inpatient vs outpatient treatment for substance dependence revisited

INPATIENT VS OUTPATIENT TREATMENT FOR SUBSTANCE DEPENDENCE REVISITED

H e l e n M. P e t t i n a t i , Ph.D. , K a t h l e e n Meye r s , M.S., J a c q u e l i n e M. J e n s e n , M.A., F r a n c e s K a p l a n , B.A.,

and B r a d l e y D. Evans , M.D.

Miller and Hester's 1986 review of inpatient versus outpatient alcohol treatment studies concluded with no "justification" for inpatient treatment. Further exam- ination of these studies revealed shortcomings such as the use of random assignment designs which excluded psychiatrically-complicated patients. Carrier Foun- dation's inpatient/outpatient study of private psychiatric patients with alcohol and/or cocaine dependence includes a patient-treatment matching design to address weaknesses in the existing literature. Patients with high psychiatric severity and/or a poor social support system are predicted to have a better outcome in inpatient treatment, while patients with low psychiatric severity and/or a good social support system may do well as outpatients without incurring the higher costs of inpatient treatment. Preliminary results from 183 inpatients and 120 outpatients indicated outpatients, regardless of level of psychiatric severity, were 4 times more likely to be early treatment failures ( ch i - square = 41.2, d f = 1~

p < .01). While the determination of long-term follow-up status of eariy treatment failures is currently underway, this finding underscores the potential risk of early treatment failure in outpatient compared to inpatient substance abuse treatment programs and the importance of addressing the issue of early attrition in conduct- ing outcome analyses.

The r i s ing costs of U.S. h e a l t h care h a v e forced s ign i f i can t c h a n g e s in i n s u r a n c e r e i m b u r s e m e n t policies. Th is is r e ad i l y seen in t he g r o w i n g n u m b e r of c u t b a c k s in m e n t a l h e a l t h and s u b s t a n c e a b u s e benef i t s . In pa r t i cu l a r , r e i m b u r s e m e n t for tess cos t ly outpa-

Address correspondence to Helen M. Pettinati, Ph.D., Carrier Foundation, Belle Mead, New Jersey 08502.

PSYCHIATRIC QUARTERLY, Vol. 64, No. 2, Summer 1993 0033-2720/93/0690-0173507.00/0 © 1993 Human Sciences Press, Inc. 173

174 PSYCHIATRIC QUARTERLY

t ient substance abuse t reatment programs is becoming common- place in comparison to reimbursement for inpatient t reatment programs (1). Despite recent findings by Walsh et al. (2) tha t blue- collar employees with alcohol dependence had a better outcome when treated in an inpatient setting compared to attending com- munity-based programs, substantial decreases in reimbursement for inpatient substance abuse t reatment have occurred. Reduc- tions in reimbursement may have resulted from empirical evi- dence, predating the Walsh et al. study, that inpatient substance abuse t reatment programs are no more effective than outpatient t reatment (3). These studies, however, are at variance with a relatively pervasive clinical perspective that inpatient care is nec- essary for a number of substance dependent patients (4). This chasm between the clinical and empirical perspectives is not well understood, and has dictated a need to reexamine the studies and their empirically-based conclusions.

The purpose of this report is twofold: 1) to further evaluate the conclusions reached from prior inpatient/outpatient research that inpatient t reatment is not justified; 2) to present an inpatient/ outpatient research study, currently ongoing at Carrier Founda- tion, tha t was designed to address prior weaknesses in the existing set of empirical studies.

PRIOR RESEARCH: INPATIENT VS OUTPATIENT

Miller and Hester (3) reviewed 34 studies which had evaluated the effectiveness of t reatment settings for alcohol dependence. Eight of these studies were uncontrolled and, therefore, had confounding factors l imiting the value of their results. Another 11 studies are not relevant since they examined the efficacy of either inpatient or outpatient t reatment alone, but did not compare the effectiveness of inpatient versus outpatient care. This left 15 studies which used an experimental paradigm and included data tha t would allow a comparison of the inpatient and outpatient alcohol t reatment programs. However, 4 of these 15 studies compared inpatient treatment programs to partial hospital programs or day treatment hospitals. Although these are sometimes referred to as %utpatient programs," they provide all-day treatment services at the same intensity as residential programs and, therefore, should not be

HELEN M. PETTINATI, ET AL. 175

intermixed or confused with what is typically considered outpa- t ient treatment. Hence, only 11 of the originally reviewed 34 studies actually address the inpatient/outpatient question. Unfor- tunately, the nature of the outpatient programs in these 11 studies, particularly the intensity, varied greatly. For example, the many forms of outpatient "treatment" included a post-detoxification wait ing list with check-in visits (5), an evaluation plus a brief advice session (6), an aftercare program (7), and a medication program (8). The presence of consistent results against inpatient t reatment across such heterogeneous t reatment conditions could have suggested to the reviewers that no significant differences exist between inpatient and outpatient substance abuse programs. However, such extreme variation in outpatient t reatment services would generally be viewed as a methodological flaw.

To briefly summarize the findings of the 11 studies, 8 failed to show significant outcome differences between inpatient and outpa- t ient treatment. However, follow-up assessment intervals ranged from 1 month to 2 years post-treatment and two studies were flawed by poor follow-up rates as low as 8% (8, 9). Furthermore, one of the studies which showed "no significant difference" between the programs (6) had a 90% relapse rate in both inpatient and outpa- t ient groups by 4 months with the remaining 10% relapsing after just 2 years (10). Unfortunately, a finding of ~no significant difference" in inpatient versus outpatient t reatment has often been related to very high relapse rates in both groups. This can signify other factors at work, such as poorly designed treatment programs, an extremely impoverished population, or grossly insensitive outcome measures. Of the remaining 3 studies, one showed a trend toward a more favorable outcome in outpatient t reatment (11), and two (12, 13) found statistical differences favoring outpatient care, although this did not hold up by the 10-month follow-up assessment in one of the studies (12).

The above-mentioned observations alone indicate that conclusions derived from prior research should be viewed cautiously. However, there is a glaring oversight in all of the studies which is likely the single, most important factor responsible for the devia- tion between the clinical and experimental perspectives on the inpatient/outpatient question. Patients clinically suited for inpatient treatment, e.g., psychiatrically-complicated substance abusers, were typically not included in these studies. The use of


random assignment, where patients who require inpatient treatment cannot be included due to ethical concerns, occurred in all 11 studies. Thus, the very patients who may require inpatient treatment, and are at risk for poor outcome in the outpatient setting, were not studied. This undoubtedly may have biased results in favor of outpatient treatment.

In addition, the almost exclusive study of alcohol dependent samples treated at public facilities limits the generalizability of the results to other substance use groups, particularly those dependent on more than one substance, and to t reatment facilities which accept private insurance. Unfortunately, patients with private insurance are most affected by changes in insurance reimbursement.

In addition, while most of the prior research has relied solely on extreme categorical outcome measures like '~complete abstinence for a one-year period," it is important to note that alcohol and drug dependence are recurrent conditions. Using complete and contin- uous abstinence as the yardstick of success will inevitably create a narrowly-defined success group and a heterogeneous failure group. Therefore, more sensitive, dimensional outcome measures are needed.

Finally, there has been an overall lack of attention to attrition rates in these studies, particularly in the early intensive phase of the t reatment program. If the two settings should have different attrition rates, long-term outcome judgments will be affected if only t reatment completers are evaluated (14).

THE CARRIER FOUNDATION PROJECT

In an attempt to address some of these concerns, a 5-year research study has been initiated to evaluate the cost-effectiveness of inpa- t ient compared to outpatient substance abuse treatment (funded by the National Institute on Alcohol Abuse and Alcoholism). The sample comes from a private, nonprofit, psychiatric t reatment setting where the majority of patients carry private insurance. In addition to evaluating alcohol abusing patients, the sample also includes patients who abuse cocaine either alone or in conjunction with alcohol.

A random assignment design was not chosen so that patients


requiring inpatient t reatment could be included in the project. Instead, an alternative experimental patient-treatment matching design is being used. ~'Patient-treatment matching is a method of choosing between alternative t reatment options based on particu- lar patient charac te r i s t ics . . , to produce more beneficial results than if matching had not been done" (15). The current study adapted a patient-treatment matching design tha t McLellan and colleagues (16, 17) had used to retrospectively evaluate the effectiveness of six substance abuse t reatment facilities. Using this design allowed them to formulate a number of variables that predicted good outcome in the respective treatments. These predictor variables were then successfully tested in a prospective study.

Carrier's Patient-Treatment Matching Design

Although the Carrier project delineates a priori matching criteria, patients are not "assigned" to inpatient or outpatient t reatment settings as was done by McLellan and colleagues (17) in their prospective study. Rather, patients are admitted to either the inpatient or outpatient program depending upon both clinical factors (assessed by clinical intake staff) and nonclinical factors (e.g, employment flexibility). Following admission to the treatment program, a comprehensive research interview is conducted and questions are asked tha t determine whether the patient has been matched or mismatched to the t reatment setting.

Figure 1 provides the two primary dimensions, i.e., psychiatric severity and social supports, that are being used in the determination of a patient 's matching status. These criteria were selected from both clinical and empirical information (17). Psychiatric severity is determined by the patient 's responses to the Symptom Checklist-90-Revised (SCL-90-R) (18). Patients scoring at or above .87 (males) or 1.47 (females) on the global severity index of the SCL-90-R (mean value of all 90 raw scores) were determined to have high psychiatric severity. This measure was chosen because it has been argued that a dimensional approach is advantageous to a categorical approach (19). Woody and colleagues (20) reported a high correlation between the SCL-90-R and the psychiatric subscale of the Addiction Severity Index (ASI) (21), the latter measure being the instrument used for determining degree of psychiatric severity in an alcohol abusing Veterans Administration population.


Figure 1

I II I III I I IIIIIIII III

RESEARCH DESIGN: i o ATCHED-TO-TREATMENT SETTING C PARING INPATIENTS TO OUTPATIENTS

I , I I IIIIII I I

PRIVATE HOSPITAL PATIENTS

, , , , ,

INPATIENT

OUTPATIENT

MATCHED

Psych Social Severity Supports

,,,,,

High QL Poor

Low ~ Good /

MISMATCHED

Psych Social Severity Supports

Low ~ Good i I

High ~ Poor

Social supports are determined by a series of questions asked of the patients at the initial interview. Essentially, patients have poor support if they live alone and have less than weekly contact with family or friends and spend most of their free time alone and/ or have serious conflicts involving physical or emotional abuse. (Of note, while only two dimensions for matching were chosen a priori, additional data are being collected for retrospective analyses of other potentially useful matching criteria.)

It is predicted tha t patients with high psychiatric severity or poor social supports will have a better outcome with inpatient rather than outpatient t reatment (as seen in Figure 1, outcome of the matched inpatient will be compared to tha t of the mis-matched outpatient). In contrast, it is predicted tha t matched outpatients who are low in psychiatric severity and have good social supports will have a similar outcome to the mismatched inpatient (as seen in Figure 1, the outcome of the matched outpatient will be compared to tha t of the mismatched inpatient). However, mismatched inpatients will have more costly t reatment than matched outpatients.

Description of Treatment Programs

Carrier Foundation's addiction treatment approach is an individu- alized, multimodal clinical approach that is well grounded in the traditional 12 step programs of Alcoholics and Narcotics Anony-


mous (AA/NA). In addition to the required addiction education and counseling, allied clinical therapy activities, and A . ~ A meetings, patients are also offered individual, marital, family, and group counseling as well as psychiatric and medical t reatment based on individual needs.

Both the inpatient and outpatient programs incorporate an intensive treatment phase and an aftercare phase. The intensive inpa- t ient phase lasts approximately 28 consecutive days. Rehabilitation occurs on a closed unit which operates on a structured three-level privilege system. Patients receive increased privileges as their con- dition improves, as assessed by staff and peers on the unit. Intensive outpatient rehabilitation consists of six consecutive weeks of treatment offered at hours convenient to patients. The aftercare phase of treatment occurs on an outpatient basis and lasts approximately 12 weeks, beginning at discharge from intensive treatment. It consists of weekly group sessions and AA/NA meetings.

Description of Patient Study Population

Thus far, 303 patients have participated in the project: 183 inpatients (60%) and 120 outpatients (40%). Patients are primarily Caucasian (85%), male (75%) and of middle socioeconomic status (55%). Their average age is 34 years, and they have completed an average of 13 years of education. The sample is relatively balanced with regard to marital status with 37% married, 41% single, and 22% divorced. Approximately half (55%) of the sample meet DSM- III-R criteria for alcohol abuse/dependence only, while the remaining patients (45%) meet DSM-III-R criteria for cocaine abuse/dependence, most of whom also have a concomitant diagnosis of alcohol dependence. (Of note, this sample includes a small percent- age - less than 7%-of patients from Somerset Medical Center, Somerville, New Jersey, whose outpatient t reatment program is comparable to Carrier's.)

Description of Research Procedures

As part of the research project, each patient 's mood, alcohol and drug use behavior, and psychosocial functioning is assessed at regular in-person research contacts using standardized interviews such as the Addiction Severity Index (ASI) (21), self-report ques- tionnaires such as the Michigan Alcoholism Screening Test-Re-


vised (MAST-R) (22), and biological measures such as breathalyzer tests and supervised urine drug screens. Typically, patients are seen weekly during intensive inpatient or outpatient treatment, monthly during aftercare, and at 3-month intervals until one year after discharge from the intensive treatment program. Weekly phone contact is made during aftercare to record each patient's hours in treatment, frequency of substance use and/or craving for substances.

Prel iminary Results: Attrit ion~Early Treatment Failure

Preliminary analyses have focused on evaluating whether there are differential attri t ion and early t reatment failure rates between settings in the intensive phase of treatment. Early treatment failures are defined as patients who meet at least one of the following criteria: 1) choose not to attend more than half of the intensive t reatment program (i.e., less than 3 weeks of outpatient or 2 weeks of inpatient); 2) leave treatment against medical advice; 3) are discharged for violating program principles (e.g., contra- band); 4) at tend treatment, but show no significant change in amount or frequency of alcohol or drug use by the end of the intensive t reatment phase.

Prel iminary analyses revealed tha t outpatients were 4 times more likely to be early t reatment failures than inpatients (chi- square = 41.2, d f = 1, p < .01). Thus, it appeared that engaging patients to stay in t reatment may be more difficult in the outpa- t ient than the inpatient setting.

Currently, an at tempt is being made to recontact t reatment failures. In addition, the high t reatment failure rate in the outpa- t ient t reatment setting has potential long-term consequences. An- alyses are underway to determine if there is any relationship between difficulty in responding to or staying with the treatment program and being mismatched to the t reatment setting. Finally, inpatients with profiles similar to mismatched outpatient treatment failures will be evaluated to assess whether the inpatient setting is more appropriate for patients with more severe profiles.

FINAL NOTE

In conclusion, given the number of shortcomings in existing research studies on the inpatient/outpatient addiction t reatment question, there is concern over the advisability of the drastic cut


backs in insurance reimbursement for inpatient substance abuse treatment. Research studies are needed, similar to the study cur~ rently presented, which allow for inclusion of patients who are clinically determined to require inpatient treatment. It is the hope of the authors that this communication wilt stimulate further scientific inquiry into this important question.

ACKNOWLEDGEMENTS

This research was supported by the National Institute on Alcohol Abuse and Alcoholism #AA07831 and Carrier Foundation. We are grateful to Gary Abraham, Wm. Dundon, isabelle Richards, and Joseph Rochford for their clinical support. Special thanks to Gina Byrnes and David Samuets for data collection, and Carolyn McGovern and Connie Osborn for preparing the manuscript.

REFERENCES

1. Culhane C: Bet W Ford: Stop punishing drug addicts. U.S. Journal of Drug and Alcohol Dependence 15(6):3, 1991.

2. Walsh DC, Hingson RW, Merrigan DM, et al: A randomized trial of treatment options for alcohol-abusing workers. The New England Journal of Medicine 325: 775-782, 1991.

3. Miller WR, Hester RK: Inpatient alcoholism treatment: Who benefits? American Psy- chologist 41:794-805, 1986.

4. Chernus LA: Clinical issues in alcoholism treatment. Social Casework: Journal of Contemporary Social Work 66:67-75, 1985.

5. Eriksen L: The effect of waiting for inpatient alcoholism treatment after detoxification: An experimental comparison between inpatient treatment and advice only. Addictive Behaviors 11:389-397, 1986.

6. Edwards G, Orford J, Egert S, et al: Alcoholism: A controlled trial of '~reatment" and "advice." Journal of Studies on Alcohol 38:1004-1031, 1977.

7. Stein LI, Newton JR, Bowman RS: Duration of hospitalization for alcoholism. Archives of General Psychiatry 32:247-253, 1975.

8. Kissin B, Platz A, Su WH: Social and psychological factors in the treatment of chronic alcoholism. Journal of Psychiatric Research 8:13-27, 1970.

9. Gallant DM, Bishop MP, Mouledoux A, et at: The revolving-door alcoholic: An impasse in the treatment of the chronic alcoholic. Archives of General Psychiatry 28:633-635, 1973.

10. Wallace J: Inpatient and °utpatient care: A balanced perspective" NAATP Review JulY/ August:18-21, 1989.

11. Edwards G, Guthrie S: A controlled trial of inpatient and outpatient treatment of alcohol dependency. Lancet 1:555-559, 1967.

12. Wilson A, White J, Lange DE: Outcome evaluation of a hospital-based alcoholism treatment programme. British Journal of Addiction 73:39-45, 1978.

13. Smart RG, Finley J, Funston, R: The effectiveness of post-detoxification referrals: Effects on later detoxification admissions, drunkenness and criminality. Drug and Alcohol Dependence 2:149-155, 1977.


14. Backland F, Lundwall L: Dropping out of treatment: A critical review. Psychological Bulletin 82:738-783, 1975.

15. Mattson ME, Allen JP: Research on matching alcoholic patients to treatments: Find- ings, issues, and implications. Journal of Addictive Diseases 11(2):33-49, 1991.

16. McLellan AT, Luborsky L, O'Brien CP, et al: Is treatment for substance abuse effective? Journal of the American Medical Association 241:1423-1428, 1982.

17. McLellan AT, Woody GE, Luborsky L, et al: Increased effectiveness of substance abuse treatment: A prospective study of patient-treatment '~matching." Journal of Nervous and Mental Disease 171:597-605, 1983.

18. Derogatis LR: SCL&0 administration scoring and procedures manual-I for the revised version. Baltimore, Johns Hopkins University School of Medicine, 1978.

19. Rounsaville BJ, Dalinsky ZS, Babor TF, et al: Psychopathology as a predictor of treatment outcome in alcoholics. Archives of General Psychiatry 44:505-513, 1987.

20. Woody GE, McLellan AT, Luborsky L, et al: Severity of psychiatric symptoms as a predictor of benefits from psychotherapy: The Veterans Administration-Penn Study. American Journal of Psychiatry 141:1172-1179, 1984.

21. McLellan AT, Luborsky L, Woody GE, et al: An improved diagnostic evaluation instrument for substance abuse patients: The Addiction Severity Index. Journal of Nervous and Mental Disease 168:26-33, 1980.

22. Selzer ML: The Michigan Alcoholism Screening Test: The quest for a new diagnostic instrument. American Journal of Psychiatry 127:1653-1658, 1972.

inpatient vs outpatient treatment for substance dependence revisited

Documents