A new version of this report is available.
This publication is provided for historical reference only and the information may be out of date.
Outcomes research is concerned not only with results of studies but also with determining what exactly should be studied. For alcohol treatment among older adults, for example, should the measure of success be treatment compliance? Amount of alcohol consumed? Level of physical health? Psychological well-being? This chapter reviews compliance studies and prospective studies on treatment for older adults and examines the measures used. Because there have been few systematic studies of alcoholism treatment outcome (Atkinson et al., 1993) or the costs of treatment (Institute of Medicine, 1990) for older adults, this chapter also applies more general studies to that population. There are virtually no outcome studies of prescription drug use treatment for older adults, so this chapter addresses alcohol use only.
The chapter also provides an overview of instruments for measuring various treatment outcomes, instruments that are more important than ever as the health care system moves toward managed care. Payers increasingly are reimbursing only treatment approaches that have been validated by outcome studies - in particular, studies that quantify resource savings. Treatment costs and reimbursement issues are discussed, and the chapter ends with recommendations for future research.
Spectrum of Alcohol Treatment Outcomes
Brief Intervention Outcomes
Randomized controlled trials in other countries have demonstrated that brief interventions can reduce alcohol use and related problems in at-risk or nondependent problem drinkers under age 65 (Saunders et al., 1993; Anderson and Scott, 1992; Persson and Magnusson, 1989; Wallace et al., 1988; Kristenson et al., 1983). (For a more complete discussion of brief interventions, see Chapter 5.) Brief intervention studies have been conducted in health care settings ranging from hospitals and primary health care locations (Chick et al., 1988; Wallace et al., 1988; Babor and Grant, 1992a; Fleming et al., 1997b) to mental health clinics (Harris and Miller, 1990). The first randomized controlled U.S. trial in community-based primary care practices, the Trial for Early Alcohol Treatment (Project TREAT), which studied adults age 65 and younger (Fleming et al., 1997b), found that brief intervention for alcohol problems in primary care patients reduced both alcohol consumption and consequences.
A study of brief physician advice with at-risk drinkers age 65 and over, Guiding Older Adult Lifestyles (Project GOAL), also found positive changes in drinking patterns of the experimental (n = 158) compared with the control group (n = 71) (Fleming et al., 1997a). At the time of the 12-month followup, there was a significant reduction in 7-day alcohol use (t = 3.77; p < .001), episodes of binge drinking (t = 2.68; p < .005), and frequency of excessive drinking (t = 2.65; p < .005). The results indicated that brief physician advice made a difference in the drinking patterns of older at-risk and problem drinkers.
Most studies of alcohol brief interventions have only included patients early in their drinking careers, explicitly excluding dependent drinkers with significant withdrawal symptoms. The rationale for this practice has been that alcohol-dependent individuals or those affected most severely by alcohol should be referred to formal specialized alcoholism treatment programs because their conditions are not likely to be amenable to a low intensity intervention (Institute of Medicine, 1990; Babor, 1994). However, only one study to date has addressed the validity of this assumption. Sanchez-Craig and colleagues found that when comparing the 12-month treatment outcomes of men who were severely dependent and men who were not, both receiving brief treatment, there were no significant differences in "successful" outcomes as measured by rates of abstinence or moderate drinking (Sanchez-Craig et al., 1991).
Alcohol Treatment Outcomes
The study of treatment outcomes for older adults who meet criteria for alcohol abuse or dependence has become a critical issue because of older adults' unique needs for targeted interventions. Because traditional residential alcoholism treatment programs generally provide services to few older adults, sample size issues have been a barrier to studying treatment outcomes for older alcoholics. The development of elder-specific programs in recent years has, however, yielded sufficient data on older alcoholics to permit more comprehensive studies of this population (Atkinson, 1995).
Previous research on alcoholism treatment in older adults can be divided into two broad categories: compliance studies and outcomes studies.
Studies of treatment compliance
Most treatment outcome research on older alcoholics has focused on compliance with treatment program expectations, in particular the patient's fulfillment of prescribed treatment activities and goals, including drinking behavior (Atkinson, 1995). Results from compliance studies have shown that age-specific programming improved treatment completion and resulted in higher rates of attendance at group meetings than did mixed-age treatment (Kofoed et al., 1987). Studies also show that older alcoholics were significantly more likely than younger patients to complete treatment (Schuckit, 1977; Wiens et al., 1982/1983). Atkinson and colleagues also found that, proportionately, twice as many older male alcoholics completed treatment than younger men (Atkinson et al., 1993).
Age of onset of alcohol problems has been a major focus of research for older adult treatment compliance studies. In one study using a matched-pairs, post hoc design, rates of completion of 6-month day treatment for 23 older men and women alcoholics (age 55 and older) whose problem drinking began before age 50 (early onset) were compared with 23 who began problem drinking after age 50 (late onset) (Schonfeld and Dupree, 1991).
In another study of 132 male alcoholic veterans age 60 and older, the sample was divided into three subgroups: early onset (age 40 and younger, n = 50), midlife onset (age 41 to 59, n = 62), and late onset (age 60 and older, n = 20) (Atkinson et al., 1990). Age of onset was related to program completion and to weekly group therapy meeting attendance, with the late onset subgroup showing the best compliance in bivariate analyses. However, a subsequent multivariate analysis of 128 men age 55 and older in alcoholism treatment found that drinking relapses during treatment were unrelated to age of onset (Atkinson et al., 1993). Furthermore, age of onset did not predict program completion but was related to attendance rate at scheduled visits (Atkinson et al., 1993). The studies on the effect of age of onset on treatment compliance have therefore yielded mixed results.
In a study of treatment matching, Rice and colleagues compared drinking outcomes for randomly assigned male and female alcoholics 3 months after beginning one of three mixed-age outpatient cognitive-behavioral treatment conditions scheduled to last for 4 months (Rice et al., 1993). The sample included 42 individuals age 50 and older, 134 patients age 30 to 49, and 53 patients age 18 to 29. There were no significant effects of age or treatment condition on treatment compliance. However, there were significant age group-by-treatment condition effects. For older patients, the number of days abstinent was greatest and the number of heavy drinking days fewest among those treated with a focus on self-efficacy rather than a focus on occupation or family issues.
Major limitations remain in the treatment compliance literature, including lack of drinking outcome data, failure to report on treatment dropouts, and variations in definitions of treatment completion. Few carefully controlled prospective treatment outcome studies, even those with sufficiently large numbers of older alcoholics, address the methodological limitations inherent in compliance studies.
Prospective studies of treatment outcomes
Although it is important to examine the factors related to completion of treatment, studies thus far have inherent selectivity bias and provide no information on treatment dropouts or on short- or long-term outcomes of treatment. Other sampling issues may limit the applicability of such studies to larger groups, such as the exclusion of women in some studies and the use of varying age cutoffs that sometimes place individuals as young as 45 in the "older" category.
Problems with previous outcome studies extend beyond sampling to study methods. The majority of studies used relatively unstructured techniques for assessing drinking patterns and alcohol-related symptoms. Furthermore, the assessment of outcomes has been narrow in focus. Most studies have dichotomized treatment outcome (abstention vs. relapse) based solely on drinking behavior. Given evidence from numerous studies that heavy or binge drinking is more strongly related to alcohol consequences than average alcohol consumption (Anda et al., 1988; Chermack et al., 1996; Kranzler et al., 1990), there may be important differences in outcome for nonabstinent individuals depending on whether binge drinking was part of the posttreatment pattern. Current recommendations include categorizing nonabstinent drinking outcomes along dimensions, such as whether drinkers ever drink to the point of intoxication (Heather and Tebbutt, 1989). Furthermore, most studies have not addressed other relevant domains that may be positively affected by treatment, such as physical and mental health status and psychological distress.
One reason for the lack of prospective treatment outcome studies is that studying older alcoholics during and after treatment is so complex. One exception is a study of 137 male veterans (age 45 to 59 years, n = 64; age 60 to 69 years, n = 62; age 70 years and older, n = 11) with alcohol problems who were randomly assigned after detoxification to age-specific treatment or standard mixed-age treatment (Kashner et al., 1992). Outcomes at 6 months and 1 year showed that elder-specific program patients compared with mixed-age group patients were 2.9 times more likely at 6 months and 2.1 times more likely at 1 year to report abstinence. The two treatment groups, however, could not be adequately compared at baseline because baseline alcohol consumption and alcohol severity data were not included in the study.
Recognizing that older individuals have been underrepresented in standard alcoholism treatment programs (Booth et al., 1992; Higuchi and Kono, 1994), as well as in treatment outcome studies (Atkinson, 1995), the Institute of Medicine published a special report calling for specific longitudinal studies focused on factors associated with more successful treatment outcomes in older adults (Institute of Medicine, 1990).
Blow and colleagues conducted a study to determine outcomes for older adults receiving specialized elder-specific inpatient alcoholism treatment (Blow et al., 1997). A range of treatment outcomes was assessed using a prospective longitudinal design. To address limitations of previous studies, this study used validated techniques to assess baseline alcohol symptoms and psychiatric comorbidity, age of onset of alcohol problems, drinking patterns, physical and emotional health functioning, and psychological distress. Followup was conducted 6 months after discharge. This study also examined a range of different drinking outcomes, including abstinence, nonbinge drinking, and binge drinking.
Adults over the age of 55 in the treatment program (n = 90) were interviewed. The physical health functioning of the sample was similar to that reported by seriously medically ill inpatients in other studies, whereas psychosocial functioning was significantly worse. Nearly one-third of the sample had one or more comorbid psychiatric disorders, with anxiety disorders and major depression most common.
Participants who completed the 6-month followup assessment (n = 68; 76 percent of the original sample) were classified into the following outcome groups: abstainers (n = 38), nonbinge drinkers (who never exceeded four drinks on any drinking day during the followup period; n = 12), binge drinkers (who had one or more days in which they consumed five or more drinks; n = 18), and noncompleters (who did not complete the 6-month followup assessment; n = 22). These groups did not differ significantly on demographic variables, pretreatment drinking patterns and symptoms, age of onset of alcohol problems, psychiatric comorbidity, or length of stay in treatment. For all groups who completed the 6-month followup, there were improvements in general health. Psychological distress decreased significantly between baseline and followup for abstainers and nonbinge drinkers. However, binge drinkers did not show a decline in psychological distress and were significantly more distressed at 6-month followup than both the abstainers and nonbinge drinkers.
Measurement of Multidimensional Outcomes for Older Adults
Consumption levels are not the only measure of success: Drinking patterns, alcohol-related problems, physical and emotional health functioning, and quality of life can also be used to assess alcohol intervention and treatment outcomes with populations of older adults. It is particularly important to use benchmarked methods to assess older adults to determine whether treatment regimens are effective. Older adults have unique issues based on changes in physical functioning, changes in tolerance to alcohol, and internal (e.g., hearing, eyesight) and external (e.g., death of spouse, retirement) losses requiring a multidimensional approach to assessment and outcome evaluation to ameliorate potential reasons for relapse or a return to hazardous drinking.
Outcome assessment is invaluable from both a management and a referral perspective. The providers of treatment, the clinicians and agencies referring patients, and patients themselves need to have information regarding the likely outcomes of treatment. Because treatment options range from brief interventions to structured outpatient and inpatient treatment programs, evaluation is recommended at varying points in the treatment process (McLellan and Durell, 1996). Initial evaluation in any setting should take place at the beginning of the intervention or treatment to obtain baseline data. McLellan and Durell recommend conducting first followup evaluations 2 weeks to 1 month after the patient leaves the inpatient setting. The short time frame reflects the need to determine if the patient is engaged in aftercare with an outpatient program to maximize the effect of inpatient treatment.
A review of the literature on patients receiving substance abuse treatment indicated that 60 to 80 percent of people who relapse do so within 3 to 4 months (McLellan et al., 1992). Older adults who comply fully with and complete the intervention or treatment, however, are more likely than younger adults to positively change their drinking behavior (Finch and Barry, 1992). Therefore, outpatient outcomes should be assessed no sooner than 3 months and possibly as long as 12 months after treatment (McLellan and Durell, 1996). For all types of intervention and treatment, ongoing outcome evaluation is important since the course of alcohol problems in older adults is dynamic and changes over time with circumstances. Additional life stressors can change the pattern of alcohol use in this age group.
In response to the rising costs of treatment and concerns about the effectiveness of alcohol treatment for both younger and older adults, the demand to evaluate and demonstrate the quality of a variety of treatment options has also grown. For the purposes of this section, outcome measurement will include methods to measure alcohol use and alcohol-related problems, physical and emotional health functioning, and quality of life and well-being.
Measures of Alcohol Use
Drinking patterns can be assessed using approximations such as average number of drinking days per week and average number of drinks per occasion or day. Two of the instruments assessing average consumption are the Alcohol Use Disorders Identification Test (AUDIT) (Babor et al., 1992b) and the Health Screening Survey (HSS) (Fleming and Barry, 1991), both of which are reproduced in Appendix B. The AUDIT, which has been validated internationally and with populations under age 65 in the United States, assesses quantity and frequency of alcohol use as well as alcohol-related problems (Babor et al., 1987; Fleming et al., 1991; Barry and Fleming, 1993; Schmidt et al., 1995).
The HSS, originally developed by Wallace and Haines and adapted by Fleming and Barry, measures average quantity and frequency of alcohol use in the previous 3 months (Wallace and Haines, 1985; Fleming and Barry, 1991) and includes parallel questions about weight, exercise, and smoking. It has been validated in people under 65 in primary care settings in the United States (Fleming and Barry, 1991) and has been used with older adults as part of a brief intervention trial (Fleming et al., 1997a).
The most accurate method used to assess current alcohol consumption is the Time Line Follow Back (TLFB) procedure. TLFB is a structured interview that uses calendar cues (e.g., holidays, family events, trips) to quantify daily alcohol use over a period of time ranging from 7 days to a number of months (Sobell et al., 1988, 1996). Researchers have used this method to obtain up to 1 year of drinking data. This method has shown high test-retest reliability in a variety of drinking populations ranging from normal drinkers to heavy drinkers to persons participating in inpatient or outpatient treatment. Fleming and colleagues used this procedure to assess 7-day alcohol use with adults age 65 and older as part of the initial assessment in a clinical trial to test the effectiveness of brief physician advice with older at-risk and problem drinkers (Fleming et al., 1997a).
Measures of Alcohol Problems
It is necessary but not sufficient to determine quantity and frequency of alcohol use for initial and followup assessments in older adults. The use of multidimensional screening and outcome instruments provides clinicians, programs, and referral agencies with measurements regarding the nature and severity of problems presented by persons who abuse alcohol.
An important multidimensional screening instrument for use specifically with older adults is the Michigan Alcoholism Screening Test - Geriatric version (MAST-G) (Blow et al., 1992a) (See Figure 4-4). This tool was developed because many of the screening measures did not identify alcoholism among older adults as reliably as among younger populations. The MAST-G was validated using criteria for alcohol dependence in the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (American Psychiatric Association, 1987) as the gold standard on 305 older adults including (1) persons currently meeting alcohol dependence criteria but not in treatment, (2) those currently in treatment, (3) those with a previous history of alcoholism but in recovery, (4) social drinkers, and (5) abstainers.
The MAST-G is a 24-item scale (sensitivity = 0.94; specificity = 0.78) in which a score of five or more "yes" responses indicates an alcohol problem. Scores do not discriminate between current and past problems, although some items address the current situation and others address problems in the past. Tolerance is not measured in light of data indicating that older adults with even low consumption can experience alcohol-related problems due to physiological changes that occur with age.
The Addiction Severity Index (ASI) (McLellan et al., 1985; McLellan et al., 1990) was developed specifically to assess over time the alcohol-related problems and the severity of symptoms of patients in treatment for alcohol and drug abuse and dependence. The ASI is a semistructured interview that provides information about aspects of the patient's life that may contribute to the substance abuse syndrome. The focus of the interview is on seven functional areas that have been shown to be affected by substance abuse: medical status, employment and support, drug use, alcohol use, legal status, family and social status, and psychiatric status. Each area is assessed individually for past and present (last 30 days) status. Each area has a 10-point interviewer-determined severity rating of lifetime problems and a multi-item composite score indicating severity of the problems in the last 30 days.
The ASI is targeted to all adult populations in substance abuse treatment or in treatment for co-occurring psychiatric and substance abuse disorders. The ASI has good interrater reliability as well as good predictive, concurrent, and discriminant validity (McLellan et al., 1985). Although it has not been widely used or validated with older patients and is not generally used with patients who are at-risk or problem drinkers in primary care or community-based settings, it is included in this review because it is a standard measure in the field and can provide important information regarding older adults in treatment settings, particularly in areas of greatest concern with this population - medical status, alcohol use, family and social status, and psychiatric status.
Measures of Physical and Emotional Health
One of the most widely used measures of physical and emotional health is the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36). This instrument was originally developed for use with adults as a 20-item scale for the Medical Outcomes Study (MOS) (Ware and Sherbourne, 1992; Tarlov et al., 1989; Stewart et al., 1988) from more detailed measures used in the Rand Health Experiment. It was subsequently expanded to 36 items that measure physical functioning, limitations in functioning due to physical health problems, social functioning, bodily pain, general mental health, limitations in role functioning due to emotional problems, vitality, and general health perceptions. The SF-36 has published norms for these various subscales over distinct age groups, including older adults (McHorney et al., 1993).
In addition to the subscales addressed in the SF-36, other measures of psychological distress are useful in alcohol outcomes assessment with older adults. The Symptom Checklist-90-Revised (SCL-90-R) (Derogatis, 1994b) is a self-report symptom inventory designed to measure psychological distress. The Brief Symptom Inventory (BSI) is a brief form of the SCL-90-R (Derogatis and Melisaratos, 1983). These tests provide an overview of a patient's symptoms and identify the level of distress that a patient is experiencing during a specific time period (e.g., "the last 7 days"). Both the SCL-90-R (90 items) and the BSI (53 items) measure nine symptom dimensions: somatization, obsessive-compulsiveness, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism. Each of the tests also has three global measures of distress, a measure of the intensity of distress, and a measure of the total number of patient symptoms.
Depending on the population being assessed, the internal consistency for the SCL-90-R ranges from 0.77 to 0.90; for the BSI, from 0.71 to 0.85. Test-retest reliability ranges from 0.80 to 0.90 for the SCL-90-R (with a 1-week interval between tests) and from 0.61 to 0.91 for the BSI. When used for outcome measurement, these measures are often administered at intake, during treatment, at discharge, and at followup intervals (Smith, 1996).
Measures of Quality of Life
Quality of life measures have most frequently been used for outcomes assessment in mental health treatment. One of the most widely used instruments is the Quality of Life Interview (QLI) (Lehman, 1988). Research suggests that quality of life, as perceived by the patient, is an important factor in maintaining optimal functioning. The quality of life measure is constructed to include a single-item measure of general well-being and seven dimensions of well-being. The constructs and dimensions in this scale are applicable to the alcohol treatment field, particularly in outcomes assessment with older adults for whom concerns about housing, leisure, family, social relationships, health, safety, and finances are salient factors in functioning and relapse.
Internal consistency reliabilities range from 0.79 to 0.88 for the life satisfaction scales and from 0.44 to 0.82 for the objective quality of life scales. Normative data are available for various subgroups of patient populations, and the life satisfaction items can be compared with national norms in the general population.
Costs of Alcohol Treatment
Outcomes studies obviously can help treatment providers and health care professionals improve treatment. They also play an important role in paying for treatment: Third-party payers want validated proof that the treatment approaches they are reimbursing actually work. The other side of this equation is the cost - to individuals, the health care system, and society at large - of alcohol-related problems. If costs of treatment can be measured against these larger costs, it is more likely that treatment will be reimbursed.
The costs of alcohol abuse and dependence are estimated to be over $100 billion a year, due in part to increased mortality, significant social costs, and health consequences (National Institute on Alcohol Abuse and Alcoholism, 1995; Brower et al., 1994; Holder and Blose, 1992; Goodman et al., 1991). Individuals who have alcohol disorders are among the highest cost users of medical care in the United States. Persons with alcohol dependence, who represent between 3 and 14 percent of the U.S. population, consume more than 15 percent of the national health care budget (Rice et al., 1993). Although a number of cost studies have examined drinking in younger adults (Holder et al., 1991; Holder and Blose, 1992; Finney and Monahan, 1996), few studies have separated the costs of alcohol disorders for older adults or even included older adults in cost analyses. In a study of Federal employees, one half of whom were over age 60, Holder and Blose analyzed 4 years of claims data (Holder and Blose, 1986). They found that alcohol treatment contributed to sustained reductions in total health care utilization and costs and that reductions in posttreatment costs appeared to be sustained into the fourth and fifth years following treatment.
In a review of studies of alcohol treatment and potential health care cost savings that included Medicare studies, Holder found that mean monthly medical care costs increased for persons with alcohol problems before initiation of treatment, declined immediately following treatment, and continued to decline 2 years following treatment (Holder, 1987). The oldest group in the study (age 65 and older) experienced the highest medical care costs and showed the least convergence to levels prior to the initiation of alcohol treatment. Reasons for this might include increased general morbidity with age and the potentially more serious health problems due to a longer period of chronic alcohol abuse or dependence.
Among all of the economic analyses of alcohol programs, there has been little work regarding the cost savings of substance abuse prevention and early intervention in managed care settings. One of the few recent studies of managed care (Holder et al., 1995) estimated that for every $10,000 spent on brief intervention for alcohol or drug abuse, $13,500 to $25,000 is saved in medical spending for the managed care provider. Gaps remain in the literature regarding the economic effectiveness and implications of brief interventions in managed care settings. Filling in the gaps is particularly important because managed care providers are challenged to provide needed services with fewer dollars.
Findings on the efficacy and cost-effectiveness of brief intervention, however, can be misleading (Heather, 1995; Peele, 1990). These reviews generally do not assess costs and needs of older adults in these settings. Methodologies across cost analysis studies have not been consistent, making comparisons more difficult. In addition, Heather points out problems in interpreting the data from brief intervention studies because brief interventions are not a homogeneous entity (they vary in length, structure, targets of intervention, and personnel responsible for delivery), and there is a distinction between treatment seekers (e.g., persons who answer ads indicating that they would like to decrease their drinking) and nontreatment seekers (e.g., individuals with regularly scheduled appointments for medical problems who receive interventions from their health care providers) (Heather, 1995). Clear delineation of the type of study and the potential audience for the research can help to alleviate problems of misinterpretation.
Most economic studies of alcohol treatment have focused on hospital inpatient and outpatient treatment for abuse and dependence (Peele, 1990; Annis, 1986). Peele's review of the literature revealed that, although hospital treatment is no more effective than outpatient treatment, reimbursement systems have often supported the more costly, medically based inpatient treatment options (Peele, 1990).
Some experts suggest that effectively treating alcoholism and reducing the social and medical consequences of alcohol disorders will yield the largest savings in a reformed American health care system (McCrady and Langenbucher, 1996).
Reimbursement Issues in The Treatment of Older Adults
The barriers to care experienced by many individuals who need intervention or treatment for problems related to their alcohol use have been of great concern to the alcohol treatment field. A further concern has been the observation that only a small minority of those who need treatment has received it (Institute of Medicine, 1990). It has not always been clear if this is due to the lack of identification and referral of those who need treatment, the lack of treatment options, or financial barriers to care. All of these barriers may affect older adults. Currently, however, the financial barriers are changing the fastest - and some of the shifts in reimbursement are alarming.
Private third-party insurers are funded through premiums paid by purchasers, with premiums adjusted based on claims made by any subscriber group. Generally, except for self-insured plans, coverage minimums and premiums are regulated by States through their insurance departments. Medicare is generally thought of as a public third-party payer for health care services. The benefits provided are authorized through legislation.
The current trend, however, is for States to turn their Medicare programs over to managed care companies. Although, since its inception in 1965, Medicare has generally covered 12 days of inpatient alcohol treatment, most managed care companies eliminate coverage for as much inpatient treatment as possible and often cut services for alcohol treatment altogether to keep costs down.
These cuts in coverage are antithetical to all that is known about treating older adults with alcohol problems. Coverage of 12 days of inpatient treatment is extremely important for older adults because they are likely to have a greater number of physical and cognitive problems than younger adults. For example, older adults often have more prolonged and severe alcohol withdrawal than younger adults (Brower et al., 1994), and participation in group treatment is more difficult for them in the early stages of treatment. They are also more likely to need more intensive outpatient care after an inpatient stay than younger adults.
Furthermore, Medicare should reimburse for alcohol prevention and early intervention efforts in primary care settings, because research indicates that early intervention programs are effective with a large proportion of older at-risk and problem drinkers. Such initiatives will save the medical community money by preventing more costly complications of heavier alcohol intake.
With ongoing changes in the delivery of alcohol treatment services from inpatient to outpatient settings coupled with the shifting reimbursement structure from fee-for-service Medicare to managed Medicare, coverage of effective treatment is increasingly uncertain. The changes in treatment venue and fee structures underscore the importance of conducting multidimensional outcomes assessments in the context of quality management. Convincing research is an important component of efforts to ensure that older adults who need intervention and treatment for alcohol problems receive the appropriate level of treatment and adequate followup. Ongoing evaluation of patient outcomes can help safeguard the health of at-risk older adults and foster the development of innovative treatment approaches to meet the needs of this vulnerable population.
Areas Requiring Future Research
As the number of older adults rises, the use of mood-altering drugs such as alcohol and tranquilizers by older adults is a growing area of concern from a clinical and research perspective. As the Baby Boom generation reaches traditional retirement age, the field of substance abuse treatment and research will be faced with both growing numbers of individuals who have alcohol-related problems and emerging problems unique to the aging population, namely a potential increased prevalence of illicit drug use and drug-related problems.
Even though the prevalence of alcohol and drug use decreases with age, alcohol and prescription drug use continue to be important health problems in the current cohort of older adults. To advance the field and address coming needs of future older adults, research needs to be focused in some specific areas. The general areas for new research initiatives are (1) alcohol and other drug consumption, (2) treatment, (3) biomedical consequences, (4) behavioral and psychological effects, and (5) special issues.
In the area of alcohol and other drug consumption, future research directions should include
- Life course variations among alcohol, illicit drug, and prescription drug use patterns
- Gender and ethnic variability
- Reasons for changes in drinking and drug use patterns with aging
- Early and late onset of alcohol and drug problems
- Health care costs for older adults maintaining abstinence compared with costs for those who reduce their consumption
- Development of valid screening instruments for illicit and prescription drug use.
Issues related to treatment have traditionally been studied in males and younger cohorts of adults. Some of the issues requiring both new and renewed study with older adults include
- Prevention and early intervention techniques
- The use of technology (e.g., computers, interactive voice recognition) in the treatment of substance abuse problems in older adults
- The effectiveness of various older adult-specific alcohol and drug treatment modalities
- Alcohol and drug withdrawal issues
- The effect of physical and psychiatric comorbidity on treatment outcomes
- Older subgroups (i.e., 60-65, 65- 70, 70-75, 75-80, 80+)
- Relationship of provider characteristics (e.g., age, similarity to client) to completion of treatment
- Risk factors for drinking and drug use relapse, including a better understanding of specific treatment needs for older adults.
Biomedical research can forge a new and important path in the understanding of alcohol use and abuse in older adults. Directions include
- The effects of alcohol and drugs on aging organisms
- Alcohol and drug medication interactions
- Physiological reasons for increased sensitivity to alcohol as people age
- Medical consequences of moderate and heavy drinking and illicit drug use
- Interactions of alcohol, nicotine, and illicit drugs.
Behavioral and psychological research initiatives may be focused in the following directions:
- Demographics relating to older adult alcohol and drug use and abuse (relationship between drinking and drug use status and employment, marital status, residence, education, and other variables specifically affecting older adults)
- Older adults' reasons for changing their drinking patterns
- Stress, coping, and adaptation, and their relationship to alcohol and drug use
- Cognitive effects of moderate and heavy drinking and drug use in this age group.
Finally, there are some special issues related to older adult substance abuse that have been the target of clinical concern and some initial research, which needs to be expanded in order to address the needs of the current and future cohorts of older adults:
- Elder abuse and neglect
- Homelessness
- Underrepresentation of older adults in treatment settings.
Researchers in gerontology, substance abuse treatment, and related fields must take the lead in providing the above information. Only with such knowledge can clinicians and policymakers improve the identification and treatment of substance use disorders among older adults. Without the information - and the response - such disorders will take a greater and greater toll on one of the most vulnerable and fastest growing sectors of the population.
Publication Details
Copyright
Publisher
Substance Abuse and Mental Health Services Administration (US), Rockville (MD)
NLM Citation
Center for Substance Abuse Treatment. Substance Abuse Among Older Adults. Rockville (MD): Substance Abuse and Mental Health Services Administration (US); 1998. (Treatment Improvement Protocol (TIP) Series, No. 26.) Chapter 6 - Outcomes and Cost Issues in Alcohol Treatment for Older Adults.