Doi:10.1016/j.jsat.2005.08.003

Journal of Substance Abuse Treatment 29 (2005) 267 – 276 Evidence-based treatment: Why, what, where, when, and how? William R. Miller, (Ph.D.)a,b,T, Joan Zweben, (Ph.D.)c,d, Wendy R. Johnson, (M.S.)a,b aDepartment of Psychology, University of New Mexico, Albuquerque, NM 87131-1161, USA bCenter on Alcoholism, Substance Abuse, and Addictions (CASAA), University of New Mexico, Albuquerque, NM 87131-1161, USA c14th Street Clinic and East Bay Community Recovery Project, Oakland, CA, USA dUniversity of California, San Francisco, CA, USA Received 19 May 2005; received in revised form 29 July 2005; accepted 10 August 2005 Research and clinical perspectives are blended in this commentary on the rapidly emerging requirement for evidence-based treatment (EBT) in substance abuse programs. Although, historically, it has not been a standard of care in behavioral health, there are sound scientific,ethical, and compassionate reasons to learn and deliver an EBT as it becomes available. This article explores a series of issues, including thefollowing: (1) Why should EBTs be used in substance abuse treatment? (2) What kinds of treatment are EBTs, and how are they determined?(3) Where can EBTs be implemented—at what levels of service delivery? (4) When should EBTs be used? and (5) How do clinicians learnEBTs? Potential pitfalls in implementing EBTs are also considered. D 2005 Elsevier Inc. All rights reserved.
Keywords: Treatment; Evidence; Efficacy; Effectiveness; Policy 1. Why use empirically supported treatments? treatment, incentives to change and update one’s practicehave been minimal. Third-party payers have reimbursed for Suppose that you have a life-threatening illness and went generic contexts of mental health and substance abuse for health care to a physician who told you, bI really don’t treatments, such as evaluation, group therapy, inpatient pay much attention to medical research. I’ve been treating treatment, and case management. What goes on behind people like you for 30 years, and I know what works.
closed doors has been left to professional judgment, and Medical research isn’t all that relevant to my practice, and, practitioners have preferred it that way.
besides, I’m too busy to read journals.Q Would you go back Standards of care are changing, however, with the or would you find yourself another doctor? recognition that not all btreatment worksQ ( Although we do expect primary care and specialist States and cities are beginning to mandate that behavioral physicians to keep current in their field and to provide us the health care providers observe the same evidence-based most current and effective treatment that science has to treatment (EBT) standards that are expected in health care offer, the same standard has not been applied in behavioral more generally. Reimbursement is gradually being tied to health care. In treating substance use disorders, practice has the delivery of EBTs and even to the outcomes of treatment.
been largely guided by whatever approach a provider was The handwriting is on the wall: Those who are not trained in or preferred. Because providers are inclined to providing empirically supported interventions are going to believe that their services are effective, and because have a harder time getting paid for their services. bAnything payment for services has not been linked to the content of There are good reasons for moving toward EBT. All treatments are not created equal. There is consistent T Corresponding author. Department of Psychology, University of evidence that the outcomes of specific approaches vary New Mexico, Albuquerque, NM 87131-1161, USA. Tel.: +1 505 925 widely in treating alcohol and other drug problems ( E-mail address: wrmiller@unm.edu (W.R. Miller).
1998; Miller, Wilbourne, & Hettema, 2003) and that 0740-5472/05/$ – see front matter D 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.jsat.2005.08.003 W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 therapists differ significantly from one another in their effect sizes, estimates of how large treatment effects really effectiveness in delivering a particular form of treatment are. Effect sizes level the playing field, being less affected by factors such as the number of clients studied. Many outcome McLellan, Woody, Luborsky, & Goehl, 1988; Najavits & studies, however, do not report the detail that is necessary; Weiss, 1994; Project MATCH Research Group, 1998).
therefore, effect sizes must be inferred from available Certain types of treatment are rather consistently found to information. Even meta-analyses can require dozens if not produce no benefit or worse yet continue to be delivered and hundreds of judgment calls in estimating effect sizes. A way of protecting against such bias is to show the detailed work by we do in treatment and how we do it.
which studies were rated and conclusions drawn.
If that is the case, then we owe it to our clients to keep abreast of what works best for whom and to provide, as much as possible, the treatment services that are most likelyto be beneficial. Courts are upholding clients’ right not only A distinction is often drawn between efficacy and to treatment but also to effective treatment. Again, this is a effectiveness. Efficacy studies evaluate the benefits of a common expectation in medicine. Physicians who provide treatment when delivered under ideal and highly controlled outdated or ineffective treatments are vulnerable to claims of conditions. Interventions in efficacy trials are often delivered malpractice. Furthermore, trends toward person-centered by highly qualified clinicians with modest caseloads who care highlight the right of clients to have sufficiently receive intensive training plus ongoing supervision and accurate information about their therapeutic options to fidelity monitoring, working within a well-funded university allow them to make informed choices about their own research clinic rather than in a busy and underfunded community practice setting. The efficacy of a treatment Finally, there are clear trends toward the integration of under such conditions does not guarantee its effectiveness substance abuse treatment with mainstream health care when implemented in real-world treatment programs. There are many fewer clinical trials on EBTs delivered under disorders frequently have concomitant mental and physical normal practice conditions in the community. The National health problems as well and are best served by integrated Institute on Drug Abuse (NIDA) Clinical Trials Network 1990; Hubbard et al., 1989; McLellan et al., 1997, 1998). As CTN/about.html) is addressing this issue, testing EBTs on the the management of substance use disorders becomes more frontlines, in community treatment programs, with regular closely integrated with health care systems, it is inevitable program staff delivering EBTs to their ordinary clients.
that EBT will increasingly become the standard of treatment.
Furthermore, multisite trials often reveal significant site- by-treatment interactions, such that an EBT works well in 1.1. Systematic reviews and meta-analyses one location but not in another. If this is true under thehighly controlled conditions of a multisite trial, surely it is Few clinicians have the time and expertise to analyze and the case that the effectiveness of an EBT will vary con- integrate hundreds of clinical trials. We therefore rely on siderably across clinical populations and programs in which knowledgeable reviewers to summarize evidence and derive it is implemented. Using an EBT is a sensible place to start, its implications for practice. Narrative reviews about what but it does not ensure effectiveness in a particular case or bstudies show,Q however, involve a substantial amount of context. Dissemination research is exploring what it takes subjective judgment, and reviewers do not always show for a treatment shown to be effective in clinical trials to be effectively delivered in community practice ( Two refinements have emerged to reduce selective or Rawson, Guydish, & Zweben, 2003).
otherwise biased presentations of reviews and to makereviewers’ work more transparent and reproducible. Thefirst of these is the systematic review, in which standard elements are examined and reported for each study and cleardecision rules are specified for reaching conclusions.
Although most practitioners would endorse the statement, Thorough literature search procedures are also used to bWe should offer our clients the best treatment we can,Q there ensure that reviewers have identified most or all of the is minimal consensus and substantial confusion as to what bbest Q means. The move toward EBT is shaping a new The second refinement is statistical meta-analysis, which definition of what constitutes optimal treatment, one that seeks to use a common yardstsick in comparing study out- moves beyond best practice guidelines developed by practi- comes. In treatment outcome studies, for example, a meta- tioner consensus. There is a test of efficacy higher than analytic review might identify for each study the best measure clinician judgment that can be found in scientific evidence.
of substance use and use this common metric for comparison.
What, then, constitutes evidence? Courts of law have A standard feature of meta-analyses is the computation of clear standards as to what is and is not admissible evidence, W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 and such guidelines, although less clear cut, are emerging receiving counseling in addition to medication, and this with regard to substance abuse treatment as well.
Broad agreement points to a hierarchy of scientific evi- Postier, & Kenny, 2004). Similarly, studies rather consis- dence, with different types constituting stronger or weaker tently indicate the importance of providing for children to demonstrations of efficacy. In most EBT systems, the attract and retain women in residential or outpatient treat- highest level of credence is given to randomized clinical ment. Such evidence-based practices represent a broader trials, the gold-standard research design of the U.S. Food perspective than EBTs alone and should be considered in and Drug Administration (FDA) for approving new pharmacotherapies. No single clinical trial is conclusive, A fourth level of evidence is found in anecdotal case of course, and evidence strength grows as the number of reports, professional opinion, and best practice guidelines well-controlled studies increases. The FDA approves new developed by clinician consensus with minimal basis in treatments based on clinical trials that are logically capable, research, often because of the absence of relevant studies.
by design, of demonstrating efficacy, taking into account the methodological quality of each study and the relative weight Addiction Medicine (2001), which offer decision rules for placing patients into various levels of treatment intensity, There is good reason for higher credence given to began as a professional consensus document; subsequent randomized clinical trials. Few other research designs can studies have shed light on the reliability, validity, and utility control for the beliefs and expectancies that can substan- of these criteria and helped refine them ( tially bias outcomes. Human beings are highly susceptible Magura et al., 2003). Professional consensus is also the to paying selective attention to information that confirms standard used to develop the Treatment Improvement preexisting beliefs and to ignoring conflicting information.
Protocols published by the U.S. Center for Substance Abuse Some clinicians assert that, bThe therapy I do is more Treatment. Although often most persuasive to practitioners, effective than any of those EBTs. I just don’t have the time this level of evidence is typically given least credence in and resources to study it.Q Well-designed randomized trials provide a persuasive, although imperfect, correction for When, then, does a treatment become an EBT? Various authorities have established different and sometimes Even in medicine, of course, many treatments that have conflicting standards for when there is enough evidence not been validated by randomized trials are provided to constitute an EBT. Part of the problem here is drawing a discrete line (EBT or not) on what is actually a from quasi-experimental studies that offer some degree of continuous dimension (amount, type, and strength of control over factors that can confound the interpretation of available evidence). Various definitions of evidence have results but fall short of the rigor of controlled clinical trials been used to generate lists of EBTs for substance use same pattern of behavior change across multiple cases or ing lists from the American Psychiatric Association, the groups given the same treatment, for example, offers some American Psychological Association, Cochrane collabora- evidence of consistency of outcomes but no basis for tions, meta-analyses, and the NIDA. It is useful to have a comparison with other treatment approaches or no treat- compilation of the strengths of evidence for (or against) ment. Some EBT reviews include such quasi-experimental different approaches to inform and demystify the dichot- omous and somewhat arbitrary decision as to which Hettema, Steele, & Miller, 2005).
treatments are evidence based and which are not.
A third tier of evidence is found in correlational studies As a starting point, we examined the conclusions of with systematic observation across cases or programs.
10 reviews of EBTs from seven research groups ( Although randomized trials represent a gold standard for Thelander, & Jonsson, 2003; Carroll, 1998; Finney & Moos, demonstrating efficacy, there are clinically important ques- 2002; Mattick & Hall, 1993; Mattick & Jarvis, 1992; tions for which this design is not optimal. When patients are McCrady, 2000; McCrady & Ziedonis, 2001; Miller & randomly assigned to different levels of treatment intensity Wilbourne, 2002; NIDA, 1999; Rawson, 1996). As shown or duration, clinical trials typically show no main effect of in some treatments are found on most lists of EBTs whereas others appear on only one or two. This illustrates clinical studies, however, rather consistently show a positive the extent to which EBT lists can differ depending on the correlation of retention in treatment with better outcomes.
procedures and criteria one uses. A further source of The latter finding has laudably discouraged the practice of variance in this list is that some reviews focused only on prematurely discharging people for the same reason that alcohol problems whereas others did only on illicit drug use.
they were admitted (drug use) and has encouraged the Given these, the amount of convergence across reviews is search for effective motivational strategies to improve encouraging, albeit inconclusive. Reviewers do not work in retention. Field studies have shown that clients undergoing isolation, and they read each other’s work. Some of these methadone maintenance programs tend to fare better when reviews are more transparent than others, showing their W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 Table 1Convergence of 10 summaries of EBTs Community reinforcement approach plus vouchers Day treatment with abstinence, contingencies, and vouchers Multidimensional family therapy for adolescents Voucher-based reinforcement therapy in methadone maintenance treatment The reviews (5) Ziedonis (2001); (6) (7) (8) (9) and (10) (2000).
(+) indicates that the review identifies the treatment as evidence based; (++), the review differentiates strong evidence base for the treatment; A, total numberof (+) ratings for the treatment across the 10 reviews.
work and clearly articulating the review procedures so that able research. There are, of course, many gaps in treatment they could be replicated. Others (e.g., simply outcome research, although not as many as was the case name EBTs without specifying the criteria and processes even 10 years ago. With more than a thousand controlled clinical trials in the literature for alcohol, tobacco, and illicit provide statistical meta-analyses to compare the absolute drug use, it is no longer defensible to say that there is impact of treatments. Other summaries rely on the limited research from which to draw any conclusion ( Farrell, & Ali, 2004). points to treatment methods Neither are treatment approaches randomly selected for with good evidence of efficacy. Like health care more testing in clinical trials. There has been concern that certain generally, substance abuse treatment can now be guided by approaches (e.g., behavioral) are favored by researchers or easier to test in randomized trials and, therefore, have the So perhaps a better question is when one would not use opportunity to accumulate greater evidence of efficacy. We an EBT. One obvious answer is, bWhen there is no EBT examined this within a review of published clinical trials on available.Q It is the case that there is currently insufficient research evidence to indicate a specific approach in treating Pharmacotherapies were tested 106 times (20%); cognitive– certain substance use disorders. The inhalation of solvents behavioral treatments, 231 times (43%); and other psycho- such as glue or gasoline is clearly a serious concern, and, at present, there is not enough research evidence to guidepractice with confidence. Should one then leave inhalantabuse untreated? Certainly not. The same is true for specific combinations of disorders. Treatment methods are beingdeveloped and tested for the combination of substance When should we use EBTs? One reasonable answer is, bWhenever possible,Q recognizing the limitations of avail- W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 This means, of course, that providers need to learn how disorder (and borderline personality dis- to deliver EBTs. Beyond the challenge of changing established practice habits, developing competence in a approach is to use EBTs that are indicated for each of the new treatment method may not be a simple matter. Studies clarifying effective dissemination methods are gaining Clinicians are often concerned that available clinical trial evidence may not apply to the population they treat.
practitioners and programs to rely on conferences, work- It is reasonable to question the generalizability of EBTs shops, and in-service trainings. Yet such one-shot methods across groups with which they have not been tested. In tend to be ineffective in changing practice behavior and the interest of internal validity, efficacy studies often increasing clinical skillfulness. In one study, reading about, exclude patients with a concomitant psychiatric diagnosis viewing videotapes, and attending a 2-day workshop or multiple substance use or dependence, medically ill resulted in a minimal increase in skill in motivational patients, and, possibly, clients less motivated for change interviewing and in no apparent change in client response, (by virtue of self-selection into a demanding trial). In although participants believed that they had developed short, clinical trials may exclude a majority of the clients seen in community practice. Furthermore, very limited providing ongoing feedback and coaching significantly clinical trial evidence is available on how to treat improved posttraining clinical proficiency in the EBT substance use disorders in specific populations who may constitute most or all of those seen in particular agencies:HIV-positive patients, Native Americans, adolescents,Hispanics, or African Americans. Although it is unrea- sonable to expect all EBTs to be tested with all popu-lations, the external validity of existing studies remains a EBTs can be implemented at various levels. At the simplest level, an individual practitioner may learn and What should one do in this case? In the absence of an provide one or more particular EBTs as part of her or his EBT for the specific population one treats, it is service delivery. The use of EBTs is quite compatible with reasonable to start with EBTs that have been supported individualizing treatment to patient needs and preferences, in multiple trials with other groups. There is no particular reason to withhold EBTs from clients based on their Teague, Bartels, & Torrey, 2003). Acquiring competence in racial/ethnic background, and to do so is a form of the delivery of EBTs is likely to enhance a practitioner’s discrimination. In the nationwide Project MATCH trial, prospects for future employment and for referrals from no differences were found in the responses of African Americans, Hispanics, and Non-Hispanic Whites or of At a program level, an agency can make a policy women and men to the three treatments tested: 12-step decision to deliver preferentially, as much as possible, treatment or prevention interventions that are based on solid scientific evidence. This is a commitment to EBT in general, as distinguished from the provision of particular EBTs.
DiClemente, & Rychtarik, 1992). Native Americans, Consider these three policy statements adopted by com- however, did show significantly better outcomes with motivational enhancement therapy (& Miller, 2005). Other studies have shown no differential The Board of Directors of the South Central response of Hispanic and Non-Hispanic clients to sub- shall operate only those mental health treatments, Whereas 25 years ago there were no EBTs for substance services, and programs for which there exists use disorders, we are now blessed with a variety of evidence in the professional literature of their evidence-based approaches. One reasonable policy, then, efficacy in their application under conditions and is to provide EBTs to most people seeking treatment for circumstances similar to those existing in the substance use disorders while also meeting their other needs communities and populations served by the center.
(Adopted in 1990 by SCCMHC, operating public seem to be responding to one EBT, there are usually other mental health and substance abuse service programs good possibilities to try. Clients can therefore be given informed choices among good options. A public treatment The ChangePoint mission is to provide the most program directed by the first author (W.R.M.) offered clients effective substance abuse, mental health, and domes- a menu of EBTs. New clients were given a description of the tic violence intervention treatments using culturally options available to them and were actively involved in competent evidence-based approaches designed to choosing the treatment they preferred.
give the best possible outcomes to clients. (Adopted W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 in 1998 by ChangePoint, which operates six treat- different conclusions depending on the range of studies examined and the methodology used to do so. Systematic The Center on Alcoholism, Substance Abuse, and review and meta-analysis necessarily invoke certain Addictions (CASAA) shall operate only those treat- assumptions and are variable and fallible processes.
ments, services, and programs for which there exists Furthermore, lists of approved methods necessarily reduce evidence of efficacy in the current scientific literature.
quite complex information to a binary decision: EBT or not.
Services with unproven efficacy will be designated as Suppose, however, that we could develop at least a short experimental procedures and offered only within the list of EBTs that have been shown to be more effective than context of appropriately designed research to deter- no treatment or alternative treatments. There are still some mine their efficacy. (Adopted in 1994 by the reasonable grounds for concern. First, such lists can change University of New Mexico CASAA, operating a substantially as new evidence emerges and, thus, should large public substance abuse treatment system.) always be considered a work in progress. Second, it isimportant to remember that the absence of efficacy studies Most EBTs are treatments and not programs; they does not constitute evidence of ineffectiveness. It may be describe specific procedures but do not prescribe the de- useful, then, to compile and inform practice by enumerating tails of the many day-to-day operational decisions within a treatment methods for which there is strong evidence of program that must be made to accommodate EBTs.
Implementing EBTs may require significant changes in some danger that EBT lists could ossify research and program philosophy, procedures, and training and hiring practice and, thus, stifle innovation.
practices. In programs where EBTs are new, this involves a Social policy regarding EBTs should take into account commitment to train or retrain clinicians to deliver EBTs.
not only scientific evidence but also the feasibility of putting New hires provide a particularly important opportunity to an EBT into practice. For example, although covert increase a program’s capability to deliver EBTs. Programs sensitization (verbal aversion therapy) has a positive balance can consider specific policies for: (1) hiring, training, or of evidence for efficacy, it is an inherently unpleasant retraining of clinicians to deliver EBTs; (2) determining the therapy for counselors and clients alike and is unlikely ever clients, problems, and situations for which EBTs apply and to be widely adopted in practice. Contingency management how to proceed clinically when no EBT is available; and (3) programs that pay clients for drug-free urine are reasonably supervising and monitoring the practice of clinicians within effective but face substantial opposition from political and the program to promote and ensure appropriate use of EBTs.
public opinion. An EBT that requires individual therapy is At a treatment system level, the complexity of EBTs is difficult to deliver if programs are reimbursed only for group greater still. A treatment system may make a policy decision counseling. It is wise to anticipate such obstacles before an for differential provision and funding of an EBT among its EBT is promoted for adoption and runs into a solid wall, multiple programs and providers. Here the vexing chal- undermining the credibility of EBT implementation more lenges include (1) defining which EBTs are to be provided generally. The identification of EBTs to be implemented in or reimbursed; (2) defining and educating programs as to practice is, thus, a process that best involves both scientists when EBTs apply, how providers are to proceed when EBTs with treatment research expertise and clinicians with are unavailable, and the conditions under which clinician wisdom about feasibility in community programs.
judgment may override the use of an established EBT; and(3) determining how to monitor and audit provider practices to ensure adherence to EBT standards.
It is true that substantial gains have been made through clinical research on substance abuse treatment. It is also important to maintain an appropriate level of humility aboutour current knowledge. Clinical scientists and practitioners One of the aims of this article is to represent both clinical are both susceptible to the tendency of nondepressed people and scientific perspectives on EBT. This involves consid- to overestimate their efficacy. In general, our treatment ering not only the strengths and advantages of EBT but also interventions show small to moderate effects and repeated the limitations and emergent problems as social policy episodes of care are the norm. Substance abuse treatment yields outcomes at least comparable with those for otherchronic conditions such as diabetes, asthma, and hyper- tension (andthere are no magic bullets to cure addiction in one acute Clinical practice ought to be guided by the best science care episode. Studies appropriate to a comprehensive and available, but there are good reasons to be cautious about continuing care model may yield insights about how blessing lists of EBTs. As is evident in well- intentioned and qualified reviewers can and do reach quite Forman, Cacciola, & Kemp, in press).
W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 5.3. Cost– effectiveness of implementing EBTs or he actually does so in routine practice. It is one thing tomandate that providers deliver EBTs and quite another to Specific treatment methods are evaluated for the extent to determine through quality assurance that they are really which they improve outcomes, relative to the cost of doing so competently. Mandates without quality assurance delivery. Similar questions can be asked at program and are likely to have more effect on verbal reports about system levels regarding the cost–effectiveness of adopting EBTs. Relative to current practice, how much better wouldclient outcomes be after implementing one or more EBTs? The costs of training and supervising new EBTs, which canbe substantial, must be weighed against the degree of As discussed, an absence of research is not proof of ineffectiveness. This raises the conundrum of what to do There is a temptation for policymakers to require the about treatment practices for which limited or no scientific delivery of EBTs without providing support for the evidence is available. Some common interventions lack substantial effort and costs required to convert programs evidence of efficacy precisely because they are more and systems to new standards of practice. Comprehensive challenging to study in a rigorous manner (e.g., group intervention at a system level, although currently underway therapies and the use of 12-step programs). Other inter- in specific areas (takes time and resources ventions remain unstudied because clinical scientists have to accomplish. Helping staff learn and competently deliver not had sufficient interest in them. For example, should art even a single EBT is likely to require far more than therapy, for which there is no single outcome study, be a providing a treatment manual and a one-shot workshop reimbursable substance abuse service unless and until it is shown to be ineffective? What about age regression mandates to implement EBTs are predisposed to fail.
Even with good training and support for implementation, The FDA standard is that the burden of proof is on the the effectiveness of an EBT in practice is not ensured.
purveyor of a treatment to show that it is effective before it Because the outcomes of EBTs vary across sites and is approved for delivery. It is an unreasonable standard that populations, it is desirable to study the impact of EBTs any intervention must be proven ineffective (itself a when implemented, which converges with increased challenging task) before delivery and funding can be denied.
demands from funding sources for outcome monitoring.
At the same time, there are pressing everyday practice This, too, requires dedicated time and effort, and funders challenges with no EBTs to provide clear guidance but for may fail to provide the requisite resources to collect reliable which communities have developed intervention methods.
outcome data. Contract funds are frequently mandated to be Research should be encouraged to evaluate community- spent for direct services only, without supporting an infra- supported approaches that are widely practiced (e.g., within structure for the quality assurance and outcome evaluation a particular population) and for which outcome knowledge needed in systemic adoption of EBTs.
5.4. Quality assurance: How do we know if an EBT is being 5.6. The National Registry of Evidence-Based Programs It is easier to determine whether certain medical A variety of federal initiatives are in process or under procedures (e.g., surgery) have been properly performed development to address some of the issues in this article.
than to verify when a substance abuse provider is actually One of the most significant is the National Registry of delivering a psychosocial EBT. The delivery of even a Evidence-Based Programs and Practices (NREPP), initiated pharmacotherapy is in doubt without tests to verify in 1998 by the U.S. Substance Abuse and Mental Health medication blood levels. Without ongoing monitoring, Services Administration (SAMHSA) and currently under individual clinicians and programs only need to report that expansion. Originally focused on substance abuse preven- they are delivering EBTs and indeed may incorrectly believe tion programming, it is broadening its scope to include interventions to treat substance abuse disorders and to author (W.R.M.) once delivered a community lecture on prevent and/or treat mental illness. The vision in expanding EBTs for alcohol problems. The following week, a local the NREPP is that the system becomes a leading national treatment program listed in a newspaper advertisement the resource for contemporary and reliable information on the services that it provided—the very same list of EBTs for scientific basis and practicality of interventions to prevent which, to our knowledge, they had received no training.
and/or treat mental and addictive disorders. Plans are It is challenging to verify, from chart review alone, underway to align technical assistance with this expansion.
whether an EBT was actually delivered. Even a taped work The SAMHSA obtained public comment on its plans for sample of a clinician providing an EBT demonstrates only expanding the NREPP and will plan to relaunch the that she or he is able to deliver the treatment, not that she W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 As of this writing (August 2005), the NREPP policies available resources. Clinicians also benefit from access to and procedures are still being formulated, but treatment and new methods, and an evidence-based standard for practice prevention programs can apply online and achieve promotes self-examination and stimulates new ways to think status as model evidence-based programs ( about care and service delivery. The introduction of new modelprograms.samhsa.gov). Although we laud the effort interventions may further generate new enthusiasm and to provide the public and funding agencies with a reliable consumer’s guide to evidence-based substance abuse Busy providers and program managers cannot be services, the devil is going to be in the details. First, a list expected to digest the entire treatment outcome literature of approved practices must be developed, which involves all and come to their own conclusions about EBTs. A the abovementioned complexities of deciding how much of seemingly simple approach for encouraging or requiring which kinds of data constitute evidence.
the use of EBTs is to develop a list of treatments that are To complicate matters further, programs are to be evidence based (and, by omission, those that are not).
registered as actually providing evidence-based practices, However, the processes and criteria for arriving at an EBT apparently based on the self-report of program admin- list are by no means straightforward and different review istrators. Even individual providers themselves are not reliable reporters of their own proficiency and delivery in Neither is it clear where the responsibility and authority lie reimbursement at stake, the second-hand assurance of Retraining providers with established habits is consid- administrators regarding providers’ adherence to evidence- erably more challenging than shaping the practice of based methods is unlikely to reflect actual practice. Quality clinicians in training. In the new wave of enthusiasm for assurance measures are imaginable, albeit complex, but no EBTs, surprisingly minimal attention and priority have been such checks have been proposed for the NREPP. No list at given to ensuring that future substance abuse treatment all would be preferable to a registry that provides the professionals will be prepared to competently deliver EBTs.
public with unreliable consumer information about pro- Special attention should be given to EBTs in any training program to prepare the next generation of clinicians to workwith substance use disorders.
Perhaps the proper attitude toward EBTs is one of respect but not reverence. Evaluating scientific evidence isa complex and evolving process. There is danger that The issues discussed here are not new. More than three funders and regulators will take action prematurely, with- out good understanding of the state of the evidence and the Most clinical psychologists I know would be outraged to practical constraints inherent in implementing worthy discover that the Food and Drug Administration allowed goals. A solid evidence base for the treatment services a new drug on the market without sufficient testing, not we provide is perhaps the best defense against extinction, only of its efficacy to cure or relieve symptoms, but also and funding agencies are understandably impatient. For the of its short term side effects and the long term effects of field of substance abuse treatment to move forward, there continued use. Many of these same psychologists, should be cooperative dialogue among the stakeholders, however, do not see anything unethical about offering with EBT implementation plans developed through services to the public. . .which could not conceivably He further warned that in the absence of an evidence base for treatment, clinicians bwill find themselves restrained from the outside (as are drug companies by the FDA) as aresult of their own failure to do what ethical and scientific The preparation of this article was supported in part by Grants U10-DA015833 and U10-DA15815 from the NIDA Clinical Trials Network; Grants TI10377, TI15802, TI16005, and TI16246 from the Center for Substance As long as psychotherapies resist pressure to produce Abuse Treatment; and Grant K05-AA00133 from the scientific evidence that they work, the economic squeeze National Institute on Alcohol Abuse and Alcoholism.
will tighten. After all, if psychotherapy is really an art, itshould be supported by the National Endowment, not byMedicare. The first to face extinction will be the longer-term therapies. . .Where it ends, though, is not clear.
It makes good sense to give priority to EBTs, particularly American Society of Addiction Medicine. (2001). Patient placement within this era of fiscal austerity. We owe it to our clients to criteria for the treatment of substance-related disorders (PPC-2R).
provide the best treatment that we can offer them within W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 Arroyo, J. A., Miller, W. R., & Tonigan, J. S. (2003). The influence of Magura, S., Staines, G., Kosanke, N., Rosenblum, A., Foote, J., DeLuca, A., Hispanic ethnicity on long-term outcome in three alcohol treatment et al. (2003). Predictive validity of the ASAP patient placement criteria modalities. Journal of Studies on Alcohol, 64, 98 – 104.
for naturalistically matched vs. mismatched alcoholism patients.
Berglund, M., Thelander, S., & Jonsson, E. (Eds.). (2003). Treating alcohol American Journal on Addictions, 12, 386 – 397.
and drug abuse: An evidence-based review. Weinheim, Germany7 Mattick, R. P., & Hall, W. (1993). An outline for the management of opioid dependence: Quality assurance project. Sydney7 University of New Carroll, K. M. (1998). Treating drug dependence: Recent advances and old South Wales, National Drug Abuse Research Centre.
truths. In W. R. Miller, & N. Heather (Eds.), Treating addictive Mattick, R. P., & Jarvis, T. J. (1992). An outline for the management of behaviors (2nd ed.). New York7 Plenum Press.
alcohol problems: Quality assurance project. Sydney7 University of Carroll, K. M., Nich, C., & Rounsaville, B. J. (1995). Differential New South Wales, National Drug Abuse Research Centre.
symptom reduction in depressed cocaine abusers treated with McCrady, B. S. (2000). Alcohol use disorders and the Division 12 Task psychotherapy and pharmacotherapy. Journal of Nervous and Mental Force of the American Psychological Association. Psychology of Addictive Behaviors, 14, 267 – 276.
Drake, R. E., & Mueser, K. T. (Eds.). (1996). Dual diagnosis of major McCrady, B. S., & Ziedonis, D. (2001). American Psychiatric Association mental illness and substance use disorder: Recent research and clinical practice guidelines for substance use disorders. Behavior Therapy, 32, implications (vol. 70). San Francisco7 Jossey-Bass.
Drake, R. E., Rosenberg, S. D., Teague, G. B., Bartels, S. J., & Torrey, McGovern, M. P., & Carroll, K. M. (2003). Evidence-based practices W. C. (2003). Fundamental principles of evidence-based medicine for substance use disorders. Psychiatric Clinics of North America, 26, applies to mental health care. Psychiatric Clinics of North America, McLellan, A. T., Grisson, G. R., Zanis, D., Randall, M., Brill, P., & Essock, S. M., Goldman, H. H., Van Tosh, L., Anthony, W. A., Appell, O’Brien, C. P. (1997). Problem-service matching in addiction treatment.
C. R., Bond, G. R., et al. (2003). Evidence-based practices: Setting the Archives of General Psychiatry, 54, 730 – 735.
context and responding to concerns. Psychiatric Clinics of North McLellan, A. T., Hagan, T. A., Levine, M., Gould, F., Meyers, K., Bencivengo, M., et al. (1998). Supplemental social services improve Finney, J. W., & Monahan, S. C. (1996). The cost–effectiveness of outcomes in public addiction treatment. Addiction, 93, 1489 – 1499.
treatment for alcoholism: A second approximation. Journal of Studies McLellan, A. T., Lewis, D. C., O’Brien, C. P., & Kleber, H. D. (2000). Drug dependence, a chronic mental illness: Implications for treatment, Finney, J. W., & Moos, R. H. (2002). Psychosocial treatments for alcohol insurance, and outcomes evaluation. Journal of the American Medical use disorders. In P. E. Nathan, & J. M. Gorman (Eds.), A guide to treatments that work (2nd ed.). London7 Oxford University Press.
McLellan, A. T., McKay, J. R., Forman, R., Cacciola, J., & Kemp, J.
Gastfriend, D. (2003). Addiction treatment matching: Research foundations (in press). Reconsidering the evaluation of addiction treatment: From of the American Society of Addiction Medicine (ASAM) criteria.
retrospective follow-up to concurrent recovery monitoring. Addiction.
Binghamton, NY7 Haworth Medical Press.
McLellan, A. T., Woody, G. E., Luborsky, L., & Goehl, L. (1988). Is the Gerstein, D. R., & Harwood, H. J. (Eds.). (1990). Treating drug problems counselor an bactive ingredientQ in substance abuse rehabilitation? An (vol. 1). Washington, DC7 National Academy Press.
examination of treatment success among four counselors. Journal of Hall, G. C. (2001). Psychotherapy research with ethnic minorities: Nervous and Mental Disease, 176, 423 – 430.
Empirical, ethical, and conceptual issues. Journal of Consulting and Miller, W. R., & Hester, R. K. (1986). Inpatient alcoholism treatment: Who Clinical Psychology, 69, 502 – 510.
benefits? American Psychologist, 41, 794 – 805.
Hanson, G. R., Leshner, A. I., & Tai, B. (2002). Putting drug abuse research Miller, W. R., & Meyers, R. J. (1995). Beyond generic criteria: Reflections to use in real-life settings. Journal of Substance Abuse Treatment, 23, on life after clinical science wins. Clinical Science, 4 – 6.
Miller, W. R., & Mount, K. A. (2001). A small study of training in Hettema, J., Steele, J., & Miller, W. R. (2005). Motivational interviewing.
motivational interviewing: Does one workshop change clinician and Annual Review of Clinical Psychology, 1, 91 – 111.
client behavior? Behavioural and Cognitive Psychotherapy, 29, Hubbard, R. L., Marsden, M. E., Rachal, J. V., Harwood, H. J., Cavanaugh, E. R., & Ginzburg, H. M. (1989). Drug abuse treatment: Miller, W. R., & Walker, D. D. (1997). Should there be aromatherapy for A national study of effectiveness. Chapel Hill, NC7 University of North addiction? Addiction, 92, 486 – 487.
Miller, W. R., & Weisner, C. (Eds.). (2002). Changing substance abuse Kadden, R., Carroll, K., Donovan, D., Cooney, N., Monti, P., Abrams, D., through health and social systems. New York7 Kluwer/Plenum.
et al. (1992). Cognitive–behavioral Coping Skills Therapy manual Miller, W. R., & Wilbourne, P. L. (2002). Mesa Grande: A methodological (vol. 3). Rockville, MD7 National Institute on Alcohol Abuse and analysis of clinical trials of treatments for alcohol use disorders.
Kranzler, H. R., Burleson, J. A., Del Boca, F. K., Babor, T. F., Korner, P., Miller, W. R., Wilbourne, P. L., & Hettema, J. (2003). What works? A Brown, J. M., et al. (1994). Buspirone treatment of anxious alcoholics: A summary of alcohol treatment outcome research. In R. K. Hester, & W. R.
placebo-controlled trial. Archives of General Psychiatry, 51, 720 – 731.
Miller (Eds.), Handbook of alcoholism treatment approaches: Effective Krauthammer, C. (1985, Dec. 29). Psychotherapy is dying from dilution.
alternatives (3rd ed., pp. 13–63). Boston, MA7 Allyn & Bacon.
Miller, W. R., Yahne, C. E., Moyers, T. B., Martinez, J., & Pirritano, M.
Linehan, M. M., Dimeff, L. A., Reynolds, S. K., Comtois, K. A., Shaw- (2004). A randomized trial of methods to help clinicians learn Welch, S., Heagerty, P., et al. (2002). Dialectical behavior therapy motivational interviewing. Journal of Consulting and Clinical Psychol- versus comprehensive validation plus 12-step for the treatment of opioid dependent women meeting criteria for borderline personality Miller, W. R., Zweben, A., DiClemente, C. C., & Rychtarik, R. C. (1992).
disorder. Drug and Alcohol Dependence, 67, 13 – 26.
Motivational Enhancement Therapy manual: A clinical research guide Ling, W., Farrell, M., & Ali, R. (2004). Cochrane systematic reviews: Time for therapists treating individuals with alcohol abuse and dependence for an introduction and appraisal. Drug and Alcohol Dependence, 73, (Vol. Project MATCH Monograph Series, Vol. 2). Rockville, MD7 National Institute on Alcohol Abuse and Alcoholism.
Luborsky, L., McLellan, A. T., Woody, G. E., O’Brien, C. P., & Auerbach, Minkoff, K. (2001). Developing standards of care for individuals with co- A. (1985). Therapist success and its determinants. Archives of General occurring psychiatric and substance use disorders. Psychiatric Services, W.R. Miller et al. / Journal of Substance Abuse Treatment 29 (2005) 267 – 276 Najavits, L. M. (2001). Seeking safety: A treatment manual for PTSD and Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and substance abuse. New York7 Guilford Press.
quasi-experimental designs for generalized causal inference. Boston7 Najavits, L. M., & Weiss, R. D. (1994). Variations in therapist effectiveness in the treatment of patients with substance use disorders: An empirical Sorensen, J. L., & Midkiff, E. E. (2002). Bridging the gap between research and drug abuse treatment. Journal of Psychoactive Drugs, 32, 379 – 382.
National Institute on Drug Abuse. (1999). Principles of drug addiction Sorensen, J. L., Rawson, R. A., Guydish, J., & Zweben, J. E. (Eds.). (2003).
treatment: A research based guide. Bethesda, MD7 National Institute on Drug abuse treatment through collaboration: Practice and research partnerships that work. Washington, DC7 American Psychological Nowinski, J., Baker, S., & Carroll, K. M. (1992). Twelve Step Facilitation Therapy manual: A clinical research guide for therapists treating Steinberg, E. P., & Luce, B. R. (2005). Evidence based? Caveat emptor! individuals with alcohol abuse and dependence. Rockville, MD7 Health Affairs (Millwood), 24, 80 – 92.
National Institute on Alcohol Abuse and Alcoholism.
Villanueva, M., Tonigan, J. S., & Miller, W. R. (2005). A retrospective study of client–treatment matching: Differential treatment response of Native three treatments for alcohol problems. Psychotherapy Research, 8, American clients in Project MATCH. (submitted for publication).
White, W. L. (2005). Treatment works: Is it time for a new slogan? Rawson, R. A. (1996). Is psychotherapy effective for substance abusers? In Addiction Professional, 3, 22 – 28.
A. M. Washton (Ed.), Psychotherapy and substance abuse: A Willenbring, M. L., Hagedorn, H. J., Postier, A. C., & Kenny, M. (2004).
practitioner’s handbook (pp. 55 – 75). New York7 Guilford Press.
Variations in evidence-based clinical practices in nine United States Rotter, J. B. (1971). On the evaluation of methods of intervening in other Veterans Administration opioid agonist therapy clinics. Drug and people’s lives. The Clinical Psychologist, 24, 1 – 2.

Source: http://www.bldtf.ie/Portals/0/Documents/Miller05%20why%20when%20what%20where%20how.pdf