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.
20 Paying for the Treatment System This chapter considers possible changes in the methods that are currently used to pay for the treatment of alcohol problems. As outlined in earlier chapters, the major sources of funding for treatment of alcohol problems are private health insurance and governmental agencies (i.e., tax-financed public insurance or public production of treatment); out-of-pocket reimbursement is a much less dominant source. Under private insurance, a person obtains coverage through the payment of a premium himself or by an employer and enters a risk pool along with others who have purchased the same coverage. Under tax-financed public insurance or public production of treatment, a person obtains coverage through an entitlement-attaining a particular status through illness. age. or a given legal or social condition. ~, by, ~ Both methods offer a "benefit plan" to those persons for whom there is coverage of medical care expenses by a third-party payer even if that coverage is not described as a benefit plan. A benefit plan is the specific description of the services that will be paid for, the providers who are eligible to be reimbursed for providing those services, the amounts that will be paid, and the mechanisms by which such services will be authorized and paid for on behalf of the plan's beneficiaries. From the committee's perspective, for example, the alcohol, drug abuse, and mental health block grant funds given by the federal government to each state for use in purchasing services is a benefit plan with minimum restrictions (e.g., that these funds cannot be used to purchase inpatient hospital treatment). The state determines who will be eligible to receive services purchased with these funds and distributes them to venders; the venders provide the state's list of approved services to eligible clients. In some states, the state passes the funds through to the county to distribute with minimum restrictions; an example is the California Division of Alcohol Programs, which provides a formula allocation of state, county, and federal funds to purchase nonhospital detoxification, residential recovery services, and nonresidential recovery services using a social model to define eligible providers and services. In other states, the block grant is passed through to counties with a very strict legislatively defined method for level of care placement; an example is the Minnesota Chemical Dependency Program Division, which provides a formula allocation to counties to purchase both hospital and nonhospital services for eligible beneficiaries meeting stringent income guidelines. (These funding mechanisms are described in greater detail in Chapters 4 and 18~. An example of a more traditional benefit plan would be that offered by a commercial insurance company that provides employers with coverage for detoxification as part of the company's basic medical plan at no extra premium cost. Such a plan might have the following elements. Detoxification is covered under the same deductible and copayment schedule as other physical illnesses. Coverage for rehabilitation is offered at an ^~^^ ~ Rehabilitation in a hospital or residential setting is covered for UD to 30 days with an additional deductible of $200 and additional premium of 51.2() ner month per enrollee. _ 1 a 20 percent copayment requirement; rehabilitation in an outpatient or day care setting or rehabilitation provided in a private practice setting by a physician or licensed clinical psychologist is covered to a maximum of $1,000 per year, and requires a 50 percent copayment. ' 1 ~, A benefit plan is, simply, a list of eligible services and venders that the plan will reimburse on behalf of its beneficiaries. To a large extent the criteria to be used for determining whether a benefit plan should provide coverage for a given condition, procedure, or provider are the same regardless of whether insurance or taxation is the source of payment. The coverage criteria are the same for Medicaid, for the categorical 464
PAYING FOR THE TREATMENT SYSTEM 465 funds administered by the state alcoholism authority, for the block grant, or for health insurance offered by a commercial firm, an HMO, or Blue Cross/Blue Shield. To set the stage for a discussion of past, present, and future financing levels, it would be useful to know what is being spent today, relative to the costs to society of all the damage caused by alcohol problems does (e.g., Harwood et al., 1984~. Unfortunately, existing data do not provide a complete picture of total spending. There are reasonably accurate estimates of spending in specialized treatment facilities (see Chapter 8), and it is possible to construct estimates of inpatient costs for persons hospitalized with a primary diagnosis of alcohol abuse and alcohol dependence. What is quite speculative, however, is the cost of treatment of people with alcohol abuse or alcohol dependence who contact the medical care system for some other complaint as well but who receive treatment for their alcohol problems without a recorded diagnosis. The cost of treatment for hospital admissions with an alcohol-related secondary diagnosis can be estimated, but what is not known is what portion of the cost of those hospital episodes represents direct treatment of the alcohol consumption problem, as opposed to the other conditions. Using 1983 data adjusted for inflation (Harwood et al., 1985; Davis, 1987), the committee estimates that the level of 1988 spending on specific formal treatment of alcohol problems in all health care settings was more than $6 billion. It was not possible to account for the cost of the informal treatment associated with secondary diagnoses that occurred in hospitals, primary care settings, and nursing homes, but it is unlikely that much of these expenditures were for treatment of the alcohol problem; probably most of the expenditure was for treatment of the primary physical problem. Though not included in previous estimates, the cost of informal treatment outside the medical sector (e.g., Alcoholics Anonymous) is estimated as a real cost (even though no money changes hands) of perhaps $1.5 billion. In total, treatment costs for alcohol problems, even using liberal estimates that will yield a high number, will probably not exceed $10 billion. In comparison with these current levels of expenditure, the estimated cost to society of alcohol problems is more than 10 times as high, generally estimated at $117 billion (Davis, 1987; U.S. Department of Health and Human Services, 1987; Gordis, 1987~. These estimates are based on a study conducted by the Research Triangle Institute (Harwood et al., 1984; U.S. Department of Health and Human Services, 1986~; an update is currently in progress, but the findings will not be available until next year. Nevertheless, it seems possible to come to the firm conclusion that the current cost of treating alcohol problems is quite small relative to the costs to society that could be avoided if alcohol problems were successfully treated. _ . . In the discussion that follows, the committee has considered the changes that will be required in terms of two options: first, the changes that are necessary to improve the current system of specialist treatment with its multiple sources of financing and variety of methods to access the system and, second, the changes that are required to pay for treatment in the proposed ideal comprehensive treatment system outlined in Chapter 13. In that system (as discussed in Chapter 13) access to the appropriate orientation, setting, and modality at each stage would be controlled through assessment, matching, outcome monitoring, and continuity assurance activities. Financing the Current Treatment System Developing and implementing the committee's ideal treatment system will take some time, but it is not unrealistic to consider implementation within a five-year period. In the meantime, the current system of treatment should be financed in a way that is more appropriate than at present and which is conducive to the emergence of the ideal system.
466 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS There is obviously a connection between the cost-effectiveness of care (and the absence of evidence for cost-effectiveness) and the certainty that coverage is desirable. Criteria for Evaluating Coverage Appropriateness Insurance appropriately covers the cost of uncertain medical care expenses. In an ideal world, the best insurance coverage would be coverage that paid only for appropriate care and nothing else. Insurance itself is costly because of the administrative cost associated with collecting premiums and paying benefits. If the cost of appropriate care is sufficiently large and sufficiently uncertain, then risk-averse people will gain from buying insurance coverage against the cost of this care. Out-of-pocket payment is appropriate when the expense is small, when the expense is more nearly certain, or when full insurance coverage would stimulate inappropriate use. These principles immediately imply that if there were any type of care which was known to be ineffective for a particular type of problem or person, that type of care should not be covered at all for that type of problem or person. When care is known to be effective (in the sense of yielding positive net benefit), coverage is appropriate if the value placed on that benefit is great enough and the expense is sufficiently large and uncertain. If there are two alternative ways of achieving the same benefit, then the benefit plan should cover only the least costly way. Finally, when a type of care is of uncertain effectiveness and there is no care that is less costly and of certain effectiveness, policy directions become murkier. In instances in which reasonable people differ about the expected benefit to be received from some type of care, coverage will obviously be appropriate for those who think the benefit is great enough. Where the objective evidence is incomplete but is nevertheless regarded as sufficiently compelling to suggest that the type of care is a reasonable gamble, one could say that coverage is, indeed, appropriate. Yet it would not be appropriate to compel someone who reads the evidence in a different way to buy coverage. In instances in which the evidence is more ambiguous, in which one's conclusion about what is a "good bet" depends on how one reads the literature and what subjective probabilities one attaches to some very uncertain relationships, then perhaps the best strategy is to label coverage for that type of care as neither appropriate nor inappropriate but "neutral." Those who are sufficiently convinced of a positive effect should buy the coverage, but policy advice should be neutral on recommending that a benefit plan include coverage for that type of care. In real insurance settings, the insurance contract cannot specify or monitor the precise set of circumstances in which benefits are to be provided. Variations in the severity of illness, in the type of person, and in individual preferences can all lead to different choices regarding the appropriate level of care. If, however, insurance pays simply on the basis that care has been rendered and costs incurred, there is a chance that more costly care, even if it is not much more effective than some other type of care, will nevertheless be rendered if both types of care are fully covered by insurance. If it is not possible to write an insurance contract (i.e., benefit plan) that specifies what is covered unequivocally, without any danger of misinterpretation, what other options are available? As discussed in Chapter 18, there are three broad strategies that have been used: (1) develop insurance that provides complete coverage for care that is usually appropriate (i.e., Effective, The least cost, given its effectiveness, and Effective enough, given the costs), less complete coverage (coverage with cost sharing through deductibles and copayments) for care that is less frequently appropriate or of uncertain appropriateness, and no coverage for care that is rarely appropriate; (2) develop coverage that nominally pays for a very wide continuum of services but requires prior approval from an insurer-employed case manager; and (3) develop very broad coverage, but set the maximum
PAYING FOR THE TREATMENT SYSTEM 467 price that suppliers will receive at a sufficiently low rate that they will tend to supply the more effective forms of care (i.e., use a reimbursement/payment strategy). These three strategies are not mutually exclusive; they will, however, be treated separately in the discussion below in order to examine the strengths and weaknesses of each. The separation also benefits the committee's development of a comprehensive coverage strategy, as well as recommendations to implement its proposed plan. Before proceeding, however, it is important to note that none of these three methods are perfect. Imposing cost sharing provisions or failing to cover some services at all necessarily means that insurance will sometimes fail to pay for care that is appropriate. The only way to avoid the Denial of needed care" (actually the "denial of payment for needed cares) is to generate more situations in which payment is made for care that is low in value or high in cost. Moreover, in practice, the use of case managers is another imperfect method. Case managers cost money; they may also lack the information needed to judge appropriateness perfectly. Sometimes it is difficult to motivate them to use properly the information they can obtain. Finally, paying providers just enough to do things right means that people who use inefficient providers or who use providers that call the insurer's "bluff" will not get the care they should. These principles help us judge alternative ways of financing treatment for alcohol problems. Clearly, if a treatment were to prove ineffective under all circumstances, or even to be equally effective but more costly than available alternatives, insurance coverage would not be appropriate. Judgments about the appropriate financing of services of known or possible effectiveness are more difficult because of uncertainties about effectiveness and the extent and need for care, the value of the effect, and the possibility of insurance-induced inappropriate use all of which need to be taken into account. These issues are considered in the following discussion of the three broad coverage strategies noted earlier in this chapter. Optimal Conventional Coverage We begin by using these principles to define what might be desirable coverage of the conventional sort, coverage without extensive case management that pays market prices for medical services. This type of insurance uses out-of-pocket payments to affect beneficiary and provider choices. If a type of care is thought to be wholly inappropriate, there is no coverage, and the person who nevertheless still wants the service must pay for it. Thus, out-of-pocket payment serves as a deterrent to inappropriate use. Coverage will be regarded as appropriate either when it covers treatment that is costly but of great benefit to the individual or if it significantly stimulates use that prevents further utilization (and cost) of other types of already insured medical care. The committee began its review by attempting to separate the types of care into those that are known to be effective, those that are of probable effectiveness, those that are of uncertain effectiveness, and those that are not known to be effective-all compared with lower cost alternatives. The guiding principle in formulating a recommended strategy is that coverage of care of uncertain (or rare) effectiveness will not be desirable-even though the lack of coverage discourages use of care that is occasionally beneficial. This review obviously required a reconsideration of the evidence on effectiveness described in Chapter 5 and Appendix B. Evaluation of the appropriateness of coverage can be no better, and no more certain, than the evidence on the effectiveness of care. Given the lack of adequate cost-effecfiveness studies comparing alternative treatments, it is not possible at this time to say definitive, which treatments should and should not be covered. It is possible to make recommendations on which strategies should be followed in providing coverage.
468 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS On effectiveness grounds, brief interventions, as described in Chapter 9, might be highly appropriate candidates for full insurance coverage. However, the cost of brief interventions is so low that such coverage would be efficient only if it stimulated use. The reason is straightforward: most people could afford the small cost of a brief intervention, and the financial risk attached to this low cost is not great. Insurance coverage would only lead to wasteful paperwork. Things are quite different if the use of brief interventions should prove to be responsive to coverage and if there is a substantial cost offset in the form of lower future medical costs. If insurance coverage of brief interventions causes many more people to seek this form of treatment, given the existence of the cost offset, it would be desirable to expand the use of this type of treatment. However, the critical empirical fact how the demand for brief intervention responds to its user price-is unknown. In addition, the evidence on the cost offset is more in the Probably a good bets than in the "sure thing" class. More important than financing per se (i.e., having insurance pay for brief interventions) would be strategies to ensure that such treatment is available for purchase by larger populations and that information about the treatment's usefulness is provided to the public at large so that they can appropriately seek to purchase brief intervention rather than the more expensive intensive interventions. Making brief interventions widely available in the larger community and within the specialized treatment system will require a reorientation of our existing treatment system, training of treatment providers to use these techniques, and extensive campaigns to inform the public of their availability. Marginal efficacy has been established for selected more intensive treatments but there has been no documentation of the marginal cost of these interventions or, more importantly, of whether the additional benefit is worth the cost. (That is, for some methods and for some cases, it is known that treatment that is more costly than brief interventions is beneficial, but it is not known whether the additional cost is justifiable.) For more intensive treatment of possible but uncertain effectiveness and for more costly treatment locations (e.g., hospitals or other residential settings), the question arises as to whether the additional benefits that may be realized justifier the additional expenditures required to achieve them. Answering this question is a task for future research. As an interim measure, the committee suggests that insurance coverage for intensive and costly services should continue to be furnished but that these services should be utilized cautiously and according to explicit guidelines developed through the consensus process recommended in Chapter 18. These observations suggested to the committee that its recommendations should be designed to encourage financing for a broad range of alternative settings and modalities at each stage in the treatment process and to permit limitation of the more costly settings to persons with conditions for which that setting is generally appropriate. Thus, with regard to the current' available system of identification and treatment, the committee recommends that public and private insurance should provide coverage for the following: (a) assessment, reassessment, and continuity assurance to facilitate matching to the appropriate level and intensity of treatment at each stage; (by brief interventions for alcohol problems, if coverage is needed to facilitate the use of the service; (c) detoxification and other forms of acute intervention in the lowest cost setting of appropru~te quality; (d) rehabilitation in the lowest cost setting of appropriate quality; (eJ maintenance in the lowest cost setting of appropriate quality; and O treatment for as long as is clinical necessary with no prospect umber of inpatient days and/or outpatient sessions as par! of desired coverage. The intent of this recommendation is to encourage financing for a broad range of alternative sites for therapy at each stage in the treatment process, a strategy which would permit limitation of the more costly sites to persons with conditions for which that site is generally appropriate. As transition occurs and financing moves toward a more ideal system in which assessment and matching become a primary guide to appropriateness, then
PAYING FOR THE TREATMENT SYSTEM 469 particular restrictions should become unnecessary. For example, given the lack of support for differential effectiveness of the length of stay in an inpatient rehabilitation program for heterogeneous groups, there seems to be no reason to include a prespecified number of inpatient days or outpatient visits as part of the desired coverage. Rather, the goal would be to provide treatment for as long as clinically necessary. The committee fisher recommends that coverage for detox~fcalion and other forms of acute intervention, rehabilitahon, awl maintenance be provided u' both social model and medical model programs. If nonmedical professionals are providing appropriate care of high quality, medical supervision should not necessarily be required for provision of the benefit. Medical consultation, however, should be available and covered by the benefit. Given the findings of studies that found no greater positive outcomes of treatment in medical model programs compared with mixed medical-social model and social model programs, a medical presence should not necessarily be required for coverage of the service. There has been enough clinical experience and research to persuade the committee that appropriate care of high quality has been and can be provided in these ~nonmedical" programs under certain conditions: when there is state licensure or professional accreditation, when there are procedures for referring persons with physical and psychiatric comorbidities to the appropriate treatment setting, and when there is matching of persons seeking treatment to the appropriate orientation, setting, and modality. From a cost-effectiveness perspective, it is desirable to implement a significant redistribution of resource use from more intensive to less intensive medical model and social model treatments. Therefore, the committee recommends that in each commur~i0, the full range of alternative treatments be established arid covered by both public and private f nancir~g mechanisms. That such a distribution of alternative treatment is currently not available can be inferred from the uneven distribution of the "types of cares reported in the 1987 NDATUS and described in Chapter 7. The committee anticipates that the availability of such alternatives as brief intervention in community settings, day care programs, residential programs, and intensive primary rehabilitation outpatient programs will result in a significant redistribution of resource utilization. The committee believes that such a redistribution is consistent with its understanding of appropriate care within current practices. There are several issues regarding the redistribution: the rate at which it would occur, the magnitude, and the way in which the distribution is to be accomplished. It is clear to the committee that, from what we now know of cost-effectiveness, some restraints should be placed on insurance coverage of the more intense hospital- and residential-based treatments. On the other hand, for a meaningful redistribution to occur, it will be necessary to provide incentives for the development of coverage for less intensive options. Changes in coverage-that is, restrictions on coverage for inpatient and residential treatment-should not be put in place prior to the development of coverage for alternatives and the establishment of quality alternative programs. Implementation of these recommendations will require availability of the full range of appropriate resources. One sort of restriction is to provide coverage but to specify the types of conditions or situations in which resources are to be utilized. Numerous criteria for determining the appropriateness of an individual's admission to a treatment program and for his or her continuing stay are now operative in insurance and government contracts with private and public sector treatment providers. As discussed in Chapter 18, there has been a proliferation of managed care programs, each with its own criteria for determining the appropriate level of care for a person being treated. As yet there is no common framework; the definitions of levels of care and the criteria used by such programs vary. Committee members held different opinions regarding appropriate criteria for the utilization of the inpatient level of care. Some members suggested that expert clinical determinations of appropriateness should be the primary guide to resource utilization.
470 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS Others maintained that sufficient data were at hand to sharply and categorically restrict inpatient coverage options for care. The committee has reviewed a number of the sets of criteria now in use and found none that it could recommend unequivocally. GINO sets of criteria the committee reviewed are examples of the efforts now under way to restrict the use of more expensive hospital and residential settings. One is the criteria developed for the implementation of a consolidated chemical dependency treatment fund, a statewide program recently instituted by the state of Minnesota (see Chapters 10 and 18~. In each county, assessors use uniform guidelines established by the state's Chemical Dependency Program Division to determine the level of care placement for persons who meet the financial eligibility requirements for coverage (Minnesota Chemical Dependency Program Division, 1987, 1989; Minnesota Department of Human Services, 1987~. Criteria have been developed for placement in either outpatient care, primary residential care in a freestanding facility or in a hospital setting, or extended care in a residential facility or a halfway house. (These components of the Minnesota continuum of care were described in Chapter 4.) The basic principles of the Minnesota criteria are that persons should be offered rehabilitation in the least restrictive setting consistent with sound clinical judgment and the availability of an appropriate program. Exceptions can be made, for example, if an outpatient program is not available within a 50-mile radius of the home of the person seeking treatment and the assessor and the individual can agree on an alternative placement. Exceptions can also be made when a culturally specific program at the appropriate level of care is not available and the assessor and individual agree on placement in a culturally specific program at a different level. The placement criteria employ an assessment of a number of factors to determine the appropriate level of care: level of chemical involvement, social and vocational functioning within the past year, physical health status, emotional health status, history of prior treatment for alcohol and other drug problems, history of specific behaviors when under the influence, and family and friends' support for treatment and achieving program goals. Assessors classic applicants for rehabilitation services into four levels of chemical involvement: (1) no apparent problem, (2) at risk of developing future problems, (3) chemical abuse, or (4) chemical dependence. To be referred for treatment, persons must be assessed as a chemical abuser or as chemically dependent. The specific criteria used are contained in Figure 20-1, which is a copy of the placement summary that documents the appropriateness of the level of care placement. To be placed in a licensed outpatient treatment program, a person must be assessed as capable of functioning in the usual community environment in spite of the existing chemical use. He or she is either assessed as chemically dependent or as a chemical abuser who has experienced either an arrest or other legal intervention related to chemical use in the past year, loss or impairment of employment or education as a result of chemical use, or the deterioration of family relationships due to chemical use. In contrast, to be placed in primary treatment in either a freestanding residential or hospital setting, the person must be assessed as chemically dependent and unable to abstain from chemical use outside a residential facility that controls access to chemicals. He or she must also be experiencing one or more of the following conditions: loss or impairment of employment or education owing to chemical use; lack of family support; an arrest or legal intervention related to chemical use in the past year; or participation in a chemical dependency treatment program within the past year. To be placed in primary treatment in a hospital setting, the person must also be assessed as either having a physical complication that requires more than detoxification or brief or episodic nursing care or having a mental disorder that requires more than brief or episodic nursing care but that does not otherwise prevent the person from participating in and benefiting from treatment.
PAYING FOR THE TREATMENT SYSTEM 471 Preliminary data from the state's first year of experience with these uniform placement criteria suggest that there has been an increased use of outpatient and halfway house programs and a decreased use of inpatient residential and hospital programs (Minnesota Chemical Dependency Program Division, 1989~. Unfortunately, there could be no direct comparison of placements under the old system and the new system because the consolidated fund mechanism has brought together for the first time in one billing system diverse funding sources that previously did not report to the state alcoholism authority. Yet the preliminary analysis is that more persons are receiving treatment at a lower per episode cost in a greater variety of licensed programs. The criteria have not yet been rig DHS-2794 (7-87) PZ-02794-02 PLACEMENT SUMMARY CHEMICAL USE ASSESSMENT GENERAL GUIDELINE: . Clients should be offered the least restrictive referral consistent with sound clinical judgment. All items checked must be clearly documented in the client file. This form should remain in the client file. DEFINITIONS: CHEMICAL ABUSE: A pattern of inappropriate and harmful use which exceeds social or legal standards of acceptability, the outcome of which includes at least three of the following: Circle those which are documented. A. Weekly use to intoxication; B. Inability to function in a social setting without becoming intoxicated; C. Driving after consuming sufficient chemicals to be considered legally impaired under Minnesota Statutes, section 1 69.1 21, whether or not an arrest takes place; D. Excessive spending on chemicals that results in an inability to meet financial obligations; E. Loss of friends due to behavior while intoxicated; or F. Chemical use that prohibits one from meeting work, school schedule, or social obligations. CHEMICAL DEPENDENCY: Must meet both PART 1 and PART 2 PART 1: either n markedly increased tolerance or ~ . .._ I I withr~r~w~l cvnr~rnm~ PART 2: a pattern of pathological use. This means compulsive use including at least three of the following. Circle those which are documented. A. Daily use required for adequate functioning; B. An inability to abstain from use; C. Repreated efforts to control or reduce excessive use; D. Binge use, such as remaining intoxicated throughout the day for at least two days at a time; E. Amnesic periods for events occurring while intoxicated; and F. Continuing use despite a serious physical disorder that the individual knows is exacerbated by continued use. LICENSED HALFWAY LICENSED OUTPATIENT LICENSED PRIMARY RESIDENTIAL LICENSED EXTENDED O Chemically Dependent O Chemically Dependent O Chemically Dependent PLUS PLUS O Unable to Abstain FOUR OR MORE OR PLUS O Prim. Resid. Past 2 Yrs ONE OR MORE O Legal in Past Year O Chemical Abuser PLUS O Loss/lmpair Voc/Educ Due to O M D Documents Physical Chemical Use Caused _ C.D. Deterior Due to C D _ Legal in Past Year Rae OR LJ Lacks Family Support O Lacks Family Support Rae O Legal in Past Year O Lost Voc/Educ Due to CD Ad Loss/lmpair Voc/Educ ~ OR Ld C.D. Treatment in Past Year O Documented Mental Disorder O Family Rel. Deterioration OR Under Control O Hospitalization due to physi- O Lacks Recognition of Need to Cal or mental disorder Change L] Exception | C: Exception C] *Exception *SEE BACK AND CHECK WHICH ' :ATEGORY: C] 1. O2. OF . O4. O5. O6. O7. O8. OOTHER (Explain) O Chemically Dependent PLUS C: Now in Outpt or was in Detox, Primary, Extended PLUS THREE OR MORE O Unable to Abstain O Lacks Family Support O Lost Voc/Educ Due to CD 3 Lacks Helpful Social Network O Documented Mental Disorder Under Control O *Exception FIGURE 20-1 Minnesota Chemical Dependence, Program Division Level of Care Placement Form.
472 BROADENING ITIE BASE FOR TREATMENT OF ALCOHOL PROBLEMS orously evaluated; although comprehensive evaluation is currently being undertaken. It should be noted that these criteria are tied to the existing state continuum of care and a county based funding and assessment mechanism as defined in Minnesota and may not be easily transportable to other states. The so-called Cleveland criteria (which are discussed in Chapters 11, 15, and 18) represent another, more easily transferable example of the kind of level of care placement guidelines that could be used to determine reimbursement availability as well as to provide clinical guidance in the matching of patients to appropriate levels of care (Hoffmann et al., 1987~. In this system, there are six levels of care, that differ in intensity, involvement of professional staff, degree of environmental control, and degree of medical and nursing care available. At each level the type of treatment is specified along with a description of the expected programmatic features of the treatment. Levels of care are ordered from the least to the most intensive and the treatment episode is conceptualized in stages in which it is possible to transfer the person from a higher to a lower level of care as treatment progresses, as well as from a lower to a higher level of care if there is deterioration or a lack of progress. Using such a stepped approach, coverage guidelines could be developed requiring less intensive approaches be tried before the more costly and intensive approaches are used. Although the Cleveland criteria utilize many of the same variables as are used by the Minnesota consolidated fund uniform criteria in making the level of care placement, the two systems differ in their specific cutoff points for each level and in their definitions of treatment settings and stages. As yet neither has been empirically validated. The lack of uniformity and of empirical validation are also characteristic of other sets of criteria currently in use and evoked concerns from the committee regarding their use-at least at present-as tools for decisions about level of care. In addition to its consideration of existing sets of criteria, the committee also reviewed a list of criteria prepared by some of its members. These criteria were designed to provide guidance for the use of the most intensive and costly forms of care, namely, hospital and residential care. The guides for coverage would be as follows: 1. Detoxification and other forms of acute intervention should be carried out in the lowest cost setting of appropriate quality. a. Detoxification should be carried out in a residential setting if, and only if, one or more of the following conditions is present: 1. potentially severe withdrawal symptoms or 2. a nonsupportive psychosocial environment. Detoxification should be carried out in a medical or psychiatric inpatient hospital setting if, and only if, one or more of the following concomitant health conditions is present: 1. withdrawal symptoms that require close medical monitoring and continuous nursing care; severe comorbid medical conditions; or severe comorbid mental illness. 2. Rehabilitation should occur in the lowest cost setting of appropriate quality. a. Rehabilitation should be carried out in a nonresidential (outpatient or day care) setting as the preferred treatment strategy. b. Rehabilitation should occur in a residential setting if, and only if, it is required by the disruption of the individual's psychosocial environment. Rehabilitation should take place in a medical or psychiatric inpatient hospital setting if, and only if, one or more of the following concomitant c.
PAYING FOR THE TREATMENT SYSTEM health conditions is present in addition to the disruption of the individual's psychosocial environment: 1. severe comorbid mental illness or 2. severe comorbid medical conditions. 473 Although the Minnesota consolidated fund criteria allow for exceptions and some committee members endorsed the continuing use of clinical judgement, other committee members felt that allowing providers to waive coverage restrictions could, in select circumstances, lead to excessive use of the inpatient option. When the clinical judgment system is used, criteria such as the two sets presented above can form the basis for postadmission audits to determine whether clinicians who are allowed to use their clinical judgement are using each level of care appropriately. Of course, in the long run, outcome monitoring would determine the appropriateness of any set of placement criteria and any exceptions allowed. As was discussed in Chapter 11, placement at the proper level of care is essentially a problem in matching, and any set of criteria must be empirically tested against the criterion of most favorable outcome and lowest marginal cost. Although the committee does not recommend a particular set of criteria for inpatient coverage in this report, it has concluded that, in general, a significant number of persons now cared for in inpatient facilities could receive appropriate care in less restrictive and less costly settings, if and when treatment in such settings were available. As discussed in Chapters 11 and 18, the committee recommends that a consensus acedia be carried out to develop a common set of deft nitions and criteria for determining appropriate type and level of care placenta at each stage of treatn~r~t, arid that the criteria that are adopted be based or available research as well as the broadest range of shared clinical e~enence. Although a number of vehicles could be designed to conduct this activity, an expert committee convened on a regular basis to review what is known about matching would seem to be the most appropriate vehicle. There was agreement within the committee that, if more alternative treatment approaches and guidelines for their use are to be placed in the existing system in order to achieve a redistribution of resource use, a transition period would be essential. Such a transition, while providing the mechanisms for redistribution (including restrictions and incentives), would take into account the overall effects of such a policy on patient care. In particular, it is imperative that the alternative forms of less intensive and less costly treatments (e.g., partial hospitalization, social setting detoxification, outpatient care) should be available before constraints are placed on more intensive treatments. The committee reflected on other experiences germane to this consideration for instance, the problems that arose with the deinstitutionalization of patients from psychiatric hospitals (IOM, 1988) and the decriminalization of public intoxication (Finn, 1985; Wittman and Madden, 1988~. In these situations less intensive or alternative services were specified but were not adequately provided in many communities. The committee wishes to emphasize that what it is recommending here is an expansion and redistribution of resource utilization and not an abrupt curtailment of existing services. These recommendations avoid, on the one hand, the current problem of a lack of realistic coverage for more cost effective outpatient alternatives and, on the other hand, the problem of inpatient use expanding to the maximum number of days for which coverage is available. Specifying in insurance documents the conditions or disorders for which certain types of treatment are appropriately covered is bound to be imperfect; consequently efforts to match persons to sites or types of treatment will add benefits to this system. As proposed, coverage does, however, cover the spectrum of treatment alternatives so that any appropriate match should carry coverage along with it. At one extreme end of the spectrum, coverage for inpatient treatment is unlimited in terms of days but is limited in terms of conditions. At the other extreme, it may not
474 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS even be necessary to pay explicitly for brief interventions: the advice and information that constitute such "treatments may already be included in a covered medical encounter or may be of such low cost that explicit coverage is unnecessary to encourage use. In both cases, there is considerable room for research to refine these indications. Moreover, these recommendations speak only to minimum desirable coverage. A buyer, especially a private buyer, who wants to provide more generous coverage would, of course, be free to do so if the cost were covered either through the premiums paid or the tax revenue available. The ideal method would be to cover treatment for alcohol problems in the same manner that coverage is provided for effective treatment of medical problems, rather than under a separate benefit with the restrictions described above. Inappropriate use would then be controlled by the same mechanisms that are available for other medical conditions, and any restrictions would become part of the utilization review standards rather than a model benefit. Therefore, the committee recommends that coverage for the treatment of alcohol problems should be subject to the same deductibles, coinsurance, limits, case management, arid utilization review as are applied to coverage of treatment for other medical conditions. Of course, as the coverage in conventional insurance changes-whether that change is a modification in cost sharing or in greater use of case management the same sort of change would be appropriate for the coverage of treatment for alcohol problems. This "uniform coverage" recommendation is most easily applicable when the insurance policy is a so-called "comprehensive policy, paying the same fraction of approved charges for all medical services after the deductible is covered. If the base plan provides different levels of coverage for different types of treatment (e.g., Medicare), coverage for treatment for alcohol problems would follow the same general rules as does coverage of treatment for physical problems. Paying for the Current Treatment System If these propositions about desirable, undesirable, and possibly desirable (or optional) insurance coverage are near the mark, how should they affect policy in private and public insurance, the two major methods of paying for treatment of alcohol problems? This question is considered in the discussion that follows. Encouragement, Subsidization, or Mar~dation of Private Health Insurance There are several ways to affect private group health insurance purchases. One way is simply to provide information to buyers about which coverage makes the most sense, which coverage does not make sense, and why. Benefits managers, whether employed by the employer or by a union, do not want to waste money and do not want to deny benefits of high value (e.g., Tsai et al., 1988~. Consequently, they may well pay attention to such encouragement, especially if the information and advice is embodied in a model policy that is carefully constructed. A second method is subsidization. Group health insurance already receives substantial federal and state/local tax subsidies because of the exclusion of premiums and benefits from taxation. For many firms (as noted earlier), especially large firms with high-wage workers, this subsidy has been sufficient to stimulate coverage. Because the size of the tax subsidy rises with the marginal tax rate on wages, the subsidy, not surprisingly, is strongly related to an employee's likelihood of receiving coverage for treatment of alcohol problems (or any coverage). The notion that the current tax subsidy is not well designed to achieve efficiency, equity, and social objectives is not new (Pauly, 1986~; the additional possibility here would be to limit tax deductibility to coverage that meets the appropriateness criterion the committee has outlined. If inpatient treatment of alcohol problems should be reserved for a restricted subset of persons with alcohol problems, there is no obvious reason to permit payments for such coverage for other individuals (should
PAYING FOR THE TREATMENT SYSTEM 475 someone wish to offer them such coverage) to escape taxation. Any taxes collected on such payments could be used to enhance the tax subsidy to appropriate coverage, especially for low-wage workers. The mandating of coverage by government is a third method of affecting private health insurance purchases. As discussed in Chapter 18, many state governments now mandate that treatment for alcohol problems be covered in any health insurance policy, or that its coverage be offered as an option. Frequently, the mandate specifically includes inpatient treatment and sometimes does not include outpatient treatments or treatment in nonmedical settings. Therefore, the commiace recommends that existing state manual and public programs be modified to be consistent with its recommend stra - . Should new mandates and public programs be enacted, they should be designed to reflect these recommendations. Those national organizations which have been involved in the development of model benefit packages (e.g., the National Council on Alcoholism, the National Association of State Alcohol and Drug Abuse Program Directors, NIAAA, the Alcohol and Drug Problems Association, the National Association of Insurance Commissioners) should be encouraged to develop a model policy format that embodies these principles. The committee, however, did not take a position on state mandates per se. On the one hand, encouraging coverage for effective treatment that provides substantial benefit is obviously desirable. On the other hand, the encouragement mandates may give to self-insurance and the possibility that other mandated benefits that push up premiums will accompany mandates for treatment for alcohol problems, possibly driving premiums out of reach of some purchasers, suggest that mandates are not always the best way to encourage coverage. In some circumstances providing persuasive information on the benefits of coverage for treatment of alcohol problems may well be preferable to new mandates. Conducting adequate research on the cost-effectiveness of alternative treatments is critical to providing such information to those who develop the specifics of financing policy; it is also necessary for any strong endorsement of mandates. The lack of research in this area is one of the causes of the continuing discomfort on the part of third-party payers with current treatment providers. Recognizing the difficult and complex nature of mandated coverage of all types, the committee is not prepared to suggest either extension or contraction of this mechanism. The committee suggests that mandates be reviewed to determine whether they support the principles outlined in this chapter (i.e., the committee's interpretation of the current research findings). Judicious use of case management, described in the next section, may be an effective way to deal with existing mandated inpatient benefits under the current treatment system. Case MarlagEd Coverage As discussed in Chapter 18, case management has become a more prominent strategy for obtaining appropriate resource use and cost containment by both public and private third-party payers (e.g., Temkin-Geser and Clark, 1988; Tsai et al., 1988~. The focus of case management has been primarily on reducing the cost of lengthy inpatient stays, and the treatment of alcohol problems within expensive hospital settings has been a major target of these efforts. If case management could be perfect, if case management could be costless, if case management could be the alter ego of ideal matching, recommending coverage policy would be simple: there should be complete coverage of a whole spectrum of services with case management determining what would actually be covered in each individual case. There is no rigorous evidence on the cost or the effects of case management, however; nor is it known how to bring about an approximation of perfection. On the positive side, it certainly seems sensible to encourage work toward a system of case managed coverage for treatment of alcohol problems, the more so because case management fits so well with the matching approach the committee believes, in
476 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS principle, to be most effective for delivering treatment. The development of model benefit packages and model case management procedures should be a high priority. Such an approach would permit the treatment of alcohol problems to be covered in the same nominal way as any other medical service, although case management for this type of care might lead to a quite different pattern of actual benefits. The committee prefers to speak of the required function and role of case managers as "continuity assurance" (see Chapter 131. ~. , This distinction is critical in that case management has become identified (primarily by providers, although also by employers and persons seeking treatment) with the effort to cut costs by reducing the care provided, rather than by developing a plan for matching a person to the most appropriate treatment regimen at each treatment stage. The committee's stance is that the addition of assessment and reassessment, coupled with the emphasis on outcome monitoring in its proposed system, makes case management a legitimate clinical endeavor, one that has been severely lacking in the treatment of alcohol problems under a unitary model of problems and treatment. The information-subsidization-mandation question also arises regarding case managed coverage. Such coverage should, if it fulfills its promise, virtually sell itself to most insurance buyers, public or private. Mandation, however, raises some additional issues. The employer or union that only adds benefits for treatment of alcohol problems only because of a mandate will presumably seek a case manager that is especially adept at limiting benefit payments. It may prove administratively difficult to deal with a mandated case-managed benefit, and the committee's previous cautions on mandation also apply here. One strategy to prevent reluctant employers or unions from limiting benefits through case management would be to set a minimum average or expected value for such benefits. Supply Side or Reimbursement Limits Reimbursement or payment policy can also be used to promote more cost-effective care. Payment on a prospective basis, payment of an amount sufficient on average to pay only for outpatient care, and other uses of _ reimbursement incentives that restrict access to certain settings may he wortnwn~e. Snecific evidence of the impact of these strategies on the treatment of alcohol problems is r r ~ _ lacking at this time. lnere~ore, it Is premature lo craze once approuc;~= us flu ~ur;~` the adoption of one or more specific mechanisms. Given the committee's recommendation that the coverage of treatment for alcohol problems be the same as that for all medical conditions, it sees no reason to exclude this treatment from the current experimentation with alternative reimbursement methods. It suggests only that there be very careful study of the specific effects of reimbursement methods on utilization and outcome. If a population of persons with alcohol problems could be identified, the payment of providers on a capitation basis would probably be desirable. Public Fu~ancir~g and Public Insurance Governments, especially state governments, currently play a major role in the financing and delivery of treatment for alcohol problems. Historically, the role has been undertaken as part of government's function of expressing the community's concern about the well-being of its less fortunate citizens. Beyond this, however, it is difficult to find a coherent articulation of the public sector's special role in financing treatment or in providing treatment for people with alcohol problems. The enormous variation across states in the extent of public involvement probably reflects this uncertainty. The committee views public provision of care (the source of care paid for through taxes) as largely an expression of the state's welfare function-a manifestation of the desire of all citizens that those who might injure themselves or others be helped if they do not have the resources to help themselves. Governmental involvement in the treatment of alcohol problems takes a variety of forms. At one extreme, some people are eligible for various kinds of government insurance that pays for some treatment of alcohol problems. (For example, Medicare
PAYING FOR THE TREATMENT SYSTEM 477 provides limited coverage and Medicaid provides a more extensive range of benefits for the categorically eligible poor and disabled, all available through providers of care who conform to the medical model of treatment.) Some states effectively administer their categorical program of treatment for alcohol problems as if it were a disease-specific insurance policy for the categorically eligible poor and disabled. State programs generally are available through a wide range of providers who use both the social and medical models. At the other extreme, some state and county systems are basically producers of treatment for alcohol problems, controlling production, volume, and cost directly by means of ownership or contracting, or both. For historical reasons, states have been major funders and providers of public treatment for alcohol problems. The federal government's role has receded in recent years, while that of local governments seems to be growing. Since more than half of all specialty treatment for alcohol problems is paid for by state and local funds, and the states and local governmental units are major providers of treatment as well, practical and political considerations suggest that the main locus of governmental control should remain with the states. To the extent it is possible, this financial obligation should then be matched with the financial control embodied in a consolidated fund which would permit more rational spending of public funds. The Minnesota Consolidated Chemical Dependency Treatment Fund is an example of mechanisms for controlling access and cost which should be followed by other state alcoholism authorities and the federal government. The federal government, in particular, needs to consider such consolidation in both its contributions to community-based service provision through the block grant, Medicare, Medicaid, and the federal employees' health insurance program and its direct operations in the Department of Defense, the Department of Veterans Affairs, and the Indian Health Service. The kind of "coverage" that would be provided by governmentally sponsored public health insurance would not differ in design from the coverage for appropriate care that was discussed earlier in this chapter. It seems to be well-established practice that such appropriate care need not be as extensive, as costly, or (often) as inefficient as what well-to-do citizens might choose to buy for themselves. The notion of appropriateness here seems quite consistent with that of the President's Commission for the Study of Ethical Problems-in Medicine and Biomedical Behavioral Research (1983~. Practical and political considerations also imply that state spending for treatment will probably remain means tested through a sliding scale of charges or subsidies (e.g., eligibility and fee level determined by income and other expenses). At some sufficiently adequate income level, the state subsidy would be terminated altogether and the person expected to pay the full cost, either directly or through private insurance. The same considerations that were applied to private insurance regarding case management and reimbursement limits are pertinent to public financing. Again, the committee asserts that the same general principles for determining optimum coverage and for controlling costs should govern both publicly and privately financed treatment for alcohol problems. Financing the Ideal Treatment System In considering methods for financing the treatment system proposed by the committee, the overall principle is that public and private insurance financing should cover a broad range of treatment alternatives so that care will be effective and worth the cost. The new method of screening for and treating alcohol problems described in Section III of this report is justified, the committee believes, on the grounds of producing large benefits relative to its cost. Moreover, intrinsic to the method is a set of devices for matching and monitoring the quality and cost of care that should limit care to that which is appropriate.
478 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS As a consequence, it should not be necessary to use out-of-pocket payments or benefit limitations as substitute methods for controlling costs. The committee recognizes that broadening the base of the spectrum of intervention responses to include more individuals with mild and moderate alcohol problems and .sharnenin~ the anex to include more outpatient treatment capacity may raise concerns that ~ -~ ~ .r ' ~ ,. . . · · ~ it is recommending vast increases In the amount or tuning neeaeu for ~r~m`;~. Although data on which to develop projections of any additional costs are not available, the committee believes that any additions can be largely paid for by the savings generated from matching persons to the most appropriate intervention strategies early in the course of development of their alcohol problems when treatment is likely to be more effective and less costly. Yet sound estimates of any additional costs to be generated or any costs to be saved cannot be developed because there are few good data available. The committee attempted to develop such projections and rejected the approach because of the paucity of sound data of costs in each of the sectors that will be impacted by the change. Instead, the committee has chosen to illustrate how the new system would be financed and what its cost-benefit would be by continuing to elaborate on the vignettes introduced earlier (see Chapter 2~. Consider, for instance, the case of Elizabeth who lives and works in the California wine country. Elizabeth was hospitalized after vomiting blood and passing out. Her high level of daily wine consumption, which was previously thought to be in conformity with the practices of her family, is now identified as a problem, and she is seen She is employed and has health by her physician as requiring specialist treatment. insurance that will cover up to 30 days of rehabilitation for alcohol problems in a general hospital; the policy has a $1,000 limit on outpatient treatment. She would probably be willing to seek treatment, and the prognosis would be good. Given the orientation of the majority of treatment programs under the current system, her presenting symptoms, and insurance coverage. referral would most likely be to a f~ed-length 28-day medically _ O supervised, hospital-based primary rehabilitation program at a cost or approximately $15,000 (based on the 1987 average cost per day of $537 for 28 days). In contrast, working from the alcohol problems perspective, as presented in Chapter 2 and Section III, Elizabeth would be seen as experiencing substantial alcohol problems and to be a good candidate for outpatient rehabilitation in a freestanding day care program with medical consultation (following her discharge from the hospital for treatment of her physical problems) at a cost of $3,900 ($130 per session for 30 sessions). ~ lnere would also be the likelihood of a substantial cost offset, (that is, a reduction in future costs for treating the alcohol-involved physical illness); consequently, the net cost would be reduced for both treatments if she entered treatment immediately. The prevalence of persons with substantial alcohol problems, like Elizabeth, is estimated at 6 percent of the adult population (18 years of age and above). The likelihood that she and others who are experiencing substantial problems will be identified and referred to treatment is unknown, but estimates range from 15 to 55 percent. Persons with problems similar to those of Elizabeth are estimated to account for approximately two-thirds of those who enter specialist treatment in hospital settings under the current system in which the full spectrum of intervention responses is not available. Assuming that a significant proportion of these individuals can be identified earlier and can be treated in lower cost residential and outpatient settings or in brief intervention settings involving no additional cost, there should be sufficient savings to pay for the cost of outcome monitoring and continuity assurance and to increase the proportion of persons who can be treated. The amount of savings depends on the proportion of current admissions that can be redirected. There is no set of studies available that can be used to develop projections. Such studies are sorely needed.
PAYING FOR THE TREATMENT SYSTEM 479 Consider Patrick, the foundry worker who enjoys drinking with his workmates in the evening and on weekends. Under the current system, Patrick's need for rehabilitation would probably not be identified despite his recent absenteeism and family history. In the proposed system, however, Patrick's moderate level of alcohol problems would be picked up through screening; the optimal referral would be to a brief intervention in the general health care system or, if available, in his company's employee assistance program. The additional cost for a brief intervention if done in the general health care system is estimated to be zero or, at most, $280 ($70 per session for 4 sessions), if done in the specialist system. These costs are in contrast to a zero explicit medical cost under the current system. Under the current system, David, the star salesman whose drinking is injuring his health, is most likely to be referred by his physician to a specialist inpatient hospital program because of his health problems, his "denials of the existence of excessive drinking, and his work environment, which is not considered to be supportive of recovery. In any event, David would probably refuse a referral to any treatment, particularly an inpatient program. More appropriate choices would be a brief intervention or a course of outpatient rehabilitation. If David's physician were trained in brief intervention and relapse prevention techniques, then this method would be appropriate and less likely to be resisted. For David, like Patrick, this brief intervention would add a cost where there had been none previously. Bringing additional persons with a moderate level of alcohol problems like David and Patrick into treatment suggests an aggregate cost increase. Estimates of the prevalence of this level of problems or of the number of such individuals now in treatment at each level of care are not available to serve as the basis for projecting aggregate costs and savings. Provided such persons can be successfully treated using brief interventions, however, any increase should be small. Yet any near term cost increase is likely to be much less than if either David or Patrick continued drinking and later needed more intensive and expensive medical, psychiatric, or alcohol problems treatment. Let us assume that Sally, the young assistant receptionist with a speech impediment, has recognized spontaneously that her daily relief drinking has become a problem and seeks assistance. Her spontaneous recognition that she had moved from drinking to relieve her anxiety to drinking for enjoyment makes her a good candidate for treatment. Under the current system, Sally most likely would be referred for a course of outpatient treatment. Under the committee's proposed system, Sally would be an excellent candidate for a brief intervention. Here, the savings in the individual case could be approximately $1,800, the difference between the cost of an outpatient rehabilitation episode and a brief intervention. Once they have been identified and referred to formal specialist treatment, the proportion of persons with mild alcohol problems like Sally who are treated in outpatient settings is estimated to be about one-third under the current 1 _' system in which the full spectrum of intervention responses recommended by the committee is not available. This proportion is projected to decrease by half under the proposed comprehensive system. Redirecting to brief interventions those persons with mild problems ~ 1_ _ _ _ 1 ~ _ 1 ~ 1 ~ ~ (e.g., belly) who are now being treated in oulpauenl programs snouts a~o you ~ ~;r savings. Consider the cases of Jimmy, who has chronic severe alcohol problems, and William, who has chronic but intermittent substantial alcohol problems. Jimmy has been through standard inpatient rehabilitation programs and Alcoholics Anonymous several times, relapsing after a period of several months on each occasion and continuing to deteriorate emotionally, physically, and socially. William has not received any formal treatment and manages to maintain a marginal work and social existence between drinking bouts. Both require extended formal treatment (maintenance and relapse prevention) of the type that does not now exist in most communities~immy in a long-term alcohol-free living facility (categorized as custodial/domiciliary in the 1987 NDATUS) and William in
480 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS an outpatient setting. There is likely to be no savings generated in either of these cases, because the activity required is not currently available. In Jimmy's case, a cost offset may occur as the lower cost supportive living environment reduces the need for repeated hospitalizations to treat DTs, pancreatitis, and other physical illnesses. Again, however, the prevalence of such cases and their costs under the current and proposed systems are only speculative at this time, given the lack of comparative data. The committee has reviewed several schemes for moving beyond these vignettes to project changes in referral patterns and utilization levels but finds that none are adequate to provide policy guidance at this time. Its review does suggest that there could be a decrease of about one-third of the current admissions to inpatient programs if the proposed comprehensive system were put in place. Redirecting those persons to day care or outpatient programs or to brief interventions could yield a substantial net saving from current levels of spending. However, the committee recognizes that any savings to be achieved in this area of funding could be absorbed by the increase in services involved in broadening the base of treatment to include those with mild and moderate alcohol problems. It is the committee's hope that the creation of alternatives and the ability to match persons to the appropriate treatment will bring additional persons with severe and substantial problems who are not now being seen into both nonspecialist brief interventions and specialist treatment. Prompt early treatment can decrease the need for later, more expensive treatment. Implementing the new treatment system described in Section III will require that benefit plans offered by all payers, both public and private, incorporate the principles outlined in these recommendations into a single strategy. Such an approach must include the use of empirically determined matching criteria that are continuously validated through outcome monitoring as the method of access to the appropriate setting and modality at each stage of treatment. Given the redirection aspect of the proposed system (i.e., providing prompt treatment in lower cost sites), the committee anticipates that the savings offered by the system would be sufficient to cover the costs of the expanded treatment options and the continuity assurance activities, as well as to attract more persons into treatment. Indeed, the committee would make the same recommendations even if there were short-term increases in costs because of the long-term reductions in social costs that can be anticipated. Therefore, the committee recommends that sufficient insurance coverage be provided to facilitate adequate access to the continuum of services included in the new system: screening, assessment, treatment, reassessment, maintenance, and continuity assurance. Because the new system will use both matching and a type of case manager to provide continuity assurance, the use of inappropriate' costly care should be less like' than under the old system. Accordingly, more generous coverage for the new system might be possible. The committee reiterates its support for and recommendation of the Uniform coverage rules here as the minimum required coveragemhat is, coverage for the treatment of alcohol problems should be governed by the same principles as coverage for physical problems. Financing the Vision: Paying for Specialized Treatment Services for the Poor and the Uninsured In the committee's view, it is clear that payment for the cost of specialized treatment services should ultimately be included in some system of comprehensive coverage for all citizens and for all conditions. For the poor, this premise means that, in an ideal setting, the payment system ought to combine the dollars available from all sources for covering treatment for alcohol problems and in turn combine those dollars with the resources available in the health care system for covering other kinds of treatments. In
PAYING FOR THE TREATMENT SYSTEM 481 line with the principle of insuring treatment for alcohol problems on the same basis and in the same way as other treatment for illnesses is insured, no distinction in coverage should be made based on the condition requiring the treatment. Yet moving the financing of alcohol problems treatment from its current state to what is ultimately recommended, even when the ideal treatment system comes into existence, is no small task. However logical the uniform payment approach, the current problem of paying for treatment for poor citizens is complicated by the fact that these citizens are either covered by Medicaid or a state or locally financed medical assistance plan or are uninsured. The problem in paying for other citizens is that some fraction of them is not insured privately or publicly as well. Minnesota's consolidated fund administered by the state alcoholism authority is one approach for fostering integration of the diverse sources of public funds that must be used to pay for the proposed comprehensive system. The committee recommends that, in addition to monitoring the progress of this exhort, demonstrations of other approaches also be undertaken and evaluated. For those low-income people covered by Medicaid, the situation is complicated because these services are covered in a variety of ways across states (Toff, 1984~. Although there may be some scope for allowing states to vary the form of coverage in accord with their circumstances and to vary the extent of coverage with state taxpayer generosity, the committee believes that there should probably be more uniformity-and a higher basic level of coverage-than at present. In a few states, the service dollars earmarked for the treatment of alcohol problems are combined with Medicaid funds when the Medicaid recipient requires specialized alcohol treatment services. This integration fits well with the committee's concept of combining alcohol treatment financing generally with insurance for other medical conditions. Outcomes are ultimately constrained by the availability of resources, but using those resources for Medicaid beneficiaries in a more integrated way would probably be more efficient and more equitable. Nevertheless there is still a long way to go. Most states have not integrated their Medicaid and their categorically funded programs for treatment of alcohol problems. In addition, many of the poor are not eligible for Medicaid. Their "default insurance" is frequently uncompensated care in a hospital's emergency room, a budgetary account which is difficult to combine with state treatment monies into a single fund in any case (Gage et al., 1988~. For people who are not poor but are uninsured, neither Medicaid nor uncompensated care are viable options. For the uninsured, both poor and nonpoor alike, the state-funded specialist treatment system is likely to be the real insurer for the treatment of alcohol problems. At least until the problem of the generally uninsured is solved, it may be best to try to make this system work. That commitment probably means more generous funding of treatment for alcohol problems so that all who need care can receive it. In addition, it probably means using a method of payment that makes use of the entire medical care treatment system, rather than relying only on specialist state-provided alcohol treatment services. It probably also means using a sliding scale of fees (low or zero for the poor but rising for others) so that a reasonable percentage of out-of-pocket revenues are generated in each state and the public treatment system does not become more attractive than private insurance coverage. The ultimate objective is clear: everyone should have some insurance and that insurance should combine alcohol treatment funding into a single policy or plan, to be supplemented by out-of-pocket payment in instances when doing so will not discourage cost-effective treatment. While we wait for the emergence of a fully rational and complete system of health care financing, an interim solution would be to build on and improve Medicaid and the state categorical funding systems now in place. For Medicaid, this strategy would mean the combination of funds into a payment package that is appropriately generous in terms of coverage and payment to providers for appropriate treatments. For
482 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS the other uninsured, it means a revision of the state financing system to produce one with better funding (i.e., funding that is more equitable across states and tied to need), more flexible ~coverage" of alternative providers in different settings, and a sliding scale of fees tailored to the responsiveness of the demand for treatment. States are featured prominently in this transitional system because they are of central importance in the current approach to paying for treatment of alcohol problems and because there does not seem to be a principle that would argue against this role. However, as discussed in Chapters 7 and 8, the states now vary substantially in their financing and administration of alcohol treatment services. Research tells us that state spending varies enormously (Institute for Health and Aging, 1986) (see Chapter 8~. It also reveals that the availability of alternative forms of treatment varies tremendously across the states (see Chapter 7~. What is not known is what such variation means for access to needed treatment; in part, the reason for this uncertainty arises from a lack of knowledge about the need for effective and economical treatment. As one means to reduce that variability, the committee recommends that the state-funded specialist systems incorporate an income-related sliding scale of fees. A federal minimum standard could be developed and made part of the federal block grant requirements. Specific priority should be set to determine the allocation of funds to free care for the poor versus partially paid programs for the nonpoor who are not covered for specific treatment of alcohol problems. One could well envision the formulation of a federal minimum standard requirement, for both Medicaid and state specialist programs (through the alcohol, drug abuse, and mental health services block grant) which ensures that benefits and eligibility do not sink below some minimal level. Above this level, states should try to choose the level of coverage and form of reimbursement that supports only the most effective programs in terms of treatment and costs. As noted earlier, however, existing knowledge does not identify that minimum level or the most effective programs. As new information becomes available, states should be strongly encouraged to structure their "categorical benefit plans" toward what are known to be the most effective forms of treatment. Any federal minimum level of coverage should also be varied as new information becomes available on the effectiveness of alternative treatment forms. The proposed expert committee described above could provide independent, objective monitoring of the data. What is known at present is that maximum effectiveness requires the matching of persons with alcohol problems to treatment methods at each stage of the treatment process. Currently, Medicaid and state categorical programs do not include the full range of effective treatment options and are not designed to encourage assessment and matching. The committee encourages greater cooperation between these programs at the federal and state levels of government to design "consolidated" trials of the vision of treatment offered in this report. One further step, which is difficult to implement in the current era of cost containment and budget stringency but worth considering for the near future, would be to test a program that provides automatic access for a year to Medicaid coverage for anyone below a specified income level who is judged to need specialized services for the treatment of alcohol problems and who continues in a treatment program. In this sense, the presence of serious alcohol problems would become one of the categories for Medicaid eligibility, and continued eligibility for Medicaid would depend on remaining in treatment. The state-funded specialist treatment system would continue to treat people who are unable to continue in the Medicaid specialist program. The attractiveness of Medicaid coverage for treatment of other medical care needs would offer some incentive to comply with and stay in the alcohol treatment program.
PAYING FOR THE TREATMENT SYSTEM 483 These considerations are crucial in developing a mechanism for financing the proposed comprehensive system that integrates the specialist financing of treatment for alcohol problems with the other major vehicle of financing treatment of medical problems for the poor and uninsured. Their importance leads the committee to recommend that Medicaid and the state-funded specialist systems should be expanded to provide integrated coverage of treatment of alcohol problems for those eligible for Medicaid and for the uninsured. This treatment should be available in a variety of medical model and social model settings and should implement the principles of the "vision" as much as is possible. As better results on the effectiveness and cost-effectiveness of alternative treatment modalities, settings, and insurance coverage become available, the committee recommends that they be incorporated into the design of these consolidated state systems. Including provisions for outcome monitoring is crucial for such feedback and refinement. REFERENCES Davis, K 1987. The organization and financing of alcohol and drug abuse services. Presented to the Annual Meeting of the Institute of Medicine, Washington, D.C., October 21. Finn, P. 1985. Decriminalization of public drunkenness: Response of the health care system. Journal of Studies on Alcohol 46:7-22. Gage, L. S., D. P. Andrulis, and V. Beers. 1987. America's Health Safety Net: A Report on the Situation of Public Hospitals in our Nation's Metropolitan Areas. Washington, D.C.: National Association of Public Hospitals. 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. Hardwood, H. J., P. Kristiansen, and J. V. Rachal. 1985. Social and economic costs of alcohol abuse and alcoholism. Issue Repon No. 2. Research Triangle Institute, Research Triangle Park, N.C. Hardwood, H. J. D. M. Napolitano, P. L. Kristiansen, and J. J. Collins. 1984. Economic Costs to Society of Alcohol and Drug Abuse and Mental Illness: 1980. Prepared for the Alcohol, Drug Abuse, and Mental Health Administration. Research Triangle Park, N. C.: Research Triangle Institute. Hoffmann, N. G., J. A. Halikas, and D. Mee-Lee. 1987. The Cleveland Admission, Discharge and Transfer Criteria: Model for Chemical Dependency Treatment Programs. Prepared for the Northern Ohio Chemical Dependency Treatment Directors Association. Cleveland: Greater Cleveland Hospital Association. Institute for Health and Aging. 1986. Review and Evaluation of Alcohol, Drug Abuse and Mental Health Services Block Grant Allotment Formulas: Final Report. Prepared for Alcohol, Drug Abuse, and Mental Health Administration. San Francisco, Calif. Institute of Medicine. 1988. Homelessness, Health, and Human Needs. Washington, D.C.: National Academy Press. Minnesota Chemical Dependency Program Division. 1987. Biennial Report to the Governor and the Minnesota Legislature. St. Paul: Minnesota Department of Human Services. Minnesota Chemical Dependency Program Division. 1989. Report to the State Legislature on the Status of the Consolidated Chemical Dependency Treatment Fund. St. Paul: Minnesota Department of Human Services. Minnesota Department of Human Services. 1987. Consolidated Chemical Dependency Treatment Fund: County and Reservation Training Manual. Saint Paul, Minn.: Minnesota Department of Human Services. Pauly, M. 1986. Taxation, health insurance, and market failure in~the medical economy. Journal of Economic Literature 24:629~75.
484 BROADENING THE BASE FOR TREATMENT OF ALCOHOL PROBLEMS President's Commission for the Study of Ethical Problems in Medicine and Biomedical Behavioral Research. 1983. Securing Access to Health Care: The Ethical Implications in the Availability of Differences. Washington, D.C.: U.S. Government Printing Office. Temkin-Geser, H., and K T. Clark. 1988. Ethnicity, gender, and utilization of mental health services in a Medicaid population. Social Sciences in Medicine 26:989-996. Toff, G. E. 1984. Mental Health Benefits under Medicaid: A Survey of the States. Washington, D.C.: Intergovernmental Health Project, George Washington University. Tsai, S. P., S. M. Reedy, E. J. Bernacki, and E. S. Lee. 1988. Effect of curtailed insurance benefits on the use of mental health care: The Tenneco plan. Medical Care 26:430~40. U.S. Department of Health and Human Services (USDHHS). 1986. Toward a National Plan to Combat Alcohol Abuse and Alcoholism. Report submitted to the United States Congress. Rockville, Md.: National Institute on Alcohol Abuse and Alcoholism. U.S. Department of Health and Human Services (USDHHS). 1987. Sixth Special Report to the U.S. Congress on Alcohol and Health. Rockville, Md.: National Institute on Alcohol Abuse and Alcoholism. Wittman, F. D., and P. A. Madden. 1988. Alcohol Recovery Programs for Homeless People: A Survey of Current Programs in the U.S. Rockville, Md.: National Institute on Alcohol Abuse and Alcoholism.
