Behavioral alcohol treatment research advances: Barriers to utilization

Behavioral alcohol treatment research advances: Barriers to utilization

Adv. Behov. Res. The-r. Vol. 9, pp. 145-164, 1987. Printed in Great Britain. All rights reserved. 01466402/87 SO.00+.50 Copyright 91987 Pergamon Jour...

1MB Sizes 2 Downloads 22 Views

Recommend Documents

No documents
Adv. Behov. Res. The-r. Vol. 9, pp. 145-164, 1987. Printed in Great Britain. All rights reserved.

01466402/87 SO.00+.50 Copyright 91987 Pergamon Journals Ltd.

BEHAVIORAL ALCOHOL TREATMENT RESEARCH ADVANCES: BARRIERS TO UTILIZATION William R. Miller University of New Mexico, USA Abstract - Treatment research in the alcoholism field has thus far had relatively little impact on public policy and professional practices. U.S. alcoholism treatment programming does not reflect advances in knowledge with regard to the effectiveness of specific approaches, treatment length and setting, or matching of clients with treatments. Four characteristics of decision-making are considered as possible causes: inertia, ignorance, selective inattention, and incentives. Four corresponding characteristics of the behavior of researchers are also discussed: failure to use effective attitude change procedures, failure to disseminate research findings, intolerance for disconfirming data, and failure to address the needs of decision-makers. Any program of research should include a specific plan to disseminate and to promote utilization of findings. Six strategies for increasing utilization are suggested: setting clear goals, choosing optimal dissemination routes, tailoring communications to the audience, providing implementation support, following up on initial communications, and teaching utilization skills. Research on the effectiveness of such utilization strategies is needed.

THE PROBLEM A common frustration among behavioral scientists is the seeming lack of impact our research has had on public policy, standard treatment practices, and social change in general. The completion of a typical clinical research project requires the devotion of years of one’s life to conceptualizing, proposing, preparing, conducting, analyzing, and writing. The most common result is an article in a professional journal, published perhaps five years after the project’s inception, read by a few dozen colleagues, having no discernible effect on professional attitudes or practices at even the local level. Regarding behavioral contributions to the alcoholism field, McCrady (1986) has observed: “Our research work is impeccable; our approaches to treatment as creative and effective as any other approach; our theory development careful and thoughtful. Unfortunately, our work is also invisible” (pp. 173-174).

Alcoholism Treatment Procedures

The examples are legion. Reid Hester and I have for seven years been reviewing the immense literature on alcoholism treatment outcome. We have 145


W. R. Miller

read, during these years, well over a thousand articles, chapters, and books on the efficacy of various treatments. From the hundreds of empirical studies, we have been able to derive several practical conclusions. First is that effective treatment procedures for alcoholism are available. Several treatment approaches have emerged as promising strategies, based on a reasonable amount of empirical support in properly controlled clinical trials (Miller and Hester, 1986a). This list includes aversion therapies, behavioral self-control training, the community reinforcement approach, marital and family therapy (either behavioral or systems), social skills training, and stress management. The list of components that constitute standard practice in U.S. alcoholism treatment, by contrast, is quite different: Alcoholics Anonymous, alcoholism education, confrontation, disulfiram, group therapy, and individual counseling. What these ‘standard’ treatments have in common is an almost complete lack of scientific support for their effectiveness. Controlled clinical trials have been reported for all of these methods, and the results have been predominantly negative. In fact, significant detrimental impact has been reported in controlled evaluations of some of these methods (Miller and Hester, 1980, 1986a). Particularly striking is the lack of overlap between these lists of scientifically supported versus utilized treatment methods. Therapeutic approaches that are reasonably well supported by scientific research share an almost total neglect by the American alcoholism treatment community. Typical treatment programs simply are not using those approaches which have been substantiated by controlled clinical trials. Instead, despite hundreds of studies conducted over the past seven decades, alcoholism treatment continues to cling primarily to scientifically unproven procedures that conform to the dominant American disease-model ideology (Miller, 1986).

Behavioral Self-Control Training

A specific example is research on behavioral self-control training (BSCT) methods for helping problem drinkers reduce their alcohol consumption to a moderate and nonproblematic level. Broadly defined, BSCT consists of teaching clients self-management strategies for modifying their own alcohol consumption (Kanfer, 1986; Miller and Mufioz, 1982). To date, at least 23 controlled and comparative studies have reported positive changes following treatments comprising various forms of BSCT. The tindings of these studies bear striking resemblance to one another, even though they have been conducted by 12 research teams working in six different nations including Australia (Caddy and Lovibond, 1976; Lovibond, 1975; Lovibond and Caddy, 1970), Canada (Alden, 1978; Sanchez-Craig, 1980; Sanchez-Craig et

Behavioral Alcohol Treatment Research Advances


al., 1984), New Zealand (Brown, 1980), Norway (Berg and Skutle, 1986) Scotland (Robertson et al., 1986), and the United States (Baker et al., 1975; Buck and Miller, 1981; Carpenter et al., 1985; Coghlan, 1979; Hedberg and Campbell, 1974; Miller, 1978; Miller et al., 1981; Miller and Taylor, 1980; Miller et al., 1980; Pomerleau et al., 1978; Schaefer, 1972; Sobell and Sobell, 1973; Vogler et al., 1975; Vogler et al., 1977a; Vogler et al., 1977b). Three other uncontrolled studies have reported little successin establishing stable nonproblem drinking outcomes among chronic alcoholics (Czypionka and Demel, 1976; Ewing and Rouse, 1976; Maxwell et al., 1974). The only controlled study to report negative findings (Foy et al., 1984) documented significantly poorer functioning in a BSCT-trained group relative to an alternative intervention, although this difference proved transient, disappearing after six months. A fifth negative report is the much-publicized and controversial critique of the Sobell and Sobell (1973) study (Pendery et al., 1982), which described relatively unfavorable long-term outcomes among patients receiving an extensive BSCT program. Pendery and her colleagues admitted, however, that controls receiving traditional treatment had fared at least as poorly. These five negative reports must be weighed against the 23 positive controlled and comparative evaluations of BSCT methods. The methodology of the positive studies has been unusually rigorous, relative to most research on alcoholism treatment (Miller and Hester, 1980). All 23 studies included control or comparison groups, 22 of which employed a randomization or matching procedure. Most included at least 12 months of follow-up, locating over 80% of clients. Most interviewed collaterals to corroborate client selfreport. Eight of the studies reported significantly better outcome in a BSCTtreated group relative to a comparison group receiving an alternative intervention (Brown, 1980; Coghlan, 1979; Lovibond, 1975; Schaefer, 1972; Sobell and Sobell, 1973), a more minimal intervention (Alden, 1978; Robertson et al., 1986) or no intervention (Buck and Miller, 1981). To be sure, each of the positive and negative studies could be criticized on methodological grounds (e.g. Caddy, 1975; Carey and Maisto, 1985; Heather and Robertson, 1983; Pendery et al., 1982; Sobell and Sobell, 1978, 1984). Nevertheless this group of studies constitutes, without question, the largest body of confirmatory findings for any existing alcoholism intervention procedure (Miller, 1983a). By comparison, the empirical support for other treatment approaches is at best modest, and for American ‘standard practice’ procedures, is almost completely lacking. Yet BSCT methods are employed by only a handful of clinics, and are typically reviled by alcoholism treatment personnel. Even distinguished scientists in the alcoholism field continue to assert that moderation training lacks adequate support, often while promoting the use of other, less validated approaches (e.g. Nathan, 1986; Vaillant, 1983). JABRT 9:2/3-G


W. R. Miller

Preventive Interventions The prevention literature similarly reflects a lack of empirical support for widely implemented strategies. Alcohol and drug education, for example, has failed dismally as a strategy for altering addictive behaviors. The vast majority of evaluative studies show some effect on knowledge, minimal effect on attitudes, and no beneficial effect on behavior (Hester and Miller, in press; Kinder et al., 1980). Several evaluations, in fact, have demonstrated significantly increasedalcohol and drug use or abuse among those exposed to a purportedly ‘preventive’ education program (Hagen et al., 1978; Salzberg and Klingberg, 1983; Stuart, 1974). In sum, treatment and prevention approaches that are now widely regarded as ‘standard practice’ and ‘state of the art’ in the U.S. alcohol field are grossly lacking in scientific support. From a review of the available controlled literature (Miller and Hester, 1986a), these standard approaches would be expected to have, at best, little or no beneficial impact on drinking behavior and its sequelae, and perhaps even a detrimental effect. Other treatment methods which have received encouragingly consistent support in controlled clinical trials remain largely unused. Treatment Setting and Intensity A second consistent trend that we have observed in the alcoholism treatment literature is a lack of superior outcomes associated with longer, more intensive, residential, and more expensive interventions. Making treatment longer, offering it in an inpatient setting, or providing more intensive staff contact simply have not been found to increase either the short-term or the long-term impact of alcoholism treatment. Randomized, controlled comparisons of traditional treatment in inpatient versus outpatient programs, longer versus shorter residential care, inpatient versus day treatment, day treatment versus outpatient, or intensive versus minimal intervention have all consistently failed to provide evidence of any absolute difference in effectiveness (Miller and Hester, 1986b). Impervious to three decades of such findings, the treatment industry continues to proliferate and promote residential treatment centers which charge from several thousand dollars to more than $50,000 for a few weeks’ stay. Most third-party payers continue to provide differentially favorable reimbursement rates for these more expensive forms of treatment, unnecessarily inflating health care costs by billions of dollars per year. For three years in a row, the Department of Health and Human Services has extended an exemption of alcoholism treatment from the diagnosis-related group (DRG) reimbursement formula system, again failing on October 1,

Behavioral Alcohol Treatment Research Advances


1986, to set limits on what providers may charge. In the absence of constraint, alcoholism treatment in the United States has become a ten billion dollar industry (Miller and Hester, 1986b). Matching of Clients with Optimal Treatments It is commonplace to affirm that alcoholism treatment should be individualized, tailored to each person’s needs. Although many more studies are needed, there is already a substantial empirical literature on clienttreatment matching (Miller and Hester, 1986c). Predictor studies point toward characteristics of positive responders to various approaches. Successful long-term abstainers, for example, tend to cluster toward the upper end on measures of severity, reporting more alcohol-related problems and symptoms. Those who succeed in maintaining long-term problem-free drinking, on the other hand, tend to occupy the opposite end of the severity continuum (Miller, 1983a; Miller and Hester, 1986c). Stronger evidence is provided by studies contrasting the effects of two or more different treatments, and studying differential predictors of prognosis within each modality. Such studies have provided support for the matching hypothesis: that clients who are appropriately matched to interventions will show a higher rate of favorable outcomes than those who are unmatched or mismatched, based on their personal characteristics. Differential efficacy of treatments has been demonstrated with matching procedures based on problem severity (McLellan et al., 1983; Orford et al., 1976), cognitive style (Karp et al., 1970; McLachlan, 1972, 1974; Thornton et al., 1977, 1981), self-esteem (Annis and Chan, 1983), and marital status (Azrin et al., 1982). There is also evidence that clients who participate in the selection of their own treatment, having options from which to choose, show more satisfaction, compliance, persistence, and success in treatment (Miller, 1985; Miller and Hester, 1986c). Actual practice in current alcoholism treatment programs, however, is to offer few choices and instead to prescribe a relatively standard set of procedures for all alcoholic clients (Costello, 1975; Orford and Hawker, 1974). Differential diagnosis and matching is the rare exception rather than a norm, with each program tending to diagnose a need for the particular services it offers (Hansen and Emrick, 1983). Optimal matching of clients to alternative treatments remains more an unfulfilled promise than an established practice. POSSIBLE CAUSES OF THE PROBLEM How can we account for what seems to be a gross failure to implement, even a disregard for, the findings of empirical research in general, and behavioral


W. R. Miller

research in particular? Why has there no been more public impact of these hundreds of clinical trials, conducted at great public expense, to which so many scientists have devoted substantial proportions of their lives and efforts? Inasmuch as causal attributions have traditionally been classified in two categories, internal and external, I shall propose a variety of hypotheses falling within these two general classes. External Attributions Inertia There is great comfort in the status quo. It is difficult to depart from a well-practiced habit, even in favor of an alternative that offers substantial benefits. This is, after all, the central common characteristic of the addictive behaviors themselves. The immediate, albeit partial, reinforcement received from the present habit overrides its longer-term adverse consequences (particularly when it is ofhers who are adversely affected). The promise of greater reinforcement from changed behavior is not sufficient to persuade the individual to abandon his or her present habit. At a social level, this same phenomenon is observable in our nation’s reluctance to convert to the metric system, and humanity’s enduring propensity to build and then use against each other an ever more deadly array of weapons. From this perspective, current entrenched alcoholism treatment practices can be understood as an addictive behavior, powerfully reinforced by short-term incentives but apparently unresponsive to long-term adverse consequences. Reinforced habit is not the only factor promoting inertia. Another is the powerful pull of conformity. Decades of social psychological research have demonstrated the power of conformity to a consensus or an authority. Otherwise rational people can be induced to swear that a longer line is really the shorter line in a pair (Asch, 195 l), to deliver potentially fatal shocks to an innocent victim (Milgram, 1963, 1974), and to behave with brutality toward peers (Haney and Zimbardo, 1977). Deviation from a widely shared consensus is difficult, and causes one to question one’s own judgment rather than the consensus itself (Asch, 195 1). I have observed this in some of my own colleagues and students, who upon completion of training have entered a traditional alcoholism treatment network. Soundly trained in behavioral science and research methodology, they rapidly ‘go native’, affirming and justifying unsubstantiated principles and practices. The force of this conformity pressure can, at times, approximate to that of a religious conversion experience (Lovern, 1982). A systems example of conformity pressure is found in the recent decision of the Raleigh Hills hospitals to abandon chemical aversion therapy as a component of their treatment. This was done despite the fact that the Raleigh Hills program had been ‘devoted

Behavioral Alcohol Treatment Research Advances


exclusively’ to aversion therapy from its very inception, and their own data and prior reports had pointed to a higher success rate than is typical for traditional treatment programs (Wiens et al., 1976). The decision was made in spite of, rather than because of, their treatment outcome data, apparently in response to marketing pressures to bring the program into closer conformity with standard treatment practices. Ironically, social accountability procedures that are intended to protect public welfare can also stifle the implementation of new research knowledge. Malpractice is commonly judged against what a profession defines as ‘standard practice’. Deviation from standard practice, then, can place one at greater risk of liability for untoward outcomes. Our local Albuquerque affiliate of the National Council on Alcoholism, for example, operates a drunk driver screening program for the metropolitan court system, evaluating thousands of DWI offenders per year and referring them to treatment services. During the first two years of its operation, this program included on its referral list several programs offering alternative treatment goals of moderation or abstinence. This is clearly justifiable, in that some of the only controlled studies reporting a specific positive treatment effect with a DWI popultion have been evaluations of BSCT interventions (Brown, 1980; Coghlan, 1979; Lovibond, 1975). In response to political pressure, however, the program later announced a policy decision to discontinue referrals to any program offering a moderation goal. The rationale provided was that the program could suffer a liability suit for untoward outcomes following referral to a moderation-oriented program. In this reasoning, if one advises abstinence and the client fails to follow treatment advice, it is the client’s responsibility. If one teaches moderation, however, and the client fails to follow treatment advice, then the program is responsible. Absurd as this argument may be, it is conceivable because of the principle of ‘standard practice’. Adhering to widely accepted treatment procedures, no matter how ineffective, provides some degree of protection. Departing from this inertia to implement less accepted (albeit well supported) treatment methods affords a degree of risk. Ignorance

A second possible hypothesis regarding the unresponsiveness of the treatment field to research findings is ignorance. The reading of research journals is, needless to say, not normative behavior among alcoholism counselors. The language and methodology of scientific research are unfamiliar to many if not most service providers. The content of continuing education for alcoholism personnel seldom provides a representative presentation of current research findings. New research knowledge may not be implemented, then, simply because it is not known by or interpreted to those who could implement it. The responsibility for this could be placed with


W. R. Miller

service providers and those who educate them, for failing to keep up with research developments. Selective inattention A third plausible explanation of intransigence toward research data assumes that people are exposed to research findings, but avoid or reject them through biases inherent in information processing. Failure to respond to new data could thus be understood as a manifestation of normal human cognitive processes. At a social policy level, these cognitive biases can have costly implications. A failure to shift policy-making paradigms in response to data “constitutes the wasting of scarce opportunities to try to influence futures while engaging in nonsense and counterproductive efforts to improve incrementally a rapidly sinking curve” (Dror, 1986, p. 144). There is a clear tendency for humans in general and policy makers in particular to maintain and defend their own current views and to ignore or discount contrary information (Dror, 1986). This phenomenon of selective inattention has been variously discussed as the avoidance of cognitive dissonance (Festinger, 1957), assumption drag (Ascher, 1978), and illusory correlation (Chapman and Chapman, 1967, 1969). Research in cognitive psychology has clarified common errors in human information processing and decision-making, including a tendency to place undue weight on early input, small samples, and salient examples (Tversky and Kahnemann, 1974). Feedback constraints can also be built into a treatment system, greatly biasing the perceptions of personnel. A clear example is that we all see each other’s treatment failures. Many who present themselves for treatment have already tried alternative approaches without success. This easily reinforces the notion that other treatment programs are relatively ineffective, since their successes are seldom encountered. Biases may also be built in such that personnel have contact only (or predominantly) with treatment successes. An example of this is the not uncommon practice of offering free aftercare or ‘booster’ sessions to any client who has remained abstinent, while announcing in advance that any client who has resumed drinking will not receive such services. Relapsed clients then are unlikely to return to the treatment center. Processes such as these can create selective inattention at a systems level. The overt biases of treatment personnel may similarly constrain the information volunteered by clients. In our long-range follow-up efforts, we have commonly heard reports from ‘abstinent’ clients of occasional drinking that has not been reported to their therapists, family, or A.A. colleagues. Fox (1967) once asserted, “Among my own approximately 3,000 patients not one has been able to achieve [moderate drinking], although almost every one of them has tried to” (p. 777). Given the nearly universal observation of at least a few moderate nonproblem drinking outcomes within any treated population

Behavioral Alcohol Treatment Research Advances


of alcoholics, even by avowed skeptics (Heather and Robertson, 1981; Helzer et al., 1985; Pendery et al., 1982; Polich et al., 1981) one wonders how a professional could encounter 3,000 consecutive cases without witnessing a single instance of controlled drinking. Such a statement testifies less to the nature of the treated population, than to the inherent or motivated constraints on the professional’s perception. This leads us to one final possible class of causes for the implementation problem. Incentives

A fourth possible cause for failure of research implementation is incentivemotivated avoidance, rejection, or suppression of new data. One such motivation is a form of denial: the desire to avoid accepting that what one has been doing is less than optimal. Beyer and Trite (1982) observed that “Resistance to change of any kind is a common phenomenon - perhaps because accepting change seems to discredit what people did before” (p. 608). Broad and Wade (1982) related the story of Dr Ignaz Semmelweis, a physician who was greatly concerned with the disease of ‘childbed fever’, which was claiming the lives of up to 30% of women and newborns in European obstetric hospitals. Hypothesizing that it resulted from contamination during obstetric examinations, Semmelweis clearly demonstrated that mortalities could be reduced dramatically if the physician would simply wash his or her hands in a chlorine solution. Armed with convincing data, he set out to reform obstetric practice and prevent the needless deaths of many women and infants. To his amazement, he found his views rejected and his colleagues unwilling to wash. Becoming increasingly impassioned, he was dismissed from his job and branded as a fanatic. Ultimately, his colleagues had him confined to an insane asylum, where he died within two weeks. It would not be until three decades (and presumably thousands of unnecessary deaths) later that Lister and Pasteur would succeed in communicating the same message, and physicians would begin washing their hands. The data were understandably difficult to accept. To admit their truth was to acknowledge that one had caused scores, perhaps hundreds, of needless deaths by failing (albeit unknowingly) to wash one’s hands. This in turn suggests another incentive for rejecting or suppressing new data: to avoid their potential embarrassing, discrediting, or reforming effects. Commissioned to evaluate a county drunk driver intervention program, Marquis (personal communication, November, 1976) succeeded in establishing a random assignment procedure whereby some offenders were sent to one of two ‘preventive’ education programs while others received only probation. The results of his evaluation revealed no significant differences in recividism rate among groups, with the highest rate of repeat offenses occurring in the standard county education program. The results of this study were never


W. R. Miller

published, however, and more than a decade later the county continues to operate its mandatory education program for drunk drivers much as before. Selective publication and representation has constrained the communication of research findings through the National Institute on Alcohol Abuse and Alcoholism (NIAAA), which publishes Alcohol and Health, a periodic report to Congress that is also widely disseminated to professionals. The NIAAA commissioned Peter Nathan, an eminent alcoholism scholar and scientist, to draft the chapter on prevention research for its fifth congressional report, and three years later to write the chapter on treatment research for the sixth (1987) report. The conclusions drawn by Nathan were apparently found to be unacceptable, for in both cases the chapters were extensively rewritten, deleting what Nathan (personal communication, October 23,1986) regarded to be key studies and their implications. Similarly, the Institute commissioned the writing of several handbooks providing guidelines for teaching alcohol and alcoholism knowledge within specific professions. Volumes for several other professions have been released, but an excellent handbook for the training of psychologists (Sobell and Sobell, 1982) was suppressed although it had already been printed, apparently because of the controversy surrounding their 1973 study (Pendery et al., 1982). Psychology remains without a corresponding handbook to guide the training of psychologists in new knowledge within the alcohol field. Utilization of available knowledge, then, can be constrained by selective censorship of research through central channels by which it is communicated. Finally, economic incentives can provide a substantial motivation for ignoring new research findings. The above-discussed research showing no advantage for intensive residential treatment clearly contravenes the vested financial interests of a now immense alcoholism treatment industry. The U.S. Congress commissioned a special study of the relative effectiveness of alcoholism treatment settings, which was completed by the Office of Technology Assessment (1983). Although the report clearly recommended against continued use of expensive residential approaches, its findings have yet to find expression in federal health care reimbursement policies (Miller and Hester, 1986b). Internal Attributions Another way to conceive of the etiology of this implementation deficit is to look not ‘out there’, but to our own behavior as researchers. If for no other reason, it is worthwhile to consider such internal attributions because they are much more readily amenable to our own change efforts. The four causal categories to be discussed here correspond directly to the four external attributions addressed above.







Failure to use effective attitude change procedures

I have argued elsewhere (Miller, 1983b, 1985) that motivation versus resistance in alcoholism treatment is not a client problem, but rather a therapist problem. Denial and resistance can be evoked, or at the very least are ineffectively addressed, by the persuasion procedures that we customarily employ in dealing with alcoholics. A similar argument could be made that our accustomed approaches to implementation are (judging from the results) egregiously ineffectual. Motivation and attitude change are, after all, psychological phenomena that have engendered mountains of theoretical and empirical work. To those of us working in the addictive behaviors, overcoming inertia is a familiar challenge in our clinical work. How have we done in applying our psychological expertise toward planned implementation? The answer, it seems to me, is obvious. Most behavioral scientists seem content to conduct their research and publish it through traditional scientific journals. The implicit assumption seems to be that once a piece of research hits the refereed journals, implementation follows automatically. Were we to apply this same reasoning to our clinical work, we would still be seeking to instill insight and assuming that behavior change would inevitably follow. The testimony of those who observe utilization processes, however, is that implementation occurs most readily (and sometimes only) when it is inrentionallyplanned(Beyer and Trite, 1982). Policy decisions are guided only to a limited extent by available data, and rationality cannot be assumed (Danziger, 1983). A salient and belief-congruent anecdote may exert more persuasive power than a bevy of methodologically sophisticated experiments, particularly if the latter are ineffectively or inaccurately interpreted to the decision-makers. Models for effective implementation planning may be found right in our own back yard. Rogers (1975), for example, has proposed a ‘protection motivation’ theory of attitude and behavior change that convincingly organizes many available data. According to his model, behavior change occurs when: (1) the decision-maker perceives a significant probability and severity of risk (or gain); and (2) the decision-maker perceives the availability of acceptable alternative behaviors that will accomplish desired changes. Alternative and overlapping expressions of the same phenomena can be found in Bandura’s (1982) self-efficacy theory, research on goal-setting (Locke et al., 1981), Leventhal’s (1970, 1971) work on fear persuasion, Deci’s (1975) research on intrinsic motivation, Miller’s (1983b, 1985) conceptualization of motivational interviewing, and methods commonly employed in sales and marketing. Social psychologists have amassed a body of knowledge on the vectors and determinants of decision-making and commitment processes (Janis and Mann, 1977). Strategies derived from this substantial base of


W. R. Miller

theory and research seem logically preferable to neglect, confrontation, or exhortation in pursuing the types of behavior change required to implement research findings in public policy and professional practices. Failure to disseminate research findings

General ignorance of research findings can be attributed in part to ineffective efforts of scientists to disseminate and interpret their own work to decision-makers. A scientist is the most likely and often the most effective advocate and interpreter for his or her own work (Beyer and Trite, 1982). It is a reasonable assumption that one’s own work will not be widely disseminated, utilized, or implemented without some specific, planned effort to aid in the process. The research journals are functionally inaccessible to many treatment providers and most decision-makers. The jargon is thick, the reading laborious, the statistics confusing, and the format distracting to those who are not trained consumers or contributors of scientific publications. It is simply unrealistic to expect that those who make and implement treatment policy will be regular readers of scientific journals. Language barriers, too, constrain the communication of research findings across international boundaries. Effective implementation of research requires as a prerequisite the formulation of clear goals and strategies for knowledge dissemination. Intolerance for disconforming data

Disbelief and rejection of findings is not limited to the potential consumers of research. Behavioral science literature in general, and the alcoholism literature in particular, is replete with examples of scientists who accurately report their findings, but then fail to accept their implications. Disconformation of an hypothesis is equally important to confirmation, and both represent good science. It is equally vital to map the promising paths and the cul-de-sacs. If we refuse to take our own data seriously, can we expect others to do so? Scientifically trained professionals are clearly subject to biases of illusory correlation, selective attention, and defensive reinterpretation of facts. Examples are not difficult to find within the field of alcoholism research. Woititz (1976) found, contrary to her expectations, that children of alcoholics who were attending Alateen meetings showed significantly lower scores on a self-esteem scale, relative to children of alcoholics not attending Alateen. Rather than accepting this as a disconfirmation of her hypothesis, she maintained that the non-Alateen children did not really manifest higher self-esteem, but were simply in greater ‘denial’. Similarly, the unexpected finding that alcoholics show internal locus of control scores has sometmes been interpreted as ‘defensive internality’, a defense against their ‘true’ lack of control (Henricksen, 1976). In research, as in clinical settings, the concept of

Behavioral Alcohol Treatment Research Advances


‘denial’ provides a convenient tool for explaining away the absence of predicted pathology. In my own research, I found it difficult to believe an early finding that a minimal self-directed intervention was as effective as therapist-directed treatment (Miller et al., 1981). I spent four years conducting three controlled replications which produced the same finding. Once I accepted this as a phenomenon, it led me in an entirely new and intriguing research direction, and it substantially altered some of my prior assumptions about treatment. Replication is an important process, and the result has been positive, but I wonder now if I might not have saved myself several years of verifying the same blind alley. Nonsignificant differences and disconfirmations of experimental hypotheses (especially of cherished ones) are important and informative elements of scientific inquiry. Intolerance for inconsistent and embarrassing data, then, is by no means uniquely characteristic of the potential consumers of scientific knowledge. The behavior of reseachers often reflects a similar reluctance to accept the implications of evidence. Failure to address the needs of decision-makers

In order to impact decision-makers and override the natural biases inherent in existing systems, new knowledge should be presented in a manner that addresses the needs and operation patterns of decision-makers. Interpretations of research findings should address questions and problems that are of interest to decision-makers. Clarity and brevity are crucial in an era of massive information flow. Furthermore, the presentation should propose practical and alternative options derived from the research findings, mindful of the powerful incentives that influence decision-making. The accustomed language, scientific formats, and prolonged discussions characteristic of research writing are unlikely to communicate effectively with those who make and carry out treatment policy. It is senselessto decry a lack of utilization if we do not present our research results in language that is comprehensible, and in a way that addresses the needs and interests of those whom we hope will utilize our findings. SOME PRACTICAL



Having said this, it would be hypocritical for me to conclude w&out proposing some clear and practical alternatives for us to follow, as behavioral scientists, in promoting the dissemination and utilization of reseach data. A general assumption underlying all of these recommendations is: anyprogram of research should include a speciJic plan to disseminate and to promote


W. R. Miller

utilization offindings. Perhaps the most common lack in behavioral science research has been the absence of any considered plan (other than publication in a research journal) to engender utilization. Setting Clear Goals

A first step in developing a utilization plan is to set clear goals. To whom should the results be disseminated? What audiences would have a natural interest in the findings? What specific action(s) follow from the findings, and who are the decision-makers in a position to implement such actions? Some of the possible audiences to reach with the findings of alcoholism treatment research include: alcoholism counselors, psychologists working with addictive behaviors, primary health care professionals, pastoral counselors, authors of major textbooks, insurance company executives, congressional and legislative representatives, employee assistance program counselors, family practitioners, judges who handle alcohol-related offenses, state planners for health and human services, alcoholism information centers, other alcoholism treatment researchers, administrators of alcoholism treatment centers, and university professors training new professionals in psychology, medicine, social work, nursing, counseling, public administration, law, etc.

Choosing Optimal Dissemination Routes

Each of these potential audiences is served by a few major information sources. Many of these are ‘trade’ journals, for which researchers are typically neither readers nor contributors. The Alcoholism Treatment Quarterly and the Journal of Substance Abuse Treatment, for example, have high readerships among front-line alcoholism treatment personnel. Employee assistance program personnel are most likely to read publications such as the EAP Digest. To reach a specific audience, it is sensible to publish in their most-read trade periodicals. These publications require a different style of writing from that appropriate to research journals. The introduction-methods-resultsdiscussion format is seldom applicable. A single article in a trade periodical is likely to reach an audience that is both larger and more in a position to implement findings, relative to the readers of research journals. Other possible dissemination routes include the annual meetings or mailing lists of organizations such as the Association of Labor-Management Administrators and Consultants on Alcoholism, the National Council on Alcoholism, and the American Personnel and Guidance Association. Specific selected mailings may be planned to reach key textbook authors, legislators, federal officials, judges, or state and local decision-makers.

Behavioral Alcohol Treatment Research Advances


Another much-neglected dissemination channel is through foreign languages. Collaborations with foreign colleagues can result in original articles, interpreting important research findings in the language and style of another nation. Although many university-trained professionals in other nations can read English, many cannot or do not make use of English language journals. Tailoring Communications Communicating with decision-makers and front-line personnel usually requires a marked departure from our accustomed styles of writing and speaking. Most likely to be read is a simple, clear, and brief report, preferably with a one-page summary. Quantitative analyses are best kept to a minimum, perhaps with references to more detailed reports. Qualitative data, illustrative examples, and practical applications are persuasive. Particularly effective are presentations that raise a question or problem of interest to the audience, and then apply research findings toward possible alternative solutions.

Providing Implementation


Assessment and treatment approaches are more likely to be implemented if support materials are available. These include resources such as therapist or client handbooks, demonstration videotapes or audiotapes, computer software, training workshops, and consultation services. Often such materials have been developed, at least in rudimentary form, in the process of conducting treatment research. Made more widely available, these can encourage and enable others to implement developed procedures.

Following Up Utilization often requires more than a single contact. The target audience may take greater notice of a communication if it requires some response from them, or will result in further exchanges. It is wise, for example, when writing to an elected representative, to ask a specific question or request a particular response to the communication. This requires additional attention and stimulates a specific response rather than a form letter. A written communication can be followed up with a call or personal visit, and the intent to do so can be mentioned in the initial contact. We have, in this regard, much to learn from our colleagues in sales and marketing. Once utilization has begun, the need for following up may be even greater.


W. R. Miller

Treatment innovation, like recovery from an addictive behavior, is a frangible state, and relapse to well-practiced prior repertoires is an ever-present danger. Strategies of relapse prevention (Marlatt and Gordon, 1985), now being developed in the treatment of addictions, might also be brought to bear on the implementation process.

Teaching Utilization

Those of us who are involved in training future potential research and clinical professionals can also encourage and require our students to consider how research can be utilized. A common assignment is for the student to choose a topic of special interest, then to locate, read, and write about research on this subject. An often overlooked but very useful addition to such an assignment is to require the student to derive and list recommendations for specific actions and policies that follow from the research findings (Beyer and Trite, 1982).

A NEED FOR RESEARCH It is habitual to end with a statement that more research is needed on the topic being addressed. With regard to the utilization of behavioral treatment research, however, it is fair to say that it would be helpful to have any empirical research addressing this problem. Although we have very extensively studied the etiology, assessment, diagnosis, prevention, and treatment of alcoholism, researchers have almost totally ignored the question of what methods succeed in inculcating the application of our research findings on these topics. The few available empirical studies of utilization (e.g. Beyer and Trite, 1978; Backer et al., 1986) point to the importance of factors such as those outlined above. There are very few empirical guidelines for designing utilization strategies at present. Those of us who do research in the modification of addictive behaviors may be particularly well-suited to conduct such studies. We are accustomed to dealing with seemingly intransigent behaviors, overcoming resistance, instilling motivation for change, and replacing overpracticed habits with novel alternative responses. Research is beginning to appear on behavioral strategies for decreasing resistance to change within individual and family therapy (e.g. Chamberlain et al., 1984; Miller, 1985; Patterson and Forgatch, 1984). Perhaps it is time we focussed our expertise on a genuinely challenging problem: how to motivate, evoke, and maintain the behaviors that constitute research utilization.

Behavioral Alcohol Treatment Research Advances


REFERENCES Alden, L. (1978) Evaluation of a preventive self-management programme for problem drinkers. Canad.

J. behav.

Sci. 10,258-263.

Annis, H. M. and Chan, D. (1983) The differential treatment model: Empirical evidence from a personality typology of adult offenders. CriminaI Justice andBehavior, 10, 159-173. Asch, S. E. (1951) Effects of group pressure upon the modification and distortion ofjudgments. In: H. Guetzkow (Ed), Groups, leadership and men. Carnegie Press, Pittsburgh. Ascher, W. (1978) Forecasting: An appraisal for policy-makers and planners. Johns Hopkins University Press, Baltimore, MD. Azrin, N. H., Sisson, R. W., Meyers, R. and Godley M. (1982) Alcoholism treatment by disulfiram and community reinforcement therapy. J. Behav. Ther. exp. Psychiaz. 13, 105-l 12. Backer, T. E., Liberman, R. P. and Kuehnel, T. G. (1986) Dissemination and adoption of innovative psychosocial interventions. J. cons&. c/in. Psycho/. 54, 11l-1 18. Baker, T. B., Udin, H. and Vogler, R. E. (1975) The effects of videotaped modeling and self-confrontation on the drinking behavior of alcoholics. Int. J. Addict. 10, 779-793. Bandura, A. (1982) Self-efficacy mechanism in human agency. Am. Psychol. 37, 122-147. Berg, G. and Skutle, A. (1986) Early intervention with problem drinkers. In: W. R. Miller and N. Heather (Eds), Treating addictive behaviors: Processes of change, pp. 205-220. Plenum Press, New York. Beyer, J. M. and Trite, H. M. (1978) Implementing change: Alcoholism policies in work organizations. Free Press, New York. Beyer, J. M. and Trite, H. M. (1982) The utilization process: A conceptual framework and synthesis of empirical findings. Admin. Sci. Q. 27, 591-622. Beyer, J. M. and Trite, H. M. (1978) ImpIementing change: Alcoholism policies in work organizations. Free Press, New York. Broad, W. and Wade, N. (1982) Betrayers of the truth: Fraud and deceit in the halls of science. Simon & Schuster, New York. Brown, R. A. (1980) Conventional education and controlled drinking education courses with convicted drunken drivers. Behav. Ther. 11,632-642. Buck, K. A. and Miller, W. R. (1981, November) Why does bibliotherapy work? A controlled study. Paper presented at the annual meeting of the Association for Advancement of Behavior Therapy, Toronto, Canada. Caddy, G. R. (1975) How not to replicate: A commentary on Maxwell etal.‘s (1974) replication of Lovibond and Caddy’s (1970) “discriminated aversive control” study. Behav. Ther. 6, 710-711. Caddy, G. R. and Lovibond, S. H. (1976) Self-regulation and discriminated aversive conditioning in the modification of alcoholics’ drinking behavior. Behav. Ther. 7,223-230. Carey, K. B. and Maisto, S. A. (1985) A review of the use of self-control techniques in the treatment of alcohol abuse. Cogn. Ther. Res. 9, 235-251. Carpenter, R. A., Lyons, C. A. and Miller, W. R. (1985) Peer-managed self-control program for prevention of alcohol abuse in American Indian high school students: A pilot evaluation study. Int. J. Addict. 20,299-301. Chamberlain, P., Patterson, G., Reid, J., Kavanagh, K. and Forgatch, M. (1984) Observation of client resistance. Behav. Ther. 15, 144-155. Chapman, L. and Chapman, J. (1967) Genesis of popular but erroneous psycho-diagnostic observations. J. abnorm. PsychoI. 72, 193-204. Chapman, L. and Chapman, J. (1969) Illusory correlation as an obstacle to the use of valid psycho-diagnostic signs. J. abnorm. Psychol. 74,271-280. Coghlan, G. R. (1979) The investigation of behavioral self-control theory and techniques in a short-term treatment of male alcohol abusers. Unpublished doctoral dissertation, State University of New York at Albany. University Microfilms No. 7918818. Costello, R. M. (1975) Alcoholism treatment and evaluation: In search of methods. Inr. J. Addict. 10,251-275.

Czypionka, A. and Demel, I. (1976) Kontrolliertes Trinken bei Alkoholkranken: Versuch einer Behandlung mit verhaltenstherapeutische Methoden. Zeitschr. k/in. Psycho!. J,92-108.


W. R. Miller

Danziger, J. N. (1983) Power is knowledge: The linkage between the political system and knowledge. In: K. E. Boulding and L. Senesh (Eds), The optimum utilization ofknowledge: Making knowledge serve human betterment, pp. 262-280. Westview Press, Boulder, CO. Deci, E. L. (1975) Intrinsic motivation. Plenum Press, New York. Dror, Y. (1986) Policymaking under adversity. Transaction Books, New Brunswick. Ewing, J. A. and Rouse, B. A. (1976) Failure of an experimental treatment program to inculcate controlled drinking in alcoholics. Br. .I. Addict. 71, 123-134. Festinger, L. (1957) A theory of cognitive dissonance. Row, Peterson, Evanston, IL. Fox, R. (1967) A multidisciplinary approach to the treatment of alcoholism. Am. .I. Psychiat. 123, 769-778. Foy, D. W., Nunn, B. L. and Rychtarik, R. G. (1984) Broad-spectrum behavioral treatment for chronic alcoholics: Effects of training controlled drinking skills. J. consult. clin. Psychol. 52,218-230. Hagen, R. E., Williams, R. L., McConnell, E. J. and Fleming, C. W. (1978) An evaluation of alcoholabuse treatment as an alternative to drivers license suspension or revocation, Report No. CAL-DMV-RSS-78-68 to the Legislature of the State of California. Haney, C. and Zimbardo, P. G. (1977) The socialization into criminality: On becoming a prisoner and a guard. In: J. L. Tapp and F. L. Levine (Eds), Law, justice and the individual in society: Psychological and legal issues, pp. 198-223. Holt, Rinehart & Winston, New York. Hansen, J. and Emrick, C. D. (1983) Whom are we calling “alcoholic” Bull. Sot. Psychol. Addict. Behav.


Heather, N. and Robertson, I. (1983). Controlled drinking (rev. ed.). Methuen, London. Hedberg, A. G. and Campbell, L. M. (1974) A comparison of four behavioral treatment approaches to alcoholism. .I. Behav. Ther. exp. Psychiat. 5251-256. Helzer, J. E., Robins, L. N., Taylor, J. R., Carey, K., Miller, R. H., Combs-Orne, T. and Farmer, A. (1985) The extent of long-term moderate drinking among alcoholics discharged from medical and psychiatric treatment facilities. N. Engl. .I. Mea! 312, 1678-1682. Henricksen, J. J. (1976) Locus of control among alcoholics: Some empirical and conceptual issues. J. Stud. Alcohol 31,908-916. Hester, R. K. and Miller, W. R. Empirical guidelines for optimal client-treatment matching. In Adolescent Drug Abuse: Analysis of TreatmentResearch. National Institute on Drug Abuse, Rockville, MD. Janis, I. L. and Mann, L. (1977) Decision making: A psychological analysis of conflict, choice, and commitment. Free Press, New York. Kanfer, F. H. (1986) Implications of a self-regulation model of therapy for treatment of addictive behaviors. In: W. R. Miller and N. Heather (Eds), Treating addictive behaviors: Processes of change, pp. 29-47. Plenum Press, New York. Karp, S. A., Kissin, B. and Hustmyer, F. E., Jr. (1970) Field dependence as a predictor of alcoholic therapy dropouts. J. Net-v. Mental Dis. 150,77-83. Kinder, B. N., Pape, N. E. and Waltish, S. (1980) Drug and alcohol education programs: A review of outcome studies. Znt. .I. Addict. 15, 1035-1054. Leventhal, H. (1970) Findings and theory in the study of fear communications. Adv. exp. social Psychol. 5, 119-186. Leventhal, H. (1971) Fear appeals and persuasion: The differentiation of a motivational construct. Am .I. Public-Health 61, 1208-1224. Locke. E. A.. Shaw. K. N.. Saari. L. M. and Latham. G. P. (1981) Goal settine and task ‘performance:‘1969-1980. Psychol. Bull. 90,125-152. \ ’ Lovern, J. (September/October, 1982) Moderation to abstinence: A confession. Alcoholism, 38-39. Lovibond, S. H. (1975) Use of behavior modification in the reduction of alcohol-related road accidents. In: T. Thompson and W. S. Dockens, 3rd (Eds), Applications of behavior modification, pp. 399-406. Academic Press, New York. Lovibond, S. H. and Caddy, G. (1970) Discriminated aversive control in the moderation of alcoholics’ drinking behavior. Behav. Ther. 1,437-444. Marlatt, G. A. and Gordon, J. R. (1985) Relopseprevention. Guilford Press. Maxwell, W., Baird, R. L., Wezl, T. and Ferguson, L. (1974) Discriminated aversion conditioning

Behavioral Alcohol Treatment Research Advances


within an alcoholic treatment program in the training of controlled drinking. Behav. Engng 2, 17-19. McCrady, B. S. (1986) Implications for behavior therapy of the changing alcoholism health care delivery system. The Behav. Therapist, 9, 171-174. McLachlan, J. F. C. (1982) Benefit from group therapy as a function of patient-therapist match on conceptual level. Psychother. Theory Rex Pratt. 9,317-323. McLachlan, J. F. C. (1974) Therapy strategies, personality orientation and recovery from alcoholism. Canad, Psychiat. Assoc. J. 19,2.5-30. McLellan, A. T., Woody, G. E., Luborsky, L., O’Brien, C. P. and Druley, K. A. (1983)Increased effectiveness of substance abuse treatment: A prospective study of patient-treatment “matching”. .I. Nerv. Mental Dis. 171, 597-605. Milgram, S. (1963) Behavioral study of obedience. J. Abnormal and sot. Psychol. 67,371-378. Milgram, S. (1974) Obedience to authority. Harper & Row, New York. Miller, W. R. (1978) Behavioral treatment of problem drinkers: A comparative outcome study of three controlled drinking therapies. J. consult. clin Psychol. 46,74-86. Miller, W. R. (1983a) Controlled drinking: A history and critical review. J. Stud. Alcohol 44, 68-83. Miller, W. R. (1983b) Motivational interviewing with problem drinkers. Behav. Psychother. 11. 147-172. Miller, W. R. (1985) Motivation for treatment: A review with special emphasis on alcoholism. Psychol.

BuIl. 98, 84-107.

Miller, W. R. (1986a) Haunted by the Zeitgeist: Reflections on contrasting treatment goals and concepts ofalcoholism in Europe and the United States. Ann. NYAcad. Sci. 472,110-129. Miller, W. R., Gribskov, C. J. and Mortell, R. L. (1981) Effectiveness ofa self-control manual for problem drinkers with and without therapist contact. Int. J. Addict. 16, 1247-1254. Miller, W. R. and Hester, R. K. (1980) Treating the problem drinker: Modern approaches. In: W. R. Miller (Ed), The addictive behaviors: Treatment of alcoholism, drug abuse, smoking, and obesity, pp. 1 I-141. Pergamon Press, New York. Miller, W. R. and Hester, R. K. (1986a)Theeffectiveness ofalcoholism treatment: What research reveals. In: W. R. Miller and N. Heather (Eds), Treating addictive behaviors: Processes of change, pp. 121-174. Plenum Press, New York. Miller, W. R. and Hester, R. K. (1986b) Inpatient alcoholism treatment: Who benefits? Am. Psychol.

41, 794-805.

Miller, W. R. and Hester, R. K. (1986~) Matching problem drinkers with optimal treatments. In: W. R. Miller and N. Heather (Eds), Treating addictive behaviors: Processes of change, pp. 175-203. Plenum Press, New York. Miller, W. R. and Murioz, R. F. (1982) How to control your drinking (rev. ed.). University of New Mexico Press, Albuquerque. Miller, W. R. and Taylor, C. A. (1980) Relative effectiveness of bibliotherapy, individual and group self-control training in the treatment of problem drinkers. Addict. Behav. 5, 13-24. Miller, W. R., Taylor, C. A. and West, J. C. (1980) Focused versus broad-spectrum behavior therapy for problem drinkers. J. consult. clin. Psychol. 48,590-601. Nathan, P. E. (1986) Outcomes of treatment for alcoholism: Current data. Ann. behav. Med. 8, (2-3), 40-46. Office of Technology Assessment, U.S. Congress (1983) The effectiveness and costs of alcoholism treatment. Author, Washington, D.C. Orford, J. and Hawker, A. (1974) An investigation of an alcoholism rehabitation halfway house: II. The complex question of client motivation. Br. J. Addict. 69, 315-323. Orford, J., Oppenheimer, E. and Edwards, G. (1976) Abstinence or control: The outcome for excessive drinkers two years after consultation. Behav. Res. Ther. 14,409-418. Patterson, G. R. and Forgatch, M. S. (1985) Therapist behavior as a determinant for client noncompliance: A paradox for the behavior modifer. J. consult. c/in. Psychol. 53,846-85 1. Pendery, M. L., Maltzman, I. M. and West, L. J. (1982) Controlled drinking by alcoholics? New findings and a reevaluation of a major affirmative study. Science 217, 169-175. Polich, J. M., Armor, D. J. and Braiker, H. B. (1981) The course of alcoholism: Four years after treatment. Wiley, New York. Pomerleau, O., Pertshuk, M., Adkins, D. and d’Aquili, E. (1978) Treatment for middle income .lkBRT9:2/3-H


W. R. Miller

problem drinkers. In: P. E. Nathan, G. A. Marlatt and T. Lsberg (Eds), Alcoholism: New directions in behavioral research and treatment, pp. 143-160. Plenum Press, New York. Robertson, I., Heather, N., Dzialdowski, A., Crawford, J. and Winton, M. (1986) A comparison of minimal versus intensive controlled drinking treatment interventions for problem drinkers. Br. J. clin. Psychol. 25, 185-194. Rogers, R. W. (1975) A protection motivation theory of fear appeals and attitude change. J. Psychol. 91,93-l 14. Salzberg, P. M. and Klingberg, C. L. (1983) The effectiveness of deferred prosecution for driving while intoxicated. J. Stud. Alcohol 44,299-306. Sanchez-Craig, M. (1980) Random assignment to abstinence or controlled drinking in a cognitive-behavioral program: Short-term effects on drinking behavior. Addict. Behav. 5, 35-39. Sanchez-Craig, M., Annis, H. M., Bornet, A. R. and MacDonald, K. R. (1984) Random assignment to abstinence and controlled drinking: Evaluation of a cognitive-behavioral program for problem drinkers. J. consult. clin. Psychol. 52,390-403. Schaefer, H. H. (1972) Twelve-month follow-up of behaviorally trained ex-alcoholic social drinkers. Behav. Ther. 3,286-289. Sobell, M. B. and Sobell, L. C. (1973) Individualized behavior therapy for alcoholics. Behav. Ther. 4,49-72. Sobell, M. B. and Sobell, L. C. (1978) Evaluating the external validity of Ewing and Rouse. Br. J. Addict.

13, 343-345.

Sobell, M. B. and Sobell, L. C. (1984) The aftermath of heresy: A response to Pendery et al.‘s (1982) critique of “Individualized behavior therapy for alcoholics”. Behav. Res. Ther. 22, 413-440. Sobell, M. B. and Sobell, L. C. (1982) Alcohol abuse curriculum guide for psychology faculty, Pre-publication copy. National Institute on Alcohol Abuse and Alcoholism, Rockville, MD. Thornton, C. C., Gottheil, E., Gellens, H. K. and Alterman, A. I. (1977) Voluntary versus involuntary abstinence in the treatment of alcoholics. J. Stud. Alcohol 38, 1740-1748. Thornton, C. C., Gottheil, E., Gellens, H. K. and Alterman, A. I. (198 1) Developmental level and treatment response in male alcoholics. In: E. Gottheil, A. T. McLellan and K. Druley (Eds), Matching patient needs and treatment methods in alcoholism and drug abuse. C. C. Thomas, Springfield, IL. Tversky, A. and Kahneman, D. (1974) Judgment under uncertainty: Heuristics and biases. Science 185, 1124-1131. Vaillant, G. E. (1983) The natural history of alcoholism: Causes, patterns, andpaths to recovery. Harvard University Press, Cambridge, MA. Vogler, R. E., Compton, J. V. and Weissbach, T. A. (1975) Integrated behavior change techniques for alcoholism. J. consult. clin. Psychol. 43, 233-243. Vogler, R. E., Weissbach, T. A. and Compton, J. V. (1977a) Learning techniques for alcohol abuse. Behav. Res. Ther. 15,31-38. Vogler, R. E., Weissbach, T. A., Compton, J. V. and Martin, G. T. (1977b) Integrated behavior change techniques for problem drinkers in the community. J. consult. c/in. Psychof. 45, 267-279. Wiens, A. N., Montague, J. R., Manaugh, T. S. and English, C. J. (1976). Pharmacological aversive counterconditioning to alcohol in a private hospital: One-year follow-up. J. Stud Alcohol 37, 1320-1324. Woititz, J. G. (1976) A study of self-esteem in children of alcoholics. Unpublished doctoral dissertation, Rutgers University. University Microfilms No. 77-13,299.