Climbing above the forest and the trees: three future directions in addiction treatment research.

Abstract

Miller & Moyers 1 have revived a debate, dating back more than four decades, about the importance of specific versus non-specific factors in addiction treatment and psychotherapy writ large 2. They conclude, quite correctly, that the two are inseparable. I fully agree that clinical research has focused on specific treatment content, to the neglect of the relational and programmatic contexts of treatment delivery. I agree that the field of behavioral intervention research needs to move beyond a hyper-focus on standardized empirically supported treatments. However, as therapist factors overall account for only 3–7% of the variance in client outcomes 3, 4, will studying the therapeutic context be sufficient to boost the effects of addiction treatments? Instead, perhaps the way beyond this long-held debate is to focus upon 1 specific techniques predicting outcomes, 3 stand-alone technologies and 4 broadening the focus to include cost-effectiveness. Few studies have linked therapist techniques with treatment outcomes, even though they have great potential to boost effect sizes 5. Recently, my colleagues and I conducted the first study of therapist techniques predicting smoking. We found that therapists’ techniques to promote awareness and observation of smoking cues, without acting on them by smoking, predicted 42–52% lower odds of smoking at the next counseling session 6. The implication is that focusing upon and refining intervention techniques aimed at awareness and observation have the potential for boosting the effect sizes of smoking cessation intervention and, perhaps, addiction treatments overall. Indeed, there is a growing literature on the study of taxonomies of behavior change techniques which has recently shown specific techniques that predict reductions in alcohol consumption 7. Therapists themselves may be a confound that, at least partly, accounts for the ‘effect’ of specific techniques. Our answer for addressing that confound is to study the therapy platforms that take the therapist out of the equation. Technology-delivered interventions offer a convenient method to do that. Stand-alone technologies such as websites and smartphone applications (apps) are increasingly becoming a primary method for intervening on addictive behaviors, including alcohol and tobacco at low cost. Technologies address a critical problem that traditional therapist-delivered interventions are unlikely to ever address adequately: reach the people who need addiction treatment the most. For example, in the United States alone, the Smokefree.gov website reaches nearly 2 million people each year and has 10% quit rates—which are more than double the effect of quitting on one's own 8. Moreover, smartphone apps for smoking cessation reached 3.2 million people in the United States alone during 2012–13 9. Conveniently, such technologies provide precise methods for studying the role of specific therapy techniques in addiction treatment outcomes. For example, in our analysis of our SmartQuit apps for smoking cessation, we found that tracking practice of therapy skills, viewing a quit plan and tracking instances of letting one's urge pass (without smoking) were associated prospectively with a higher odds of quitting 10. Such analysis illustrates the value of technology for focusing upon specific techniques that, if shown promising, can be used in interventions—delivered either with or without a therapist. Finally, I would argue that the focus on the factors (whether specific or non-specific) that predict outcomes is too narrow. Cost-effectiveness needs to be a new focus. Metrics such as the cost per successful outcome (e.g. abstinence), health-care savings and life years gained need to be taken into account. Beyond effect sizes, these are the hard data that payers (e.g. insurance companies, governments) and health-care administrators need in order to decide to spend precious time and resources to adopt a new treatment. Cost-effectiveness analysis can indicate 2 whether a therapy might yield the same outcomes as another therapy but cost less to implement, or 3 whether a therapist is equally effective as another therapist but costs less to employ. In sum, I am glad to see Miller & Moyers revitalize the debate over specific versus non-specific factors in psychotherapy. The way out of this long-held debate is to climb above it. None.