Developing Indicators of Access to Care: Waiting Lists for Drug Abuse Treatment
The epidemic of acquired immune deficiency syndrome (AIDS) and the most recent version of the "war on drugs" have drawn increased attention to the issue of "waiting lists": lists of persons who have applied for drug abuse treatment but for whom a treatment position is not presently available. These waiting lists are generally seen as a measure of unmet demand for drug abuse treatment. For instance, the expansion of drug abuse treatment to the point where waiting lists need not occur was one of the primary recommendations of the Presidential Commission on the Human Immunodeficiency Virus Epidemic (hereafter the Presidential Commission; 1988). The National Commission on AIDS 3 has also advocated sufficient expansion of drug abuse treatment programming to prevent waiting lists (see the Commission's press release of September 26, 1989).
There is no consensus, however, on the issue of waiting lists. As discussed later in this paper, some experts in the drug abuse field contend that current waiting lists do not represent unmet demand for drug abuse treatment but rather poor referral mechanisms among currently operating drug abuse treatment programs. Other experts contend that current waiting lists greatly underestimate the unmet demand for treatment. Still others concede that waiting lists exist but nonetheless argue that treatment programs should not be expanded to the point where waiting lists would not occur.
This paper examines the empirical literature and expert opinions on waiting lists for drug abuse treatment and critiques the concept of waiting lists as a measure of unmet demand for treatment. Data from the early 1970s, when waiting lists for drug abuse treatment were also used as a measure of unmet treatment demand (U.S. House of Representatives, 1972), are included to provide some historical perspective. The limited data on the relationship between the waiting experience and subsequent client experience in drug abuse treatment are also reviewed. Finally, some unpublished data on the behavior of persons while on waiting lists are presented.
The development of techniques such as meta-analysis has greatly increased the analytic power of reviews of the scientific literature. Unfortunately, such techniques are not appropriate for examining the literature on waiting lists for drug abuse treatment. First, the number of studies addressing this issue is quite small. As one research group interested in the topic noted, "In spite of the increasing reports of the need to employ waiting lists, and their significance for the user and the treatment program, there is a virtual absence of research in this area, with the exception of those studies using waiting list controls" (Brown et al., 1989). Fewer than 15 studies were found through computerized literature searches in the MEDLARS, DIALOG, and PsychINFO 4 data bases with "waiting list" as a keyword. Most of the studies appeared in at least two of these three data bases, suggesting considerable overlap among them; such duplication also implied that additional computer searching was not likely to lead to many other articles or books. Interestingly, neither of the two most recent studies of the number of persons on waiting lists throughout the country was in these computerized literature data bases.
Many of the studies in the data bases were not relevant to this paper. A number of them recruited research subjects from among persons already enrolled in drug abuse treatment and then randomly assigned them to immediate
entry into an additional treatment component or to a waiting list control group that received the additional treatment component after the experimental group (e.g., Henik and Domino, 1975; Ingram and Salzberg, 1990). This is a powerful research design for assessing additional components to standard drug abuse treatment, but because the waiting list controls are already receiving some form of standard treatment, the data from these studies are not relevant to the question of waiting lists for initial entry into treatment.
The computer searches did identify a number of waiting list control studies for alcoholism treatment, smoking cessation treatment, and various types of psychotherapy. After a brief review of a number of these studies, however, the decision was made not to discuss them here. Attempting to bridge the many differences in specific research methods, as well as the differences in substantive content, would have meant a loss of focus for the present work. Only one study, involving subjects with both alcohol and other drug problems, was included because of its potential relevance to the argument that waiting lists can perform a useful screening function.
A telephone survey of eight experts in drug abuse treatment was conducted to try to find additional studies that were not in the computerized literature data bases. Some additional sources of useful information surfaced, but waiting list information was usually tangential to the main thrust of these studies, so it is not surprising that they were not coded for "waiting list" in the data bases. Although both the Presidential Commission on the Human Immunodeficiency Virus Epidemic and the National Commission on AIDS held hearings on waiting lists for drug abuse treatment, and the testimony at these hearings provided a good range of expert opinions on the subject, these hearings also illustrate the relative lack of scientific data on the subject.
This paper draws, as well, on the past personal experience of one of its authors (D. C. Des Jarlais) as a research scientist for the Division of Substance Abuse Services of the state of New York. Because of the size of the state drug abuse treatment system and the concentration of programs in New York City, waiting lists and the demand for treatment have been administratively studied in New York for almost 20 years. New York attempts to both eliminate double-counting in its waiting list compilation and to distinguish between those persons on waiting lists who are not yet in any treatment program and those persons already in treatment but waiting to transfer to another program. (As discussed later in this paper, there are also persons who operate drug abuse treatment programs in New York who believe that the waiting lists are artificial.)
Finally, this paper contains some unpublished data on the behavior of persons listed as waiting to enter drug abuse treatment. These data were collected as part of a study of an "interim" methadone clinic funded by the
Centers for Disease Control (CDC; Yancovitz et al., 1991). They were the only data that the authors could locate that included longitudinal urinalysis results from persons on a waiting list for entry into drug abuse treatment.
Because of the difficulties that we encountered in finding scientific data on waiting lists for drug abuse treatment, this paper is closer to an essay than the series of meta-analyses that we would have liked to perform. The paper, therefore, expresses our personal opinions rather than conclusions based on a substantial literature relevant to the topic.
THE NUMBER OF PERSONS ON WAITING LISTS
The National Association of State Alcohol and Drug Abuse Directors (NASADAD) conducts occasional surveys of its members regarding waiting lists for drug abuse treatment. The 1989 survey covered 43 states and showed a total of 66,000 persons on waiting lists (Institute of Medicine, 1990). NASADAD's series of surveys is one of the few nationwide studies to date of waiting lists for drug abuse treatment. Yet although the NASADAD information is clearly useful, there are severe methodological limitations to these data. First, the definition of the term waiting list is not standardized across states (and probably is not standardized within many states). Some programs may include an applicant on a waiting list after a simple telephone contact, others after a face-to-face contact, and still others only after a preadmission determination of program eligibility has been made. There is probably even more variation regarding when programs remove applicants from a waiting list. Some programs remove applicants immediately if they do not respond to one attempted contact by letter or telephone; others do so only after repeated attempts at contact. Other programs remove applicants after a fixed period of time following the first attempt at contact, whereas others wait until one (or more) scheduled intake appointments have been missed. Some programs remove applicants from a waiting list at regular intervals; others do so sporadically. Moreover, the extent to which corrections for double-counting (when a single applicant is on waiting lists for more than one program) or for transfers (when an applicant on one program's waiting list is currently receiving treatment at another program) were made at the state level, before the data were submitted to NASADAD, is not known.
The U.S. Conference of Mayors (1987) conducted a survey of 42 cities in 1987 and found waiting lists for drug abuse treatment in three-quarters of those municipalities. The average duration of time on waiting lists ranged from 7 to 26 weeks. This survey cannot be used to estimate the total number of persons on waiting lists nationwide, but it does confirm the other studies indicating the widespread use of waiting lists throughout the country.
The National Institute on Drug Abuse (NIDA) recently completed a survey of waiting lists at the request of the Office of National Drug Control Policy (Anita Lewis Gadzuk, public health analyst, Office of Applied Studies, Substance Abuse and Mental Health Services Administration, personal communication, March 1991). Known as the Drug Services Research Survey (DSRS), this study differed from previous NASADAD studies in several ways. First, for-profit as well as publicly funded programs were included. Second, information on program capacity, program utilization, and program costs was also obtained. Third, the DSRS utilized a sampling technique rather than attempting to obtain information from all programs. Questionnaires were mailed to 1,183 programs, followed by telephone interviews to complete the questionnaire. Site visits to a subsample of 120 programs were then conducted, and 20 client records were examined at each site. The DSRS study, however, did not eliminate double-counting or waiting list transfers from its data collection.
Preliminary data from the DSRS study suggest that at the time that the study was conducted, there were 530,000 treatment slots in the United States and 107,000 persons on waiting lists. Sixty-three percent of persons who had entered treatment spent one month or less on a waiting list; 37 percent waited longer than one month. The average utilization rate for all programs was 90 percent. Great variation in utilization rates was seen, however, ranging from 56 percent for inpatient programs to 97 percent for methadone maintenance programs.
No certain explanations were found for the differences between the NASADAD and the DSRS results (e.g., 66,000 versus 107,000, respectively, as the total number of persons on waiting lists). The two surveys differed not only in the time at which they were conducted (i.e., 1989 for the NASADAD survey versus 1990 for the DSRS) but also in the sampling frame and the methodology of data collection employed. As discussed later, the large difference in the size of the waiting lists is probably not attributable solely to inclusion of the for-profit drug treatment programs in the DSRS study.
''THERE ARE NO REAL WAITING LISTS FOR TREATMENT"
Even though both the NASADAD and the DSRS studies showed large numbers of persons on waiting lists for drug abuse treatment in the United States, some experts argue that these waiting lists do not represent unmet demand for treatment per se. Prominent among these is Dr. Beny Primm, who previously headed the Addiction Research and Treatment Corporation (ARTC), a large methadone maintenance treatment program in New York City. He is currently head of the Office of Treatment Improvement, a component agency of the federal Alcohol, Drug Abuse, and Mental Health
Administration. The major premise of Dr. Primm's argument is that waiting lists represent the drug treatment system's inability to efficiently refer persons applying for treatment to programs that have unused capacity. When he was at the ARTC and the program did not have places for new applicants, he found that calling other programs in the city would almost always identify a program with an open treatment slot (B. Primm, personal communication, March 1991). This was during a time, the late 1980s, when the official waiting list of persons applying for treatment in New York City averaged more than 1,000 individuals. 5
The argument that the present treatment waiting lists represent poor treatment referral systems (and not unmet demand) is clearly a minority position—but it does deserve consideration. It is possible to reconcile Primm's experience at the ARTC with the waiting lists reported by other programs in New York. There are approximately 35,000 long-term drug abuse treatment positions in New York City, with a turnover of probably 20,000 persons per year. 6 Thus, diligent referral work can often locate an open position in a treatment program. Sometimes this slot will be found at a program that has openings and no current waiting list. Often, the open position will be one that had been assigned to a person on a waiting list for a "full" program but the applicant had failed to show up for the intake procedures. (The question of waiting list applicants who fail to appear for intake will be discussed in some detail later.)
Street outreach programs in New York City also report that they receive many requests for assistance in securing a place in a treatment program and that diligent referral work can locate an opening for many of these individuals (Centers for Disease Control, 1990; Des Jarlais 1989; Friedman et al., in press; Jackson and Rotkiewicz, 1987). These outreach programs also report, however, that the unmet demand for treatment is real and that the ability to find an opening for an individual client is quite a different matter from the ability to place the large numbers of persons who evidently want to enter drug abuse treatment. Within a large, high-turnover drug treatment system such as that in New York City, it is quite possible that better referral work could place a substantial number of additional individuals in treatment programs; yet it is also true that the waiting lists still reflect a substantial unmet need for treatment.
At another level of analysis, quite good evidence exists that waiting lists and unused treatment capacity currently coexist in the United States. There are a large number of for-profit treatment programs in the nation, and many of them clearly have unused capacity. The DSRS study found that the average utilization rate among for-profit programs was only 57 percent. Indeed, the utilization rates of these programs are so low that many of them are in danger of closing (Korcok, 1991). However, within current drug abuse treatment systems in the United States, it would not be possible to reduce waiting lists by simply referring applicants from publicly funded programs to for-profit programs. The for-profit programs typically require that their clients have either private health insurance or the ability to pay the substantial fees that these programs charge. 7 McAuliffe (1990) has described the results of the two-tier (publicly and privately funded) structure of programs as having "rationed treatment to lower-income addicts seeking care."
The two parallel systems—of publicly funded and for-profit drug abuse treatment programs—are in many ways similar to the general provision of health care in the United States, and a detailed analysis of what would have to happen in order for persons on waiting lists for publicly funded programs to be accommodated within for-profit programs is beyond the scope of this paper. We merely note that integration of the programs would require a major philosophical change in the way drug abuse treatment is funded in this country. It would also require an open examination of the social class and ethnic/racial antagonisms that the present two-tier structure conceals.
WAITING LISTS UNDERESTIMATE UNMET DEMAND
The preceding section presented the arguments surrounding the notion that current waiting lists overestimate the unmet demand for drug abuse treatment. Yet there are also those who argue that current waiting lists substantially underestimate the unmet demand for treatment. These arguments are based on experience with the rapid, large-scale expansion of drug abuse treatment and also on the finding by Watters and colleagues (1986) that nearly half of the out-of-treatment drug users they interviewed said that they would enter treatment “tomorrow” if a position were available.
From 1971 through 1974, the New York City Health Department opened 40 methadone maintenance treatment clinics, and more than 22,500 persons were admitted for treatment (Newman, 1977). Applications for treatment were accepted prior to the opening of the clinics (thus creating waiting lists), and the zip code of the applicant's residence was included in the basic information collected in the application, permitting analyses by geographic area. Newman compared the number of new applicants from zip codes that already had methadone clinics (prior to the opening of the Department of Health clinic) with the number of new applicants from zip codes in which the Department of Health clinic was the first in the area (Newman, 1977, p. 110). In the areas with a preexisting clinic, the average number of new applications to the Health Department clinic nevertheless remained relatively constant over time: from 274 in the second month prior to the opening of the Health Department clinic, to 235 in the month immediately preceding opening, to 251 in the month of opening, to 284 in the month after opening, to 231 in the second month after opening. (The clinics provided services to between 250 and 300 patients.) If the waiting list of applications that was established prior to the opening of the clinics had, indeed, represented all of the unmet need for treatment, then there should have been a significant drop in the number of applications after the opening of the clinics. Instead, the number of new applications per month remained relatively constant, suggesting that the opening of the clinic itself brought forth many new applications from persons who would not have applied without the perception that treatment would actually be available.
This potential effect of the perception of treatment availability was even more dramatic in the areas in which there were no preexisting methadone maintenance clinics. In those areas, the number of new applications dramatically increased with the opening of the Health Department clinics: from an average of 279 in the second month preceding opening, to 349 in the month preceding opening, to 706 in the month of opening, to 752 in the month after opening, to 864 in the second month after opening. For these areas without prior methadone treatment, the start of actual provision of treatment was followed by large-scale increases in the number of new applicants. In the light of these results, the preopening waiting list was clearly an underestimate of the actual unmet demand for treatment.
A more recent example involving New York City occurred during the spring of 1988. At that time, the Beth Israel methadone maintenance program had an unduplicated waiting list of approximately 500 persons who had applied but were not currently receiving methadone maintenance treatment. Funds were obtained to add an additional 500 treatment positions to the more than 8,000 treatment positions then in the program. The intent was to reduce or eliminate the waiting list. The 500 new patients were admitted within a period of three months. But only three months later, the
waiting list had again stabilized at approximately 500 persons. Hence, the opening of a substantial number of new treatment positions had not led to a permanent reduction in the waiting list but instead brought out new applications from persons who would not have applied without the perception that more treatment was being made available.
The argument that waiting lists underestimate the true unmet demand for treatment can be most easily understood by considering an analogy between waiting lists and the official unemployment rate. Just as waiting lists are composed of persons who are seeking but have not secured treatment, the officially unemployed are persons who are seeking but have not secured employment. Yet in addition to persons who are officially unemployed, there are also "discouraged workers" who might want employment but who have stopped seeking it because they do not expect to be able to find a job. Because they are not actively seeking employment, these discouraged workers are not included in official unemployment calculations. Nevertheless, if large-scale sources of employment develop in their communities, many discouraged workers apply for the positions. Similarly, the opening of new drug treatment positions may lead many persons with drug abuse problems to apply for treatment even though they were not previously on waiting lists.
The analogy between waiting lists and official unemployment figures can be carried at least one step further. If the new jobs are particularly attractive, one would expect larger numbers of discouraged workers to apply for them than might apply for less attractive jobs. So, too, if new treatment programs are particularly attractive, one would expect larger numbers of "discouraged persons with drug problems" to apply for the new treatment positions. Methadone maintenance was a particularly attractive type of treatment for heroin addicts, one that induced many more persons to seek treatment than were on waiting lists prior to its development. A chemotherapy that would both relieve cravings for, and block the effects of, cocaine might be a particularly attractive type of treatment that could attract many more persons than are currently on waiting lists for cocaine treatment.
SHOULD THERE BE WAITING LISTS?
To summarize, some experts believe that current waiting lists accurately reflect unmet demand, others believe that current lists greatly overestimate such demand, and still others believe that current lists greatly underestimate it. Nevertheless, all of these groups tend to agree that the presence of waiting lists is undesirable. There are, however, still other experts in the United States who argue that meaningful waiting lists do exist but that it is desirable to have them.
This argument is not usually presented in terms of the desirability of
waiting lists but in opposition to the concept of "treatment on demand." Treatment on demand is not a well-defined term in the drug abuse field; at a minimum, however, it implies a treatment system in which a person would be able to receive treatment immediately after applying. There are two components of the arguments against treatment-on-demand (Kleber, 1990). 8 The first is cost. Kleber has estimated that a true treatment-on-demand system would require that the programs in the system operate at no more than 95 percent of capacity to ensure absorption of any unexpected surge in the number of applications. Estimating the increased costs of operating a drug abuse treatment system in which the programs operated at no more than 95 percent of capacity is beyond the scope of this paper, but it is worthwhile to note that Kleber's formulation could provide a useful empirical standard for assessing unmet demand for treatment independent of the actual number of persons on waiting lists.
Standardized definitions of the capacities of drug abuse treatment programs and of who is enrolled in a treatment program 9 are less than ideal, but clearly they are much easier to formulate than a standardized definition of who is on a waiting list. In addition, the problems of double-counting and transfers do not arise. The factor of perceived treatment availability leading to more new applicants for treatment could also be incorporated into this framework—provided that information about the immediate availability of treatment was disseminated to persons with drug abuse problems. Currently, many programs in the United States operate above 95 percent of capacity (some are operating at above 100 percent of official capacity) so that, by this standard, there is clearly a situation of unmet demand for treatment in the country.
The second argument against treatment on demand concerns the use of waiting lists for motivational screening of applicants. One of the major problems in current forms of drug abuse treatment is the high percentage of persons who drop out of treatment before completion. In addition, up to 50 percent of all applicants never actually enter treatment, and in some types of treatment, up to 50 percent of those who do enter drop out within the first three months (Hubbard et al., 1989; Newman, 1977; Simpson et al., 1978). The long-term effectiveness of treatment in reducing drug abuse—and in reducing HIV risk behaviors (Ball et al., 1988)—is strongly related to the time spent in treatment (Hubbard et al., 1989; Simpson et al., 1978).
Applicants who fail to enter treatment or entrants who drop out shortly after entering represent groups for whom being in treatment provides little or no long-term benefit.
There are undoubtedly many reasons for some applicants' decision not to enter treatment or to drop out so quickly after entering; one frequently cited by clinicians is that these are the people who were never "sufficiently motivated" to actually enter and/or remain in treatment. If being on a waiting list indeed serves to screen out applicants who are not sufficiently motivated to enter and remain in a treatment program, then the waiting list would be serving a positive function by maximizing the effectiveness of scarce drug abuse treatment resources. Consideration of this motivational screening argument leads us to the small number of empirical studies of the behavior of drug users while on waiting lists and the effect on subsequent treatment experience of being on a waiting list.
EMPIRICAL STUDIES OF WAITING LIST BEHAVIOR
As noted earlier, it was quite difficult to find empirical studies of waiting list behavior, and most of the waiting list studies that used controls involved subjects who were already in some form of drug abuse treatment. Moreover, none of the few studies that are directly relevant to the topic of drug user behavior (while not in treatment and on a waiting list) were able to obtain representative samples of the persons on waiting lists at the time. Thus, the limited findings must be interpreted with considerable caution; they do suggest, however, that the topic is quite complex.
Brown and colleagues (1989) conducted a study of 29 persons on a waiting list for a residential treatment program in Baltimore that specialized in treatment of cocaine abuse. The 29 respondents interviewed were recruited from among 50 persons on the waiting list who could actually be contacted, after it was found that less than 50 percent of all persons with whom contact was attempted could actually be reached. Comparisons were made between 16 persons who had been on the waiting list for one to three months, and 13 persons who had been on the waiting list for four to six months. Being on the waiting list for the longer period of time was associated with more criminal justice system involvement and more pressure from others to enter treatment. Forty-eight percent of the total group of subjects had reduced their drug use while on the waiting list, but 59 percent were pessimistic about their ability to remain free of drug use-related problems. Eighty-seven percent of the 23 intravenous drug users in the study reported having changed their behavior to reduce the risk of AIDS, with safer injection as the primary form of risk reduction. A majority of the subjects (52 percent) reported that their interest in entering treatment had decreased since being on the waiting lists. The authors noted that the subjects whom they
were able to reach were those who were probably doing relatively well and that any deleterious effects of not being able to enter treatment were probably greater for the waiting list persons whom they were not able to contact.
Patch and colleagues (1973) conducted a study of heroin users who were on a waiting list for methadone maintenance treatment in Boston. The subjects were on the waiting list for periods ranging from 18 months to 2 years. There were high rates of death, incarceration, and family separation among subjects during the time that they were on the waiting lists. The very length of time on the waiting list in this study makes it difficult to draw causal inferences about being on a waiting list and the observed outcomes. Without reapplication to the program or some form of continued contact between the program and subjects, it is not clear what (if anything) being on the waiting list meant to the subjects over such an extended period. This is of particular concern, given the number of studies cited earlier that found that approximately half the persons on waiting lists do not enter treatment when a treatment position opens up and the program attempts to contact them. It is quite possible that, if time on the waiting list had been relatively short (e.g., several weeks to a month), half of the subjects in this study would have been removed from the waiting list without actually entering the program to which they originally applied. Even with this limitation, this study still should be taken as an important caution against allowing long waiting lists to develop.
Gunne and colleagues (Gunne and Gronbladh, 1984) conducted a randomized assignment study of heroin injectors applying for methadone maintenance treatment in Sweden. Thirty-four subjects were randomly assigned to either acceptance into methadone treatment or to a control/no-methadone treatment condition. (The study was ethically justifiable because it would not have been possible to exceed the official capacity of the clinic.) Even though the study involved a small number of subjects, the results were quite dramatic: 76 percent of the treatment group were considered successfully rehabilitated at follow-up versus only 6 percent of the control group. None of the treatment subjects had died, compared with 5 of the 17 control subjects. Because the follow-up period covered more than two years for the subjects, however, this study should be interpreted in terms of denying treatment to persons with heroin addiction problems rather than merely delaying such treatment by putting a person on a waiting list.
Addenbrooke and Rathod (1990) directly examined the relationship between time on a waiting list and later retention in treatment for 130 drug users referred to the Substance Abuse Project at Crawley Hospital, West Sussex, England. The researchers were not testing the motivational screening hypothesis but rather its opposite—that quick entry into treatment would increase motivation and lead to higher retention rates. Ninety of these individuals were accepted for treatment and had a clearly documented date
of initial referral. Alcohol use was the primary problem for these individuals, with 69 of the 90 reporting problems associated with alcohol use only. Forty-four of the subjects were accepted into treatment within one week of referral, and 46 were accepted after a longer period. The mean duration of treatment was longer for the quick-entry-into-treatment group (median duration of treatment = 2.9 months) than for those who delayed entry into treatment (median duration of treatment = 1.6 months), but this difference was not statistically significant according to the Mann-Whitney U test. Subjects who had problems with drugs other than alcohol (but no problems with alcohol) stayed in treatment for a shorter period than subjects who had trouble with alcohol only and with alcohol plus other drug use-related problems (medians of 1 and 2.1 months, respectively). In this study, a small number of subjects had problems with drugs other than alcohol, and the relationship between time from referral and retention in treatment was not presented separately for this subgroup. One therefore cannot extrapolate these results to the population of illicit drug users who are the concern of this paper. Clearly, however, the paper does not provide support for the use of waiting lists for motivational screening to increase the cost-effectiveness of drug abuse treatment.
Grenier (1985) conducted a study that used waiting list controls to assess the effectiveness of an adolescent drug abuse treatment program. (Because the persons of interest were minors and because they had not signed informed consent documents, the data were actually collected from their parents.) Only a minority of the persons on the waiting list could be contacted for data collection—27 out of 74—although full cooperation was obtained from all persons who were reached. There was some evidence of improvement among the adolescents on the waiting list. After excluding those who had received other treatment, 14 percent were classified as abstinent from mood-altering drugs and a total of 43 percent as "improved." (The abstinence rate among persons who had received treatment in this particular program was significantly higher—66 percent.) Although this study suggests that the program had a positive influence, the "motivational screening" effect might have undermined the comparison. If the persons who received treatment had to undergo a waiting period prior to entering treatment, there might have been a selection bias toward "more treatment motivation'' in this group.
Yancovitz and colleagues (1991) conducted a study of a limited-service "interim" methadone maintenance clinic in New York City. Subjects were recruited from the waiting list of the Beth Israel Medical Center methadone maintenance program and randomly assigned to either immediate entry into the interim clinic or to continuation on the waiting list. Interviews and urinalyses were performed for both groups of subjects.
The study had several limitations. Because the subjects were all volunteers,
they cannot be considered a representative sample of persons on the general waiting list for the methadone maintenance program. Moreover, because they were already on a waiting list when they entered the study, any immediate effects of being placed on a waiting list had already occurred for both the experimental and control groups. Yet despite these limitations, this is the only study that could be located that had actual random assignment, multiple time-point follow-up, and urinalysis results. (The study's analyses of self-reported drug use indicated systematic biases.)
At the group level, there was little change in drug use over the one-month follow-up for waiting list controls. Sixty-two percent of the subjects had evidence of heroin use in their urine sample at their entry into the study, and 60 percent had evidence at follow-up. Seventy-one percent had evidence of cocaine use in their urine sample at entry; 70 percent showed such evidence at follow-up. Twenty-six percent had evidence of unprescribed methadone in their urine sample at entry; 37 percent had it at follow-up. This last difference was a statistically significant increase and, given the risk of AIDS when illicit drugs are injected, can be considered some evidence for "improvement" while on the waiting list. (The experimental treatment group showed a highly significant reduction in heroin use and a nonsignificant trend toward a reduction in cocaine use.)
At the individual level, considerable variation occurred over time among the waiting list control subjects. For 26 percent of the subjects, the heroin urinalysis results differed from entry to follow-up. For 31 percent of the subjects, the cocaine urinalysis results differed from entry to follow-up. For 33 percent of the subjects, the methadone urinalysis results differed from entry to follow-up. Although urinalysis detects only recent drug usage (approximately two days for cocaine and one week for heroin and methadone), these results suggest considerable variation over time in drug use by individual subjects in a waiting list condition.
After one month in the waiting list control condition for this study, subjects were then transferred into the interim clinic experimental condition. Eventual enrollment in regular methadone treatment was compared for the group that was immediately assigned to experimental treatment versus the group that remained on a waiting list for an additional month. The subsequent enrollment in regular treatment was significantly higher—72 percent—for the group that had been immediately assigned to treatment than for the group that remained on the waiting list for an additional month—56 percent.
In summary, there are very few studies of the actual behavior of persons on waiting lists. All of these studies have major limitations; in particular, none utilized a representative sample of persons on waiting lists. Nevertheless, these few studies are consistent on several points. First, the behavior of persons on the waiting list is not frozen at the level of behavior
observed when the person applied for treatment. There is considerable individual variation over time while subjects are on the waiting list, and the general group direction appears to be toward modest improvement. As expected, the studies that compared being on a waiting list with actually being in drug abuse treatment showed more positive results associated with being in treatment. No evidence could be found that time on a waiting list leads to positive motivational screening in such a way that a longer time waiting would lead to better treatment outcomes for those who do enter treatment. If anything, there is some slight evidence that being placed on a waiting list may have a generally discouraging effect, with a potential net loss of treatment effectiveness.
The diversity of expert opinion about waiting lists for drug abuse treatment in the United States is not surprising, given the small amount of empirical research on either the administrative aspects of waiting lists or the behavior of persons while on waiting lists. The discussion in this paper must therefore be a mixture of cautions regarding use of the currently available data and of suggestions for needed additional research.
Current estimates of the number of persons on waiting lists should not be considered accurate quantitative measures of the unmet demand for drug abuse treatment. The unsolved problems of the same individual being on different waiting lists, and of individuals wanting to transfer, in themselves preclude using these estimates as measures of actual unmet demand. Nevertheless, the number of programs that have waiting lists and the number of persons on those lists demonstrate that the present drug abuse treatment system is not effectively meeting the demand for treatment in this country. Some of the unmet demand for treatment could be alleviated by better referral mechanisms among programs. Reviewing studies on the effectiveness of referral systems for drug abuse treatment was beyond the scope of this paper, but caution is necessary here. Referral systems themselves consume scarce resources. Giving a referral to a drug user may satisfy a service provider's need to do something, but it does not even guarantee that the drug user will actually enter that program, much less remain in it. In addition, at a motivational level, drug users may actually do better in programs that they have indicated that they wish to attend, rather than in those that happen to have openings at a given time.
The currently unmeasured number of drug users who desire treatment but do not apply because they do not expect to be taken into treatment needs to be addressed (e.g., Watters et al., 1986). Following the analogy of official unemployment rates, which fail to account for "discouraged" workers, the number of discouraged drug users who would like to enter (but are
not seeking) treatment is likely to be greatest when the waiting lists are especially long.
A three-part definition of the desired situation, in which demand for treatment is effectively met, can be proposed:
All programs in the system normally operate at 95 percent (or less) of capacity. Capacity is increased for those programs that approach 100 percent utilization.
Drug users in the community know that they can be accepted into the treatment program of their choice as soon as they apply.
There are no barriers to treatment program entry that would inhibit drug users from applying. Such barriers might include the need for (private or public) health insurance, the need for cash payments, the lack of child care, and the limited treatment modalities available in some cities.
Comparisons of the extent to which different communities are meeting the need for drug abuse treatment would have to include all of these components. Data on the first and third criteria could usually be obtained from program records or interviews with staff, although the latter would preferably also include interviews with users. Data on the second would require interviews with drug users in the community. Fortunately, a number of research projects are currently interviewing large numbers of drug users not in treatment. The National AIDS Demonstration Research studies and the Drug Use Forecasting system could, at very little additional cost, collect data on drug users' perceptions of the availability of treatment.
Finally, given that at present the country appears to be tolerating large (but undetermined) numbers of persons on waiting lists, more research is critically needed on what happens to these drug users. Little is known about the effects of being placed on a waiting list, and almost nothing is known about why so many drug users do not enter treatment when a position becomes available. The new studies should employ better methods than those currently in use. For example, it may not be possible to obtain a perfectly representative sample of persons on a waiting list, but it is surely possible to come much closer to this objective than current studies have done. Larger sample sizes are also necessary to allow examination of the possible differential effects of being on a waiting list, considering such variables as age, gender, ethnicity, and history of drug use.
In preparing this paper, we have examined opinions and research on waiting lists for drug abuse treatment in the United States. We have conducted research on waiting list behavior and in this paper call for more and better studies on the topic. We also find ourselves deeply troubled by the
ethics of performing research on people who need and are awaiting treatment—unless that research is tied to efforts to help them get that treatment. Debating the meaning of waiting lists, without a good faith commitment to provide treatment for all who need it, appears to be even less ethically justifiable.
Addenbrooke, W. M., and Rathod, N. H. 1990. Relationship between waiting time and retention in treatment amongst substance abusers. Drug and Alcohol Dependence 26:255–264.
Ball, J. C., Lange, W. R., Myers, C. P., and Friedman, S. R. 1988. Reducing the risk of AIDS through methadone maintenance treatment. Journal of Health and Social Behavior 29:214–226.
Brown, B. S., Hickey, J. E., Chung, A. S., Craig, R. D., et al. 1989. The functioning of individuals on a drug abuse treatment waiting list. American Journal of Drug and Alcohol Abuse 15:261–274.
Centers for Disease Control. 1990. Update: Reducing HIV transmission in intravenous drug-users not in drug treatment—United States. Morbidity and Mortality Weekly Report 39:529, 535–538.
Des Jarlais, D. C. 1989. AIDS prevention programs for intravenous drug users: Diversity and evolution. International Review of Psychiatry 1:101–108.
Friedman, S. R., Neaigus, A., Jose, B., et al. In press. Behavioral outcomes of organizing drug injectors against AIDS. Proceedings of the Second Annual Research Conference of the National AIDS Demonstration Research Projects, Rockville, Md.: National Institute on Drug Abuse.
Grenier, C. 1985. Treatment effectiveness in an adolescent chemical dependency treatment program: A quasi-experimental design. International Journal of the Addictions 20:381–391.
Gunne, L., and Gronbladh, L. 1984. The Swedish Methadone Program. Pp. 205–213 in Social and Medical Aspects of Drug Abuse. G. Serban, ed. New York: Spectrum Publications.
Henik, W., and Domino, G. 1975. Alterations in future time perspective in heroin addicts. Journal of Clinical Psychology 31:557–564.
Hubbard, R. L., Marsden, M. E., Rachal, J. V., Harwood, H. J., Cavanaugh, E. R., and Ginzburg, H. M. 1989. Drug Abuse Treatment: A National Study of Effectiveness. Chapel Hill and London: University of North Carolina Press .
Ingram, J. A., and Salzberg, H. C. 1990. Effects of in vivo behavioral rehearsal on the learning of assertive behaviors with a substance abusing population. Addictive Behaviors 15:189–194.
Institute of Medicine. 1990. Treating Drug Problems. D. R. Gerstein and H. J. Harwood, eds. Washington, D.C.: National Academy Press.
Jackson, J., and Rotkiewicz, L. 1987. A Coupon Program: AIDS Education and Drug Treatment. Paper presented at the Third International Conference on AIDS, Washington, D.C., June 4.
Kleber, H. G. 1990. Testimony before the National Commission on AIDS, March 15, 1990.
Korcok, M. 1991. Private addiction treatment faces closings, sell-offs, cuts. American Medical News, May 8.
McAuliffe, W. E. 1990. Health care policy issues in the drug abuser treatment field. Journal of Health and Political Policy Law, 15:357–385.
Newman, R. G. 1977. Methadone Treatment in Narcotic Addiction. New York: Academic Press.
Patch, V. D., Fisch, A., Levine, M. E., et al. 1973. A mortality study of waiting list patients at the Boston City Hospital methadone maintenance clinic. Pp. 523–529 in Fifth National Conference on Methadone Treatment Proceedings. New York.
Presidential Commission on the Human Immunodeficiency Virus Epidemic. 1988. Final Report of the Presidential Commission on the Human Immunodeficiency Virus Epidemic. Washington, D.C.: U.S. Government Printing Office.
Simpson, D. D., Savage, L. J., and Sells, S. B. 1978. Data Book on Drug Treatment Outcomes: Follow-up Study of 1969–1972 Admissions to the Drug Abuse Reporting Program (DARP). Report No. 78-10. Fort Worth, Tex.: Institute of Behavior Research, Texas Christian University.
U.S. Conference of Mayors. 1987. The Anti-Drug Abuse Act of 1986: Its Impact in Cities One Year After Enactment. Washington, D.C.: The Conference.
U.S. House of Representatives. 1972. Narcotic Addiction Treatment and Rehabilitation Programs in New York City. Report to Subcommittee No. 4, Committee on the Judiciary. Washington, D.C.: U.S. House of Representatives.
Watters, J. K., Iura, D. M., and Iura, K. W. 1986. AIDS Prevention and Education Services to Intravenous Drug Users Through the Midcity Consortium to Combat AIDS: Administrative Report on the First Six Months. San Francisco: Midcity Consortium.
Yancovitz, S. R., Des Jarlais, D. C., Peyser, N. P., et al. 1991. A randomized trial of an interim methadone maintenance clinic. American Journal of Public Health 81:1185–1191.