A variety of other surveys and data systems collect data on substance use and mental illness. It is useful to consider the results of these other studies when discussing the National Survey on Drug Use and Health (NSDUH) data. In doing this, it is important to understand the methodological differences between the different surveys and the impact that these differences could have on estimates of the presence of substance use and mental illness. This appendix briefly describes several of these other data systems and where possible presents comparisons between NSDUH results and results from the other surveys for 2002, 2003, and 2004 or other recent time periods. For some comparisons, NSDUH estimates were generated to be consistent with the data collection periods or groups surveyed in other studies. In addition, this appendix describes surveys of populations not covered by NSDUH.
In-depth comparisons of the methodologies of the three major federally sponsored national surveys of youth substance use have been done. In 1997, a comparison between the National Household Survey on Drug Abuse1 (NHSDA) and Monitoring the Future (MTF) was published (Gfroerer, Wright, & Kopstein, 1997). In 2000, a series of papers comparing different aspects of the NHSDA, MTF, and the Youth Risk Behavior Survey (YRBS) was commissioned by the U.S. Department of Health and Human Services (DHHS). Under contract with the Office of the Assistant Secretary for Planning and Evaluation, Westat, Inc., identified and funded several experts in survey methods to prepare these papers. The papers were published in the Journal of Drug Issues (Hennessy & Ginsberg, 2001). The major findings of these studies were as follows:
These findings suggest that differences in survey methodology may affect comparisons of prevalence estimates among youths from various surveys. This appendix investigates the similarities and differences among rates from NSDUH and other related surveys. Descriptions of the other surveys are provided when they are first discussed in the appendix.
The Monitoring the Future (MTF) study is a national survey that tracks drug use trends and related attitudes among America's adolescents. This survey is conducted annually by the Institute for Social Research at the University of Michigan through a grant awarded by the National Institute on Drug Abuse (NIDA). The MTF and NSDUH are the Federal Government's largest and primary tools for tracking youth substance use. The MTF is composed of three substudies: (a) an annual survey of high school seniors initiated in 1975; (b) ongoing panel studies of representative samples from each graduating class that have been conducted by mail since 1976; and (c) annual surveys of 8th and 10th graders initiated in 1991. In 2002, for all three grades combined, 394 public and private schools and about 43,700 students were in the sample. In 2003, for all three grades combined, 48,500 students in 392 public and private schools were in the sample, and in 2004, nearly 49,500 students in 406 secondary schools were surveyed. The students complete a self-administered, machine-readable questionnaire during a regular class period (Johnston, O'Malley, & Bachman, 2003a; Johnston, O'Malley, Bachman, & Schulenberg, 2004c; Johnston, O'Malley, Bachman, & Schulenberg, 2005).
Comparisons between the MTF estimates and estimates based on students sampled in NSDUH generally have shown NSDUH substance use prevalence levels to be lower than MTF estimates, with differences tending to be more pronounced for 8th graders. The lower prevalences in NSDUH may be due to more underreporting in the household setting as compared with the MTF school setting. However, MTF does not survey dropouts, a group that NSDUH has shown to have higher rates of illicit drug use (Gfroerer et al., 1997). In 2003 and 2004, for most comparisons of estimates of lifetime, past year, and past month prevalence of use of marijuana, cocaine, and inhalants among 8th, 10th, and 12th graders between NSDUH and MTF, NSDUH estimates were lower (see Table E.1 at the end of this appendix). However, both surveys showed that use of these three illicit drugs was stable for most measures between 2003 and 2004. Exceptions were a decrease in past year inhalants use among 8th graders in NSDUH and a decrease in past month marijuana use and an increase in lifetime inhalants use among 8th graders in MTF.
The Youth Risk Behavior Survey (YRBS) is a component of the Centers for Disease Control and Prevention's (CDC's) Youth Risk Behavior Surveillance System (YRBSS), which measures the prevalence of six priority health risk behavior categories: (a) behaviors that contribute to unintentional and intentional injuries; (b) tobacco use; (c) alcohol and other drug use; (d) sexual behaviors that contribute to unintended pregnancy and sexually transmitted diseases (STDs); (e) unhealthy dietary behaviors; and (f) physical inactivity. The YRBSS includes national, State, territorial, and local school-based surveys of high school students conducted every 2 years. The latest YRBS was conducted in 2003 (CDC, 2005b). The 2003 national school-based survey used a three-stage cluster sample design to produce a nationally representative sample of students in grades 9 through 12. The 2003 State and local surveys used a two-stage cluster sample design to produce representative samples of students in grades 9 through 12 in their jurisdictions. The 2003 national YRBS sample included 15,214 students in grades 9 through 12 in the 50 States and the District of Columbia. The national survey and all of the State and local surveys were conducted during the spring of 2003, with the exception of Hawaii. The Hawaii surveys were conducted in the fall of 2003. The students completed a self-administered, machine-readable questionnaire during a regular class period.
In general, this school-based survey has found higher rates of alcohol, cigarette, marijuana, cocaine, and inhalant use for youths than those found in NSDUH (Table E.2) (Grunbaum et al., 2004). To examine estimates that are comparable to YRBS, NSDUH estimates are based on data collected in the first 6 months of the survey year. The prevalence of illicit drug use was generally much higher in the YRBS than in NSDUH. For example, past month marijuana use was 22.4 percent in the 2003 national YRBS compared with 13.2 percent for persons in grades 9 through 12 in January-June of the 2003 NSDUH and 12.1 percent in January-June of the 2004 NSDUH. This is likely due to the differences in study design (school-based vs. home-based).
The National Longitudinal Study of Adolescent Health (Add Health) was conducted to measure the effects of family, peer group, school, neighborhood, religious institution, and community influences on health risks, such as tobacco, drug, and alcohol use. Initiated in 1994 under a grant from the National Institute of Child Health and Human Development (NICHD) with cofunding from 17 other Federal agencies, Add Health is the largest, most comprehensive survey of adolescents ever undertaken. Data at the individual, family, school, and community levels were collected in two waves between 1994 and 1996. In Wave 1 (conducted in 1994-95), roughly 90,000 students from grades 7 through 12 at 144 schools around the United States answered brief, machine-readable questionnaires during a regular class period. Interviews also were conducted with about 20,000 students and their parents in the students' homes using a combined computer-assisted personal interviewing (CAPI) and audio computer-assisted self-interviewing (ACASI) design. In Wave 2, students were interviewed a second time in their homes. In 2001 and 2002, the original Add Health respondents, now aged 18 to 26, were re-interviewed in a third wave to investigate the influence that adolescence has on young adulthood. Identifying information was obtained from participants in order to track them over time.
Survey results from the first two waves indicated that nearly one fourth of teenagers had ever smoked marijuana. Nearly 7 percent of 7th and 8th graders used marijuana at least once in the past month as did 15.7 percent of 9th through 12th graders (Resnick et al., 1997). In the 2003 NSDUH, 19.6 percent of youths aged 12 to 17 had ever used marijuana, and 7.9 percent were past month users; in 2004, these percentages were 19.0 and 7.6 percent, respectively.
The Partnership Attitude Tracking Study (PATS) is an ongoing national research study that tracks drug use and drug-related attitudes among children, teenagers, and their parents. It is sponsored by the Partnership for a Drug-Free America (PDFA). In the 2002 PATS, 7,084 teenagers in grades 7 through 12 completed self-administered, machine-readable questionnaires during a regular class period with their teacher remaining in the room (PDFA, 2005a). For the first time in 2002, PATS included questions on prescription drug abuse. The 2002 PATS found that 20 percent of youths in grades 7 to 12 had ever used prescription pain killers without a doctor's prescription, 19 percent of adolescents reported lifetime use of inhalants, and 40 percent reported lifetime use of marijuana. In 2003, 7,270 youths completed the survey, and prevalence rates remained very similar to rates in 2002 (PDFA, 2005a). The 2003 PATS found that 21 percent of youths in grades 7 to 12 had ever used prescription drugs, 18 percent had used inhalants, and 39 percent reported using marijuana in their lifetime (PDFA, 2003). The 2004 PATS was conducted with 7,314 youths in grades 7 through 12 and found that 19 percent had used inhalants and 37 percent had used marijuana at least once in their lifetime (PDFA, 2005b). NSDUH reported notably lower prevalence rates than PATS. For youths aged 12 to 17, NSDUH estimated the rate of lifetime prescription pain reliever use to be 11.2 percent in both 2002 and 2003 and 11.4 percent in 2004; lifetime inhalant use was 10.5 percent in the 2002, 10.7 percent in 2003, and 11.0 percent in 2004; and lifetime marijuana use was 20.6 percent in 2002, 19.6 percent in 2003, and 19.0 percent in 2004. The major difference in these prevalence estimates is likely to be due to the different study designs. The youth portion of PATS is a school-based survey, which may elicit more reporting of sensitive behaviors than the home-based NSDUH. In addition, the PATS survey is conducted with a sample of students in the 7th through 12th grades, which is a slightly older sample than that of the NSDUH 12- to 17-year-old sample.
The National Survey of Parents and Youth (NSPY) is sponsored by the National Institute on Drug Abuse (NIDA) to evaluate the Office of National Drug Control Policy's (ONDCP's) National Youth Anti-Drug Media Campaign. The survey is specifically designed to evaluate Phase III of the campaign, covering the period between September 1999 and June 2003. Data collection provides estimates of trends in drug use between 2000 and the first half of 2003, as well as changes between 2002 and 2003.
In Phase I (Waves 1 through 3 of data collection), a sample of youths aged 9 to 18 and their parents were recruited to participate in the in-home audio computer-assisted self-interviewing (ACASI) survey. In Phase II (Waves 4 through 7 of data collection), the respondents from Phase I participated in two additional interviews at intervals of 6 to 24 months. In December 2003, ONDCP released the sixth semiannual report of findings that contained data from all three phases (Hornik et al., 2003).
Wave 5's data were collected between January and June 2002 and included 4,040 youths and 2,882 parents. Wave 6's data were collected between July and December 2002 and included 2,267 youths and 1,640 parents. An average of the estimates from Waves 5 and 6 showed that the past year rate of marijuana use among 12 to 18 year olds was 16.4 percent. The corresponding 2002 NSDUH estimate for past year marijuana use among youths aged 12 to 18 was 18.4 percent.
Wave 7's data were collected between January and June 2003 and included 3,587 youths and 2,621 parents. The two surveys produced similar estimates for youths (see Table E.3). For example, the latest wave of NSPY data indicated that 16.7 percent of youths aged 12 to 18 had used marijuana in the past year, and the 2003 NSDUH yielded an estimate of 18.1 percent among this age group for this time period. One explanation for the similarity in estimates is that both surveys used ACASI.
In past waves of NSPY data collection, parents also have been asked about their drug use behaviors; however, parental use was not asked in the Wave 5 or Wave 7 data collections. Lifetime use of marijuana among parents was 53.7 percent in 2001, and past month use was 3.4 percent. According to the full-year data of NSDUH, lifetime use of marijuana among adults aged 18 or older was 42.7 percent in 2002, 43.1 percent in 2003, and 42.7 percent in 2004; past month use was 6.0 percent in 2002 and 2003 and 5.9 percent in 2004.
The National Longitudinal Alcohol Epidemiologic Survey (NLAES) was conducted in 1991 and 1992 by the U.S. Bureau of the Census for the National Institute on Alcohol Abuse and Alcoholism (NIAAA). Face-to-face interviewer-administered interviews were conducted with 42,862 respondents aged 18 or older in the contiguous United States. Despite the survey name, the design was cross-sectional. The National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) was conducted in 2001 and 2002, also by the U.S. Bureau of the Census for NIAAA, using a computerized interviewer-administered interview. The NESARC sample was designed to make inferences for persons aged 18 or older in the civilian, noninstitutionalized population of the United States, including Alaska, Hawaii, and the District of Columbia, and including persons living in noninstitutional group quarters. NESARC is designed to be a longitudinal survey. The first wave was conducted in 2001 and 2002, with a final sample size of 43,093 respondents aged 18 or older. The second wave (2004 and 2005) is currently under way (Grant, Kaplan, Shepard, & Moore, 2003).
Over the decade from 1992 to 2002, the prevalence of past year marijuana use among adults remained about the same in the two surveys, at about 4.0 percent (Compton, Grant, Colliver, Glantz, & Stinson, 2004). The rate of past year marijuana use among adults was 10.1 percent in both the 2003 and 2004 NSDUHs. The discrepancy between the estimates is likely due to the differences in administration; NSDUH is self-administered, and the NLAES and NESARC are interviewer-administered.
The National Health Interview Survey (NHIS) is a continuing nationwide sample survey that collects data using personal household interviews through an interviewer-administered computer-assisted personal interviewing (CAPI) system. The survey is sponsored by the National Center for Health Statistics (NCHS) and provides national estimates of selected health measures. The data presented are from January through September of 2003 and 2004. In the NHIS, current smokers are defined as those who smoke daily, smoked on 1 or more days in the past month, or quit smoking fewer than 30 days ago (for those who smoked 100 or more cigarettes in their lifetime). The survey estimated in 2003 that 21.6 percent of the population were reporting current cigarette smoking (23.7 percent among males and 19.4 percent among females) (NCHS, 2005). The rates were slightly lower in 2004, with 20.7 percent of the population reporting current cigarette smoking (23.3 percent among males and 18.3 percent among females).
In NSDUH, current cigarette smoking is defined as any use in the past month. The full-year NSDUH rate was 27.0 percent in 2003 and 26.4 percent in 2004 for those aged 18 or older. Although the two surveys employ different methodologies, NSDUH still produces higher estimates when using the NHIS definition. For example, when using a definition similar to the NHIS definition and data from January through September, NSDUH estimates 25.1 percent of adults aged 18 or older were current smokers in 2003 and 24.7 percent of adults were current smokers in 2004. See Table E.4 for a comparison of smoking rates between these two surveys by age and gender.
The NHIS defines excessive alcohol drinkers as those who consumed an amount greater than or equal to five drinks in 1 day at least once during the past 12 months. The NHIS rate for excessive alcohol consumption among those aged 18 or older was 18.7 percent in 2004, down slightly from 19.4 percent in 2003 (NCHS, 2005). NSDUH defines heavy alcohol use as having five or more drinks on the same occasion on at least 5 different days in the past 30 days. The full-year NSDUH rates for heavy drinking among those aged 18 or older were 7.3 percent in 2003 and 7.4 percent in 2004.
Compared with NSDUH estimates, MTF estimates of cigarette use are higher among 8th graders, about the same among 10th graders, and somewhat lower among 12th graders. However, both surveys showed slight, but not necessarily significant, decreases in smoking in most grade levels between 2003 and 2004. For example, among 10th graders, there was a significant decrease in lifetime smoking estimates from 43.0 percent in 2003 to 40.7 percent in 2004 according to the MTF. The NSDUH lifetime smoking rates for 10th graders decreased significantly from 43.2 percent in 2003 to 38.3 percent in 2004. See Table E.1 for a comparison of the MTF and NSDUH cigarette use estimates by grade level.
Rates of alcohol consumption are higher overall in the MTF sample compared with NSDUH. In addition, the trends of alcohol consumption differed for the two surveys. Although the NSDUH showed both increases and decreases between 2003 and 2004, only the decrease in lifetime alcohol use among 10th graders was statistically significant; lifetime, past year, and past month alcohol use was more stable among 8th, 10th, and 12th graders in the MTF sample. Both surveys indicate a varying pattern of alcohol consumption by grade level. Table E.1 shows how the MTF estimates of alcohol use compare with NSDUH estimates.
As seen with illicit drug use, the YRBS estimates of cigarette and alcohol consumption were higher than the NSDUH estimates. According to YRBS data, in 2003, 58.4 percent of high school students had tried cigarettes and 21.9 percent of students had smoked cigarettes during the past 30 days (CDC, 2005b). Using only data from January through June, the 2003 NSDUH rates were 46.0 percent for lifetime cigarette use and 20.2 percent for past month cigarette use among students in the 9th through 12th grade, and in 2004 the NSDUH rates were 43.7 percent for lifetime cigarette use and 19.8 percent for past month cigarette use.
Past month alcohol use among 9th to 12th graders in the YRBS was 47.1 percent in the 2001 survey and 44.9 percent in 2003. In contrast, January-June data from NSDUH showed a past month alcohol use rate of 29.1 percent in both 2003 and 2004 among 9th to 12th graders. Lifetime alcohol use rates among students were 63.1 percent using January-June NSDUH data in 2003 and 60.8 percent in 2004, while they were 74.9 percent in the YRBS in 2003.
Data from PATS show that the prevalence of past month cigarette for adolescents in grades 7 through 12 was 28 percent in 2002, 26 percent in 2003, and 23 percent in 2004 (PDFA, 2003, 2005a, 2005b). The NSDUH prevalence of past month cigarette smoking among youths aged 12 to 17 was 13.0 percent in 2002, 12.2 percent in 2003, and 11.9 percent in 2004. Again, the lower prevalence estimates in NSDUH are likely due to its home-based study design and slightly younger age group.
Even though the PATS estimates were higher than the NSDUH estimates, both surveys showed relatively steady rates of drinking among youths from 2002 to 2004. PATS found that 53 percent of teenagers reported past year alcohol use in 2002, 51 percent reported past year use in 2003, and 50 percent reported past year use in 2004. This compares with NSDUH estimates of 34.6 percent of youths aged 12 to 17 reporting past year use in 2002, 34.3 percent in 2003, and 33.9 percent in 2004. The 2002 PATS also found that 36 percent of teenagers reported past month alcohol use and 30 percent reported binge drinking in 2002. In 2003, an estimated 34 percent of youths used alcohol in the past month, while 29 percent reported binge drinking. In 2004, about 33 percent of youths used alcohol in the past month and 28 percent reported binge drinking. In comparison, the 2002 NSDUH rates for past month alcohol use and binge drinking for 12 to 17 year olds were 17.6 and 10.7 percent, respectively. For the 2003 NSDUH, 17.7 percent of youths reported past month alcohol use, and 10.6 reported binge drinking. In 2004, the NSDUH rates for past month alcohol use and binge drinking were 17.6 and 11.1 percent, respectively.
The Behavioral Risk Factor Surveillance System (BRFSS) is an annual, State-based telephone survey of the civilian, noninstitutionalized adult population aged 18 or older and is sponsored by the CDC. In 2002, 2003, and 2004, BRFSS collected data from all 50 States, the District of Columbia, Puerto Rico, the U.S. Virgin Islands, and Guam using a computer-assisted telephone interviewing (CATI) design. BRFSS collects information on access to health care, health status indicators, health risk behaviors (including cigarette and alcohol use), and the use of clinical preventive services by State, and national data are calculated using a median score across States. Although both surveys looked at the percentage of adults who reported having five or more alcoholic drinks on at least one occasion in the past month, the median binge alcohol use rates among adults in the BRFSS sample (16.1 percent in 2002, 16.4 percent in 2003, and 14.9 percent in 2004) (CDC, 2005a) were lower than among adults aged 18 or older in the NSDUH sample (24.0 percent in 2003 and 24.1 percent in 2004). The median percentages of adults who had five or more drinks on an occasion at least five times in the past month, or heavy drinking, were similar among the BRFSS sample, 7.0 percent in 2001 and 7.3 percent in 2002 (CDC, 2002, 2005a), and the NSDUH sample, 7.3 percent in the 2003 NSDUH and 7.4 percent in 2004. Comparisons of estimates of adult binge drinking from BRFSS and NSDUH showed that BRFSS estimates were considerably lower than NSDUH estimates for the total United States and most States, even among demographic subgroups; however, the differences were not statistically significant. The use of ACASI in NSDUH, which is considered to be more anonymous and yields higher reporting of sensitive behaviors, was offered as an explanation for the lower rates in BRFSS (Miller et al., 2004).
Results from the 1994-95 National Longitudinal Study of Adolescent Health indicated that nearly 3.2 percent of 7th and 8th graders smoked six or more cigarettes a day, as did 12.8 percent of 9th through 12th graders (Resnick et al., 1997). In addition, the Add Health study found that 7.3 percent of 7th and 8th graders used alcohol on 2 or more days in the past month, as did 23.1 percent of 9th through 12th graders.
Past waves of the NSPY collected information on cigarette and alcohol use, but Wave 5 in 2002 did not. In 2001, this survey estimated that 34.9 percent of youths aged 12 to 18 had used cigarettes at some point in their lifetime, and past month cigarette use was 11.7 percent. The 2002 NSDUH rates of lifetime and past month cigarette use for youths aged 12 to 18 were 38.8 and 16.7 percent, respectively; these rates were 36.2 and 15.8 percent, respectively, in 2003 and 35.1 and 15.6 percent in 2004.
In 2001, the NSPY estimated that 45.9 percent of youths aged 12 to 18 had used alcohol at some point in their lifetime, and the estimate for past month use was 36.5 percent for the same age group. The 2002 NSDUH rates for lifetime and past month alcohol use were 49.1 and 22.2 percent, respectively; the 2003 rates were 49.0 and 22.2 percent; and the 2004 rates were 47.7 and 22.4 percent.
These NSDUH estimates for cigarette and alcohol use in 2002, 2003, and 2004 are based on data collected from January through June to reflect the same data collection period as the NSPY.
The Harvard School of Public Health's College Alcohol Study (CAS) is an ongoing survey of students at 4-year colleges and universities in 40 States. The study surveyed a random sample of students at the same colleges in 1993, 1997, 1999, and 2001. The schools and students were selected to provide nationally representative samples of schools and students. In 1993, a national sample of 195 colleges was selected from the American Council on Education's list of accredited 4-year colleges by using probability proportionate to size of enrollment; of the 195 colleges, 140 agreed to participate, for a school-level response rate of 72 percent (Wechsler, Dowdall, Davenport, & Castillo, 1995). Of these 140 colleges, 130 participated in 1997, 128 in 1999, and 120 in 2001. Student-level response rates to the two-stage mail survey were 70 percent in 1993, 59 percent in 1997 and 1999, and 52 percent in 2001. The researchers provided a short survey to nonrespondents in order to better weight the data (Wechsler et al., 2002).
The 2001 survey found that the overall rate of binge drinking was 44.4 percent. The CAS defined binge drinking as the consumption of five or more drinks in a row for men and four or more drinks in a row for women. The study found that 22.8 percent of the students binge drank frequently and that 19.3 percent did not drink at all. The 2003 NSDUH binge drinking rate among full-time undergraduates aged 18 to 22 was 43.5 percent, and the 2004 estimate was 43.4 percent. It is useful to note that NSDUH defines binge drinking as five or more drinks in a row on at least one occasion in the past month for both men and women. Despite using different definitions of binge drinking, the CAS estimate and the NSDUH estimate are similar, but it is important to note that the two studies were conducted in different time periods.
The National Comorbidity Survey (NCS) was sponsored by the National Institute of Mental Health (NIMH), the National Institute on Drug Abuse (NIDA), and the W.T. Grant Foundation. It was designed to measure the prevalence of the illnesses in the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) (American Psychiatric Association [APA], 1987) in the general population. The first wave of the NCS was a household survey collecting data from 8,098 respondents aged 15 to 54. These responses were weighted to produce nationally representative estimates. A random sample of 4,414 respondents also were administered an additional module that captured information on nicotine dependence. The interviews took place between 1990 and 1992. The NCS used a modified version of the Composite International Diagnostic Interview (the UM-CIDI) to generate DSM-III-R diagnoses.
There have been several recent extensions to the original NCS, including a 10-year follow-up of the baseline sample (NCS-II), a replication study conducted in 2001 and 2002 with a newly recruited nationally representative sample of 9,282 respondents aged 18 or older (NCS-R), and an adolescent sample with a targeted recruitment of more than 10,000 adolescents (NCS-A) along with their parents and teachers. The NCS-R used an updated version of the CIDI that was designed to capture diagnoses using current DSM-IV criteria (APA, 1994). It should be noted that in several recent NCS-R studies (Kessler et al., 2005a; Kessler, Chiu, Demler, & Walters, 2005b), the diagnosis for abuse also includes those who meet the diagnosis for dependence. In contrast, NSDUH follows DSM-IV guidelines and measures abuse and dependence separately. To make the NCS definition of abuse comparable with that of NSDUH, the rate for dependence must be subtracted from the rate for abuse.
Estimates from the NCS-R indicated that 3.1 percent of persons were alcohol abusers and 1.3 percent were dependent on alcohol in the past year (Kessler et al., 2005b). Excluding those who met the criteria for dependence from those who met the criteria for abuse according to the NCS-R, the resulting rate indicated that 1.8 percent had abused alcohol in the past year. According to the 2002 NSDUH, 4.3 percent of persons aged 18 or older were alcohol abusers in the past year, and 3.7 percent were dependent on alcohol; 7.9 percent were dependent on or abused alcohol. Comparable rates for alcohol abuse, dependence, and abuse or dependence from the 2003 NSDUH were 4.4 percent, 3.3 percent, and 7.7 percent, respectively; in 2004, these rates were 4.3 percent, 3.6 percent, and 8.0 percent, respectively. Therefore, the past year estimate for those with either alcohol abuse and/or dependence from the NCS-R (3.1 percent) is lower than the estimate from the 2002 NSDUH (7.9 percent).
Based on the NCS-R, an estimated 1.4 percent of persons aged 18 or older met the criteria for abuse of illicit drugs, and 0.4 percent met the criteria for dependence on illicit drugs in the past year. In the 2002 NSDUH, 0.9 percent abused illicit drugs, and 1.8 percent were dependent on illicit drugs; 2.7 percent were dependent on or abused illicit drugs. Comparable rates for illicit drug abuse, dependence, and abuse or dependence from the 2003 NSDUH were 0.9 percent, 1.7 percent, and 2.6 percent, respectively; in 2004, these rates were 0.9 percent, 1.9 percent, and 2.8 percent, respectively. This latter rate is higher than the corresponding estimate in the NCS-R (1.4 percent) for substance abuse, which also includes those with dependence. Similarly, NCS-R indicated that 3.8 percent were dependent on or abused alcohol or illicit drugs in the past year compared with 9.4 percent based on the 2002 NSDUH; this rate was 9.1 percent in the 2003 NSDUH and 9.4 percent in the 2004 NSDUH.
The NLAES and NESARC included an extensive set of questions designed to assess the presence of symptoms of alcohol and drug abuse and dependence in persons' lifetimes and during the prior 12 months, based on the criteria from the DSM-IV (APA, 1994). The 1991-92 NLAES found that 7.4 percent of adults were abusing or dependent on alcohol (Grant, 1995). In the 2001-02 NESARC, the rate of alcohol abuse among adults was 4.7 percent, and the rate of alcohol dependence was 3.8 percent. Between 1992 and 2002, the prevalence of alcohol abuse increased and the prevalence of dependence declined (Grant et al., 2004). In 2002, NSDUH found that 7.9 percent of adults were abusing or dependent on alcohol; in 2003, this rate was 7.7 percent, and in 2004 the rate was 8.0 percent. The NLAES and NESARC also found that the prevalence of marijuana dependence or abuse among adults increased from 1.2 percent in 1992 to 1.5 percent in 2002 (Compton et al., 2004). In comparison, the 2002 NSDUH found that 2.7 percent of adults were dependent on or abusing some illicit drug; the rates were 2.6 percent in 2003 and 2.8 percent in 2004. The 2002 and 2003 NSDUHs both estimated that 3.2 million adults (1.5 percent) were dependent on or abusing marijuana; this estimate was 3.5 million (1.6 percent) in 2004. Although the estimates from these two surveys are relatively close, one should note that they were conducted using different methodologies.
The Epidemiologic Catchment Area (ECA) Study (1981-83) was the first survey to administer a structured psychiatric interview and provided population-based estimates of psychiatric disorders. The prevalence rates were estimated by collecting data from households and group quarters (e.g., prisons, nursing homes, mental hospitals) in five local catchment areas (Baltimore, Los Angeles, New Haven, North Carolina, and St. Louis) that had been previously designated as Community Mental Health Center catchment areas. There were three waves of data collection with 20,861 respondents; the first and third waves were interviewer-assisted personal interviews, and the second wave was a telephone interview conducted with household participants only (Eaton et al., 1984). The ECA utilized the Diagnostic Interview Schedule (DIS), a structured clinical instrument that can be used by nonclinically trained interviewers to generate DSM-III (APA, 1980) diagnoses of psychiatric and substance use disorders. A supplemental sample of institutional settings, such as nursing homes, psychiatric hospitals, and prisons, also was included to capture those respondents with a high probability of having a mental disorder.
The National Comorbidity Survey (NCS) was conducted in response to the limitations of the ECA and a broader need to produce nationally representative data on psychiatric conditions. The first wave of the survey revealed that 48.7 percent of the population had at least one axis I or II disorder in their lifetime. This percentage can be broken down into the following: 21 percent with one disorder, 13 percent with two disorders, and 14 percent with three or more disorders (Kessler, 1994). In 2004, NSDUH estimated that 9.9 percent of persons aged 18 or older were classified as having serious psychological distress (SPD) in the past year (see Section B.4.4 in Appendix B for SPD's comparability with SMI).
Preliminary findings from the NCS-R study in 2001-02 indicate that the lifetime and past year prevalence of major depressive disorder was 16.2 and 6.6 percent, respectively. A large percentage of those with lifetime (71.1 percent) and 12-month (78.5 percent) psychiatric disorders also had at least one additional DSM-defined psychiatric disorder, suggesting that the burden of mental disorders is pervasive in the general population (Kessler et al., 2003a, 2003b). In 2004, NSDUH estimated that 14.8 percent of adults experienced a major depressive episode in their lifetime, and 8.0 percent experienced one in the past year.
NESARC was sponsored by the National Institute on Alcohol Abuse and Alcoholism (NIAAA) and designed to be a longitudinal study with the first wave of data (43,093) collected between 2001 and 2002. An additional wave (2004 and 2005) is currently under way. The study contains comprehensive assessments of drug use, abuse, and dependence, as well as associated mental disorders. Extensive data on the utilization of treatment programs and medical care also are being collected. NESARC is a representative sample of the noninstitutionalized population aged 18 or older residing in the contiguous United States, the District of Columbia, or Hawaii. The design also oversampled young adults aged 18 to 24 and minorities to increase precision, ensure adequate cell sizes for variables with low event rates, and ensure representation of major racial/ethnic categories. DSM-IV diagnoses of major mental disorders were generated using the Alcohol Use Disorder and Associated Disabilities Interview Schedule-version 4 (AUDADIS-IV), which is a structured diagnostic interview that captures major DSM-IV axis I and axis II disorders. The 12-month prevalence estimates show that 9.2 percent of respondents experienced an independent (i.e., not attributable to illness or substance use) mood disorder in the past year, whereas 11.0 percent experienced an independent anxiety disorder (Grant et al., 2004).
The NSPY, described above, is distinct in that it measures drug use and attitudes among youths as young as 9 years. The earlier NSPY results showed that youths aged 9 to 11 were strongly opposed to marijuana use. Wave 3 of the survey estimated that only 0.3 percent of youths aged 9 to 11 had used marijuana in the past year. The corresponding rates for Waves 1 and 2 were 0.8 and 0.0 percent, respectively (ONDCP, 2005).
The Washington, DC, Metropolitan Area Drug Study (DC*MADS) was designed (a) to estimate the prevalence, correlates, and consequences of drug abuse among all types of people residing in one metropolitan area of the country during one period of time with a special focus on populations who were underrepresented or unrepresented in household surveys and (b) to develop a methodological model for similar types of research in other metropolitan areas of the country. Sponsored by NIDA and conducted from 1989 to 1995 by RTI International and Westat, Inc., the project included 11 separate but coordinated studies that focused on different population subgroups (e.g., homeless people, institutionalized individuals, adult and juvenile offenders, new mothers, drug use treatment clients) or different aspects of the drug abuse problem (e.g., adverse consequences of drug abuse). DC*MADS provided a replicable methodological approach for developing representative estimates of the prevalence of drug abuse among all population subgroups, regardless of their residential setting, in a metropolitan area. The key population domains in DC*MADS were homeless people, institutionalized persons, and the household population.
A major finding of DC*MADS was that, when data are aggregated for populations from each of the three domains, the overall prevalence estimates for the use of drugs differ only marginally from those that would be obtained from the household population alone (i.e., from NSDUH), largely because the other populations are very small compared with the household population. However, a somewhat different picture emerged when the numbers of drug users were examined. Adding in the nonhousehold populations resulted in an increase of approximately 14,000 illicit drugs users compared with the corresponding estimates for the household population. About 25 percent of past year crack users, 20 percent of past year heroin users, and one third of past year needle users were found in the nonhousehold population (Bray & Marsden, 1999).
The 2002 DoD Survey of Health Related Behaviors among Military Personnel was the 8th in a series of studies conducted since 1980. The sample consisted of 12,756 active-duty Armed Forces personnel worldwide who anonymously completed self-administered questionnaires that assessed substance use and other health behaviors. For the total DoD, during 30 days prior to the survey, heavy alcohol use declined from 20.8 percent in 1980 to 15.4 percent in 1998 and increased significantly to 18.1 percent in 2002; past month cigarette smoking decreased from 51.0 percent in 1980 to 29.9 percent in 1998 and increased significantly to 33.8 percent in 2002; and past month use of any illicit drugs declined from 27.6 percent in 1980 to 2.7 percent in 1998 and also showed a nonsignificant change in 2002 to 3.4 percent (Bray et al., 1999, 2003). In 2002, military personnel had significantly higher rates of heavy alcohol use than their civilian counterparts (16.9 vs. 11.2 percent) when demographic differences between the military and civilian populations were taken into account (civilian data were drawn from the 2001 NSDUH and adjusted to reflect demographic characteristics of the military). Differences in military and civilian heavy alcohol use rates were largest for men aged 18 to 25. Among this age group, the military rate was nearly twice as high as the adjusted civilian rate (32.2 vs. 17.8 percent). Military personnel showed similar rates of cigarette use (31.6 vs. 31.1 percent) compared with civilians. Rates of illicit drug use in the military were significantly lower than those observed for the comparable civilian population when demographic differences between the military and civilian populations were taken into account (3.3 vs. 12.1 percent).
The 1997 Survey of Inmates in State and Federal Correctional Facilities sampled inmates from a universe of 1,409 State prisons and 127 Federal Prisons for the Bureau of Justice Statistics (BJS). Systematic random sampling was used to select the inmates for the computer-assisted personal interviews. The final numbers interviewed were 14,285 State prisoners and 4,041 Federal prisoners. Among other items, these surveys collected information on the use of drugs in the month before the offense for convicted inmates. Women in State prisons (62.4 percent) were more likely than men (56.1 percent) to have used drugs in the month before the offense (BJS, 1999, 2000). Women also were more likely to have committed their offense while under the influence of drugs (40.4 vs. 32.1 percent of male prisoners). Among Federal prisoners, men (45.4 percent) were more likely than women (36.7 percent) to have used drugs in the past month. Male and female Federal prisoners were equally likely to report the influence of drugs during their offense (22.7 percent of male and 19.3 percent of female prisoners). The survey results indicate substantially higher rates of drug use among State and Federal prisoners as compared with the household population.
|Drug/Current Grade Level||SURVEY/TIME PERIOD|
|Lifetime||Past Year||Past Month||Lifetime||Past Year||Past Month|
|-- Not available.
a Difference between 2003 estimate and 2004 estimate is statistically significant at the .05 level.
b Difference between 2003 estimate and 2004 estimate is statistically significant at the .01 level.
MTF = Monitoring the Future.
Sources: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2003 and 2004.
The Monitoring the Future Study, University of Michigan, 2003 and 2004.
|Past Month Use||44.9||--||29.1||29.1|
|Past Month Use||21.9||--||20.2||19.8|
|Past Month Use||22.4||--||13.2||12.1|
|Past Month Use||4.1||--||1.2||1.0|
|Past Month Use||3.9||--||0.9||1.0|
|-- Not available. The YRBS is only conducted every 2 years.
YRBS = Youth Risk Behavior Survey.
Sources: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, January-June for 2003 and 2004.
Centers for Disease Control and Prevention, Youth Risk Behavior Survey, 2003.
|Use Measure||Age Group||Percent Reporting Use|
|Past Year||12 to 13||3.3||4.0||--||3.1||2.3||2.8|
|14 to 16||17.0||18.3||--||19.1||19.5||17.4|
|12 to 18||16.4||16.7||--||18.4||18.1||17.1|
|Past Month||12 to 13||1.1||1.8||--||1.4||0.9||1.1|
|14 to 16||8.3||8.2||--||9.4||9.7||9.0|
|12 to 18||8.9||7.9||--||9.8||9.9||9.1|
|-- Not available.
1NSPY estimates for 2002 are averages of data from Wave 5 (collected between January and June 2002) and Wave 6 (collected between July and December 2002).
NSPY = National Survey of Parents and Youth.
Sources: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2002, 2003, and 2004.
National Institute on Drug Abuse, National Survey of Parents and Youth, 2002, JanuaryJune 2003.
|Gender/Age||NHIS (January September)||NSDUH (January September)|
|18 to 44||25.3||23.5||31.1||30.5|
|45 to 64||21.6||22.3||22.5||23.5|
|65 or Older||9.4||8.5||10.6||8.8|
|18 to 44||28.1||25.8||34.2||33.6|
|45 to 64||23.3||25.1||24.9||25.5|
|65 or Older||10.5||10.0||10.9||10.5|
|18 to 44||22.5||21.3||28.2||27.4|
|45 to 64||20.1||19.6||20.2||21.6|
|65 or Older||8.5||7.3||10.3||7.5|
|Note: For the NHIS, past month cigarette use is defined as currently smoking daily or smoking on 1 or more days in the past month or quitting smoking fewer than 30 days ago (for those who smoked 100+ cigarettes in the lifetime). The analysis excluded those with unknown use status (about 1 percent each year). For NSDUH, past month cigarette use is defined as having smoked in the past month. For comparison purposes, the NSDUH definition was adjusted to include those who had smoked in the past month and smoked at least 100 cigarettes in their lifetime.
NHIS = National Health Interview Survey.
Sources: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2003 and 2004.
National Center for Health Statistics, National Health Interview Survey, 2003 and 2004.
1 Beginning with the 2002 survey year, the survey name was changed from the National Household Survey on Drug Abuse (NHSDA) to the National Survey on Drug Use and Health (NSDUH).
This page was last updated on June 16, 2008.