Skip To Content Table Of Contents
Click for DHHS Home Page
Click for the SAMHSA Home Page
Click for the OAS Drug Abuse Statistics Home Page
Click for What's New
Click for Recent Reports and HighlightsClick for Information by Topic Click for OAS Data Systems and more Pubs Click for Data on Specific Drugs of Use Click for Short Reports and Facts Click for Frequently Asked Questions Click for Publications Click to send OAS Comments, Questions and Requests Click for OAS Home Page Click for Substance Abuse and Mental Health Services Administration Home Page Click to Search Our Site

2001 National Household  Survey on Drug Abuse

5. Trends in Initiation of Substance Use

Estimates of substance use incidence, or initiation (i.e., the number of new users during a given year), provide another measure of the Nation's substance use problem. Where prevalence estimates describe the extent of current use of substances, incidence data describe emerging patterns of use, particularly among young people. In the past, increases and decreases in incidence have usually been followed by corresponding changes in the prevalence of use, particularly among youths.

The incidence estimates in this report are based on National Household Survey on Drug Abuse (NHSDA) data from 1999, 2000, and 2001 collected with computer-assisted interviewing methods. These data should not be compared with previously published NHSDA data based on paper-and-pencil interviewing methods. Not only is the mode of data collection different for the incidence estimates produced prior to the 1999 NHSDA, but the estimation methodology has been revised as well. The estimation methodology is described in Appendix B in Volume II and summarized below.

The incidence estimates are based on the NHSDA questions on age at first use, year and month of first use for recent initiates, the respondent's date of birth, and the interview date. Using this information along with editing and imputation when necessary, an exact date of first use is determined for each substance used by each respondent. For age-specific incidence rates, the period of exposure is defined for each respondent and age group for the time that the respondent was in the age group during the calendar year. Incidents of first use also are classified by year of occurrence and age at the date of first use. By applying sample weights to incidents of first use, estimates of the number of new users of each substance for each year are developed. These estimates include new users at any age (including those younger than age 12) and also are shown for two specific age groups—youths aged 12 to 17 and young adults aged 18 to 25. In addition, the average age of new users in each year and age-specific rates of first use are estimated.

The incidence rates are presented in this report as the number of new users per 1,000 potential new users because they indicate the rate of new use among persons who have not yet used the substance (i.e., potential new users). More precisely, the rates are actually the number of new users per 1,000 person-years of exposure. This measure is widely used in describing the incidence of disease. The numerator of each rate is the number of persons in the age group who first used the substance in the year, while the denominator is the person-time exposure measured in thousands of years. Each person's drug-specific exposure time ends on the date of first use of the respective drug. For age-specific estimates, exposure is limited to time during the year that the person was in the age group. Persons who first used the substance in a prior year have zero exposure to first use in the current year, and persons who still have never used the substance by the end of the current year had a full year of exposure to risk.

Because the incidence estimates are based on retrospective reports of age at first substance use by survey respondents interviewed during 1999, 2000, and 2001, they may be subject to several sources of bias. These include bias due to differential mortality of users and nonusers of each substance, bias due to memory errors (recall decay and telescoping), and underreporting bias due to social acceptability and fear of disclosure. See Appendix B in Volume II for a discussion of these biases. As explained in Appendix B, it is possible that some of these biases, particularly telescoping and underreporting because of fear of disclosure, may be affecting estimates for the most recent years more significantly. To account for this bias in the interpretation of the trends, a more stringent standard for determining statistical significance involving estimates from the most recent years (1997 and later) is used in this chapter. Differences are reported to be statistically significant only if they differ at the =.01 level. The usual standard in the rest of the report is the =.05 level. This is an arbitrary standard that provides some protection against incorrect conclusions in the face of potential biases that can fluctuate and even change direction from year to year. A more thorough analysis of the problem will be conducted in the future.

Because the incidence estimates are based on retrospective reports of age at first use, the most recent year available for these estimates is 2000, based on the 2001 NHSDA. Estimates for the year 2000 are based only on data from the 2001 survey, estimates for the year 1999 are based only on data from the 2000 and 2001 surveys, and estimates for earlier years are based on the combined 1999 to 2001 data. For two of the measures, first alcohol use and first cigarette use, initiation before age 12 is common. A 2-year lag in reporting for "all ages" estimates is applied for these measures because the NHSDA sample does not cover youths under age 12. The 2-year lag ensures that initiation at ages 10 and 11 is captured in the estimation.


















Table Of Contents

This page was last updated on June 16, 2008.