Broadening the Base of Treatment for Alcohol Problems (1990)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

11 Matching Selecting patients proper) Is one of treatment's tricks; For whether one does well or not Depends on whom one picks. Anonymous In Chapter 5 of this report the principal conclusion drawn from a review of the available evidence on the efficacy of treatment was that there was no single treatment approach that was effective for all persons with alcohol problems. The chapter went on to state that "reason for optimism in the treatment of alcohol problems lies in the range of promising alternatives that are available, each of which may be optimal for different types of individuals.H Research data corroborating these conclusions are extensively reviewed in Appendix B. Matching that is, selecting from among available alternatives the treatment or treatments that are most likely to facilitate a positive outcome in a particular individual is a necessary consequence of these conclusions. Because it has recently received much attention (cf. Annis, 1987; Gordis, 1987; Holden, 1987), matching in the treatment of alcohol problems may appear to be a new concern. In fact, it has been an object of attention for some time. The first major review of the treatment of alcohol problems (Bowman and Jellinek, 1941) stressed the need for matching. The first major experimental study of matching in the treatment of alcohol problems, the Winter VA Hospital study, was initiated in 1950 (Wallerstein, 1956, 1957~. Matching was advocated in the December 1971 First Special Report to the U.S. Congress on Alcohol and Health (USDHEW, 1971), and has also been endorsed in subsequent reports in this series. However, there has been a recent acceleration of interest in matching. Perhaps (as was suggested in the introduction) the passage of time has been required to appreciate the varieties of individuals and problems involved and to develop differing forms of treatment. In recent years there have been multiple reviews of matching in the treatment of alcohol problems (Gibbs and Flanagan, 1977; Pattison, 1978, 1979; Gibbs, 1980; Glaser, 1980; McLellan et al., 1980; Gottheil et al., 1981; Skinner, 1981; Solomon, 1981; Finney and Moos, 1986; Longabaugh, 1986; Miller and Hester, 1986; Annis, 1987, 1988; IOM, 1989~. Nor has interest been restricted to the United States and Canada (cf. Matakas et al., 1978; Lindstrom, 1986; Anokhina et al., 1987~. Developments in the treatment of alcohol problems do not occur in a vacuum. As is suggested by the bit of medical student doggerel collected some three decades ago that is the epigraph to this chapter, matching has long been an integral part of medical practice. Differential diagnosis followed by specific treatment is a pattern of clinical activity that dates back to the Hippocratic corpus (Veith, 1964), and remains the hallmark of contemporary medical therapeutics. Consider, for example, the selection of the appropriate varieties, the appropriate sequences, and the appropriate intensities of surgery, radiation, and chemotherapy in the contemporary management of cancers or the careful selection of particular antibiotics to deal with specific infections. Yet medicine is hardly alone in its use of matching. In education it has become important to recognize "the differential effectiveness of [educational] approaches on different kinds of students" (Hunt, 1971:1~. In corrections (in fulfillment of the "object all sublime" of Gilbert and Sullivan's Mika~lo-"to make the punishment fit the crimes) it has 279

280 BROADENING THE BASE OF TENT FOR ALCOHOL PROBLEMS been recognized that "treatment methods will relate specifically to the goals for the various offender subgroups" (Warren, 1969:48~. Psychology has a long history of appreciation of the therapeutic significance of individual differences (Kiesler, 1966; Paul, 1967; Abramowitz et al., 1974; Berzins, 1977; Beutler, 1979), and a growing interest has been developing in psychiatry (Frances et al., 1984; Clarkin and Perry, 1987~. Thus, matching in the treatment of alcoholproblems~s appropriated viewed rotas a unique and id~o~yncra~cdevelopmen~butas a particular app~atiorl of a general strategy in human therapeutics. ~ , it, , Studies of Matching in the Treatment of Alcohol Problems A brief description of two studies will demonstrate the effect of matching on the outcome of treatment for alcohol problems. One study involved incarcerated offenders with a history of alcohol problems treated with a highly confrontational type of group therapy (Annie, 1979~. A retrospective analysis of outcome results from this randomized controlled trial indicated a significant difference in the effect of treatment depending on the self-image of the individual. Persons who entered treatment with a positive self-image achieved outcomes that were significantly better than those of control subjects who were not treated; however, those who entered treatment with a negative self-image did significantly worse than controls. For example, they were reconvicted of crimes more frequently than those who were not treated at all to a significant degree (Annie and Chan, 1983~. The implications of such a study for the future conduct of treatment are straightforward, assuming further validation of the findings. Pursuing them briefly will provide an example of how results from matching studies might influence subsequent clinical practice. The therapy under consideration cannot reasonably be given to all individuals because it will harm some of them (cf. Chapter 6~. Rather, individuals should be carefully assessed on the relevant variables)-in this case, self-image-prior to a decision regarding treatment. Those who have a positive self-image can be treated by this method with a reasonable expectation of good results; those who do not should be provided with alternative treatments. In another study the relevant individual variable was "conceptual level," a complex construct derived from educational theory (cf. Hunt, 1971~. It was hypothesized that individuals with alcohol problems who operated at a high conceptual level (i.e., were independent, empathic, and cognitively complex) would do better in relatively unstructured therapeutic situations, whereas those who operated at a low conceptual level (i.e., were impulsive, poorly socialized, and cognitively simple) would respond more favorably to directive therapists and structured therapies. The hypothesis was tested simultaneously with respect to initial treatment and aftercare. Those who were correctly matched to either treatment or aftercare did better than those who were mismatched; however, those who were matched to both did strikingly better than those who were matched to neither (77 percent positive results as compared with 38 percent) (McLachlan, 1972, 1974~. Again, the application in practice is quite straightforward, but it involves an extension of the prior example that is instructive. Implementing the findings from this study in practice would involve assessing not only potential clients but also potential therapists and therapies; the properties of both need to be determined to create the appropriate match. This was true as well of the prior example, but there only a single therapeutic approach was considered. Here, multiple approaches as well as multiple therapists would need to be evaluated. In addition, the importance of matching to "after- care," as well as to initial treatment, is stressed. The results of these and other studies of matching in the treatment of alcohol problems are summarized in Table 11-1 (slightly modified from Annis, 1988~. All of the studies listed involve (a) a reliable matching variable that can distinguish between at least

MATCHING 281 two types of individuals with alcohol problems; (b) at least two well-deffned treatment conditions; (c) assignment (preferably random) of each type of individual to each treatment condition; (d) an adequate post-treatment follow-up period; and (e) objective, reliable measurers) of treatment impact. These elements are considered desirable methodological criteria for the investigation of matching phenomena (Annie, 1988; cf. Hayes et al., 1987~. The studies indicate that matchingindividuals to specific treatments on. a varied of variables including demographic factors, psychiatric diagnoses, personality factors, severity of alcohol problems, and antecedents to drinking has the poter'fial to improve treatment outcome sigr~if~car~. These are not the only variables that may be useful but simply those that have been studied to date utilizing an adequate methodology. Nor are these studies the only kind that are capable of shedding light on possible matching effects. Rather, there are a variety of methodologies that can be used to illuminate such effects. For example, simple outcome monitoring often demonstrates a close association between some particular variable and successful outcome. Of course, this method does not definitively prove that a matching effect has occurred, but it is certainly suggestive. As a next step, the variable (or variables) identified from outcome monitoring as associated with positive outcome can be used as a guide to the matching of a subsequent group of individuals to treatment. The outcomes from the subsequent group can then be compared with those of the previous group. A higher proportion of positive outcomes suggests that (other factors being equal between the two trials) an effective match has been identified. This "bootstrapping" strategy has been used effectively in the treatment of persons with alcohol and drug problems (McLellan et al., 1980, 1983a). It does not involve the random assignment of individuals to treatments (although such a finding could be tested in a randomized controlled trial as well) but instead is a methodology that is more readily implemented in clinical treatment settings (cf. Chapter 5~. It does involve obtaining systematic knowledge of treatment outcomes. Although this practice is not a common one at present, it is rapidly becoming a major requirement for the certification of treatment programs (Schroeder, 1987; Commission on Accreditation of Rehabilitation Facilities, 1988; see also Chapter 12~. Beyond studies that investigate ways in which researchers and treatment personnel may attempt to match patients to treatment programs are studies suggesting that patients may be attempting to match themselves to the most appropriate treatments. For example, there is evidence that patients entering treatment expect to be matched. A careful study of informed consent (Appelbaum et al., 1983), in which elaborate explanations of random assignment to treatment were provided, found that prospective patients persisted in believing, despite what they had been told, that they would in fact be assigned to what- ever treatment the experimenters had concluded was most suitable for them. The implication of the study is that the selection of treatment according to the principle of matching is generally assumed to be an integral element of therapeutics. This view is also held by some therapists (Angel!, 1984; Taylor et al., 1984; Imber et al., 1986) and some ethicists (Marquis, 1983~. It has been shown repeatedly that the populations of different treatment programs for alcohol differ significantly from one another (Pattison et al., 1969, 1973; Bromet et al., 1976, 1977; Kern et al., 1978; Skinner and Shoffner, 1978; Finney and Moos, 1979; Skinner, 1981~. To some extent, the differences may be due to such factors as eligibility requirements (it is hardly surprising to find more males and more veterans in a Veterans Administration treatment program than in other programs). Yet differential eligibility requirements and other administrative and practical factors do not appear to provide a complete explanation for the population differences that have been observed among various treatment programs. A belief on the part of individuals that certain programs may serve them better than others-a belief in matching may contribute importantly to these consistent findings. In programs that are targeted specif

282 BROADENING TIE BASE OF TREATMENT FOR ALCOHOL PROBLEMS TABLE 11-1 Findings from Selected Studies on Matching Study Matching Variable(s) Description of Match . Demographic variables Azrin et al. (1982) Mayer and Myerson (1971) Kissin et al. (1970) Psychiatric diagnosis (plus alcohol problems) Wallerstein (1957) Marital status Social stability Social and psychological stability Various diagnoses Tomsovic and Edwards Schizophrenia (1970)  Merry et al. (1976) Depression Reynolds et al. (1977) McLellan et al. (1980, 1983) Kadden et al. (1990) Psychiatric severity Global psychopathology, sociopathy, and neuropsychological impairment Single persons required behavior therapy in addition to disulfiram for favorable outcome, whereas married persons did well on disulfiram alone. Disfulfiram was associated with successful outcomes in persons with low social stability but not in those with high social stability. Persons with both social and psychological stability did best in psychotherapy; those with social stability only did better in drug therapy; those with neither did best in an inpatient program. Those diagnosed as compulsive characters did better on disulfiram; those who were passive dependent did better in group hypnotherapy; those who were less aggressive did better in conditioned reflex therapy; those with a "positive attachment factor" did better in milieu therapy. Persons with alcohol problems but without schizophrenia did better than controls who did not receive lysergic acid diethylamide (LSD); persons with both alcohol problems and schizophrenia who were given LSD did worse than controls. Depressed persons with alcohol problems had  successful outcomes on lithium compared with controls, whereas nondepressed persons with alcohol problems given lithium did worse than controls. Persons with high severity did poorly in both inpatient and outpatient treatment; those with low severity did well in both; and those with intermediate severity had outcomes that were sensitive to careful matching to either setting. Following inpatient treatment, those high in global psychopathology and sociopathy did better in a relapse prevention group; those high in neuropsychological impairment did better in interactional group psychotherapy.

MATCHING TABLE 11-1 (continued) 283 Study Matching Variable(s) Description of Match Personality facto" McLachlan (1972, 1974) Conceptual level Annis and Chan (1983) Self-image Hartman et al. (19~) Locus of control Severity of alcohol problem(s) Edwards et al. (1977) Orford et al. (1976) Sokolow et al. (1980) Lyons et al. (1982) Polich et al. (1981) Antecedents to drinldng Rosenberg (1979) Annis and Davis (1989) SOURCE: Modified from Annis (1988~.  High conceptual level individuals did best with nondireetive therapists and unstructured aftereare; low conceptual level individuals did best with directive therapists and structured aftercare. Persons with a positive self-image benefited from a highly confrontational form of group therapy, whereas those with a negative self- image did worse than no-treatment controls. Persons with an internal locus of control did better in brief, unstructured therapy; those with an external locus of control did better in more structured and intensive treatment conditions. Jellinek classification Sex and severity Alcohol dependence, age, and marital status Anxiety related to drinking Situational factors Gamma alcoholics (severe, dependent) did best in intensive treatment; nongamma alcoholics did best with simple advice. Females did better in treatment programs that had a medical orientation; males with less severe problems did best in programs with a rehabilitation orientation. Older (over 40) married individuals with high levels of physical and psychological dependence were less likely to relapse if abstinent; younger unmarried individuals with low levels of dependence were less likely to relapse with nonproblem drinking. Those with high levels of anxiety related to drinking were markedly improved with relaxation training; those with low levels of anxiety were not improved. Those who could relate their risk of drinking to specific stressful situations were highly successful with relapse prevention; those who perceived a generalized risk of drinking across all stressful situations had much less favorable outcomes with relapse prevention.

284 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS ically at special populations (e.g., women, blacks, Indians), the belief that these programs may serve them better is likely to be responsible for the presence of most of the individuals in the program. Programs for special populations are further discussed in Section IV of this report. Examples of Matching Programs There are alternatives to defining matching as the sole responsibility of those seeking treatment. Matching can be carried out by treatment programs themselves. Indeed, this practice may be advisable because matches elected by individuals seeking treatment may speak more to the attractiveness of the treatment than to its probable efficacy and because their judgment may be constrained by the pressures of the problems for which they are seeking relief. It is also highly unlikely that individuals seeking treatment will be fully conversant with all of the available therapeutic alternatives The foregoing is not to say, however, that individual preferences should not be consulted, a matter that will be discussed in more detail later in this chapter. When queried, many programs will state that they engage in matching on a regular basis (cf. Maisto and Nirenberg, 1986~. However, what is often meant by this response is (a) that they screen potential admissions to their program and do not accept all applicants, and (b) that although they provide the same principal treatment to all persons whom they accept, they also provide supplementary or ancillary services on an individualized basis. Although such practices are commendable they do not, in the committee's view, constitute matching. Matching involves varying the principal treatment approach utilized from one individual to another in accord with a preconceived and explicit plan. For example, the Winter VA Hospital study involved four completely different therapeutic approaches (disulfiram, conditioned reflex therapy, group hypnosis, or milieu therapy) (Wallerstein, 1956, 1957~. The frequency with which programs match to different primary therapies is currently unknown because few studies have examined the question. From an impressionistic standpoint it is not done frequently. Some data suggest a reason may be that few programs offer more than a single major therapeutic option. In a study of all treatment programs for alcohol problems in one state, it was found that only 5 percent of all programs offered two or more such options. These were large programs and serviced approximately 15 percent of all of the individuals in treatment in the state at that time. However, this meant that 85 percent of all individuals in treatment could not have been offered alternative options because there was also evidence that cross-referral between programs was nonexistent (Glaser et al., 1978~. There are programs that have made matching a fundamental part of their clinical operations. As of January 1, 1988, pretreatment assessment and matching to the appropriate level of care are mandated by law in the state of Minnesota for all individuals whose treatment involves the expenditure of public funds (the relevant portions of the law are quoted in Chapter 10~. Based on an advisory report issued in 1978 (Marshman et al., 1978), the province of Ontario in Canada has fostered the development of assessment centers that were independent of treatment operations and thus were necessarily engaged in matching. There is evidence that these centers have achieved high levels of community acceptance (Ogborne, Dwyer, and Ekdahl, 1984) and are evaluated as producing positive results (Malta, Rush, Gavin et al., 1985~. By 1987 there were some 35 such centers throughout the province (Rush, 1988~. The Brookf~eld Clinics, a private treatment program for alcohol and drug problems that operates at several sites in the state of Michigan, has for some time based its operations on matching (MacDonnell, 1981; O'Dwyer, 1984, 1988~. Another development

MATCHING 285 in the private sector may signal the arrival there of matching on a wholesale basis: the phenomenal rise of utilization management (Lewis, 1988; Korcok, 1988~. At least as currently practiced, utilization management is primarily a mechanism to control costs, but it may also serve to direct individuals to programs that are more effective for them as well as being less expensive. To date, there is some evidence that utilization management is an effective cost-containment strategy (Feldstein et al., 1988~; evidence that superior or even equivalent therapeutic results have been produced is not available. A program tested in the Clinical Institute of Ontario's Addiction Research Foundation, the Core-Shell Treatment Program, attempted to match all individuals seeking treatment for alcohol and drug problems to the most appropriate treatment (cf. Glaser et al., 1984~. The central position of matching in the program is illustrated in Figure 11-1. At its center, splayed arrows indicate the multiple matching options that were utilized (Skinner, 1984~. To recapitulate: the matching of particular individuals to specific treatments that are more likely than others to produce positive results has had wide therapeutic application. It has also been a matter of interest in the treatment of alcohol problems for some time. In a large number of studies conforming to specific methodological standards, a variety of variables have been shown to have value in the matching of individuals to more effective treatments for their alcohol problems. Reviews of these and other studies cited earlier have identified many additional variables that might be useful in matching. Persons entering treatment may expect to be matched to the most appropriate intervention and may engage in efforts at matching on their own. Several clinical treatment programs have been developed recently that use the process of matching as their basis for providing services; in addition, the rapid growth of utilization management may herald the arrival of matching on a very broad basis, indeed (although nothing is currently known about the outcome as opposed to the cost containment effects of utilization management). Nevertheless, matching can be a complex matter (cf. Finney and Moos, 1986~. Few studies are beyond criticism; some of the treatments dealt with in the matching literature are no longer in use, and none of the existing studies has been replicated to date. Further research on matching is indicated and in fact is being actively pursued. In September 1989 the participants in a multisite collaborative study of matching were selected from a peer-reviewed competition implemented by NIAAA Clinical application, however, is not in all instances and under all circumstances necessarily contingent on further research. Research often performs the function of corroborating the validity of clinical practice. Leo illustrative examples can be provided from the history of medicine. Dr. Edward Jenner (1749-1823), who paid attention to the local belief that dai~ymaids who contracted cowpox were immune to smallpox, introduced the practice of inoculation in 1796. Dr. William Withering (1741-1799) paid attention to the local belief that foxglove was effective against dropsy and introduced it into clinical medicine in 1785. Both proved to be major advances in therapeutics that were adopted and widely used, to good effect, long before the introduction of controlled trials into medicine. Both were based on attentiveness to therapeutic possibilities and empirical observation. Two caveats must, of course, be posted. The first is that not everything that is claimed to be therapeutic is in fact therapeutic. Another outcome of research is to demonstrate that treatments believed to be therapeutic are not; regrettably, this seems to be a more frequent outcome than the success that followed Jenner's and Withering's innovations. The second caveat, which follows from the first, is that the need for and the results of research cannot simply be dismissed. The foregoing discussion should not be read as advocating an Anything goes philosophy in treatment. Rather, the committee ad

286 Iz o tr w tc (J _ r? O ~ . . 1 - O rO~ ~ O i° <Zr o z ozl \0/ | ~ - U) O O 1 ~z 1 .~a ~°~r O ~ O O 7 z Z : 1 ~E ~. \ ~ ~ / \\\\ / / /~ ol  Ir----:jo<~o§OO 1~ O I 111 cn Ul i a: llJ I ~ I ~ Z ~ I o O ~ ~ ~ O Z ~ O ~ ~ O 0 z Z t° ~ ° a _ N  > - _ ~ J Z ~ ~ Z ~ C) ~, 1 1  ~L U) UJ~mOZO c`' I O U. I U. ~ I ~ ~ ~ m u' ~_ _ ~_ I I Z O: LU _ () ~ O Z ~o 0 Ul U] ~ O C~) ~ I . ~ Z C~ 111 O Z Q I~J :£ ~O Z O ~ -I O X t' ILI O ~ O , U~ C~ ~ ~ (/) O t) _ N 1 _ _~ /-O\ /IL ~ ~ ~ \ ,: >  z _____ 1 C) U) UJ tr - UJ ~ ~U ~ ~ I LLi O tl: (L Ul ( · Il C: 1 _1 -~ 1 ll 1 ~ O ~ 1 i2ti ~ ~ ~ U.l<)O~-< 1 1 o C) UJ o _ <:  O tn I ~n iu o IL o 5 ~r o a3 z ~ . o . . ~ Z ~ o UJ o ~ Z ~ IL o) z oo .= ~_ Ct ou 5: Q) - = ._ Ct ec ._ S o  a' C) S 4_ ~: ._ k Ct - - _` oo

MATCHING 287 vacates a balanced approach in which both practical experience and research findings are carefully weighed to find a reasonable compromise course for the present, always subject to review and reconsideration in the face of additional evidence. At the same time the obligation to provide the best possible treatment in the light of current knowledge must also be recognized. Given that no existing treatment for alcohol problems is universally effective, attempts to match individuals to treatments that are optimal for their particular needs seems to be a practical necessity even at the present time. How may matching be earned out in a responsible manner prior to the completion of the major research efforts that are only now beginning to get under way? In much the same way as clinical medicine has always proceeded: cautiously, empirically, and collectively. Matching is an issue of central importance around which researchers and clini- cians can come together to determine what might best be done riven the current state of knowledge. c, c, An appropriate model is the expert committee mechanism that has been employed to good effect by the World Health Organization. Where steps must be taken in an area that has been less than fully illuminated by research, a convening authority brings together a group of persons that is highly knowledgeable to share their information and to put forward, often in writing, their best advice on a particular subject. The committee considers matching to be an ideal focus for such an expert committee because it is the central process in the committee's vision and implies both a need for pretreatment assessment and for the determination of outcome. One matter that such a committee might consider is the further specification of additional elements of treatment. In Chapter 10 the committee discussed the specification of problems and individuals through the process of assessment. To match individuals to treatment, however, additional kinds of specification are required: a specification of treatments; a specification of matching guidelines; and a specification of outcomes. If fully specified individuals with fully specified alcohol problems are matched by fully specified matching criteria to fully specified treatment programs, and if the results of these treatment matches are fully specified, sufficient information will be at hand to refine the process so that during the next iteration better results are produced. As noted above, this ~bootstrapping" strategy has already been shown to be effective for persons with alcohol and drug problems (McLellan et al., 1980, 1983a). A long-range advantage of adopting this strategy is that, when the results of matching research do become available, just such specifications of the relevant components of treatment will be required for their implementation. Hence, the committee discusses the remaining kinds of specifications in the following pages: those of treatments and of matching guidelines in this chapter and that of outcome in the next. Specification of Treatment If individuals seeking treatment are to be matched to particular treatments, there must be detailed knowledge of the treatments that are to be provided. Otherwise, how can a choice be made among them? This requirement is familiar from the practice of referral, for which similar knowledge is required. Often, such knowledge is gained by visiting a program and observing it in action before making referrals. Although this is an excellent strategy, it is problematic in several ways. The time and expense involved may be considerable, and sampling problems may exist; that is, it is sometimes difficult to know whether what one observes under such circumstances is representative of the program's activities. Moreover, although most programs offer only a single kind of treatment, some  programs offer many potentially effective interventions in what has sometimes been referred

288 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS to as a Smorgasbord approach.. As also occurs at table, there is a strong impression that individuals obtaining treatment in these programs often sample a small amount of each intervention, to a degree known only to themselves and not documented in the written record of treatment. Under these circumstances the treatments provided are often not well characterized, and it is not possible to determine what their individual contributions might be to the outcome of treatment. Reading the written descriptions of the treatment provided is a more efficient alternative than site-visiting all treatment programs, and many such descriptions exist. Often, however, they are neither fully explicit nor consistent with one another. For example, they will indicate that individual psychotherapy is provided but neglect to specify what kind (according to one estimate, there are more than 140 varieties [Karasu, 1977~. Alternatively, one description will contain an explicit statement of the goals of the treatment, whereas another will assume that the goals are understood and omit them. It would be a happy circumstance if the capability to describe programs was as advanced as the capability to describe individuals and their problem~if, in a word, the assessment of treatment programs was well understood. Unfortunately, this is not the case (see, however, Moos and Daniels, 1967; Moos, 1968; Moos et al., 1973; Moos, 1974~. Research studies often involve the development of treatment manuals that fully embody all aspects of the therapeutic interventions that are being tested, so as to be able to specify exactly what the treatment was that did or did not produce positive results. Sometimes these manuals are retranscribed into a form appropriate for clinical application (cf. Sanchez-Craig, 1984~. Embodying what is done in treatment in a well-specified treatment manual is a practice that treatment programs might well emulate. To be maximally useful, the content of such treatment descriptions should be reasonably uniform. That is, it would be important to achieve consensus on those critical dimensions along which programs need to be characterized and on the methods for specifying those dimensions. Following the achievement of consensus the resultant descriptive paradigm could be generally applied to all treatment programs. In Chapter 3, four dimensions are discussed as general descriptors of treatment programs: the philosophy and orientation of the program (e.g., medical model, social model); the stage of the alcohol problem at which the treatment is directed (e.g., acute intervention, rehabilitation, maintenance); the setting of the program (e.g., inpatient, outpatient, residential), and the modality (e.g., disulfiram, cognitive behavior therapy). All of these descriptors may be important in achieving an appropriate match. Other dimensions that may need to be considered both for matching purposes and to achieve a comprehensive description include an explicit statement of the objectives or goals of the treatment; the criteria for assignment to the treatment and the target population or kind of problem at which the treatment is directed; its length; its intensity; its cost; and a reasonably detailed description of what is expected to occur during the course of treatment. The general application of a set of uniform descriptive dimensions would create a gazetteer of available treatments that would be of considerable value. For example, if most treatment programs offer only a small complement of interventions, and if they engage in careful matching, they are likely to find it necessary to refer a large number of persons to programs that provide alternative interventions. But how are they to know whether these other programs constitute a better match in terms of the treatments they offer? Only if there is a relatively detailed and uniform method of program description is it likely that an appropriate referral can be made. Conversely, such descriptions make it possible to utilize electively the full complement of interventions that may be available on a local or regional basis. Descriptions of treatments constructed along the lines of common dimensions are at present not plentiful, but they do exist. The appendix to this chapter reproduces the so-called Cleveland criteria as an example (Hoffmann et al., 1987~. The criteria were

MATCHING 289 developed on the basis of a comprehensive review of both the relevant literature and of "clinical criteria developed by various treatment providers, state agencies, and professional groups" (p. 2~. The criteria in the appendix are for the admission of adults to what are termed Levels of care." For each such level (e.g., mutual/self-help, low intensity outpatient treatment, intensive outpatient treatment, etc.) a Programmatic description" is provided along six dimensions (1) setting, (2) support systems, (3) staff, (4) therapies, (5) assess- ments/treatment plan review, and (6) documentation. Although the committee has some problems with the concept of "levels of care," the specification of each component of the treatment system along the same set of descriptive dimensions is an important contribution to the specification of treatment. Another example of treatment specifications are those of the Brookfield Clinics, which were mentioned earlier. They provide an intervention rationale, general objectives, criteria for assignment, and a detailed outline of each session for all interventions provided by the program (P. O'Dwyer, Brookfield Clinics, personal communication, January, 1988~. The potential utility of uniform descriptions goes well beyond matching. Those seeking treatment, for example, might well wish to know the content of the proposed treatment; with rising consumer consciousness in therapeutics, this is increasingly likely. Third-party payers providing reimbursement for treatment might wish to know for what they are paying. For these and other reasons, such explicit descriptions may soon be required. Specification of Matching Guidelines Even assuming that adequate specification of individuals, problems, and treatments might be accomplished, there is an additional need to specify how to connect them in the most appropriate manner that is, for specifying matching guidelines. Several methods can be used to guide the selection of optimal treatment for a particular individual. Far from being mutually exclusive, they are potentially complementary, even if occasionally contradictory. Ideal matching guidelines would take them all into account. The methods include therapist selection of the optimal intervention; patient selection; selection on the basis of the most prominent problem or problems; selection on the basis of theory; selection on the basis of research; and selection on the basis of empirical data about outcome. Therapist Selection .~_., , ,. ~_ _ It seems likely that many therapists select treatments for individuals differentially on the basis of their experience and knowledge. The committee does not doubt that this is in many respects a valid procedure and believes that the knowledge of therapists ought to be tapped systematically in creating guidelines for matching. Unfortunately, there are no data that establish the validity of therapist selection of treatment; the committee could tif~ no st~'dies of treatment for alcohol problems that compare outcomes from reatments selected by therapists with outcomes from treatments selected in other ways. The lack of data does not necessarily mean that the method lacks validity but only that it has not been studied. Even though therapists may select treatments for their patients, experience suggests  that they do not commonly make explicit the reasons why one treatment was selected in preference to another. However, matching guidelines can be examined carefully only if they are made completely explicit. As well, only if they are explicit can they be shared and ap

290 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS Matching Patient Needs and Treatment Methods Listed are a series of statements related to specific treatment needs and goals. Read each statement carefully and decide how important this treatment goal ~ for you. Circle the appropriate response on the accompanying answer sheet. 1. I need a structured inpatient program to learn how to live without taking drugs. 2. I want to make changes in my life-style that I can continue back in the community. 3. I want to learn how to cope more effectively with my urges to drink. 4. I need someone who will know me and help in sorting out my situation. S. I need to learn better ways of coping with my problems. 6. I want to argue less frequently with my family. 7. I want to learn how to communicate more honestly with my husbandlwife. 8. I want to learn better ways of talking with others in a group. 9. I want to learn how to be tense less often. 10. I need help in organizing a more effective job search. 11. I want to learn had to enjoy my free time without using drugs or alcohol. 12. I want to learn bow to manage my financial affairs. 13. I want to learn how to improve my physical fitness. 14. I want to take Antabuse or Temposil and stop drinking completely. 15. I would like to learn more about attending Alcoholics Anonymous (A^) meetings. 16. I want to stop my illegal use of narcotics. 17. I want to become free of narcotics within six months. 18. I want to stop drinking completely. 19. I want to stop my nonmedical use of drugs. 20. I want to work on my drug problems in an inpatient program with people my own age. 21. I want to understand the conditions under which I tend to drinlc 22. I need a half-way house where I can learn  bow to handle my drinking problems. 23. I would like to have one counsellor I can count on to discuss my problems and concerns. 25. 29. 30. 31. 37. 38e 39. 40. 41. 42. 24. I want an inpatient program that will help me deal with my drinking problems at work. I want a therapist to help me develop a greater awareness of my alcohol (drug) dependence. 26. I want family oounselling to help us deal with problems around my drinking or drug use. 27. I want a residential program where I can learn how to get along with others. 28. I need counselling with my husbandAvife  regarding problems related to my alcohol or drug use. I want to learn how to feel at ease with other people. I want to be able to improve my sleeping patterns. I want to explore new career-educational possibilities. 32. I want to learn how to enjoy my free time along with other responsibilities. 33. I need long-term (six months or more) methadone support. 34. I want a program where I can gradually decrease my dependence on, narcotics. 35. I want to stay in a half-way house and develop new interests outside of drinking. 36. I need someone to help me with day-to4ay problems like finding a place to stay. I want an inpatient program that will involve my employer. I want a therapist to help me look beneath the surface and find better ways of handling my problems. I need a temporary change (inpatient program) to start working on my alcohol- drug problems. I want to learn how to relax in everyday situations. I want to learn wears of making my work more satisfying. I want to learn haw to spend free time out of my home in the community. FIGURE 11-2 Treatment Goals Inventory for the Core-Shell Treatment Program (Glaser and Skinner, 1981:312- 313~. Each goal related to a particular treatment offered as part of the program. By indicating which of these goals were theim, individuals indicated which of the available treatments was a potential match for themselves.

MATCHING 291 plied in other program settings. This requirement applies not only to guidelines used by therapists, but to matching guidelines derived from any source. Patient Selection Some of the problems involved in exclusive dependence on patient selection of major treatment interventions have been noted already. Those seeking treatment may select programs on the grounds of their attractiveness rather than their probable efficacy, may make their choice under the constraint of their problems, and may be unaware of possible alternatives. Nevertheless, reliance upon self-selection-the so-called "cafeteria plans-has been recommended (Ewing, 1977~. It would certainly be unwise as well as unnecessary to exclude personal preference from the treatment selection process. Attrition from treatment programs characteristically has been high (cf. Baekeland and Lundwall, 1977; Silberfeld and Glaser, 1978; Brandsma et al., 1980~. Obviously this will continue to be the case unless a very considerable degree of customer satisfaction can be generated. Involving the person who is to undergo treatment in the treatment selection process may contribute importantly to customer satisfaction and to involvement in treatment. Treatment seekers have been known to make shrewd and interesting choices of therapy (Obitz, 1975~. What seems to be required, then, are methods to elicit individual therapeutic preferences on a systematic basis and to combine them with other methods of selecting treatment. Many possible ways of doing this are available. For example, persons seeking treatment may be most comfortable in treatment programs whose fundamental beliefs about alcohol problems they share. Treatment programs differ on these beliefs; some assign the responsibility for developing alcohol problems and for dealing with them to the person seeking treatment, and others do not (Brickman et al., 1982~. Accordingly, one can systematically assess such beliefs in both potential clients and potential treatment programs prior to treatment and take them into account in the matching process. Another method that has been used is the construction of a treatment goals inventory (Glaser and Skinner, 1981~. Each of the therapeutic programs in one system of care was asked to provide three statements in simple terms of what its goals might be. These goals were then organized into a checklist. Individuals being considered for treatment were asked, first, to check off all goals they felt were personally relevant and, second, to select their top five goals in rank order. Since each goal statement was tied to a specific treatment program, the goal choices could be used as an important guideline for matching. Figure 11-2 of this chapter reproduces this goal inventory. These examples do not exhaust the possible methods for tapping patient preference. Educational programs are another possibility that has been utilized (Bohnen, 1984~. Prospective patients have been shown audiovisual presentations of different therapists providing treatment and have been asked to make their selection on this basis (Obitz, 1975~. With some ingenuity, it should be quite possible to develop a broad repertory of methods for involving the individual significantly in the selection of treatment, with the hope thereby of assisting him or her to "buy into the treatment process. Problem-Oriented Selection When persons with alcohol problems are assessed, they may be found to have a variety of other problems that may contribute in a crucial manner to their alcohol problems. For some of these problems, specific therapeutic measures are available. For example, if it were felt that an individual's drinking was closely related to his or her marital

292 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS problems, marital therapy might be indicated. A sample decision sequence of this kind is provided in Figure 11-3 (Skinner, 1981~. Reconsider Marital Therapy 1 Key Variables 1. Treatment Goals 2. Marital Satisfaction 3. Social Stability Index Assignment Rules I. Treatment Goals (at least one) 1. I want to communicate more honestly with my spouse. 2. I want help in problems Ninth my spouse arising from my drinking (drug) problem. 3. I want to learn hoer to make my marriage hap- pier for both Or us. II. clarity Satisfaction Scale score In the dissatisfaction range (acorn leas than 54) III. Social Stability Index score above 7 IV. Spouse is milling to participate No 1 1 Decision Schema for the OPD Marital Therapy Program. Eligible for I Marital Therapy I  Considers Other Programs FIGURE 11-3 An example of problem-or~ented selection of treatment (Glaser and Skinner, 1981:319~. The selection of the treatment goals of marital therapy and a low score on a marital satisfaction scale were key elements in eligibility for marital therapy on an outpatient basis. Willingness of the spouse to participate and a high social stability score, the latter indicating a reasonable likelihood of sustaining attendance in outpatient treatment, also contributed to eligibility. In addition, the figure demonstrates the contribution of several kinds of selection guidelines to an ultimate decision regarding treatment selection. The key assessment variables are a treatment goals inventory, a marital satisfaction scale, and a social stability index. To be eligible for outpatient marital therapy, four assignment rules must be satisfied. The person seeking treatment must have selected a marital therapy goal statement as one of their top five goals on the treatment goals inventory; must have scored within the dissatisfaction range on the marital satisfaction scale; must have a reasonable level of social support on the social stability scale to support an outpatient intervention; and must have a spouse who is willing to participate. Exclusion criteria include the presence of a major medical or psychiatric disorder. A series of similar problem-oriented decision rules may be developed for a whole range of problem-oriented interventions. In our present state of knowledge, rules of this kind are in large measure arbitrary and lack research validation. Still, they are fully explic- it and unambiguous; were they to be used, there could be little question about why a particular individual was provided with a particular treatment. They constitute a reasonable point of departure in matching. Because they have been made explicit, they can be tested and modified by subsequent experience.

MATCHING Theory-driven Selection 293 Theories of human behavior can be utilized as guidelines for the selection of treatment. Conceptual-level theory, for example, has been used to construct a matching paradigm that subsequently proved to have validity (McLachlan 1972, 1974; see also the description above). Another example is the use of Bandura's self-eff~cacy theory to derive both a therapeutic approach and a method of matching (Annie, 1986a; Annis et al., 1987; Annis and Davis, 1988; Annis and Davis, 1989~. Theories have the advantage not only of making disparate facts meaningful but of transcending them by leaping from the particular to the universal. Whether such a leap is justified, however, may be problematic; and, in view of the diversity of individuals and problems in the treatment of alcohol problems, universal approaches can only be viewed with some skepticism. A claim that theory-based methods of matching are superior to others is difficult to sustain. Nevertheless, as one among several approaches to the construction of matching guidelines, theory has an important place and one of demonstrated utility. Research-dr~ven Selection As is noted earlier, the committee favors the conduct of further research on matching. What has been learned to date is summarized in Table 11-1. Further research is under way, and no doubt it will make an important contribution. The committee has nevertheless given its reasons for wishing to proceed cautiously and empirically with matching without awaiting the results of further research. As discussed, the principal reason is that, because research carried out to date suggests there is no single treatment method that is effective for all persons seeking assistance, matching is a current necessity. But there are other reasons as well. Many of the limitations of randomized controlled trials that are discussed in Chapter 5 may apply to the resultant research and particularly to the problem of its generalizability. Research conducted primarily or exclusively in academic settings on highly selected subjects treated with comparatively exotic interventions may not be immediately translatable into routine clinical settings. The probable generalizability of research results could be greatly enhanced through careful study design. Matching research, as it is presently conceptualized, seems to be directed at the elucidation of universal matching guidelines. It may be questioned, however, whether matching guidelines are likely to prove to be universal. Even robust guidelines are likely to be affected by local program conditions (e.g., staffing patterns, unavoidable local variations in the delivery of treatments, specific local traditions and other circumstances; and most guidelines are not likely to be robust. For these and other reasons, research should perhaps be viewed as an important but not an exclusive source of matching guide- lines, and the need to evaluate and possibly modify research guidelines when they are applied to particular programs should be accepted as a fact of clinical life. Data-driven Selection In Chapter 10 the development of a clinical data base was discussed as a way to provide guidelines for matching to various interventions. To recapitulate that discussion briefly, such a data base contains information on individuals, on their problems, on the treatments they have been provided, and on the outcomes that have ensued. When a subsequent individual approaches treatment, those in the data base having similar

294 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS characteristics can be readily identified. The known responses of these similar individuals to treatment can then guide the selection of treatment for other individuals. It was pointed out that this process mimics one of the ways in which an experienced clinician operates. A facet of experience is knowledge about response to various therapeutic approaches and the prediction of future success based on that experience. Thus, in a certain sense this discussion has come full circle and is again ad- dressing selection on the basis of clinical experience. However, the computers that are now commonly the vehicle for clinical data bases can retain and access a larger amount of specific information than can most individuals and (through modern communications techniques) are potentially more widely available for consultation. Tumor registries are perhaps the most familiar of clinical data bases, although they have not been widely used in the manner described above (Laszlo, 1985~. Data bases in other areas have been used in this way (Fries, 1976), and a recommendation has been made (Ellwood, 1988) that, in future, the selection of treatment in medical practice should be largely based on this method. To date, there are no examples of the use of a clinical data base to select treatment for individuals with alcohol problems, but there is no reason in theory why this could not be done. Such a tactic would have many advantages. One is that it would potentially take into account the uniqueness of local circumstances. However desirable it may be to standardize treatment interventions, and however close it may be possible to come in achieving such standardization, the goal will never be fully achieved. Each program will retain some unique characteristics that will differentiate it from other programs, even of the same kind. For one thing, different people will staff the program. For another, different people will seek treatment there. For yet another, each program will develop its own history, ambience, procedural methods, and so forth. Other methods of matching (e.g., theory-driven, problem-driven, and research-driven methods) tacitly assume that all programs of a given type are the same; accordingly, they develop assignment rules that are assumed to be universally applicable. An empirically based selection strategy does not necessarily make such an assumption. Rather, by deriving its guidelines from actual outcome experience on a reiterative basis, it takes account of individual program, staff, population, and other differences insofar as they are affecting out- come. In the time-honored phrase, an empirically derived matching strategy deals with treatment "in our hands," and not with all treatment. It is not possible to be certain at this point, in the absence of actual experience, how decisive an advantage this gives to guide- lines constructed in this manner. But it may be a substantial advantage. In summary, there are several approaches to the construction of guidelines for the matching of individuals to treatment programs. Therapist experience, individual preference, the demands of particular problems, theories, research information, and accumulations of empirical outcome data can all be consulted. None of these approaches needs to be viewed as having exclusive validity in comparison with the others, although each has its potential particular advantages. All can be combined to create an explicit and reasonable set of ini- tial guidelines that can then be continuously reappraised and modified. Eventually, the committee believes, a highly effective set of matching guidelines can evolve. Some Treatment Controversies as Matching Problems In addition to being a method for the potential improvement of treatment outcomes, matching is a perspective that may help to resolve controversies that arise from time to time in the treatment of alcohol problems. Such controversies sometimes arise from sharply polarized thinking about a given issue. Thinking of this kind has been aptly parodied and paraphrased in the statement that "There are two ways of approaching this

MATCHING 295 problem: my way and the wrong ways (cf. Cook, 1985). If approaches of this kind can accurately be characterized as either/or approaches, matching can be said to constitute a both/and approach. A simple analogy to the giving of directions may be instructive. One way of giving directions is to assume that there is only one way to get from point A to point B. An alternative is to assume that there are many ways to get to point Small roads eventually lead to Romen-and that what is required is to choose the route best suited to the needs of the individual traveler. Some will wish to get where they are going as directly as possible, while others will prefer the scenic route. Neither group will embark if they cannot take their preferred route. Two recent areas of discussion may be cited which may benefit from being viewed as problems in matching. One is the issue of inpatient versus outpatient treatment. discussed in the previous chapter. Rather than thinking of either form of treatment as having exclusive validity, one can instead suppose that each is especially valid for some . , _ , . .. . . ~. persons with some kinds of problems (cf. Annis, 1986b). The issue then becomes one of developing appropriate matching guidelines that can successfully discriminate who should go where. The so-called Cleveland criteria, noted earlier in the chapter, are a recent example of such guidelines (Hoffmann et al., 1987~. At present, they lack empirical validation. But they seem a reasonable place to start and embody the view that no one setting or "level of cares is appropriate for all, while at the same time there may be specific indications for each. Another active area of discussion has been centered on the appropriate goals of treatment for alcohol problems with respect to the acceptable level of consumption. All are agreed that a reduction in the consumption of alcohol is a principal goal of the treatment of persons with alcohol problems. Some, however, feel that only a reduction to zero consumption is acceptable, whereas others are willing to accept a lesser degree of reduction, depending on individual circumstances. A matching perspective would suggest that zero consumption is an appropriate goal for some individuals with some kinds of problems but is not necessarily appropriate for others. There is evidence (see Table 11-1) that zero consumption may be a more appropriate, if not always realizable, goal for most persons with substantial or severe alcohol problems; something short of zero consumption, however, may be acceptable for most persons with mild or moderate alcohol problems. (It is likely that appropriate matching here, as elsewhere, will require the use of other information in addition to that regarding severity of the problem for example, age, sex, cultural background, previous history, and so forth.) Whatever turns out to be the case-and at this point it is a matter that requires further investigation-the goal must always be to achieve for the individual the optimum possible outcome for him or her, whether or not that outcome happens to be an ideal outcome. For example, one recent set of guidelines stated (in part) the following: . . The [Addiction Research] Foundation strongly advocates that total abstinence is the appropriate treatment goal for persons with severe alcohol Or `1r'~a nrohl~.m~ IrrP.~n~.otive. Of the. Fn''n~l~tinn'~ no~itinn in this VA ~A ~ PRAVDA - ~O · · · A&~~ VA ~ ~ V - IVAN ~ VAN 84& RABAT matter, those with serious alcohol problems who refuse to accept abstinence, despite the consequences and our advice, still require treatment. In these circumstances, the role of the Foundation's therapists is to  negotiate treatment plans that will reduce the severity of damage. The Foundation will not refuse treatment or care to a client because he or she is either unwilling or incapable of accepting the Foundation's advice related to abstinence as a treatment goal. (Addiction Research Foundation, 1988:2)

296 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS The committee commends the position of the Addiction Research Foundation. Its implementation may require a high degree of flexibility. Uniform goals for all individuals may be a simpler approach, but the notion of different goals for different individuals seems more consistent with the heterogeneity of alcohol problems and the individuals who manifest them. Consistent with its approach on other issues, the committee is prepared to accept differential goals for treatment. To summarize, matching is not only a useful technique for enhancing the effectiveness of treatment but casts new light on at least some treatment controversies. Two such controversies are those that have been presented as "inpatient versus outpatient treatments and Controlled drinking versus abstinence." A matching perspective, considered in the context of the broad spectrum of alcohol problems and of those who manifest such problems, moots the adversarial nature of these controversies. Conclusions and Recommendations Because the accumulated evidence from treatment outcome studies to date indicates that no single treatment approach is effective for all persons with alcohol problems, the effort to match each person seeking help with that treatment that is most likely to facilitate a positive outcome is now a practical necessity. Some research has been carried out that substantiates the value of matching and provides an indication of how to proceed in this regard. More research, however, would be helpful and is in fact going forward. However, the committee believes that matching can be carried out at the present time if care is taken to specify the multiple parameters and components of the treatment situation. Individuals and their problems must be specified through a process of comprehensive assessment (see Chapter 10~. The treatments to which they are being matched must be specified through the development of explicit and detailed written descriptions, such as those embodied in the treatment manuals that are a feature of some research endeavors; one cannot match to treatments of unspecified content, any more than one can determine the outcome of such treatments when provided in unspecified amounts. The guidelines used to make the match must be specified both in order to be examined and, should they prove effective, in order to be used by others. The outcomes of treatments must be specified for a variety of reasons (see Chapter 12), among them the need to test critically the matching guidelines. Under these circumstances it is possible to learn whether a positive match has been effected and to proceed accordingly, either utilizing the proposed guidelines if results have been satisfactory or modifying them if they have not been satisfactory. Ideally, this approach would be undertaken as a cooperative effort among many treatment programs so that each could learn from the other. Such an undertaking would facilitate the research effort, and would also enable the results of research to be incorporated into clinical practice without undue delay once they are available. To facilitate the development and application' Or matching, it is recommended that ar' expert committee be convened on a regular basis (perhaps once every two years) to focus or' what is currents thrown about matching individuals with alcohol problems to appropriate treatments. The committee should consist of representatives of major organizations and disciplines involved in the treatment of alcohol problems, and should include both researchers and clinicians. A written report advising the field on how best to proceed with matching, based both on available research and current clinical experience, should be the expected product of each meeting.

MATCHING 297 REFERENCES Abramowitz, C. V., S. I. Abramowitz, H. B. Roback, and C. Jackson. 1974. Differential effectiveness of directive and nondirective group therapies as a function of client internal external control. Journal of Consulting and Clinical Psychology 42:849-853. Addiction Research Foundation. 1988. The position of the Addiction Research Foundation on reduced substance use as a treatment goal. Approved by Program Policy Committee August 3. Toronto, Ontario, Canada. Angell, M. 1984. Patients' preference in randomized clinical trials. New England Journal of Medicine 310:1385-1387. Annis, H. M. 1979. Group treatment of incarcerated offenders with alcohol and drug problems: a controlled evaluation. Canadian Journal of Criminology 21:3-15. Annis, H. M. 1986a. A relapse prevention model for treatment of alcoholics. Pp. 407-433 in Treating Addictive Behaviors: Processes of Change, W. R. Miller and N. Heather, eds. New York: Plenum Press. Annis, H. M. 1986b. Is inpatient rehabilitation of the alcoholic cost effective? Con position. Advances in Alcohol and Substance Abuse 5:175-190. Annis, H. M. 1987. Effective treatment for drug and alcohol problems: What do we know? Presented at the Annual Meeting of the Institute of Medicine, Washington, D.C., October 21. Annis, H. M. 1988. Optimal treatment for alcoholism and drug dependencies. Presented to Kaiser Permanente, Southern Permanente Medical Group, Los Angeles, California, March 30. Annis, H. M., and D. Chan. 1983. The differential treatment model: Empirical evidence from a personality typology of adult offenders. Criminal Justice and Behavior 10:159-173. Annis, H. M., and C. M. Davis. 1988. Assessment of expectancies. Pp. 84-111 in Assessment of Addictive Behaviors, D. M. Donovan and G. A. Marlatt, eds. New York: The Guilford Press. Annis, H. M., and C. M. Davis. 1989. Relapse prevention. Pp. 170-182 in Handbook of Alcoholism Treatment Approaches: Effective Alternatives, R. K Hester and W. R. Miller, eds. New York: Pergamon Press. Annis, H. M., J. M. Graham, and C. S. Davis. 1987. Inventory of Drinking Situations Users Guide. Toronto: Addiction Research Foundation. Anokhina, I. P., N. N. Ivanets, Y. V. Burov, and B. M. Kogan. 1987. Alcohol and alcohol problems research 13. U.S.S.R. British Journal of Addiction 82:23-30. Appelbaum, P. S., L. H. Roth, and C. Lidz. 1983. The therapeutic misconception: Informed consent in psychiatric research. International Journal of Law and Psychiatry 5:319-329. Azrin, N. H., R. W. Sisson, R. Meyers, and N. Godley. 1982. Alcoholism treatment by disulfiram and community reinforcement therapy. Journal of Behavior Therapy and Experimental Psychiatry 13:105-112. Baekeland, F., and L. K Lundwall. 1977. Engaging the alcoholic in treatment and keeping him there. Pp. 161-195 in Treatment and Rehabilitation of the Chronic Alcoholic. Vol. 5 of the Biology of Alcoholism. B. Kissin and H. Begleiter, eds. New York: Plenum Press.  Berzins, J. I. 1977. Therapist-patient matching. Pp. 222-251 in Effective Psychotherapy: A Handbook of Research, A. S. Gurman and A. M. Razin, eds. New York: Pergamon Press. Beutler, L. E. 1979. Toward speciBc psychological therapies for specific conditions. Journal of Consulting and Clinical Psychology 47:882-897. Bohnen, E. 1984. Background to the Core-Shell client education programme. Pp. 154-168 in A System of Health Care Delivery, vol. 1, F. B. Glaser, H. M. Annis, H. A. Skinner, S. Pearlman, R. L. Segal, B. Sisson, A. C. Ogborne, E. Bohnen, P. Gazda, and T. Zimmerman. Toronto: Addiction Research Foundation. Bowman, K M., and E. M. Jellinek. 1941. Alcohol addiction and its treatment. Quarterly Journal of Studies on Alcohol 2:98-176.

298 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS Brandsma, J. M., M. C. Maultsby, Jr., and R. J. Welch. 1980. Outpatient Treatment of Alcoholism: A Review and Comparative Study. Baltimore, Md.: University Park Press. Brickman, P., V. C. Rabinowitz, J. Karuza, Jr., D. Coates, E. Cohn, and L. Kidder. 1982. Models of helping and coping. American Psychologist 37:368-384. Bromet, E., R. H. Moos, and F. Bliss. 1976. The social climate of alcoholism treatment programs. Archives of General Psychiatry 33:910-916. Bromet, E., R. H. Moos, F. Bliss, and C. Wuthmann. 1977. Posttreatment functioning of alcoholic patients: its relation to program participation. Journal of Consulting and Clinical Psychology 45:829~42. Clarkin, J. F., and S. W. Perry. 1987. Differential therapeutics. Pp. 327-441 in American Psychiatric Association Annual Review, vol. 6, R. E. Hales and A. J. Frances, eds. Washington, D.C.: American Psychiatric Press, Inc. Commission on Accreditation of Rehabilitation Facilities. 1988. Program Evaluation in Alcoholism and Drug Abuse Treatment Programs. Tucson, Ariz.: Commission on Accreditation of Rehabilitation Facilities. Cook, D. R. 1985. Craftsman vs. professional: Analysis of the controlled drinking controversy. Journal of Studies on Alcohol 46:433-442. Edwards, G., J. Orford, S. Egert, S. Guthrie, A. Hawker, C. Hensman, M. Mitcheson, E. Oppenheimer, and C. Taylor. 1977. Alcoholism: A controlled trial of "treatment" and "advice." Journal of Studies on Alcohol 38:1004-1031. Ellwood, P. M. 1988. Outcomes management: A technology of patient experience. New England Journal of Medicine 318:1549-1556. Ewing, J. A. 1977. Matching therapy and patients: The cafeteria plan. British Journal of Addiction 72:13-18. Feldstein, P. J., T. M. Wickizer, and J. R. C. Wheeler. 1988. Private cost containment: The effects of utilization review programs on health care use and expenditures. New England Journal of Medicine 318:1310-1314. Finney, J. W., and R. H. Moos. 1979. Treatment and outcome for empirical subtypes of alcoholic patients. Journal of Consulting and Clinical Psychology 47:25-38. Finney, J. W., and R. H. Moos. 1986. Matching patients to treatments: Conceptual and methodological issues. Journal of Studies on Alcohol 47:122-134. Frances, A., J. Clarkin, and S. Perry. 1984. Differential Therapeutics in Psychiatry: The Art and Science of Treatment Selection. New York: Brunner/Mazel. Fries, J. F. 1976. A data bank for the clinician? New England Journal of Medicine 294:1400-1402. Gibbs, L. E. 1980. A classification of alcoholics relevant to type-specific treatment. International Journal of the Addictions 15:461488. Gibbs, L. E., and J. Flanagan. 1977. Prognostic indicators of alcoholism treatment outcome. International Journal of the Addictions 12:1097-1141. Glaser, F. B. 1980. Anybody got a match? Treatment research and the matching hypothesis. Pp. 178-196 in Alcoholism Treatment in Transition, G. Edwards and M. Grant, eds. London: Croom Helm.  Glaser, F. B., and H. A. Skinner. 1981. Matching in the real world: A practical approach. Pp. 295-324 in Matching Patient Needs and Treatment Methods in Alcoholism and Drug Abuse, E. A. Gottheil, A. T. McLellan, and K Druley, eds. Springfield, Ill.: Charles C. Thomas. Glaser, F. B., S. W. Greenberg, and M. Barrett. 1978. A Systems Approach to Alcohol Treatment. Toronto: ARF Books. Glaser, F. B., H. M. Annis, H. A. Skinner, S. Pearlman, R. L. Segal, B. Sisson, A. C. Ogborne, E. Bohnen, P. Gazda, and T. Zimmerman. 1984. A System of Health Care Delive~y, 3 vol. Toronto: Addiction Research Foundation.

MATCHING 299 Gordis, E. 1987. Accessible and affordable health care for alcoholism and related problems: Strategy for cost containment. Journal of Studies on Alcohol 48:579-585. Gottheil, E. An, ~ T. McLellan, and K A. Druley, eds. 1981. Matching Patient Needs and Treatment Methods in Alcoholism and Drug Abuse. Springfield, Ill.: Charles C. Thomas. Hartman, L., M. Krywonis, and E. Morrison. 1988. Psychological factors and health-related behavior change: Preliminary findings from a controlled clinical trial. Canadian Family Physician 34:1045-1050. Hayes, S. C., R. O. Nelson, and R. B. Jarrett. 1987. The treatment utility of assessment: A functional approach to evaluating assessment quality. American Psychologist 42:963-974. Hoffmann, N. G., J. A. Halikas, and D. Mee-Lee. 1987. The Cleveland Admission, Discharge, and Transfer Criteria: Model for Chemical Dependency Treatment Programs. Cleveland, Ohio: The Greater Cleveland Hospital Association. Holden, C. 1987. Alcoholism and the medical cost crunch. Science 235:1132-1133. Hunt, D. E. 1971. Matching Models in Education: The Coordination of Teaching Methods with Student Characteristics. Toronto: Ontario Institute for Studies in Education. Imber, S. D., L. M. Glanz, I. Elkin, S. M. Sotsly, J. L. Boyer, and W. R. Leber. 1986. Ethical issues in psychotherapy research: problems in a collaborative clinical trials study. American Psychologist 41:137-146. Institute of Medicine (IOM). 1989. Prevention and Treatment of Alcohol Problems: Research Opportunities. Washington, D.C.: National Academy Press. Kadden, R. M., N. L. Cooney, H. Getter, and M. D. Litt. 1990. Matching alcoholics to coping skills or interactional therapies: post treatment results. Journal of Consulting and Clinical Psychology 57:698-704. Karasu, T. B. 1977. Psychotherapies: An overview. American Journal of Psychiatry 134:851-863. Kern, J. C., W. Schmelter, and M. Fanelli. 1978. A comparison of three alcoholism treatment populations: Implications for treatment. Journal of Studies on Alcohol 39:785-792. Kiesler, D. J. 1966. Some myths of psychotherapy research and the search for a paradigm. Psychological Bulletin 65:110-136. Kissin, B., A. Platz, and W. H. Su. 1970. Social and psychological factors in the treatment of chronic alcoholism. Journal of Psychiatric Research 8:13-27. Korcok, M. 1988. Managed Care and Chemical Dependency: A Troubled Relationship. Providence, R.I.: Manisses Communications Group, Inc. Laszlo, J. 1985. Health registry and clinical data base technology: With special emphasis on cancer registries. Journal of Chronic Diseases 38:67-78. I ewis, J. S. 1988. Growth in managed care forcing providers to adjust. The Alcoholism Report 16(24):1. Lindstrom, L. 1986. Val av Behandlung for Alkoholism. Malmo: Liber Forlag.  Longabaugh, R. 1986. The matching hypothesis: Theoretical and empirical status. Presented at the Annual Meeting of the American Psychological Association, Washington, D.C., August 24. Lyons, J. P., J. Welte, J. Brown, L. Sokolow, and G. Hynes. 1982. Variation in alcoholism treatment orientation: Differential impact upon specific subpopulations. Alcoholism: Clinical and Experimental Research 6:333-343. MacDonnell, F. J. 1981. How effective are our current methodologies for the treatment of alcoholism? EAP Digest 2(1):32-35. Maisto, S. ^, and T. D. Nirenberg. 1986. The relationship between assessment and alcohol treatment. Presented at the Annual Meeting of the American Psychological Association, Washington, D.C., August 24.

300 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS Malta, A. K, B. Rush, M. Gavin, and G. Cooper. 1985. A eommunity-eentred alcoholism assessment/treatment service: A descriptive study. Canadian Journal of Psychiatry 30:3543. Marquis, D. 1983. Leaving therapy to ehanee. Hastings Center Report 13(4):4047. Marshman, J., R. D. Fraser, C. Lambert, A. C. Ogborne, S. J. Saunders, P. W. Humphries, D. W. Maedonald, J. G. Rankin, and W. Schmidt. 1978. The Treatment of Alcoholics: An Ontario Perspective. Toronto: Addiction Researeh Foundation. Matakas, F., H. Koester, and B. Leidner. 1978. Welehe Behandlung fur welehe Alkoholiker. Psyehiatrisehe Praxis 5:143-152. Mayer, J., and D. J. Myerson. 1971. Outpatient treatment of alcoholics: Effects of status, stability, and nature of treatment. Quarterly Journal of Studies on Aleohol 32:620-627. MeLaehlan, J. F. C. 1972. Benefit from group therapy as a function of patient-therapist match on conceptual level. Psychotherapy: Theory, Researeh, and Praetiee 9:317-323. MeLaehlan, J. F. C. 1974. Therapy strategies, personality orientation, and recovery from alcoholism. Canadian Psychiatric Association Journal 19:25-30. MeLellan, A. T., C. P. O'Brien, R. Kron, A. I. Alterman, and K A. Druley. 1980. Matching substance abuse patients to appropriate treatment methods: A conceptual and methodological approach. Drug and Alcohol Dependence 5:189-195. McLellan, A. T., G. E. Woody, L. Luborsky, C. P. O'Brien, and K A. Druley. 1983a. Increased effectiveness of Cohen ~h'~.~ treatment A nro6neetive study of Datient-treatment "matching." Journal of Nelvous and Mental rev .~ Avis ~ ^^ y^~r~~~ ~ -- r Diseases 171:597 605. McLellan, A. T., L. Luborsky, G. E. Woody, C. P. O'Brien, and K A. Druley. 1983b. Predicting response to alcohol and drug abuse treatments: Role of psychiatric severity. Archives of General Psychiatry 40:620-25. Merry, J., C. M. Reynolds, J. Bailey, and A. Coppen. 1976. Prophylactic treatment of alcoholism by lithium carbonate: A controlled study. Laneet 2:481-482. Miller, W. R., and R. K Hester. 1986. Matching problem drinkers with optimal treatments. Pp. 175-203 in Treating Addictive Behaviors: Processes of Change, W. R. Miller and N. Heather, eds. New York: Plenum Press. Moos, R. H. 1968. Differential effects of ward settings on psychiatric patients: A replication and extension. Journal of Nervous and Mental Diseases 147:386-393. Moos, R. H. 1974. Evaluating Treatment Environments: A Social Ecological Approach. New York: John Wiley and Sons. Moos, R. H., and D. N. Daniels. 1967. Differential effects of ward settings on psychiatric staff. Archives of General Psychiatry 17:75-82. Moos, R. H., R. Shelton, and C. Petty. 1973. Perceived ward climate and treatment outcome. Journal of Abnormal Psychology 82:291-298. Obitz, F. W. 1975. Alcoholics' perceptions of selected counseling techniques. British Journal of Addiction 70:187-191.  O'Dwyer, P. 1984. Cost effective rehabilitation: A process of matching. EAP Digest 4(2):33-34. O'Dwyer, P. 1988. Executive summary: Assessment procedures currently used by alcohol treatment programs. Prepared for the IOM Committee for the Study of Treatment and Rehabilitation Se~viees for Alcoholism and Aleohol Abuse. Ogborne, A. C., D. Dwyer, and A. Ekdahl. 1984. The Niagara Aleohol and Drug Assessment Service: Referral Patterns, Client Characteristics, and Community Reactions. Toronto: Addiction Researeh Foundation. Orford, J., E. Oppenheimer, and G. Edwards. 1976. Abstinence or control: The outcome for excessive drinkers two years after consultation. Behavior Researeh and Therapy 14:409-418.

MATCHING 301 Pattison, E. M. 1978. Differential approaches to multiple problems associated with alcoholism. Contemporary Drug Problems 7:265-309. Pattison, E. M. 1979. The selection of treatment modalities for the alcoholic patient. Pp. 126-227 in The Diagnosis and Treatment of Aleoholism, J. H. Mendelson and N. K Mello, eds. New York: MeGraw-Hill Book Company. Pattison, E. M., R. Coe, and R. J. Rhodes. 1969. Evaluation of alcoholism treatment: A comparison of three facilities. Archives of General Psychiatry 20:478 488. Pattison, E. M., R. Coe, and H. O. Doerr. 1973. Population variation among alcoholism treatment facilities. International Journal of the Addictions 8:199-229. Paul, G. L. 1967. Strategy of outcome research in psychotherapy. Journal of Consulting Psychology 31:109-118. Polieh, J. M., D. J. Armor, and H. B. Braiker. 1981. The Course of Aleoholism: Four Years After Treatment. New York: John Wiley and Sons. Reynolds, C. M., J. Merry, and A. Coppen. 1977. Prophylactic treatment of alcoholism by lithium carbonate: An initial report. Aleoholism: Clinical and Experimental Research 1:109-111. Rosenberg, S. D. 1979. Relaxation Training and a Differential Assessment of Aleoholism. Ann Arbor, Mich.: Dissertation Abstracts International (University Microfilms No. 8004362~. Rush, B. 1988. Executive summary: Assessment procedures and specialized assessment in Ontario. Prepared for the IOM Committee for the Study of Treatment and Rehabilitation Programs for Alcoholism and Alcohol Abuse. Sanehez-Craig, M. 1984. Therapist's Manual for Secondary Prevention of Alcohol Problems: Procedures for Teaching Moderate Drinking and Abstinence. Toronto: Addiction Research Foundation. Schroeder, S. A. 1987. Outcome assessment 70 years later: Are we ready? New England Journal of Medicine 31eS:160-162. Silberfeld, M., and F. B. Glaser. 1978. Use of the life table method in determining attrition from treatment. Journal of Studies on Alcohol 39:1582-1590. Skinner, H. A. 1981. "Different strokes for different folks": Differential treatment for alcohol abuse. Pp. 349-367 in Evaluation of the Alcoholic: Implications for Research, Theory, and Treatment, R. E. Meyer, T. F. Babor, B. C. Glucek, J. H. Jaffe, J. E. O'Brien, and J. R. Stabenau, eds. Washington, D.C.: U. S. Government Printing Office. Skinner, H. A. 1984. An overview of the Core-Shell Treatment System. Pp. 17-26 in A System of Health Care Delivery, vol. 1, F. B. Glaser, H. M. Annis, H. A. Skinner, S. Pearlman, R. L. Segal, B. Sisson, A. C. Ogborne, E. Bohnen, P. Gazda, and T. Zimmerman. Toronto: Addiction Research Foundation. Skinner, H. A., and K R. Shoffner. 1978. Sex differences and addiction: A comparison of male and female clients at the Clinical Institute. Toronto: Addiction Research Foundation. Sokolow, L., J. Welte, G. Hynes, and J. Lyons. 1980. Treatment-related differences between female and male alcoholics. Focus on Women-Journal of Addictions and Health 1:42-56. Solomon, S. D. 1981. Tailoring Alcoholism Therapy to Client Needs. Washington, D.C.: U. S. Government  Printing Office. Taylor, K M., R. Margolese, and C. L. Soskolne. 1984. Physicians' reasons for not entering eligible patients in a randomized clinical trial of surgery for breast cancer. New England Journal of Medicine 310:1363-1367. Tomsovie, M., and R. V. Edwards. 1970. Lysergide treatment of schizophrenic and nonschizophrenic alcoholics: A controlled evaluation. Quarterly Journal of Studies on Alcohol 31:932-949. U.S. Department of Health, Education, and Welfare. 1971. First Special Report to the U. S. Congress on Alcohol and Health. Roekville, Md.: National Institute on Alcohol Abuse and Alcoholism. Veith, I. 1964. The infancy of psychiatry. Bulletin of the Menninger Clinic 28:186-197.

302 BROADENING lilE BASE OF TREATMENT FOR ALCOHOL PROBLEMS Wallerstein, R. S. 1956. Comparative study of treatment methods for chronic alcoholism: The alcoholism research project at Winter VA Hospital. American Journal of Psychiatry 113:228-233. Wallerstein, R. S. 1957. Hospital Treatment of Alcoholism: A Comparative, Experimental Study. New York: Basic Books. Warren, M. Q. 1969. The case for differential treatment of delinquents. Annals of the American Academy of Political and Social Science 381:47-59.

Appendix: Excerpts From The Cleveland Admission' Discharge and Transfer Criteria The Cleveland Adm~sswr~, Discharge and Transfer Criteria are guidelines for use by treatment providers in documenting placement decisions, by third party payers in monitoring placement, and by researchers in evaluating treatment. The guidelines were prepared for the Northern Ohio Chemical Dependency Treatment Directors Association by the CATOR/Ramsey Clinic and published by the Greater Cleveland Hospital Association (Hoffman et al., 1987:3-33~. In their original form the Cleveland Criteria include criteria for both adults and adolescents. Only portions of the adult criteria are reproduced in this appendix. The criteria are presented as guidelines to be used carefully with consideration of both individual and combinations of symptoms when making placement decisions. Exceptions based on extenuating circumstances (e.g., admission to a higher level than that specified in the guidelines) require presentation of further justification. Overview of Adult Admission Criteria Diagnosis In general all individuals accepted for the treatment of chemical dependency in these levels of care are expected to have met diagnostic criteria for a psychoactive substance use disorder as defined by DSM-III, DSM-III-R or other standardized and widely accepted criteria. Exceptions to this would be individuals who have experienced only a few problems or adverse consequences as a result of their use and who would benefit from further assessment or low levels of therapeutic involvement. It is assumed that diagnostic assessment typically can be accomplished in two to four hours of contact time with adult patients. This may need to be extended in some unusual cases or where the collection of collateral information is neeessa~y. Levels of Care Levels of care have been identified which reflect the general consensus of a wide range of previously developed criteria available to this project. The type and extent of treatment services necessary define each level, not the specific site of the services. Staffing and local factors also contribute more to costs and charges than simply the site of the program. These criteria assume not only that the levels of care are available, but also that they are reimbursed at an adequate rate. Numerous illustrations can be found where reimbursement policies or logistical and economic factors effectively preclude provision of services at a given level of care. It is assumed that clinical parameters and standards of care, no economic policies, are the driving force for the appropriate treatment placement. The criteria are designed to place individuals into an appropriately intensive treatment level in accordance with appropriate standards of clinical practice. Movement between levels typically is expected to be from a higher, or more intense, level to a lower level as the patient's treatment progress warrants. Any clinical criteria should be considered as guidelines which must evolve with advances in clinical procedures and with accumulating empirical evidence. Placement Criteria Consideration Iwo factors must be considered in making treatment placement decisions which override the patient 303

304 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS treatment match determined by these criteria: (1) prior treatment failure and (2) availability of the criteria- selected level of care. A treatment failure at any given level of care indicates the need for treatment at a higher level of care, unless the patient is uncooperative or would not benefit from such a treatment placement. These criteria attempt to minimize the occurrence of failures by initially placing individuals into an appropriately intense level of care. However, the explicit incorporation of documented prior treatment failures provides for a correction mechanism when a failure occurs. The second consideration is the availability of the optimal level of care indicated by the criteria. If that level is not available, the next higher level of care available should be utilized rather than a less adequate lower level of treatment. Both of these considerations should be relatively automatic and not require extensive justification for utilization review or reimbursement. While these criteria are intended to be as inclusive as possible, unique clinical presentation or extenuating circumstances may require some flexibility in the application of the criteria to insure the safety and welfare of the patient. Dimensions for Making Placement Decisions The following seven dimensions have been identified as the general assessment dimensions used nationally in treatment placement criteria and guidelines. These dimensions are analogous to the categories in the routine medical "Review of Systems." 1. 2. 3. 4. 6. 7. A. Brief Description of Treatment Level Acute alcohol and/or drug intoxication and/or withdrawal potential; Physical complications; Psychiatric complications; Life areas impairments; Treatment acceptance/resistance; Loss of control/relapse crisis; Recovery environment. LEVEL I: MUTUAUSELF-HELP This level of care encompasses services provided by organizations such as Alcoholics Anonymous, Narcotics Anonymous, Cocaine Anonymous, Alateen, Emotions Anonymous and other mutual self-help groups. It may also include other formal social support or peer groups that do not rely on other recovering persons to provide mutual/self-help support and therapeutic milieu. The principal criterion for admission to mutual/self help groups is that the individual wishes to stop using alcohol and/or other mood altering substances. There is no role for formal diagnostic determinations at this level; however, individuals may seek to reflect on their substance use and its consequences (a self- assessment). B. Programmatic Description  1. Sexing Any appropriate setting for group meetings and peer counseling. Support systems No direct support required, but a referral list or mechanism for referral to treatment and other services is highly recommended. S~,9 No professional staff required. Volunteers typically provide administrative services. Therapies Peer interaction/mutual help. Assessments/trea~merd plan review Self/peer assessment.

APPENDIX 6. Documentation None required. C. Brief Description of Typical Patient 305 Individuals who feel they may have an alcohol/drug use problem may seek the advice and counsel of peers based on trust, respect, and privacy in a mutual self-help setting. Patients who require professional assessment or treatment services would not be appropriate for Level I care alone. D. Dimensional Admission Criteria suggested. No formal admission criteria are required for this level; however, the following specifications are Acute alcohol and/or drug intoxication ardor potentu~l withdrawal a. History of recent or current drinking and/or drug use may be evident; b. No indication of serious physical complications due to current intoxication or withdrawal. Physical conditions and complications a. Physical state is sufficiently stable to permit participation in mutual help treatment. Psychiatric conditions or complications a. Any anxiety, guilt, and/or depression present is (are) related to chemical use rather than another psychiatric condition; , , . b. Mental state sufficiently stable to permit participation in mutual self help; c. Sufficient cognitive functioning to understand the requirements of mutual help. 4. Life areas impairment a. b. Subjective report of problems related to use; Evidence of adverse consequences (vocational, social, or legal) related to use may be evident. 5. Treatment acceptance/resistance a. b. Willingness to cooperate and attend activities; Feels that life areas impairment is associated with alcohol/drug use or wishes to use mutual self help to self-evaluate status of use. 6. Loss of co~rol/relapse crises a. Acknowledges personal problem with use. 7. Recovery endowment a. Sufficient environmental stability to indicate safe and adequate living conditions and support of recovery goals.  LEVEL II: LOW INTENSITY OUTPATIENT TREATMENT Brief Description of Treatment Level Low Intensity Outpatient Treatment is the provision of outpatient services by appropriately trained professionals. This level of care involves weekly sessions usually supplemented by involvement in Mutual/Self Help. As in primary treatment, intensity typically does not exceed 10 contact hours per week over a period of one to three months. This level of care may be utilized in maintenance services for persons who have recently completed a more intensive level of care. Such maintenance treatment typically invokes no more than four hours per week for a period of up to one year. This level of care also may be appropriate for protracted evaluation of patients who require some additional time to make a commitment to a more intensive recovery effort.

306 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS B. Programmatic Description Sewing Any appropriate professional setting. 2. 3. 4. Support systems Contractual availability of specialized professional consultation and supervision. Staff Any appropriately trained personnel, e.g., certified chemical dependency counselors, licensed mental health professionals, etc. Therapies Focused counseling and/or medical monitoring. 5. Assessment/treatment plan review Documented initial treatment plan with update at least every 30 days. 6. Documentation Problem oriented progress notes at each visit. C. D. Brief Description of Typical Patient For admission to this level of care the patient should have few or no continuing symptoms of withdrawal or intoxication; manifest only stable (if any) physical or psychiatric conditions; demonstrate sufficient motivation and supportive environmental factors to participate in a low intensity, low control treatment program. A typical patient at this level may have completed a more intensive program but still requires maintenance services. Patients inappropriate for Level II include those who are at risk for imminent relapse without more intensive services and/or those who cannot stay focused on recovery efforts and goals because the severity of any of the criteria dimensions distracts them from following through with a low intensity plan. Dimensional Admission Criteria  seven dimensions: Admission to this level of care requires meeting the specifications in all of the following Acute alcohol and/or drug intoxication and/or potential withdrawal a. Discharge referral from a higher level of care or; b. One of the following: (1) (2) (3) 2. Physical conditions or complications History of recent or current drinking and/or drug use; Mild symptoms related to withdrawal such as: headaches, insomnia vague somatic complaints; History of withdrawal syndrome with withdrawal symptoms that can safely be managed without medical intervention. a. Any physical conditions, if present, are sufficiently stable to permit participation in outpatient treatment. Psychiatric conditions or complications~aU of the following: a. b. c. Significant anxiety, guilt, and/or depression, if present, are related to chemical dependency problems rather than another psychiatric condition; Mental state is sufficiently stable to permit participation in the outpatient treatment program; Sufficient comprehension to understand the materials presented. Life areas impairment-one of the following: a. Absence from work as a consequence of substance use; b. Occupational difficulties as a consequence of substance use; c. Legal difficulties as a consequence of substance use;

APPENDIX 6. 7. 307 e. Deteriorating job performance as a consequence of substance use; Damage of personal relationships as a consequence of substance use; Documentation of substance use great enough to damage social functioning as a consequence of substance use; Patient has completed a more intensive program but still requires services until recovery efforts result in additional stability. 5. Treatment acceptance/resistance~all of the following: Willingness to participate in the treatment program and attend all scheduled activities; Evidence that the patient has restorative potential (e.g., accepts treatment; absence of organic mental disorder that would prevent improvement); Willing to work toward recovery goals with minimal supervision. Loss of Control/Relapse Crzses-eitherof the following: a. Recent history of use despite commitment not to use; b. Patient is able to maintain short term abstinence goals with support and therapeutic contact. Recovery environmental of the following: a. b. c. Sufficiently supportive psychosocial environment for low intensity outpatient treatment (e.g., significant others in agreement with recovery efforts; supportive work or legal coercion); Lack of environmental impediments to recovery (e.g., adequate transportation to program, accessibility of support meeting locations);  Adequate environmental stability to indicate safe adequate living conditions and support of recovery goals (e.g., spouse is not active alcoholic). LEVEL V: MEDICALLY SUPERVISED INTENSIVE INPATIENT TREATMENT A. Brief Description of Treatment Level This level of care utilizes a multidisciplinary staff for patients whose physical, psychiatric, and/or psychosocial problems are severe enough to require inpatient services. Twenty-four hour observation, monitoring, and treatment should be available; however, the full resources of an acute care general hospital system are not necessary. The treatment is specific to chemical dependency, but the multidisciplinary team and availability of support services allows for the conjoint treatment of coexisting physical and/or psychiatric conditionals) which could jeopardize recovery. B. Programmatic Description 4. Setting A free-standing residential facility or specialty hospital with a chemical dependency treatment unit. General hospitals may also have units at this level of care, but the full resources of such a medical system are not essential for this level of care. Support systems Availability of specialized professional consultation and supervision; direct affiliation with more intensive levels of care. Stay Multidisciplinary team of appropriately trained professionals; 24 hour availability of physician with direct patient management as necessary; 24 hour skilled nursing care and observation (16 hours of nursing coverage if based in a JCAH approved general or psychiatric hospital where supplemental nursing coverage is available); counseling services are available on site 16 hours per day. Therapies A multidisciplinary team will be available to meet the individual needs of each patient; at least four skilled treatment services will be provided per day for at least five days per week. Skilled treatment services include but are not limited to: psychotherapy, family

308 C. D. BONING ME BASE OF EVENT FOR ALCOHOL PROBLEMS 5. therapy, individual and groups counseling, educational groups, occupational and recreational therapy. Assessments/treatment plan review Documented initial treatment plan; ongoing reassessment; weekly update of treatment plan by multidisciplinary treatment team. Documentation Daily problem orientation progress notes. Brief Description of Typical Patient Admission to this level of care requires at least one of the following: a significant likelihood of the development of a withdrawal syndrome; previous history of having failed at attempts at outpatient withdrawal; the presence of a medical condition serious enough to warrant inpatient management; the presence of isolated medical symptoms of concern; external mandates for inpatient treatment; a recent history of inability to function without some externally applied behavioral controls; and significant denial of the severity of his/her own addiction. In addition, high motivation for a multifaceted intensive inpatient treatment program does not preclude admission to an inpatient setting provided other criteria are met. In addition, loss of control or impending relapse crises is suitable for the level of care. Detrimental recovery environmental factors likely to prevent a patient from maintaining treatment progress merits admission to this level of treatment. Dimensional Admission Criteria Admission to this level of care requires the presence of one or more symptoms from at least one of the following seven dimensions. 1. Acute alcohol and/or drug intoxication and/or potentu~l withdrawal-at lead one of the following:  c. f. Patient is in need of 24 hour observation and/or skilled nursing care; History of significant multiple alcohol/drug ingestion during present episode; Failure of ambulatory detoxification; Inability to continue in, or lack of availability of, ambulatory alcohol/drug detoxification program; Motor and/or gait incoordination; For opioid withdrawal-the presence of two or more withdrawal symptoms warrants admission: Rhinorrhea; lacrimation; mydriasis; piloerection; bone pain; diarrhea. 2. Physical condihons or complicailons-at lead one of the following: b. Physical conditions related to the excessive use of alcohol and/or drugs (e.g., metabolic abnormalities, which are not life threatening but impair physiological functioning, severe enough to warrant inpatient treatment, such as unstable diabetes); Documentation of alcohol/drug use great enough to damage physical health. 3. Psychiatric conditions or complications-at Cast one of the following: a. Significant depression but no active suicidal ideation; c. Recent death or other significant personal loss causing daily distress and distraction from recovery efforts; Overly complaint and pathologically dependent personality disorders that interfere with recovery without direct intervention e.g., individual who needs intensive therapy interventions to improve recovery chance); Histrionic display of emotions or hostile dependency with attempts to have others take responsibility. Life areas impairment-at least one of the following: a. Violent behavior while intoxicated; b. The patient has demonstrated an inability to sustain independent functioning without a controlled environment; Legal system (e.g., probation officer, courts, etc.) mandates the patient to participate in an inpatient treatment program.

APPENDIX 6. 309 [Note: a court order or legal mandate resulting from criminal behavior or civil commitment should not require a treatment program to accept a patient that the staff feels is inappropriate for that program or facility. Likewise, adequate reimbursement for treatment dictated by such mandate should be the responsibility of the court.!; Treatment mandated by employee assistance referral program. Treahnen' acceptance/resistance~at lead one of the following: a. Absence from three or more scheduled outpatient treatment sessions without adequate explanation, arrangement, or documentation; Significant denial and minimization of the effects of chemical dependency which would prevent the patient from following through with rehabilitative care in a lesser setting; Loss of control/relapse cr~sis~at least one of the following: a. Currently unable to effectively control chemical use at the time of evaluation; b. Requires the continuous use of alcohol and/or other drugs in order to function c. d. e. adequately; If abstinent, the patient is experiencing an acute crisis and feels himself going out of control and possibly reactivating his/her addiction and outpatient services have failed to improve the crisis; Demonstrated inability to remain substance free for at least five days; Failure to maintain abstinence while engaged in an outpatient program. 7. Recover environments least one of the following: a. b. c.  d. e. g 1. j. Introduction in treatment: Actual or threatened major losses (e.g., death, divorce, job change); Severe isolation or withdrawal from social contacts; Lives in an environment (social and interpersonal network) in which treatment is unlikely to succeed (e.g., a chaotic family, rife with interpersonal conflict, which undermines patient's efforts to change); Seriously impaired social, family or occupational observation/care in a structured inpatient treatment program (i.e., patient is unable to abstain from the use of chemicals, and this condition and associated behaviors result in the patient's inability to function on the job or in the home, in even a limited capacity); Patient's family and/or significant others are opposed to his/her treatment efforts and not willing to participate in recovery program; Family members and/or significant others living with the patient manifest current substance use disorders, and are likely to undermine the patient's recovery; Living situation not conducive to life style change necessary for recovery; Logistic impediments (e.g., distance from treatment facility, mobility limitations, lack of drivers license, etc.) preclude participation in outpatient treatment services; Danger of physical, sexual, and /or severe emotional attack or victimization in his/her current environment will make recovery unlikely without removing the individual from the environment; The patient is employed in an occupation where his/her continued employment could jeopardize public or personal safety in the event that he/she resumed use as a consequence of treatment failure. ADULT DISCHARGE & TRANSFER CRITERIA These Discharge & Transfer Criteria are measured along several dimensions of patient participation Dimension 1 (Acceptance/Awareness) is predicated on the patient's ability to accept treatment and his or her awareness of the need for treatment.

310 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS Dimension 2 (Cooperation) assesses the patient's ability to cooperate with treatment. Dimension 3 (Follow-Through) assesses the patient's ability to follow the designed treatment plan. Dimension 4 (Medical Stability) assesses the stability and control of any existing biomedical problems. Dimension 5 (Psychological Stability) assesses the stability and control of any existing psychiatric or psychological problems. Dimension 6 (Environmental Support) assesses the extent to which the patient's external environment is supportive of treatment goals. These Discharge & Transfer Criteria dimensions are compatible with but not identical to the Admission dimensions. The Admission Criteria dimensions also involve historical information which is less relevant to the status evaluation for discharge or transfer. In this conceptualization of treatment the traditional discharge may be considered as a transfer to mutual/self-help networks (Level I) and/or aftercare, or maintenance, services (Level II). While Level I or Level II care may also be primary treatment for select cases, these levels would typically provide additional assistance after a more intensive treatment experience. The current conceptualization explicitly deals with these less intensive services on their own merit rather than as adjuncts to more intensive programs. Exceptions to Transfer Criteria In some cases a patient's conditions may have improved so as to warrant transfer to a lower level of care, but a transfer which preserves the appropriate continuity of care is not available. In other cases it may be more efficient to retain the patient on the higher level of care for an additional brief period and effect a full discharge (or transfer to Level I or II) earlier than would be possible with a transfer to an intermediate level of care. Under these conditions the transfer guidelines appropriately may be overridden. Transfer to a Higher Level of Care Transfer to a higher level of care is warranted if any of the following apply: There is deterioration in any of the transfer criteria dimensions. 2. There is lack of progress in any of the transfer criteria dimensions. 3. There is identification of additional chemical dependency related problems. and all of the following apply: 1. There is evidence of restorative potential in the areas involved. 2. Greater intensity and/or range of services would positively impact the problem areas involved. 3. The required service(s) is (are) not available at the current level of care. Transfer to a Lower Level of Care In general, transfer to a lower level of care is warranted if all of the following apply:  4. There is sufficient improvement across all the transfer criteria dimensions; There is sufficient progress in all appropriate dimensions; There is sufficient resolution of problems; Lower intensity services will continue to positively impact remaining problem areas; Such services are available and can be provided while maintaining the continuity of care. A more detailed and explicit set of guidelines for transfer and discharge are provided in the following section. A discharge could be viewed as a transfer from the current level to either Level I or Level II without any intermediate level services. Levels are presented from the highest to the lowest.

APPENDS LEVEL V: MEDICALLY SUPERVISED INTENSIVE INPATIENT TREATMENT All of the following must be present for a patient to be transferred from intensive inpatient treatment to a lesser level of care. 311 Acceptance/Awareness Patient's awareness and acceptance of an addiction problem and commitment to definitive treatment is sufficient to expect treatment compliance in a less structured setting. Cooperation Patient's cooperation and acceptance of treatment is high enough that the need for motivating strategies has diminished sufficiently to reduce counseling observation, and availability of outpatient services to at least five hours of contact time per day for at least five hours of contact time per day for at least four days all week. 3. 5. 6. I Follow-Through Patient is capable of following a specific and complete post-inpatient recovery treatment plan. The patient's integration of therapeutic gains is established enough that transfer to a lesser level of care does not substantially risk reactivating the patient's addiction. Medical Stability Patient's biomedical problems, if any, have diminished or stabilized to the extent that daily availability of skilled medical/nursing care is no longer necessary. Psychological Stability Patient's psychological/psychiatric problems have diminished in acuity to the extend that daily availability of skilled psychiatric care is no longer necessary. Environmental Support Patient's social system and significant others are supportive of recovery to the extent that the patient can adhere to a post-inpatient treatment plan without substantial risk or reactivating hislher addiction.  LEVEL II: LOW INTENSITY OUTPATIENT TREATMENT All of the following must be present for a patient to be transferred from Outpatient Treatment to a lesser level of care. 1. Acceptance/Awareness Patient's awareness and acceptance of an addiction problem and commitment to recovery is sufficient to expect maintenance of a self-directed recovery plan. 2. Cooperation Patient's cooperation and acceptance of treatment is high enough that the need for motivating strategies has diminished sufficiently to allow contact to be at the patient's discretion and experienced need. 3. Follow-Through Patient is capable of following a self-motivating recovery plan involving mutual/self-help support groups. The patient's integration of therapeutic gains is established enough that transition to a lesser level of care does not substantially risk reactivating the patient's addiction. 4. Medical Stability Patient's biomedical problems, if any, have diminished or stabilized to the extent that they can be managed through outpatient appointments at the patient's discretion as the need arises.

312 BROADENING THE BASE OF TREATMENT FOR ALCOHOL PROBLEMS 5. Psychological Stability Patient's psychological/psychiatric problems have diminished or stabilized to the extent that they can be managed through outpatient appointments at the patient's discretion as the need arises. 6. Environmental Support Patient's social system and significant others are supportive of recovery to the extent that the patient can adhere to a self-directed treatment plan without substantial risk or reactivating hisser addiction.