New light on the maturing out hypothesisin opioid dependence


The data for the maturing out hypothesis


Author: J. F. MADDUX , D. P. DESMOND
Pages: 15 to 25
Creation Date: 1980/01/01

New light on the maturing out hypothesisin opioid dependence

J. F. MADDUX Professor, Department of Psychiatry, University of Texas, San Antonio, Texas, USA
D. P. DESMOND Senior Research Associate, Department of Psychiatry, University of Texas, San Antonio, Texas, USA

In 1962 Winick (1) reported that a large number of narcotic addicts known to the Federal Bureau of Narcotics apparently ceased narcotic use while in their 30s. He speculated that they "matured out" of addiction. The maturing out hypothesis has often been mentioned in the literature concerned with the course of opioid dependence. It is discussed in a recent survey of heroin addiction (2), and it appears in a well-known contemporary textbook of psychiatry (3). During the 18 years since Winick proposed the hypothesis, several studies of chronic opioid users have contributed information about the frequency and age of onset of prolonged abstinence and about the factors associated with it. In addition, we have collected information about prolonged abstinence in a longitudinal study of careers of 248 opioid users in San Antonio. In this report we shall examine the data from which the maturing out hypothesis was developed, assess the findings from subsequent studies which seem relevant to the maturing out hypothesis, and evaluate our own data with respect to the hypothesis.

The data for the maturing out hypothesis

Winick studied the age and duration of addiction of 7,234 addicts originally reported to the Federal Bureau of Narcotics in 1955, but not reported again during a five-year period through 1960. These persons were called "inactive addicts." According to Winick, experience had shown that it was almost impossible for a regular narcotic user to avoid coming to the attention of authorities during a two-year period. Therefore, inactive status seemed equivalent to cessation of drug use, with the exception of an uncertain number of subjects who died.

Winick noted a substantial concentration of addicts becoming inactive in their 30s. The distribution of his sample of inactive addicts by 10-year age groups is shown in table 1. Forty-five per cent were in the age group 30-39. The mean age of the inactive group was 35. From these data, however, we cannot infer that the addicts became inactive at a higher rate in their 30s than in other age groups. The base population of addicts from which the inactive subjects came may have had a high proportion of persons in their 30s in 1960, and consequently a high proportion of inactive addicts in their 30s simply reflected the age group distribution of the base population. Winick was aware of this problem; he showed that the age group distribution of the total active addict population in 1959 differed significantly from that of the inactive addicts, who were somewhat older than the active addicts. This difference is difficult to interpret. Some difference in this direction would be expected if the two groups had similar age distributions when originally reported, because the ages of the addicts inactive in 1960, five years after being reported, were compared with ages of the addicts active in 1959, approximately one to four years after being reported.

Table 1

Age of addicts becoming inactive on 31 December 1960 (Adapted from Winick (1))

Age group



Under 20
2313 32
3245 45
40 and over
1670 23
7234 100

Winick speculated that maturing out of addiction might be a function not only of age, but also of length of addiction. The mean length of addiction in the inactive group was 8.6 years. Using the same data base, he reported in a second paper (4) that age at onset of addiction and length of addiction were inversely correlated. That is to say, the earlier addiction started, the longer it lasted, and the later it started, the shorter it lasted. Thus age rather than length of addiction seemed to predict cessation of drug use. His speculation with respect to length of addiction was not confirmed.

The maturing out hypothesis consisted of more than a trend to cessation of drug use within a specified age group. Winick proposed a psychodynamic explanation. He speculated that the addicts begin taking heroin as a method of coping with the challenges and problems of early adulthood. Then, some years later, as a result of some process of emotional homeostasis, the stresses and strains of life become sufficiently stabilized so that the addict can face them without the support provided by narcotics. Winick did not have information about the emotional experience of his subjects.

Winick also addressed the question of the frequency of maturing out. Sixty-five per cent of 16,725 addicts originally reported during 1953 and 1954 were not reported again by 1959. Without citing his source of information, Winick stated that some 60 per cent of the patients admitted to the former Public Health Service Hospitals at Lexington, Kentucky, and Fort Worth, Texas, never returned. Perhaps he knew of Lowry's report (5) that 64 per cent of 17,741 patients admitted to the Lexington hospital during the years 1935 through 1952 had not been readmitted by 1955. From these data he speculated that two-thirds of addicts matured out of their addiction.

Subsequent follow-up studies and our San Antonio study have produced information related to three important features of the maturing out hypothesis: (1) the frequency of maturing out, (2) the age of maturing out, and (3) the psychodynamic explanation.

Frequency of maturing out

Hunt and Odoroff (6) found that 90 per cent of 1,881 patients discharged from the Lexington hospital became readdicted during a 1-4-year follow-up. Only 7 per cent remained voluntarily abstinent, that is, abstinent and not in an institution, throughout the follow-up period. Duvall and associates (7) studied a smaller sample of 453 patients in which every subject was followed for five years. They found that 97 per cent resumed opioid use at some time during the five years, but, with passage of years, increasing numbers became abstinent. By the fifth year, 25 per cent were voluntarily abstinent. In a follow-up of 267 persons identified by Baltimore police as narcotic abusers, Nurco and associates (8) found 35 per cent voluntarily abstinent at the time of interview. In a follow-up study of 266 Kentucky addicts after hospital treatment, O'Donnell (9) found that only 11 per cent of the subjects remained abstinent from the time of admission to the Lexington hospital to interview or death, a mean period of 10 years. However, at the time of interview, 43 per cent of the living subjects were abstinent. O'Donnell also found increasing abstinence with age.

These studies indicated that only a small percentage, 3-11 per cent, remained continuously abstinent for periods 5-10 years after hospital treatment, but higher percentages, 25-43 per cent, were found abstinent at the time of follow-up interview.

Winick used a follow-up period of five years. He assumed, however, that an addict might use an opioid drug for nearly two years without being apprehended. Therefore, continuous abstinence for three years or longer might be considered equivalent to maturing out. In three follow-up studies the frequency of three-year continuous abstinence was reported.

Ball and Snarr (10) found that 19 per cent of 108 former addict patients interviewed in Puerto Rico had been continuously abstinent for three years or longer at the time of interview. Vaillant (11) reported that 20 years after hospital treatment, 40 per cent of 100 opioid addicts had been continuously abstinent for three years or longer. He found increasing abstinence with passage of years. In their follow-up of Viet Nam veterans, Robins and associates (12,13) found that 39 per cent of 281 enlisted men who considered themselves addicted to narcotics in Viet Nam said that they had not used narcotics for three years after return to the United States.

In a study resembling the Winick study, Snow (14) found that 19 per cent of addicts reported to the New York City Narcotics Register during 1964 became inactive by the end of 1968. The inactive group consisted of those not reported again by the end of 1968, but excluded subjects known to be dead, arrested on a narcotics charge, committed for treatment, or institutionalized for more than two years during the study period.

We have followed the careers of 248 opioid users in San Antonio who were treated at the former Public Health Service Hospital in Fort Worth during the years 1964-1967. The methods have been previously described (15,16). The subjects began opioid use at the mean age of 18; the range was 13-46. All subjects used heroin at some time, and 95 per cent said that heroin was the principal opioid drug used. The opioid use histories varied markedly, from less than one year of daily use followed by over nine drug-free years to daily use for over 20 years except during periods of incarceration.

Figure 1 shows the opioid use status of the 248 subjects in percentages in July of each year from 1966 through 1975. The mean age of the subjects in 1965 was 28, and in 1975, if all were alive, would have been 38. The figure shows a moderate increase in the percentage abstinent, from 13 per cent in 1966 to 21 per cent in 1975. This trend corresponds to the findings of Duvall and associates, O'Donnell, and Vaillant. The figure also shows a marked increase, from zero to 19 per cent, in the percentage maintained on methadone. The data suggest that the decrease in the percentage using an illicit opioid drug was due primarily to provision of methadone maintenance.

By 1979, 54 (22 per cent) of our subjects had achieved three or more years of continuous abstinence, and they remained abstinent at the time of the most recent interview. An additional 27 (11 per cent) also achieved three or more years of abstinence but they later resumed heroin use. For seven of these 27, the period of abstinence occurred before the onset of daily use.

Thus the follow-up studies of addicts indicate that a minority, 19-40 per cent, appeared to achieve continuous abstinence of 3 years or longer. Vaillant called this "stable abstinence", but, as some of our subjects have demonstrated, three years of abstinence does not signify lifetime abstinence. If, nonetheless, we equate three-year abstinence with "maturing out", then much less than the two thirds suggested by Winick mature out, even when followed for 20 years or more after treatment.

Figure 1

Full size image: 20 kB, Figure 1

Opioid drug use status of 248 subjects in July of each year, 1966-1975. The percentage of subjects in each status is represented by the space between lines. Some percentages do not total 100 due to rounding

Winick, however, did not follow patients after treatment; his subjects were persons identified as addicts who were reported to the Federal Bureau of Narcotics. During the years 1967 into 1972, 44 of our 248 San Antonio subjects were reported to the Bureau of Narcotics and Dangerous Drugs (the BNDD, successor to the Bureau of Narcotics). A five-year follow-up of these subjects from the time of report would more closely resemble the Winick study than does follow-up after hospital treatment. Examination of the drug use histories of the 44 subjects revealed that only two (5 per cent) were abstinent three years or more during the five years after being reported.

Our study has also yielded data relevant to the assumption that it is almost impossible for a regular user to avoid coming to the attention of authorities during a two-year period. As noted, 44 of our 248 subjects were reported to the BNDD during a period somewhat longer than five years, 1967 into 1972; 204 were not reported. We reviewed the opioid drug use histories of these 204 unreported subjects to identify those who used heroin regularly for two years or more during approximately the same period, January 1, 1967, through December 31, 1971. We classified the person as a regular user if he used on a daily or near daily basis, and he did not have a period of hospitalization or incarceration exceeding 30 days. Thirty-six subjects met these criteria for regular use for two years. Thus it appears possible for an opioid user to use regularly for two years without being reported to the BNDD, though only a minority (15 per cent) of our subjects did so.

The marked variations in frequency of three-year abstinence reported from the follow-up studies can be attributed in part to differences in the study groups, to differences in duration of the follow-up period, and to differences in follow-up methodology. For example, the Viet Nam addicts differed markedly from the San Antonio addicts. The Viet Nam addicts began opioid drug use in the special environment of military service in Viet Nam, where drug use was highly prevalent and drugs were cheap and easily available; most of the men used by smoking, and they had used for less than a year when they returned to the United States. In contrast, the San Antonio addicts began their opioid drug use in their home environment, they used by intravenous injection, and they had used for a median of eight years when identified as research subjects. Vaillant may have found more three-year abstinence than others because he followed his subjects for 20 years. Some of Vaillant's high frequency of abstinence may also be attributed to his follow-up methodology. His 20-year data were derived primarily from institutional records. The San Antonio data, in contrast, were derived not only from agency records, but from repeated interviews with subjects during the follow-up period. Our experience has indicated that closer scrutiny of the longitudinal behaviour reveals more opioid drug use and less abstinence.

Age of maturing out

Both the Hunt study and the Duvall study found that male subjects aged 30 and over were abstinent more frequently than those under 30. O'Donnell's Kentucky addicts had a mean age of onset of addiction of approximately 30 and a mean age of first admission to the Lexington hospital of approximately 42. On the average, therefore, they had not matured out during their 30s. At the time of follow-up interview, the mean age of the Puerto Rican addicts studied by Ball and Snarr was 33. The 21 subjects abstinent for three years had ceased opioid use at the mean age of 30. If we assume a normal distribution of the ages, then approximately half ceased opioid use in their 20s. Half of Vaillant's New York addicts also began their three-year abstinence in their 20s.

Bess, Janus and Rifkin (17) studied 17 ex-heroin addicts who had used heroin for 2-24 years, and then had been continuously abstinent for two or more years at the time of study. Twelve (71 per cent) of the 17 ceased heroin use under the age of 30. Brill (18) studied 31 ex-addicts who had been abstinent for one or more years. They used heroin for a median of nine years, and they ceased use at the median age of 26.

Schasre (19) interviewed 40 persons who had used heroin for a median of five months (range 2-24) but who had been abstinent for 8-16 years when interviewed. The age of first use ranged from 13 to 17. Since the maximum age of first use was 17 and the maximum duration of use was two years, all subjects ceased heroin use under the age of 20. Schasre did not state how many of the 40 became compulsive daily users. Some probably stopped using before they became addicted.

At the time of the three-year follow-up interview, the Viet Nam veterans had an average age of 24. If the abstinent former addicts had the same age distribution as the entire sample, then their three-year abstinence began at the average age of 21.

Snow found that 46 per cent of his inactive group fell in the age range 28-37, but almost the same percentage, 44 per cent, of his active group fell in the same age range. It appears from his data that the incidence of becoming inactive in the base population increased with each older group.

These studies show that opioid drug users stop using at widely varying ages, from less than 20 to 40 and over. The studies do not demonstrate a high rate of onset of enduring abstinence in any particular age group, but some of them point to increasing abstinence of three years or longer with advancing age.

We stated that 54 of our 248 subjects achieved three or more years of abstinence by 1979. The 54 subjects began opioid use at the mean age of 19 (range 13-40), and had a mean of 15 years (range 1-30) from first use to onset of abstinence. At the latest interview, the mean duration of continuous abstinence was seven years (range 3-14). Table 2 shows the distribution of subjects by age of onset of abstinence. We see, in remarkable parallel to Winick's data (table 1), that 48 per cent became abstinent in their 30s. As we have noted, however, the percentage distribution of the abstinent group by age of onset of abstinence does not yield information about rates of onset in the base population. For each of 11 years, 1965 through 1975, we computed the annual rate of onset of three-year abstinence per 1,000 subjects by 10-year age groups. Then we computed the mean age-group rates for the 11 years. The results are shown in table 3. We excluded the age groups 10-19 and 50-59 because during the 11 years only small numbers were in these age groups. The highest rate appears not in the 30-39 group, but in the 40-49 group. Perhaps of greater importance is the trend toward increasing rates of three-year abstinence with advancing age. Possibly, if we can continue to follow our subjects as they grow older, we will find a yet higher incidence of onset of abstinence in the 50-59 age group.

Table 2

Age of onset of three-year abstinence of 54 former chronic opioid users

Age group



1 2
14 26
26 48
12 22
1 2
54 100
Mean: 34
Median: 33

Table 3

Mean annual onsets of three-year abstinence per 1,000 opioid users for 11 years, 1965 through 1975, by 10-year age groups


Annual onsets of abstinence per 1,000

All ages

Psychodynamic explanation

The studies which we have cited not only provided evidence about the frequency of prolonged abstinence and the age of onset; they also, in varying degree, offered explanations of the abstinence.

O'Donnell concluded that, consequent to changed prescribing practices, reduced availability of opioid drugs produced most of the abstinence among the Kentucky addicts. His abstinent subjects, however, attributed their abstinence to varied personal characteristics and events: will-power, responsibility to family, recognition of loss of control of drug use, disappearance of medical need for drugs, absence of addict associates, and the effect of hospitalization or therapy. These reasons for abstinence do not clearly point to personal maturation, but they do not exclude it. Like O'Donnell, Robins considered environmental change to be the principal factor in the abstinence of the Viet Nam addicts.

Bess, Janus and Rifkin found that traumatic events commonly preceded the renunciation of heroin use. These included impending prostitution, the threat of jail, disgust with their lives, a brother's death of overdose, and giving up a baby for adoption. The authors did not cite personal maturation as a factor in the abstinence. Brill also identified unpleasant features of the addictive life as significant factors prompting abstinence. He called these the "pushes" toward abstinence. However, he also identified "pulls" to abstinence, which consisted of the attractions of conventional values, the influence of relatives and friends, and interactions with treatment personnel. Often Brill's subjects became abstinent after they had reached a low point of loss and failure which he called "rock bottom". He did not cite evidence of maturing out. Apparently he believed that the increasing maladaptiveness of the addictive life provided a better explanation of cessation of drug use.

Schasre's subjects all began and ceased their heroin use during adolescence. Nearly half(18) of his 40 subjects stopped using after adverse experience arising from heroin use. The adverse experiences included recognition of physical dependence, conviction of a friend on narcotic charges, death of a friend from overdose, and arrest of the subject. In the remaining 22 subjects, reduced availability of heroin seemed to be the major factor in the onset of abstinence. The heroin source was arrested or left town, or the subject moved out of the neighbourhood in which he began the use of heroin. Personal maturation apparently played no part in the onset of abstinence among these subjects. They simply stopped using after some adverse experience or after they lost their heroin supply.

Vaillant (20) found that compulsory supervision (usually probation or parole), a substitute addiction (often alcohol), and establishment of a stable non-parental relationship appeared to facilitate abstinence. Although he did not use the specific term "maturing out", Vaillant noted that the journey from addiction to abstinence resembled adolescent maturation. Ball and Snarr also suggested maturation as a concept of the change from addiction to abstinence. They concluded that in the sense of terminating a drug-centred way of life and assuming a legitimate role in society, their abstinent Puerto Rican subjects matured out of their dependence on drugs. Thus of six published studies, only two offered maturation as an explanation of abstinence.

When our San Antonio subjects attempted to explain how they became abstinent, some cited single important events preceding the abstinence, but most described a combination of events, some external and some internal.

Some cited adverse events associated with compulsive opioid use. Friends died from overdoses, and our subjects stopped using because they did not want this to happen to them. Others stopped following one or more convictions for law violation. Some of the latter accumulated many years of prison, of loss, and of failure, and perhaps found the "rock bottom" cited by Brill. Seven of these became abstinent following a religious conversion.

Some attributed their abstinence to the effects of advancing years, but not to maturity in the sense of improved coping or a higher level of development. They said that they were "tired of running and hustling", or "too old to hustle", or "too old to be part of the rat race". Perhaps of similar significance but not expressed as fatigue, others said that heroin lost its appeal, or "at my age you find the thrill is gone".

One subject said that physical pain due to gallstones prompted his heroin use; when the pain was relieved, he no longer needed or desired heroin. Similarly, one or more of O'Donnell's subjects attributed the abstinence to disappearance of medical need for drugs. In contrast, however, three of our subjects attributed their abstinence to physical illness or injury resulting in chronic disability. These subjects became unable to conduct the daily activities required to maintain their drug dependence.

Some of the comments suggested an awareness of maturing. One subject said "It was time I grew up". Another said "Hustling on the streets and using drugs is kid stuff". Other comments implied maturing, if we conceive of heroin use as an immature activity, and commitment to personal relationships as more mature. Thus one subject gave up heroin because "I wanted to be someone my daughter could look up to". Another said "I did it for my children. My boy remembers when I got busted. I want to be a better example to him". Some cited a new investment, not only in people, but also in things. One Subject said, "I've got too many things going for me to go back to heroin. My kids are in school, I'm paying off my house, and I've bought a truck". Others attributed their abstinence partly or wholly to the help of a wife, to peers at work, or to therapy.

In addition to these explanations offered by subjects, we observed other conditions which probably facilitated their abstinence. These included relocation away from their usual source of drugs, evangelical religious participation, employment with a drug abuse treatment agency, probation or parole, and alcohol substitution. Over half (52 per cent) of the 23 subjects who substituted alcohol for opioid drugs had arrests for drunkenness or driving while intoxicated, or hospital treatment of alcohol dependence. The frequency of the conditions which probably facilitated abstinence is shown in table 4. These conditions probably do not signify maturation, except that moving away from the source of drugs, or working regularly for a drug abuse treatment agency could in part arise from increased maturity.

Table 4

Frequency of five conditions which probably facilitated three-year abstinence in 54 former addicts




Relocation away from usual source of drugs
19 36
Evangelical religious participation
13 25
Employment with drug abuse treatment agency
10 19
Probation or parole for 1 year or more
22 42
Alcohol substitution
23 43

Writing in 1974 about the careers of chronic heroin users, Winick (21) suggested five categories of reasons for abstinence: external circumstances, relationships jeopardized by drug use, weariness, personality and insight, and incapacitating physical problems. These reasons overlap but perhaps do not include all of the reasons offered in the studies we have cited. Winick asserted that some of the reasons may, in combination, develop into a process of maturing out. Without clearly stating it, he implied that maturing out was equivalent to permanent abstinence.


At a time when opioid addiction was generally thought to be a lifetime affliction, Winick offered the startling suggestion that perhaps two-thirds of the addicts mature out of their addiction approximately five years after being reported for opioid drug use. Subsequent studies have shown that he overestimated, but they have also shown that a substantial minority of addicts achieve continuous abstinence enduring for three years or longer. The onset of abstinence of three years or longer does not seem concentrated in the 30-39 age group. Instead, it gradually increases in frequency with advancing age. No study has followed a cohort of addicts until death; consequently the lifetime incidence of prolonged abstinence is unknown.

Several problems make it difficult to assess the importance of maturing in the process of becoming abstinent. First, although opioid dependence seems to inhibit the normal maturing process, some maturing probably occurs while the person is dependent on an opioid drug. Maturing may in some degree have facilitated the abstinence of persons who ostensibly stopped using primarily because of outside events. Second, the influence of maturing may vary with the severity and duration of opioid drug dependence. The person who stops using after many years of compulsive daily use probably experiences more personal change in the process of cessation than does the person who stops after less than a year of compulsive daily use. Third, Winick and other authors seem at times to equate maturing out of addiction with prolonged abstinence. If we conceive of chronic opioid drug use as an immature coping device, then any enduring abstinence can be called "maturing out".

In interviews, former addicts have offered an intriguing array of explanations for their abstinence, many of which do not point to maturing. Some apparently stopped using because of reduced availability of drugs, some because of adverse consequences of opioid drug use to themselves or others, and some because they became tired of the struggle to support their habits. Maturing out seems insufficient as a general explanation for all prolonged abstinence. While the "maturing out" concept has had considerable value in stimulating thought and research, its vague and global quality limits its usefulness. Present knowledge of prolonged abstinence can be summed up as follows:

A minority of chronic opioid users have periods of prolonged abstinence lasting for three years or longer. The lifetime frequency of such abstinence is unknown because no one has followed a cohort of opioid users throughout their careers until death. The incidence of three-year abstinence increases with age. Diverse conditions and motives seem to prompt and maintain the abstinence. Continuous abstinence for three years or longer does not ensure lifetime abstinence, for some of those abstinent for three years resume opioid drug use.



C. Winick, "Maturing out of narcotic addiction", Bulletin on Narcotics , vol. 14, No. 1 (1962), pp. 1-7.


J. J. Platt and C. Labate, Heroin Addiction, Theory, Research and Treatment (New York, John Wiley, 1976), pp. 193-194, 198.


A. M. Freedman, "Drug dependence", in Comprehensive Textbook of Psychiatry , 2nd ed., A.M. Freedman, H.I. Kaplan and B.J. Sadock, eds. (Baltimore, Williams and Wilkins, 1975), p. 1308.


C. Winick, "The life cycle of the narcotic addict and of addiction", Bulletin on Narcotics , vol. 16, No. I (1964), pp. 1-11.


J. V. Lowry, "Hospital treatment of the narcotic addict", Federal Probation , vol. 20, 1956, pp. 42-51.


G. H. Hunt and M. E. Odoroff, "Follow-up study of narcotic drug addicts after hospitalization", Public Health Reports , vol. 77, 1962, pp. 41-54.


H. J. Duvall, B. Z. Lock and L. Brill, "Follow-up study of narcotic drug addicts five years after hospitalization", Public Health Reports , vol. 78, 1963, pp. 185-193.


D. N. Nurco and others, "Study addicts over time: methodology and preliminary findings", American Journal of Drug and Alcohol Abuse , vol. 2, 1975, pp. 183-196.


J. A. O'Donnell, Narcotic Addicts in Kentucky (Washington, DC, Government Printing Office, 1969).


J. C. Ball and R. W. Snarr, "A test of the maturation hypothesis with respect to opiate addiction", Bulletin on Narcotics , vol. 21, 1969, pp. 9-13.


G. E. Vaillant, "A 20-year follow-up of New York narcotic addicts", Archives of General Psychiatry , vol. 29, 1978, pp. 237-241.


L.N. Robins, The Vietnam Drug User Returns , Monograph Series A, No. 2 (Washington, DC, Government Printing Office, Special Action Office, 1974).


L. N. Robins and others, "Vietnam veterans three years after Vietnam: How our study changed our view of heroin", in Proceedings of the 39th Annual Scientific Meeting of the Committee on Problems of Drug Dependence (Cambridge, Mass., 1977), pp. 24-40.


M. Snow, "Maturing out of narcotic addiction in New York City", International Journal of Addictions , vol. 8, 1973, pp. 921-938.


J. F. Maddux and D.P. Desmond, "Obtaining life history information about opioid users", International Journal of Addictions , vol. 1, 1974, pp. 181-198.


J. F. Maddux and D. P. Desmond, "Reliability and validity of information from chronic heroin users", Journal of Psychiatric Research , vol. 12, 1975, pp. 87-95.


B. Bess, S. Janus and A. Rifkin, "Factors in successful narcotics renunciation", American Journal of Psychiatry, vol. 128, 1972, pp. 851-865.


L. Brill, The De-addiction Process (Springfield, Ill., C. C. Thomas, 1972).


R. Schasre, "Cessation patterns among neophyte heroin users", International Journal of Addictions , vol. 1, 1966, pp. 23-32.


G. E. Vaillant, "A twelve-year follow-up of New York addicts: some characteristics and determinants of abstinence", American Journal of Psychiatry , vol. 123, 1966, pp. 5-584.


C. Winick, "Some aspects of careers of chronic heroin users", in Drug Use-Epidemiological and Sociological Approaches , E. Josephson and E. E. Carroll, eds. (New York, John Wiley, 1974), pp. 105-128.