NIH grant focuses on cost effectiveness

PULLMAN – Just how successful and cost-effective are community-based, substance abuse prevention programs? It’s a question that plagues politicians and policy makers, especially in a time of shrinking resources and increasing demand for services.

Now, thanks to a two-year, $400,000 grant from the National Institutes of Health, researchers at WSU are working on a new way to assess the effectiveness of those programs in the real world. That tool, they say, could help save time and money.

“These statistical techniques could help us make better use of limited resources at the local, state and national level,” said professor Robert Rosenman, associate director of WSU’s School of Economic Sciences.

Laura Griner Hill, a faculty member in the WSU Department of Human Development, agreed.

She conducts research on WSU Extension’s Strengthening Families Program for Parents and Youth 10 – 14, a national program delivered throughout Washington by WSU Extension faculty. That program will be the focus of the economic analysis techniques being developed.

Other members of the research team include professor Ron Mittelhammer, director of the School of Economic Sciences, and assistant professor Bidisha Mandal, an extension economist specializing in health issues.

“There is research evidence that prevention programs work, and teens who attend them are less likely to use alcohol, tobacco and drugs,” Hill said. “What we want to know, though, is do substance abuse prevention programs like Strengthening Families really work in the real world, and if they do work, what’s the true value to society?”

Traditionally, the effectiveness of prevention programs has been measured based on randomized clinical trials, using objective measures with representative samples. But, Rosenman and Hill noted, in the real world, prevention program participants choose or “self select” to attend. The question is whether or how that impacts actual program effectiveness. How well do research results translate to actual community results?

For example, maybe the only families entering a substance abuse prevention program are those already healthy and whose young teens are not likely to start drinking. In that case, the program would not be a good use of resources.

Conversely, maybe families worried about their teen’s risky behaviors are more likely to come to the program. In that case, the program might be even more effective than research has suggested, providing a greater return on investment.

“These selection effects alter the value of the program so that as randomized clinical trials are translated to real-world implementation, new measurement is needed,” said Hill.

The randomized clinical trials are not designed to detect those subtleties, according to Rosenman and Hill, and, as a result, can’t provide a true picture of effectiveness.

Their project will use new statistical tools to analyze existing outcomes data from the Strengthening Families program to determine how self-selection affects program impact.

“The purpose of the grant isn’t to evaluate the long-term effectiveness of Strengthening Families, but rather to develop and test new methods of evaluation. An important benefit of our work is that the evaluation of SFP and other family interventions will improve,” Rosenman said.