Type of Pre-Treatment Income and Response to Treatment in Methadone Maintenance


Among the major goals of psychiatric treatment in general, and drug abuse rehabilitation in particular, is the development of the patient’s ability to support himself in a socially desirable manner. Operationally, this means reducing dependence on public assistance and illegal income, while increasing employment earnings. This goal, in particular, may be central to the overall rehabilitation effort and the maintenance of treatment gains in the areas of drug abuse and psychosocial adjustment.

It must be admitted that treatments for drug abuse have had limited success in achieving the goal of patient self-support (). However, given the variations within the drug-abusing population, it is possible that treatment programs which are only moderately effective with the total population, might be particularly successful (or unsuccessful) with specific subgroups of the population. In this regard, we felt that the amount and sources of patient’s pre-treatment income might serve as predictors of post-treatment adjustment in the area of self-support. Specifically, we had noticed three distinctly different sources of income within our patient population: earnings from employment, money from public assistance or other support agencies, and illegal income. We were, therefore, interested in the relationships between these three sources of income within the population and the extent to which these relationships changed following drug abuse treatment.


Subjects – Subjects were 165 male narcotic-dependent veterans admitted to the methadone maintenance program of the Drug Dependence Treatment Service (DDTS) at the Philadelphia VA Medical Center. This sample included all patients who were admitted during the period July 1, 1978 to June 30, 1979 and who remained in treatment for at least one month.

Data Collection Instrument – The instrument used to provide both the admission and six-month data was the Addiction Severity Index (ASI) which is described in detail in another article within this volume (Veterans Administration 1979).


We initially focused on those questions from the admission ASI which identified the amount and sources of the patient’s net income during the 30 days prior to the start of treatment. It should be noted that these responses were self-reports, and while some of the figures were estimates, validity checks were built into the ASI and we have found this self-report data to be generally quite accurate (). An examination of the data suggested that the sources of monthly income could be condensed into three categories: employment, support, and illegal.

Employment income was defined as the net cash value of salaries earned from full or parttime work from either regular or unreported “odd jobs.” Unemployment compensation was not included. Eighty-nine patients (54 percent) reported some income from employment at the time of admission.

Support income was defined as the total value of monies derived from public assistance (DPA), Social Security, pension, health benefit plan, and from family or friends. Only cash income was included, thus room and board were not considered in this total. Ninety-five patients (58 percent) reported some income from support sources (generally DPA) at the time of admission.

Illegal income was estimated as the net worth of cash and goods received from criminal activity. The majority of illegal income was derived from burglaries, theft, robbery, or drug sales. Eighty-three patients (50 percent) reported some income from illegal activity.

Twenty-one patients (12 percent) reported income from only one of these sources, 66 (40 percent) reported two sources, and 44 (27 percent) reported three sources of income. The mean income for the total population during the month preceding admission was $510 (S.D.=640) with an average of $230 (S.D.=612) from illegal sources, $206 (S.D.=418) from employment, and $74 (S.D.=216) from social support. However, the population mean and standard deviations for each of these sources indicated considerable population variance in the amount and sources of income, suggesting that the population might be comprised of several distinct subgroups having more homogeneous types of income.

The admission income data indicated that 43 patients had had virtually no income during the month preceding drug treatment, due to hospitalization or incarceration. An inspection of the individual income data on the remaining patients suggested that most subjects received the majority of their monthly income from one of the three major income sources. We therefore attempted to divide the remaining patients based on the source of at least 45 percent of their total monthly income. This division resulted in 30 patients who received at least 45 percent of their monthly income from employment, 42 who received primarily support, and 36 patients who reported primarily illegal income. An additional group of 14 patients comprised a heterogenous group of subjects who did not fit into the prior categories for various reasons. Thus, in the remainder of the paper we will focus upon the main income groups since they presented the clearest income picture and were representative of three basic economic subtypes within the population.

The admission status of each of the three main income groups is shown in table 1 in terms of demographic, background and treatment history measures. Although the groups were rather similar with regard to several of the demographic and educational variables, there were clear differences among them in most of the comparison items. It is important to note that the total monthly incomes within each of the three groups were approximately normally distributed with small standard deviations, thus quite representative of the income patterns of the individuals within each group.


Population Change – The pre-post comparisons (paired t-test) for the three major groups, as well as the total population (N=165) are summarized in table 2. The table presents mean values for each of the major groups on the ASI problem severity ratings, the days of patient-reported problems, and the amounts of the three income sources. The comparisons for the total population are presented in the last columns of table 2, and as can be seen, there were significant reductions in problem severity in the areas of drug abuse (p < .01) , employment, and criminality (p < .05) . The subjective reports of the patients also showed a significant (p < .01) reduction in the average number of days of drug use, from 20 in the month prior to admission to 13 in the month prior to follow-up. Similar reductions in “problem days” are seen in all areas, significantly so in the areas of unemployment and family and social problems. Finally, an examination of the population income figures indicates a considerable increase (57 percent) in the total income from employment.

Changes in Income Groups – The pre-post comparisons for the three major groups in the present study reveal the disparity of posttreatment change within the population. While these three groups were roughly alike in terms of their patterns of pretreatment problem severity, only the Employment and Illegal groups showed evidence of significant improvement by the time of follow-up. Both the Employment and Illegal groups reported significantly fewer days of drug use and unemployment, as well as fewer days of family and social problems. Significantly fewer criminal acts were shown by the Illegal group, and fewer days of psychological problems were reported by the Employment group. The improvement for these groups was most evident in the significant reduction of illegal income with corresponding increases in employment earnings.

The performance of the Support group stands in marked contrast to the results for the other two groups. The patients in the Support group showed little criminal activity or illegal income at either comparison point, but they also showed little evidence of employment or income from earnings. In addition, there were no significant changes shown in the other indicants of psychosocial adjustment. Thus, methadone maintenance was not associated with posttreatment improvement in any area for this group.


From a clinical perspective, two points were evident from the data. First, specific improvement in the area of patient self-support was highly related to general improvement in personal and psychosocial adjustment. While it was not possible to determine the causal agent in this relationship, the initial reduction of illicit drug use may have made increased employment and decreased criminality possible. It seems likely, however, that the maintenance of reduced drug use and increased psychosocial adjustment may have relied heavily upon the development of adequate and stable source of employment.

A second point which was clear from the data was that patients’ pre-treatment source of income was a powerful predictor of treatment benefit. With regard to the actual outcomes of the study groups, the results were somewhat surprising. The exceptional improvement shown by the Illegal group was gratifying considering the reluctance on the part of many programs to accept criminal justice system referrals. The absolute lack of improvement in the Support group stands in direct contrast to the results for the Employment and Illegal groups. We were struck by the stability of the Support group over time, the increase in support income from admission to follow-up, and the general absence of employment or illegal income in the majority of these patients at both admission and follow-up. The subjects in this group may be characterized as the “lethargic” patients who apparently have little involvement with active pursuit of income from either employment or illegal sources. It is important to note that the poor employment record of this group was apparently not due to more physical disability nor to less education and technical training (). Rather, these data suggest that many of the patients in the Support group may be characterologically dependent upon government sources of income, as well as physically dependent upon opiates. This characterological or institutional dependence may be a major reason why the Support group showed no evidence of improvement in any area. It may be that these patients require a program of intensive vocational and motivational development as an adjunct to existing methadone maintenance. Regardless of the potential reasons for their performance, it seems clear that the program of methadone maintenance and counseling which was associated with significant improvement in the majority of this population was not effective with this particular group of patients. It remains to be seen if these results are generalizable to other drug abuse treatment programs and to nonaddict populations. Nonetheless, the methodology presented here is relevant to the evaluation of drug abuse treatments, and the results may be useful in the clinical management of other rehabilitation programs.

A. Thomas McLellan, Ph.D.; John C. Ball, Ph.D.; Lawrence Rosen, Ph.D.; Charles P. O’Brien, M.D., Ph.D.

Selections from the book: “Problems of Drug Dependence, 1980: Proceedings of the 42nd Annual Scientific Meeting, the Committee on Problems of Drug Dependence, Inc.” Louis S. Harris, Ph.D., ed. Comprehensive assemblage of ongoing research on drug abuse, addiction, and new compounds. National Institute on Drug Abuse Research Monograph 34, February 1981.