1998 National Household Survey on Drug Abuse
The survey covers residents of households, noninstitutional group quarters (e.g., shelters, rooming houses, dormitories), and civilians living on military bases. Persons excluded from the survey include homeless people who do not use shelters, active military personnel, and residents of institutional group quarters, such as jails and hospitals. Appendix 3 describes surveys that cover populations not included in the NHSDA sampling frame.
The 1998 NHSDA employed a multistage area probability sample which resulted in interviews with 25,500 persons. This survey was conducted from January through December 1998. Response rates for household screening and for interviewing were 93.0 percent and 77.0 percent, respectively. The sample design oversampled blacks, Hispanics, and young people, to improve the accuracy of estimates for those populations. In addition, residents of Arizona and California were oversampled to provide direct survey estimates for these state populations. This expansion of the sample was initiated in April, 1997 to measure the impact of propositions related to illicit drugs that were passed by voters in these States in 1996.
The household interview takes an average of about one hour to complete and uses self-administered answer sheets for the most sensitive portions of the interview. Use of these self-administered answer sheets is designed to maximize honest reporting of illicit drug use and other sensitive behaviors. The self-administered answer sheet includes questions on the recency and frequency of use of various licit and illicit drugs, opinions about drugs, problems associated with drug use, perceived need and demand for drug abuse treatment, and drug abuse treatment experience. Also collected in interviewer-administered portions of the 1998 questionnaire are data on demographic characteristics, employment, education, income, health status, health insurance, utilization of services, and access to health care.
A major change in the NHSDA design is being implemented during 1999. This change will significantly affect the results that will be reported next year. Beginning in 1999, data are being collected using computer assisted interviewing, including audio computer assisted self interviewing in place of the paper answer sheets used in 1998 and earlier. In addition, the NHSDA sample was expanded in 1999 to conduct 70,000 interviews per year, with the capability of providing state-level estimates of substance use prevalence.
When the new questionnaire was introduced in 1994, a supplemental sample was selected for use with the old methodology (i.e., identical to 1993). This provided the capability to assess the impact of the new questionnaire and to measure the effects of the change in methodology. Analyses of the 1994 data have shown that the new methodology had a minimal effect on some estimates, but the effect on others was substantial. A separate SAMHSA report provides details on the development of the new questionnaire and the impact of the new methodology on substance use estimates (SAMHSA 1996a).
Because of the change in methodology in 1994, many of the estimates from the 1993 and earlier NHSDAs are not directly comparable to estimates from the 1994 and later NHSDAs. Since it is important to describe long-term trends in drug use accurately, an adjustment procedure was developed and applied to the pre-1994 estimates. This adjustment uses the 1994 supplemental sample to estimate the magnitude of the impact of the new methodology for each drug category. A description of the adjustment method is given in Appendix 2.
Readers need to be aware that all 1979-93 data shown in this report are different from previously published NHSDA estimates for 1979-93. Because the adjustments were developed from sample survey data, they are subject to sampling error and, therefore, may in some cases introduce additional variation into trends. This is particularly true for estimates of rare behaviors and for small subgroups.
The tables and analyses focus primarily on recent trends, from 1997 to 1998. Long term trends are also presented and discussed, but due to the limitations of the procedure used to adjust for the differences between the pre- and post-1994 methodology, it is not possible to analyze these data in as much detail. As indicated in the tables, statistical significance testing was done for comparisons between 1998 and prior years. Significance levels are indicated in the tables, and allchanges described in the text as increases or decreases were tested and found to be significant at least at the .05 level, unless otherwise stated.
Tables and text present prevalence measures in terms of both the number of drug users and the rate of drug use in the population. Tables show estimates of drug use prevalence in lifetime (i.e., ever used), past year, and past month. The analysis focuses primarily on past month use, which is also referred to as "current use."
Data are presented for three major racial/ethnic groups: whites, blacks, and Hispanics. A fourth category, "Other," includes Asian and Pacific Islanders, American Indians and Alaskan Natives, and other groups. It should be noted that the category "white" includes only non-Hispanic whites, the category "black" includes only non-Hispanic blacks, and the category "Hispanic" includes Hispanics of any race.
Data are also presented for four U.S. geographic
regions and nine geographic divisions within these regions. These regions
and divisions include the following groups of States:
Northeast Region - New England
Division: Maine, New Hampshire, Vermont, Massachusetts, Rhode Island,
Connecticut; Mid-Atlantic Division: New York, New Jersey, Pennsylvania.
The tables also present data by population density. For this variable, large metropolitan areas are defined as Metropolitan Statistical Areas (MSAs) with a population of 1 million or more. Small metropolitan areas are MSAs with a population of less than 1 million. Nonmetropolitan areas are areas outside of MSAs. For some tables, small metropolitan areas are further classified as having either less than or greater than 250,000 population, and nonmetropolitan areas are classified as either urban or rural. In this report, all rural estimates are based on only nonmetropolitan rural areas. Tables showing data for the more detailed population density categories and the geographic divisions are based on the combined 1997 and 1998 NHSDA data sets, to ensure large enough sample sizes for these categories. For 1993 and later estimates, 1990 Census data and 1990 MSA classifications were used to determine population density. For 1992 estimates, 1990 Census counts and 1984 MSA classifications were used.
Other than presenting results by age group and other basic demographic characteristics, no attempt is made in this report to control for potentially confounding factors that might help explain the observed differences. This point is particularly salient with respect to race/ethnicity,which tends to be highly associated with socioeconomic characteristics. The cross-sectional nature of the data limits the capability to infer causal relationships. Nevertheless, the data presented in this report are useful for indicating demographic subgroups with relatively high (or low) rates of drug use, regardless of what the underlying reasons for those differences might be.
The Office of Applied Studies has also
conducted more specialized, in-depth analyses using NHSDA data on specific
substance abuse issues (see list of references in Appendix 4). Recent studies
either completed or in progress include:
This page was last updated on June 01, 2008.