 403 252,940 83.62% 402 340 247,286 86.06%
New York 2,182 1,674 2,352,294 76.70% 2,218 1,701 2,191,460 76.54% 1,399 1,062 2,204,778 74.39%
North Carolina 474 404 1,096,473 85.11% 438 365 1,101,838 83.46% 626 516 1,161,827 83.03%
North Dakota 495 415 90,131 84.87% 497 397 82,751 78.48% 393 319 88,056 81.64%
Ohio 2,134 1,696 1,382,707 79.58% 2,130 1,697 1,449,529 80.13% 1,026 799 1,394,953 77.07%
Oklahoma 523 407 474,162 76.65% 601 482 497,668 80.37% 356 270 451,557 73.23%
Oregon 457 391 462,560 85.86% 458 372 456,806 80.22% 462 369 449,656 81.64%
Pennsylvania 1,980 1,620 1,506,219 82.23% 1,967 1,623 1,484,560 82.08% 1,007 829 1,451,933 81.73%
Rhode Island 460 399 127,152 87.11% 508 430 139,658 85.71% 434 339 129,450 77.79%
South Carolina 496 414 537,771 83.64% 507 411 539,469 81.31% 398 323 542,758 81.86%
South Dakota 444 378 101,364 85.52% 506 425 103,606 82.80% 433 359 109,010 82.94%
Tennessee 439 378 731,381 84.81% 495 425 773,131 85.93% 371 298 768,150 81.06%
Texas 2,002 1,690 3,407,153 84.28% 1,968 1,591 3,455,065 80.21% 1,521 1,223 3,470,196 80.39%
Utah 434 386 396,005 88.78% 511 434 420,269 85.51% 376 327 433,820 87.10%
Vermont 495 405 73,666 81.68% 459 376 70,271 81.84% 407 325 72,041 80.91%
Virginia 484 416 952,855 85.21% 502 421 933,932 85.76% 657 542 947,201 83.34%
Washington 516 419 825,920 81.17% 503 417 880,808 84.01% 385 309 858,442 80.73%
West Virginia 582 496 228,456 84.92% 522 417 204,238 80.45% 443 317 199,369 71.92%
Wisconsin 448 381 667,867 84.74% 490 388 669,485 77.88% 467 347 662,055 72.36%
Wyoming 479 412 71,089 84.48% 546 441 68,002 81.58% 390 312 61,784 78.08%
Table C.12 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates among Individuals Aged 12 to 20, by State: 2012-2013 and 2013-2014
State 2012-2013
Total
Selected
2012-2013
Total
Responded
2012-2013
Population
Estimate
2012-2013
Weighted
Interview
Response
Rate
2013-2014
Total
Selected
2013-2014
Total
Responded
2013-2014
Population
Estimate
2013-2014
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 weighted response rates, the 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 individual response rates. The population estimate is the average of the population across the 2 years.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2012, 2013, and 2014.
Total U.S. 75,211 61,713 38,146,266 82.14% 66,769 53,834 38,033,795 80.67%
Northeast 15,505 12,477 6,513,218 79.82% 13,483 10,695 6,441,162 78.40%
Midwest 21,140 17,208 8,185,232 81.35% 17,449 13,867 8,166,243 78.66%
South 22,691 18,821 14,067,213 82.70% 20,952 17,074 14,073,644 81.43%
West 15,875 13,207 9,380,603 83.62% 14,885 12,198 9,352,747 82.84%
Alabama 966 805 577,538 82.00% 872 727 567,708 83.35%
Alaska 931 735 93,588 79.06% 957 713 91,189 73.59%
Arizona 1,029 852 816,835 82.33% 901 736 806,479 81.99%
Arkansas 1,007 796 352,254 78.75% 862 685 343,396 78.80%
California 4,086 3,413 5,013,681 83.66% 4,011 3,337 4,960,999 83.09%
Colorado 951 788 602,080 83.58% 907 732 617,970 81.42%
Connecticut 1,054 858 438,613 82.09% 983 774 430,123 79.53%
Delaware 953 786 103,775 82.52% 904 737 104,396 80.58%
District of Columbia 950 838 59,338 87.97% 794 682 53,503 85.74%
Florida 3,909 3,223 2,118,475 82.12% 3,319 2,714 2,084,470 81.94%
Georgia 980 802 1,294,072 82.22% 1,133 911 1,248,584 80.79%
Hawaii 1,008 804 145,938 79.41% 906 733 146,331 81.12%
Idaho 998 839 204,204 85.06% 886 727 209,977 82.54%
Illinois 4,084 3,219 1,562,393 79.19% 3,064 2,348 1,566,409 76.68%
Indiana 970 785 803,600 79.84% 910 719 802,087 77.76%
Iowa 969 800 359,648 81.81% 879 701 386,230 79.32%
Kansas 1,007 847 370,112 84.27% 890 711 350,919 80.10%
Kentucky 1,002 822 506,408 81.76% 930 754 521,960 80.76%
Louisiana 938 794 563,920 83.93% 944 778 586,004 81.60%
Maine 1,027 881 146,350 86.01% 888 729 143,591 80.88%
Maryland 943 775 654,590 81.68% 939 746 668,943 78.44%
Massachusetts 1,019 805 743,502 78.72% 988 780 791,819 78.67%
Michigan 4,046 3,292 1,245,219 81.54% 3,069 2,440 1,209,818 78.30%
Minnesota 927 804 628,319 86.44% 879 734 637,365 83.93%
Mississippi 1,010 863 370,048 85.31% 850 739 371,479 87.05%
Missouri 979 819 707,538 82.83% 872 716 704,482 81.30%
Montana 1,072 871 121,910 81.44% 935 745 114,821 79.77%
Nebraska 1,014 865 234,682 85.20% 944 767 229,211 80.23%
Nevada 960 834 341,475 87.42% 872 751 340,076 86.67%
New Hampshire 1,155 916 177,412 80.53% 998 779 158,101 78.58%
New Jersey 981 789 1,034,700 81.28% 1,227 948 1,045,452 77.94%
New Mexico 936 799 250,162 84.89% 879 743 250,113 84.84%
New York 4,400 3,375 2,271,877 76.62% 3,617 2,763 2,198,119 75.46%
North Carolina 912 769 1,099,156 84.28% 1,064 881 1,131,833 83.24%
North Dakota 992 812 86,441 81.66% 890 716 85,403 80.09%
Ohio 4,264 3,393 1,416,118 79.86% 3,156 2,496 1,422,241 78.63%
Oklahoma 1,124 889 485,915 78.58% 957 752 474,613 77.05%
Oregon 915 763 459,683 83.01% 920 741 453,231 80.91%
Pennsylvania 3,947 3,243 1,495,389 82.16% 2,974 2,452 1,468,246 81.91%
Rhode Island 968 829 133,405 86.40% 942 769 134,554 81.79%
South Carolina 1,003 825 538,620 82.45% 905 734 541,114 81.59%
South Dakota 950 803 102,485 84.13% 939 784 106,308 82.87%
Tennessee 934 803 752,256 85.38% 866 723 770,640 83.48%
Texas 3,970 3,281 3,431,109 82.24% 3,489 2,814 3,462,630 80.30%
Utah 945 820 408,137 87.10% 887 761 427,044 86.31%
Vermont 954 781 71,968 81.76% 866 701 71,156 81.37%
Virginia 986 837 943,393 85.48% 1,159 963 940,566 84.53%
Washington 1,019 836 853,364 82.64% 888 726 869,625 82.38%
West Virginia 1,104 913 216,347 82.84% 965 734 201,803 76.23%
Wisconsin 938 769 668,676 81.25% 957 735 665,770 75.06%
Wyoming 1,025 853 69,545 83.04% 936 753 64,893 79.88%
Table C.13 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates among Adults Aged 18 or Older, by State: 2012, 2013, and 2014
State 2012
Total
Selected
2012
Total
Responded
2012
Population
Estimate
2012
Weighted
Interview
Response
Rate
2013
Total
Selected
2013
Total
Responded
2013
Population
Estimate
2013
Weighted
Interview
Response
Rate
2014
Total
Selected
2014
Total
Responded
2014
Population
Estimate
2014
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, 2012, 2013, and 2014.
Total U.S. 60,509 45,817 235,124,274 72.00% 61,112 45,306 237,498,837 70.61% 70,248 50,855 240,248,111 70.28%
Northeast 12,788 9,352 42,937,539 68.59% 12,634 9,100 43,200,918 67.70% 13,970 9,723 43,475,540 66.57%
Midwest 16,766 12,743 50,508,549 73.29% 17,112 12,602 50,816,624 70.61% 16,534 11,906 51,090,556 70.44%
South 17,987 13,913 87,067,845 73.21% 18,390 13,878 88,156,610 72.35% 22,982 16,957 89,432,946 71.51%
West 12,968 9,809 54,610,340 71.60% 12,976 9,726 55,324,685 70.09% 16,762 12,269 56,249,069 71.05%
Alabama 803 623 3,621,189 73.90% 775 578 3,642,350 67.91% 990 733 3,661,065 70.74%
Alaska 772 596 516,839 73.05% 758 587 517,089 74.74% 1,021 694 520,976 67.87%
Arizona 773 610 4,823,495 76.18% 774 559 4,901,704 67.86% 999 741 5,000,562 73.63%
Arkansas 818 601 2,186,878 68.89% 866 653 2,198,214 72.67% 954 715 2,207,272 72.07%
California 3,370 2,449 28,284,885 68.90% 3,374 2,466 28,644,204 68.82% 5,030 3,549 29,136,282 68.68%
Colorado 812 608 3,861,324 73.85% 851 626 3,934,150 70.24% 1,035 752 4,014,421 72.22%
Connecticut 900 676 2,744,379 71.67% 807 577 2,758,083 68.93% 1,103 724 2,769,930 63.56%
Delaware 734 586 696,760 79.66% 779 581 706,947 71.27% 934 687 715,829 73.17%
District of Columbia 763 633 513,289 79.99% 768 580 524,960 74.63% 946 702 533,345 72.06%
Florida 3,160 2,351 14,999,230 69.34% 3,385 2,493 15,212,136 70.67% 3,325 2,462 15,523,521 69.21%
Georgia 800 598 7,212,572 72.11% 735 561 7,298,705 71.87% 1,566 1,182 7,399,085 73.93%
Hawaii 908 654 1,033,888 68.36% 872 618 1,038,681 65.50% 1,027 719 1,052,542 70.56%
Idaho 747 576 1,148,607 76.93% 826 627 1,163,811 74.54% 991 754 1,182,290 74.54%
Illinois 3,354 2,438 9,628,889 69.74% 3,475 2,358 9,674,009 64.56% 2,739 1,839 9,710,545 66.51%
Indiana 841 640 4,850,837 72.01% 799 602 4,889,478 70.78% 980 718 4,919,244 71.40%
Iowa 764 586 2,309,284 73.90% 807 613 2,324,742 70.53% 972 709 2,340,310 71.09%
Kansas 721 569 2,099,601 76.67% 796 591 2,106,246 72.33% 1,021 769 2,119,391 73.37%
Kentucky 800 609 3,267,986 72.62% 794 604 3,292,759 72.57% 965 689 3,313,413 68.02%
Louisiana 770 609 3,377,799 76.40% 790 606 3,406,196 72.72% 990 737 3,431,217 72.65%
Maine 775 633 1,049,900 78.59% 735 598 1,053,674 77.84% 972 744 1,057,725 75.29%
Maryland 744 592 4,447,458 74.85% 808 623 4,491,106 76.42% 967 709 4,533,230 71.33%
Massachusetts 873 646 5,168,136 70.62% 870 612 5,222,444 68.82% 1,099 732 5,281,244 65.28%
Michigan 3,161 2,477 7,509,825 75.11% 3,228 2,442 7,544,022 72.00% 2,500 1,821 7,579,361 70.38%
Minnesota 729 578 4,046,322 80.26% 791 619 4,084,784 76.42% 957 715 4,118,701 74.87%
Mississippi 716 588 2,171,602 78.33% 711 581 2,182,497 78.14% 908 693 2,193,918 75.62%
Missouri 782 603 4,511,506 73.10% 825 615 4,538,072 72.25% 922 695 4,563,701 74.97%
Montana 721 560 768,234 76.98% 783 596 776,451 73.89% 1,003 755 783,681 71.77%
Nebraska 848 662 1,363,924 71.68% 756 589 1,375,718 73.48% 962 696 1,386,201 72.80%
Nevada 801 613 2,057,758 74.47% 782 622 2,090,821 73.20% 1,009 737 2,137,932 71.60%
New Hampshire 854 645 1,031,559 72.83% 850 649 1,037,592 75.97% 950 674 1,045,117 67.96%
New Jersey 806 607 6,732,336 72.63% 858 620 6,773,350 67.83% 1,650 1,145 6,822,800 69.14%
New Mexico 769 589 1,533,828 72.67% 828 625 1,540,178 72.40% 864 700 1,546,626 79.79%
New York 3,703 2,487 15,065,487 63.25% 3,563 2,334 15,172,768 62.31% 3,775 2,467 15,282,323 63.02%
North Carolina 763 619 7,246,727 74.56% 793 614 7,345,522 74.76% 1,495 1,153 7,441,918 76.00%
North Dakota 785 586 528,614 72.53% 889 648 543,737 67.93% 959 741 554,778 76.94%
Ohio 3,199 2,390 8,711,861 71.96% 3,192 2,348 8,753,095 70.18% 2,573 1,807 8,786,823 68.81%
Oklahoma 804 605 2,793,790 71.76% 827 604 2,822,475 67.42% 1,019 739 2,845,419 68.34%
Oregon 854 653 3,000,702 75.56% 772 598 3,036,213 76.46% 966 708 3,074,556 72.04%
Pennsylvania 3,280 2,411 9,831,482 69.58% 3,377 2,517 9,863,670 72.18% 2,448 1,780 9,890,761 69.72%
Rhode Island 811 647 818,100 76.97% 795 592 821,462 70.88% 1,009 741 826,484 71.83%
South Carolina 786 621 3,541,570 74.46% 742 589 3,591,886 75.96% 1,013 759 3,645,209 74.51%
South Dakota 797 613 611,740 75.34% 747 585 619,853 76.03% 975 730 625,589 74.25%
Tennessee 806 666 4,857,966 80.57% 750 577 4,902,455 71.95% 909 708 4,951,776 78.47%
Texas 3,140 2,379 18,573,333 72.01% 3,339 2,465 18,911,482 71.04% 3,444 2,454 19,348,218 68.93%
Utah 780 639 1,942,347 82.23% 779 612 1,979,244 73.43% 906 730 2,014,221 79.69%
Vermont 786 600 496,163 73.23% 779 601 497,875 76.52% 964 716 499,157 73.19%
Virginia 722 572 6,116,656 75.62% 754 571 6,182,639 75.56% 1,544 1,148 6,246,649 72.13%
Washington 850 627 5,207,324 70.94% 822 603 5,266,752 70.21% 969 721 5,348,826 73.56%
West Virginia 858 661 1,443,040 72.82% 774 598 1,444,283 76.05% 1,013 687 1,441,863 67.30%
Wisconsin 785 601 4,336,147 74.41% 807 592 4,362,867 72.94% 974 666 4,385,912 68.63%
Wyoming 811 635 431,108 76.66% 755 587 435,387 78.48% 942 709 436,156 73.66%
Table C.14 – Sample Sizes, Weighted Interview Response Rates, and Population Estimates among Adults Aged 18 or Older, by State: 2012-2013 and 2013-2014
State 2012-2013
Total
Selected
2012-2013
Total
Responded
2012-2013
Population
Estimate
2012-2013
Weighted
Interview
Response
Rate
2013-2014
Total
Selected
2013-2014
Total
Responded
2013-2014
Population
Estimate
2013-2014
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 weighted response rates, the 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 individual response rates. The population estimate is the average of the population across the 2 years.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2012, 2013, and 2014
Total U.S. 121,621 91,123 236,311,555 71.31% 131,360 96,161 238,873,474 70.45%
Northeast 25,422 18,452 43,069,229 68.15% 26,604 18,823 43,338,229 67.13%
Midwest 33,878 25,345 50,662,587 71.95% 33,646 24,508 50,953,590 70.52%
South 36,377 27,791 87,612,227 72.78% 41,372 30,835 88,794,778 71.93%
West 25,944 19,535 54,967,513 70.85% 29,738 21,995 55,786,877 70.58%
Alabama 1,578 1,201 3,631,769 70.77% 1,765 1,311 3,651,708 69.33%
Alaska 1,530 1,183 516,964 73.87% 1,779 1,281 519,033 71.23%
Arizona 1,547 1,169 4,862,599 72.12% 1,773 1,300 4,951,133 70.86%
Arkansas 1,684 1,254 2,192,546 70.81% 1,820 1,368 2,202,743 72.37%
California 6,744 4,915 28,464,544 68.86% 8,404 6,015 28,890,243 68.74%
Colorado 1,663 1,234 3,897,737 72.04% 1,886 1,378 3,974,285 71.21%
Connecticut 1,707 1,253 2,751,231 70.35% 1,910 1,301 2,764,007 66.24%
Delaware 1,513 1,167 701,853 75.61% 1,713 1,268 711,388 72.22%
District of Columbia 1,531 1,213 519,124 77.35% 1,714 1,282 529,152 73.34%
Florida 6,545 4,844 15,105,683 70.03% 6,710 4,955 15,367,828 69.95%
Georgia 1,535 1,159 7,255,639 72.00% 2,301 1,743 7,348,895 72.94%
Hawaii 1,780 1,272 1,036,284 66.93% 1,899 1,337 1,045,611 68.03%
Idaho 1,573 1,203 1,156,209 75.67% 1,817 1,381 1,173,050 74.54%
Illinois 6,829 4,796 9,651,449 67.15% 6,214 4,197 9,692,277 65.55%
Indiana 1,640 1,242 4,870,158 71.43% 1,779 1,320 4,904,361 71.10%
Iowa 1,571 1,199 2,317,013 72.17% 1,779 1,322 2,332,526 70.80%
Kansas 1,517 1,160 2,102,923 74.48% 1,817 1,360 2,112,819 72.85%
Kentucky 1,594 1,213 3,280,373 72.59% 1,759 1,293 3,303,086 70.27%
Louisiana 1,560 1,215 3,391,997 74.50% 1,780 1,343 3,418,706 72.68%
Maine 1,510 1,231 1,051,787 78.21% 1,707 1,342 1,055,699 76.56%
Maryland 1,552 1,215 4,469,282 75.66% 1,775 1,332 4,512,168 73.92%
Massachusetts 1,743 1,258 5,195,290 69.73% 1,969 1,344 5,251,844 67.00%
Michigan 6,389 4,919 7,526,924 73.56% 5,728 4,263 7,561,692 71.17%
Minnesota 1,520 1,197 4,065,553 78.26% 1,748 1,334 4,101,742 75.65%
Mississippi 1,427 1,169 2,177,049 78.23% 1,619 1,274 2,188,207 76.86%
Missouri 1,607 1,218 4,524,789 72.67% 1,747 1,310 4,550,886 73.61%
Montana 1,504 1,156 772,343 75.34% 1,786 1,351 780,066 72.86%
Nebraska 1,604 1,251 1,369,821 72.53% 1,718 1,285 1,380,960 73.13%
Nevada 1,583 1,235 2,074,289 73.83% 1,791 1,359 2,114,376 72.39%
New Hampshire 1,704 1,294 1,034,575 74.40% 1,800 1,323 1,041,354 71.85%
New Jersey 1,664 1,227 6,752,843 70.22% 2,508 1,765 6,798,075 68.49%
New Mexico 1,597 1,214 1,537,003 72.53% 1,692 1,325 1,543,402 75.99%
New York 7,266 4,821 15,119,127 62.78% 7,338 4,801 15,227,546 62.66%
North Carolina 1,556 1,233 7,296,125 74.66% 2,288 1,767 7,393,720 75.38%
North Dakota 1,674 1,234 536,176 70.13% 1,848 1,389 549,258 72.29%
Ohio 6,391 4,738 8,732,478 71.07% 5,765 4,155 8,769,959 69.49%
Oklahoma 1,631 1,209 2,808,132 69.54% 1,846 1,343 2,833,947 67.87%
Oregon 1,626 1,251 3,018,457 75.99% 1,738 1,306 3,055,384 74.23%
Pennsylvania 6,657 4,928 9,847,576 70.86% 5,825 4,297 9,877,215 70.93%
Rhode Island 1,606 1,239 819,781 73.87% 1,804 1,333 823,973 71.33%
South Carolina 1,528 1,210 3,566,728 75.22% 1,755 1,348 3,618,547 75.22%
South Dakota 1,544 1,198 615,796 75.69% 1,722 1,315 622,721 75.15%
Tennessee 1,556 1,243 4,880,211 76.17% 1,659 1,285 4,927,115 75.16%
Texas 6,479 4,844 18,742,407 71.52% 6,783 4,919 19,129,850 69.98%
Utah 1,559 1,251 1,960,796 77.89% 1,685 1,342 1,996,733 76.74%
Vermont 1,565 1,201 497,019 74.82% 1,743 1,317 498,516 74.81%
Virginia 1,476 1,143 6,149,648 75.59% 2,298 1,719 6,214,644 73.77%
Washington 1,672 1,230 5,237,038 70.58% 1,791 1,324 5,307,789 71.84%
West Virginia 1,632 1,259 1,443,661 74.41% 1,787 1,285 1,443,073 71.61%
Wisconsin 1,592 1,193 4,349,507 73.71% 1,781 1,258 4,374,390 70.73%
Wyoming 1,566 1,222 433,248 77.56% 1,697 1,296 435,772 76.07%
Table C.15 – Outcomes, by Survey Year, for Which Small Area Estimates Are Available
Measure 2002-
2003
2003-
2004
2004-
2005
2005-
2006
2006-
2007
2007-
2008
2008-
2009
2009-
2010
2010-
2011
2011-
2012
2012-
2013
2013-
2014
X = available; -- = not available.
1 Estimates for these outcomes were not included in the 2002-2003 state report (Wright & Sathe, 2005), but the 2002-2003 estimates are included in the 2003-2004 state report as part of the comparison tables (see Wright & Sathe, 2006). However, the Bayesian confidence intervals associated with these were not published.
2 Estimates for SPD in the years 2002-2003 and 2003-2004 are not comparable with the 2004-2005 SPD estimates. For more details, see Section A.7 in Appendix A of the 2004-2005 state report (Wright, Sathe, & Spagnola, 2007). Note that, in 2002-2003, SPD was referred to as "serious mental illness."
3 Questions that were used to determine an MDE were added in 2004. Note that the adult MDE estimates shown in the 2004-2005 report are not comparable with the adult MDE estimates for later years.
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2002-2014.
Illicit Drug Use in the Past Month X X X X X X X X X X X X
Marijuana Use in the Past Year X X X X X X X X X X X X
Marijuana Use in the Past Month X X X X X X X X X X X X
Perceptions of Great Risk of Smoking Marijuana Once a Month X X X X X X X X X X X X
First Use of Marijuana (Marijuana Incidence) X X X X X X X X X X X X
Illicit Drug Use Other Than Marijuana in the Past Month X X X X X X X X X X X X
Cocaine Use in the Past Year X X X X X X X X X X X X
Nonmedical Use of Pain Relievers in the Past Year --1 X X X X X X X X X X X
Alcohol Use in the Past Month X X X X X X X X X X X X
Underage Past Month Use of Alcohol --1 X X X X X X X X X X X
Binge Alcohol Use in the Past Month X X X X X X X X X X X X
Underage Past Month Binge Alcohol Use --1 X X X X X X X X X X X
Perceptions of Great Risk of Having Five or More Drinks of an Alcoholic Beverage
   Once or Twice a Week
X X X X X X X X X X X X
Tobacco Product Use in the Past Month X X X X X X X X X X X X
Cigarette Use in the Past Month X X X X X X X X X X X X
Perceptions of Great Risk of Smoking One or More Packs of Cigarettes per Day X X X X X X X X X X X X
Alcohol Dependence or Abuse in the Past Year X X X X X X X X X X X X
Alcohol Dependence in the Past Year X X X X X X X X X X X X
Illicit Drug Dependence or Abuse in the Past Year X X X X X X X X X X X X
Illicit Drug Dependence in the Past Year X X X X X X X X X X X X
Dependence or Abuse of Illicit Drugs or Alcohol in the Past Year X X X X X X X X X X X X
Needing But Not Receiving Treatment for Illicit Drug Use in the Past Year X X X X X X X X X X X X
Needing But Not Receiving Treatment for Alcohol Use in the Past Year X X X X X X X X X X X X
Serious Psychological Distress (SPD) in the Past Year2 X X X -- -- -- -- -- -- -- -- --
Had at Least One Major Depressive Episode (MDE) in the Past Year3 -- -- X X X X X X X X X X
Serious Mental Illness (SMI) in the Past Year -- -- -- -- -- -- X X X X X X
Any Mental Illness (AMI) in the Past Year -- -- -- -- -- -- X X X X X X
Had Serious Thoughts of Suicide in the Past Year -- -- -- -- -- -- X X X X X X
Table C.16 – Outcomes, by Age Groups, for Which Small Area Estimates Are Available
Measure Age Group
12+ 12-17 12-20 18-25 26+ 18+
X = available; -- = not available.
NOTE: For details on which years small area estimates are available for these outcomes, see Table C.15.
NOTE: Tables containing 18 or older estimates were first presented with the 2005-2006 small area estimation (SAE) tables.
NOTE: Estimates for those aged 18 to 25, 26 or older, and 18 or older are available for all outcomes.
1 There are minor wording differences in the questions for the adult and adolescent MDE modules. Therefore, data from youths aged 12 to 17 were not combined with data from adults aged 18 or older to get an overall MDE estimate (12 or older).
Source: SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health, 2002-2014.
Illicit Drug Use in the Past Month X X -- X X X
Marijuana Use in the Past Year X X -- X X X
Marijuana Use in the Past Month X X -- X X X
Perceptions of Great Risk of Smoking Marijuana Once a Month X X -- X X X
First Use of Marijuana (Marijuana Incidence) X X -- X X X
Illicit Drug Use Other Than Marijuana in the Past Month X X -- X X X
Cocaine Use in the Past Year X X -- X X X
Nonmedical Use of Pain Relievers in the Past Year X X -- X X X
Alcohol Use in the Past Month X X X X X X
Binge Alcohol Use in the Past Month X X X X X X
Perceptions of Great Risk of Having Five or More Drinks of an Alcoholic Beverage
   Once or Twice a Week
X X -- X X X
Tobacco Product Use in the Past Month X X -- X X X
Cigarette Use in the Past Month X X -- X X X
Perceptions of Great Risk of Smoking One or More Packs of Cigarettes per Day X X -- X X X
Alcohol Dependence or Abuse in the Past Year X X -- X X X
Alcohol Dependence in the Past Year X X -- X X X
Illicit Drug Dependence or Abuse in the Past Year X X -- X X X
Illicit Drug Dependence in the Past Year X X -- X X X
Dependence or Abuse of Illicit Drugs or Alcohol in the Past Year X X -- X X X
Needing But Not Receiving Treatment for Illicit Drug Use in the Past Year X X -- X X X
Needing But Not Receiving Treatment for Alcohol Use in the Past Year X X -- X X X
Serious Psychological Distress (SPD) in the Past Year -- -- -- X X X
Had at Least One Major Depressive Episode (MDE) in the Past Year1 -- X -- X X X
Serious Mental Illness (SMI) in the Past Year -- -- -- X X X
Any Mental Illness (AMI) in the Past Year -- -- -- X X X
Had Serious Thoughts of Suicide in the Past Year -- -- -- X X X
Table C.17 – Summary of Milestones Implemented in the SAE Production Process, 2002-2012
SAE Production Milestone Years for Which Pooled 2-Year Small Area Estimates Were Published
2002-
2003
2003-
2004
2004-
2005
2005-
2006
2006-
2007
2007-
2008
2008-
2009
2009-
2010
2010-
2011
2011-
2012
2012-
2013
2013-
2014
check mark = SAE production milestone implemented; -- = SAE production milestone not implemented; AMI = any mental illness; MDE = major depressive episode; NSDUH = National Survey on Drug Use and Health; SAE = small area estimation; SMI = serious mental illness.
1 The weight used for 2010 was based on projections from the 2000 census control totals, and the 2011 weight was based on projections from the 2010 census control totals. For SMI and AMI, the weights used for both years were based on the 2010 census control totals.
2 Variable selection was done using 2002-2003 NSDUH data for all outcomes with the following exception: For SMI, AMI, suicidal thoughts in the past year, and MDE, variable selection was done using 2008-2009 NSDUH data. Note that the 2005-2006, 2006-2007, and 2007-2008 MDE small area estimates were based on the variable selection done in 2008-2009.
3 For all outcomes except SMI and AMI, the 2010-2011 small area estimates were produced based on 2002-2003 variable selection (see footnote 2 for an exception). For SMI and AMI, variable selection was done using 2010-2011 NSDUH data.
4 When new variable selection was done using 2010-2011 NSDUH data, one source of predictor data was revised: The American Community Survey (ACS) estimates were used in place of 2000 long-form census estimates, which resulted in dropping several predictors and adding several new predictors.
5 The 2005-2006 through 2008-2009 small area estimates were revised and republished with falsified data removed. For more information, see Section A.7 of "2011-2012 NSDUH: Guide to State Tables and Summary of Small Area Estimation Methodology" at http://www.samhsa.gov/data/.
6 The 2008-2009, 2009-2010, and 2010-2011 small area estimates were revised and republished based on the new SMI and AMI variables. These new variables will continue to be used to produce SMI and AMI small area estimates. For more information, see Section B.11.1 of the document mentioned in this table's footnote 5.
7 An adjusted MDE variable was created for 2005-2008 that is comparable with the 2009-2013 MDE variables. Hence, MDE small area estimates were produced using the adjusted variable. For more information, see Section B.11.3 of the document mentioned in this table's footnote 5.
Weights Based on Projections from 2000 Census Control Totals check mark check mark check mark check mark check mark check mark check mark check mark check mark1 -- -- --
Weights Based on Projections from 2010 Census Control Totals -- -- -- -- -- -- -- -- check mark1 check mark check mark check mark
Small Area Estimates Produced Based on Variable Selection Done Using
   2002-2003 Data2
check mark check mark check mark check mark check mark check mark check mark check mark check mark3 -- -- --
Small Area Estimates Produced Based on Variable Selection Done Using
   2010-2011 Data4
-- -- -- -- -- -- -- -- check mark3 check mark check mark check mark
Small Area Estimates Reproduced Using Data Omitting Falsified Data5 -- -- -- check mark check mark check mark check mark -- -- -- -- --
SMI and AMI Small Area Estimates Based on Updated 2013 Model6 -- -- -- -- -- -- check mark check mark check mark check mark check mark check mark
MDE Small Area Estimates Based on Adjusted MDE Variable7 -- -- -- check mark check mark check mark check mark -- -- -- -- --

Section D: References

Center for Behavioral Health Statistics and Quality. (2007). 2005 National Survey on Drug Use and Health: Methodological Resource Book (Section 20, Methamphetamine analysis report). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Center for Behavioral Health Statistics and Quality. (2014). 2012 National Survey on Drug Use and Health: Methodological Resource Book. Rockville, MD: Substance Abuse and Mental Health Services Administration.

Center for Behavioral Health Statistics and Quality. (2015a). 2013 National Survey on Drug Use and Health: Methodological Resource Book. Rockville, MD: Substance Abuse and Mental Health Services Administration.

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

Center for Behavioral Health Statistics and Quality. (in press). 2014 National Survey on Drug Use and Health: Methodological Resource Book. Rockville, MD: Substance Abuse and Mental Health Services Administration.

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

Ghosh, M. (1992). Constrained Bayes estimation with applications. Journal of the American Statistical Association, 87, 533-540.

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

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

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

Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys (Wiley Series in Probability and Mathematical Statistics: Applied Probability and Statistics). New York, NY: John Wiley & Sons.

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

Scheuren, F. (2004, June). What is a survey (2nd ed.). Retrieved September 30, 2015, from https://www.whatisasurvey.info/overview.htm

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

Singh, A. C., & Folsom, R. E. (2001, April 11-14). Hierarchical Bayes calibrated domain estimation via Metropolis-Hastings Step in MCMC with application to small areas. Presented at the International Conference on Small Area Estimation and Related Topics, Potomac, MD.

Wright, D. (2003a). State estimates of substance use from the 2001 National Household Survey on Drug Abuse: Volume I. Findings (HHS Publication No. SMA 03-3775, NHSDA Series H-19). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies.

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

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

Wright, D., & Sathe, N. (2006). State estimates of substance use from the 2003-2004 National Surveys on Drug Use and Health (HHS Publication No. SMA 06-4142, NSDUH Series H-29). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies.

Wright, D., Sathe, N., & Spagnola, K. (2007). State estimates of substance use from the 2004-2005 National Surveys on Drug Use and Health (HHS Publication No. SMA 07-4235, NSDUH Series H-31). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies.

Section E: List of Contributors

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

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

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


End Notes

1 See http://www.samhsa.gov/data/.

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

3 National small area estimates = Population-weighted averages of state-level small area estimates.

4 The census region-level estimates in the tables are population-weighted aggregates of the state estimates. The national estimates, however, are benchmarked to exactly match the design-based estimates.

5 At http://www.samhsa.gov/data/, see Tables 1 to 26 in "2013-2014 NSDUHs: Model-Based Estimated Totals (in Thousands) (50 States and the District of Columbia)."

6 Note that in the 2004-2005 NSDUH state report and prior reports, the term "prediction interval" (PI) was used 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 report estimates; thus, "prediction interval" was dropped and replaced with "Bayesian confidence interval."

7 For MDE, estimates for individuals 12 or older are not included. For AMI, SMI, and thoughts of suicide, estimates for youths aged 12 to17 and individuals aged 12 or older are not included.

8 At http://www.samhsa.gov/data/, see "2011-2012 National Surveys on Drug Use and Health: Guide to State Tables and Summary of Small Area Estimation Methodology."

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

10 The SAE expert panel, convened in April 2002, had six members: Dr. William Bell of the U.S. Bureau of the Census; Partha Lahiri, Professor of the Joint Program in Survey Methodology at the University of Maryland at College Park; Professor Balgobin Nandram of Worcester Polytechnic Institute; Wesley Schaible, formerly Associate Commissioner for Research and Evaluation at the Bureau of Labor Statistics; Professor J. N. K. Rao of Carleton University; and Professor Alan Zaslavsky of Harvard University.

11 At http://www.samhsa.gov/data/, see "2013-2014 NSDUH: Model-Based Prevalence Estimates (50 States and the District of Columbia)" (Tables 1 to 26, by Age Group).

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

13 To increase the precision of the estimated random effects at the within-state level, three SSRs were grouped together. California had 12 grouped SSRs; Florida, New York, and Texas each had 10 grouped SSRs; Illinois, Michigan, Ohio, and Pennsylvania each had 8 grouped SSRs; Georgia, New Jersey, North Carolina, and Virginia each had 5 grouped SSRs; and the rest of the states and the District of Columbia each had 4 grouped SSRs. Note that these 250 grouped SSRs were used on both the 2013 and 2014 samples.

14 For details on how the average annual rate of marijuana (incidence of marijuana) is calculated, see Section B.8 of "2011-2012 National Surveys on Drug Use and Health: Guide to State Tables and Summary of Small Area Estimation Methodology" at http://www.samhsa.gov/data/.

15 This file is available at http://www.samhsa.gov/data/.

16 See Table 9 of the "2013-2014 NSDUH: Model-Based Prevalence Estimates (50 States and the District of Columbia)" at http://www.samhsa.gov/data/.

17 See Table 9 of "2013-2014 NSDUHs: Model-Based Estimated Totals (in Thousands) (50 States and the District of Columbia)" at http://www.samhsa.gov/data/.


Long Descriptions—Equations

Section B.1

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

Long description end. Return to Equation 1.

Section B.4

Long description, Equation 2. Lower sub s and a is defined as the exponent of capital L sub s and a divided by the sum of 1 and the exponent of capital L sub s and a. And upper sub s and a is defined as the exponent of capital U sub s and a divided by the sum of 1 and the exponent of capital U sub s and a.

Long description end. Return to Equation 2.

Long description, Equation 3. Capital L sub s and a is defined as the difference of two quantities. The first quantity is the natural logarithm of the ratio of Theta sub s and a and 1 minus Theta sub s and a. The second quantity is the product of 1.96 and the square root of MSE sub s and a, which is the mean squared error for state-s and age group-a.

Long description end. Return to Equation 3.

Long description, Equation 4. Capital U sub s and a is defined as the sum of two quantities. The first quantity is the natural logarithm of the ratio of Theta sub s and a and 1 minus Theta sub s and a. The second quantity is the product of 1.96 and the square root of MSE sub s and a, which is the mean squared error for state-s and age group-a.

Long description end. Return to Equation 4.

Long description, Equation 5. The mean squared error, MSE sub s and a, is defined as the sum of two quantities. The first quantity is the square of the difference of two parts. Part 1 is defined as the natural logarithm of the ratio of capital P sub s and a and 1 minus capital P sub s and a. Part 2 is defined as the natural logarithm of the ratio of Theta sub s and a and 1 minus Theta sub s and a. The second quantity is the posterior variance of the natural logarithm of the ratio of capital P sub s and a and 1 minus capital P sub s and a.

Long description end. Return to Equation 5.

Section B.6

Long description, Equation 6. The covariance between the natural logarithm of Theta 1 hat and the natural logarithm of Theta 2 hat is equal to the correlation between the natural logarithm of Theta 1 hat and the natural logarithm of Theta 2 hat multiplied by the square root of the product of the variance v of the natural logarithm of Theta 1 hat and the variance v of the natural logarithm of Theta 2 hat.

Long description end. Return to Equation 6.

Long description, Equation 7. Variance v of the natural logarithm of Theta sub i is equal to the square of quantity q. Quantity q is calculated as the difference between capital U sub i and capital L sub i divided by 2 times 1.96, where i takes values 1 and 2.

Long description end. Return to Equation 7.

Long description, Equation 8. Capital U sub 1 is defined as the natural logarithm of the ratio of 0.1678 and 1 minus 0.1678, which is negative 1.6013. Capital L sub 1 is defined as the natural logarithm of the ratio of 0.1214 and 1 minus 0.1214, which is negative 1.9792.

Long description end. Return to Equation 8.

Long description, Equation 9. Capital U sub 2 is defined as the natural logarithm of the ratio of 0.1313 and 1 minus 0.1313, which is negative 1.8895. Capital L sub 2 is defined as the natural logarithm of the ratio of 0.0900 and 1 minus 0.0900, which is negative 2.3136.

Long description end. Return to Equation 9.

Long description, Equation 10. The estimate of the log-odds ratio, lor hat sub a, is defined as the natural logarithm of the ratio of two quantities. The numerator of the ratio is p 2 sub a divided by 1 minus p 2 sub a. The denominator of the ratio is p 1 sub a divided by 1 minus p 1 sub a, where p1 sub a is 0.1431 and p 2 sub a is 0.1089. The estimate lor hat sub a is calculated to be negative 0.3122.

Long description end. Return to Equation 10.

Long description, Equation 11. The variance v of the natural logarithm of Theta 1 hat is equal to the square of quantity q. Quantity q is calculated as the difference between capital U sub 1 and capital L sub 1 divided by the product of 2 and 1.96. Here, capital U sub 1 is negative 1.6013, and capital L sub 1 is negative 1.9792. Hence, the variance v of the natural logarithm of Theta 1 hat is calculated to be 0.00930.

Long description end. Return to Equation 11.

Long description, Equation 12. The variance v of the natural logarithm of Theta 2 hat is equal to the square of quantity q. Quantity q is calculated as the difference between capital U sub 2 and capital L sub 2 divided by the product of 2 and 1.96. Here, capital U sub 2 is negative 1.8895, and capital L sub 2 is negative 2.3136. Hence, the variance v of the natural logarithm of Theta 2 hat is calculated to be 0.01171.

Long description end. Return to Equation 12.

Long description, Equation 13. Quantity z is the estimate of the log-odds ratio, lor hat sub a, divided by the square root of the sum of the variance v of the natural logarithm of Theta 1 hat and the variance v of the natural logarithm of Theta 2 hat, where lor hat sub a is negative 0.3122, the variance v of the natural logarithm of Theta 1 hat is 0.00930, and the variance v of the natural logarithm of Theta 2 hat is 0.01171. The statistic z is calculated to be negative 2.1546.

Long description end. Return to Equation 13.

Go to Top of Page