Wilson M. Compton, M.D, M.P.E.
Director, Division of Epidemiology, Services and Prevention Research
National Institute on Drug Abuse
National Institutes of Health
Department of Health and Human Services

June 20, 2002

Statement for the Record

Mr. Chairman and members of the Subcommittee, thank you for inviting the National Institute on Drug Abuse (NIDA) to participate in this important hearing, which will provide you with background information about the Office of National Drug Control Policy's (ONDCP) National Youth Anti-Drug Media Campaign, as well as the most recent findings from the Campaign's Phase III evaluation. I am Dr. Wilson Compton, the new Director of NIDA's Division of Epidemiology, Services and Prevention Research, and I am pleased to be able to represent NIDA here this morning. It is my Division that manages the contract to Westat and its subcontractors for the evaluation of Phase III of the Campaign.

My brief comments today will focus on NIDA's role in the development and design of the Phase III evaluation for the Media Campaign. I would like to emphasize at the onset that science, particularly studies in the communications and behavioral research area, plays a critical role in the overall design and implementation of the evaluation. It is precisely because of NIDA's research support and expertise which covers all aspects of drug abuse and addiction, that NIDA was asked by ONDCP to independently evaluate the impact of the Campaign on both parents and children. I will not focus on the specific evaluation results this morning, but will leave that for the representatives from Westat and the Annenberg School for Communication from the University of Pennsylvania, the NIDA contractors who are actually responsible for implementing the evaluation.

It was in January of 1998 that NIDA was asked by the ONDCP to measure the outcomes and impact of the National Youth Anti-Drug Media Campaign on parents and their children. This request was in response to legislation, which required ONDCP to establish "a system to measure outcomes of success of the national media campaign." The Media Campaign itself has been conducted in three phases. Phase I of the Media Campaign was launched January to July 1998 as a 12-city pilot. Phase II, was a nationwide extension of the pilot, running from July 1998 to August 1999. Phase III of the Campaign began in September 1999, and it was during this phase that NIDA's evaluation was launched.

As a first step in planning its evaluation, NIDA convened leading experts in design, measurement, statistical sampling, prevention and communications research to advise on the best methodology for accomplishing this task. We issued a Request for Proposals to implement the evaluation based on the design suggested by the expert consultants. Subsequently, in the Fall of 1998, NIDA awarded a contract through a fair, open and competitive process to the nationally known health survey research company, Westat. Westat teamed with communication research experts at the Annenberg School for Communication at the University of Pennsylvania, and substance abuse experts at the National Development and Research Institute to assemble a team of nationally recognized experts to lead various aspects of the evaluation.

The overall objective of the Phase III evaluation, which covers the final four years of the Campaign, is to measure the extent to which the Campaign impacts the knowledge, beliefs, attitudes, and behaviors of parents and their children in regard to illegal drug use. The evaluation is also providing information on overall cumulative changes in these factors related to youth drug abuse as well as exposure to Campaign media messages.

Because the Media Campaign is only one piece of a larger effort that involves a multitude of federal, state and local agencies, and public and private organizations that are working to prevent and reduce drug use in this country, it is extremely challenging to determine the exact causes for any changes in drug abuse rates. To help ensure that changes can be attributed to the Campaign, the evaluation was designed to go well beyond the analysis of trends from existing data. Long-standing national surveys such as NIDA's own Monitoring the Future Survey and the Substance Abuse and Mental Health Services Administration's National Household Survey on Drug Abuse will continue to be used to monitor overall drug use trends, but a new survey had to be developed for the evaluation to be as rigorous and accurate as possible. The new survey called the National Survey of Parents and Youth (NSPY) now serves as the primary tool for the evaluation.

Since most of the country was already exposed to the early components of the national Campaign, it was recognized in planning the Phase III evaluation that there would be no opportunity to use a standard approach in which pre-exposure baseline assessments are conducted. To address this problem, the panel of evaluation experts recommended a design that relied on a dose-response analysis. This design compares youth and their parents who report no exposure to the Campaign to those who report different levels of exposure (i.e. no exposure compared to a little exposure, moderate exposure, or a lot of exposure).

The design for this evaluation is quite complex, and is strongly grounded in models derived from science. In fact, NIDA and its contractor/subcontractors were complimented in the July 2000 US General Accounting Office Report for "succeeding in designing a comprehensive theoretical model and analytic strategy for determining the impact of the Campaign on several drug intermediate and outcome variables."

As I mentioned earlier, as part of the evaluation design, the researchers developed the National Survey of Parents and Youth (NSPY). The NSPY uses reliable state-of-the art computer assisted technology to survey both parents and youths in the privacy of their own homes. Interviews are conducted in both English and Spanish. Almost 14,000 participants from 90 locations across the country were recruited to participate in the study. This nationally representative sample characterizes the target audience for the Media Campaign, the approximately 40 million youth aged 9 to 18 at the start of the evaluation, and 43 million of their parents.

The NSPY is unique in many ways. The fact that the NSPY questionnaires are administered in the respondents' homes using touch-screen laptop computers provided by the evaluators is quite progressive. Surveying parents and youth from the same household is another outstanding feature, which improves our ability to measure how parents and children respond to Campaign messages, and how parents' exposure affects their offspring's attitudes and drug use. For instance, in response to the Campaign messages to parents, when parents start talking to their children more about drug abuse, how does that affect their children's attitudes and behavior? How do those attitudes and behaviors change over time?

The questionnaires were designed to be different for the teens (aged 12-18), children (aged 9-11), and parents. For each group, the NSPY helps to assess changes in the attitudes, beliefs and behaviors that are targeted by the Campaign and to determine if these changes can be attributed to the Campaign. This allows the results of the evaluation to be looked at for parents and children separately, or together. For example, both the parent and the youth questionnaires include measurement of their exposure to Media Campaign messages and other anti-drug messages from various sources including television, print, and the internet as well as sources such as school anti-drug education. By using laptop computers, evaluators can actually show the study participant the media ads that are currently running in the Campaign and ask if the parent or child had seen a particular ad, an important innovative aspect of the survey.

We are not just interested in exposure to the Campaign. We are also trying to determine and measure other factors, such as whether parents talk to their children about not using drugs, whether children's attitudes about marijuana are positive or negative, what the parents' beliefs are on monitoring their child's activities, whether teens think their friends approve or disapprove of drug use, and whether or not children start using drugs.

Both youths and one parent or caregiver from the same household are being interviewed multiple times over the course of the evaluation period. Overall, data collection is divided into seven semiannual waves over a 3-1/2-year period, with each wave followed by a report. The study sample was recruited in the first three successive Waves, with 40 percent in the first Wave, 30 percent in the second, and 30 percent in the third. The subsequent waves go back and re-interview these same participants. We have just reported on Wave 4 of the evaluation, so there will be three more waves upon which data will be collected and reported.

Of those recruited and sampled in Wave 1, 2,478 youth and 1,752 parents were re-interviewed in Wave 4. It is these initial findings that are based on longitudinal interviews with the study participants from Wave 1, representing 40% of the sample that are currently being reported. These findings are interim, and we may find different results once the study concludes and all the data are analyzed. Behavior changes can take a very long time.

In addition to the actual reported findings that will emerge from this independent state-of-the-art evaluation once the study is completed, we will have learned much more. We will gain new knowledge about how to measure and explain changes in drug-related knowledge, attitudes, beliefs and behaviors in both children and parents, and we will gain an improved understanding of the critical factors involved in changing people's behaviors. After completion of the project, the data will be released in a public use dataset so that researchers all over the country can use this tremendous knowledge base for future endeavors, which rely on science to improve drug abuse prevention efforts.

[Testimony Index]