Science Watch

A new study suggests that a little extra planning could lead a treatment outcome study to predict not only whether a treatment works, but how and when it works. That kind of specific knowledge could help researchers to develop techniques tailored to different phases of treatment.

The study, by University of Pittsburgh psychologist Saul Shiffman, PhD, and his colleagues and published in the April Journal of Consulting and Clinical Psychology (Vol. 74, No. 2), uses an analysis of nicotine-patch treatment for smoking cessation as an example of how to identify a treatment's mechanisms of action. The key, the researchers say, is to not only examine a treatment's final outcome-the question of does it work-but to also measure treatment milestones, such as smoking lapses-to learn when and how it works.

"Understanding mechanisms of action will allow more rational prescribing of medications and behavioral methods," says Frank Vocci, PhD, director of the National Institute on Drug Abuse's division of pharmacotherapies and medical consequences of drug abuse. "If you know that different treatments operate by different mechanisms, you can use them in combination to treat different aspects of a disorder."

Shiffman's study shows that, by taking a more in-depth perspective, treatment outcome studies can begin to tease apart these mechanisms. "I really think he's on to something," says Vocci.

What's more, says University of Vermont addiction researcher John Hughes, MD, Shiffman's study "provides a nice methodology that others can easily replicate."

The power's in the details

Although Shiffman was interested in the ultimate outcome of this smoking-cessation study, he also wanted to demonstrate that giving study participants electronic diaries and conducting a more complex statistical analysis of the data they input could yield better insights not only on a treatment's effectiveness, but also when and how it works.

As in any standard smoking-cessation outcome study, Shiffman and his colleagues randomly split a group of ready-to-quit smokers into two groups: They gave one group of 188 smokers high-dose nicotine patches to use while they attempted to quit and the other group of 136 smokers a placebo patch.

The researchers tracked study participants more closely than in typical treatment studies by using handheld computers that collected continuous data by prompting participants with questions about treatment use and smoking lapses. This allowed them to examine when the treatment worked well and when it failed by identifying when people quit, when they lapsed and took a smoke, how long a lapse lasted and whether that lapse turned into a true relapse. They defined a lapse as smoking after a period of abstinence and relapse both by the field standard of smoking seven days in a row as well as a more conservative measure of at least five cigarettes a day for three days running.

They then used a statistical technique called "survival analysis" to predict the risk of these events over time, for example, the risk of lapses and relapses for people wearing a nicotine patch, compared with people wearing a placebo patch.

"This method tells you how the treatment affected each milestone," explains Shiffman.

The methodology is more intensive and somewhat more expensive than standard treatment-outcome studies because of the amount of data collected and the use of handheld computers. But the benefits can be enormous, as demonstrated by this study's results: Just like many other standard nicotine-patch studies, this study found that high-dose nicotine patches helped more people quit smoking than the placebo patch did. But it also found another, unexpected benefit of the nicotine patch-people using the high-dose nicotine patch were three to five times less likely to move from a smoking lapse to full-blown relapse than people using the placebo.

This important finding wouldn't have shown up in traditional studies, which typically measure only the final treatment outcome and don't collect data on midpoint milestones, says Hughes. In fact, current clinical practice and Food and Drug Administration guidelines encourage people to stop using the patch if they lapse. In contrast, this new method shows that it is very important to continue on the patch after a lapse, says Hughes.

Shiffman's findings not only contradict current clinical practice, but also shed light on how nicotine patches might work, says Vocci. They may block nicotine's "priming effect"-the way one dose of a drug primes people who lapse to want more and pushes them toward relapse.

"Studies in animals and humans show that a drug-dependent person is most at risk for relapse when primed," says Vocci. "If the patch really blocks this, it's a true advance in our understanding."

Hughes wonders if the patch may work by blocking the pathway in the brain that allows the nicotine in cigarettes from having its usual effect. This study points the way for future research to investigate this possibility.

Insight into future treatments

Some may ask why more researchers haven't done these kinds of studies before. It appears to be a simple case of not thinking of it, says Vocci. But now that someone has, he expects more studies will follow.

"We've known for years that the majority of smokers will lapse after they quit, but we never looked closely at the pattern of lapsing and the movement from lapsing to relapsing," says Vocci.

Adds Shiffman: "Without this refined analysis, we wouldn't know that people should stay on the [patch] even after they lapse. If we dig deeper into why and how nicotine replacement works, it will provide insight into developing other treatments."

In fact, notes Shiffman, there are currently no clinical treatments to help smokers who lapse while trying to quit. Even in most treatment-outcome studies, lapsers are considered failures. Better understanding how the patch and other nicotine-replacement products work might provide hope for these people.

It's critical that treatment research moves toward studies that shed light on mechanisms, particularly if researchers hope to improve clinicians' ability to treat complex disorders such as addiction, says psychologist Raymond Niaura, PhD, a smoking-cessation researcher at Brown University Medical School. These disorders will likely respond best to a toolbox approach-with different techniques for different phases of treatment.

Shiffman's study provides a map for the field to begin down this road, he says.

 Beth Azar is a writer in Portland, Ore.