A survey of a representative sample of addicts prescribed heroin at London clinics


Conditions of drug use
Stable addicts


Pages: 13 to 22
Creation Date: 1970/01/01

A survey of a representative sample of addicts prescribed heroin at London clinics *

Addiction Research Unit, Institute of Psychiatry, 101 Denmark Hill, London, S.E.5


A representative sample of heroin addicts being prescribed heroin was selected from London clinics. One hundred and eleven (86.7%) of the sample selected were interviewed between March and November 1969.

Eighty-four (76%) were male and twenty-seven (24%) were female. The mean age was 25.0 years, with 69% being aged 25 and under. Thirty-nine per cent were currently in full-time employment and current employment status was significantly related to previous work pattern. Eighty-four per cent reported that in the month prior to interview they used drugs other than those prescribed for them by the clinic and 80% regularly used unsterile injection techniques. Thirty-four per cent reported criminal activities other than those covered by the Drugs Acts during the three months prior to interview. Fifty-one per cent had at some time been given in-patient treatment for withdrawal from drugs, 36% had at some time been given hospital treatment for physical complications associated with drug use; the most frequent complication reported was abscesses (46%).

Current employment status and the degree of involvement with other addicts was significantly related to other areas of behaviour. It appears that there are some addicts who are "stable", but these are a small proportion of the total population investigated here. Evidence concerning the validity of the interview data is presented.

* A shorter report about the work described in this article appeared earlier this year in The Lancet, (vol. 1: 7657, p. 1163-66, 1970).


To date reports on the epidemiology of heroin abuse in the United Kingdom have either given basic information on the total population of known users [ 2] , [ 14] , or have presented the results of more detailed investigations of special sub-groups [ 3] , [ 5] - [ 7] , [ 9] , [ 11] . There is clearly a need for further information on the population at large, and the study reported here set out to fulfil this need.

As would be the case in the development of research into any new phenomenon, such as widespread drug abuse, early studies aim to describe the people involved in the problem; to estimate the incidence of certain morbid factors associated with the phenomenon, and to tabulate the incidence of, what are assumed to be, likely predisposing events, e.g. parental separation, childhood behaviour, etc. As knowledge increases and research techniques become more sophisticated, one can expect to find reports which go beyond mere description, and attempt to offer some explanation for the phenomenon in question. Moreover, one would expect that our view of the phenomenon itself would change and that it will no longer be seen as an all or nothing affair, nor would the population of people defined as manifesting the phenomenon be regarded as homogeneous. People are no longer regarded as either delinquent or not, nor is delinquency viewed as a single syndrome. Similarly one cannot talk of drug abuse without specifying the drug or drugs in question, and it seems reasonable to suppose that the abusers of a particular drug would not be all of the same kind, nor would they suffer the same consequence of their use. Research workers on both sides of the Atlantic have recognized the need to look for different types of drug users, although few studies have made differences between users, their main focus of attention. The report by Meyer et al [ 10] is a notable exception. Few British reports have so far presented the simplest kinds of cross-tabulation of the variables examined.

The study described here began with the explicit assumption that not all addicts are the same, and was designed to examine the nature and extent of inter-addict differences. Of particular interest were differences with respect to an individual's current mode of living. Our thinking was influenced in part by the concept of the "stable" addict. The assumption that such a person exists has been implicit in the British response to the problem of addiction. It is embodied in the report of the Rolleston Committee of 1926 [ 8] which defined the circumstances for the long-term prescription of opiates to addicts to be such that the addict "while capable of leading a useful and fairly normal life so long as he takes a certain non-progressive quantity, usually small, of the drug of addiction, ceases to be able to do so when the regular allowance is withdrawn". When this policy was formulated, the majority of English addicts were people who had become addicted through medical treatment and for whom drug use did not prohibit (and may indeed have helped) their living a "fairly normal life". To what extent the ideal of a stabilized addict holds up with the newer, younger addicts, who have contributed so significantly to the rising incidence of heroin addiction in this country [ 2] , is something the study was designed to investigate.

The results presented in this paper are derived from an interview study of a representative sample of addicts prescribed heroin at treatment clinics in the London area. In the main, we present only the distribution of the subjects on key variables. The results of some cross-tabulations are presented in the last section. A subsequent paper will deal with types of addicts in more detail.

The sample

Since April 1968 the prescription of heroin for the treatment of addiction has, as a result of the 1967 Dangerous Drug Act, been restricted to certain doctors who are specially licensed by the Home Secretary. In the London area the majority of licensed doctors are attached to teaching hospitals and prescribe at addiction treatment clinics. Ideally addicts from all parts of the country would be sampled, but since the majority of addicts attending clinics were, at the time of planning the study, in the London region (904 out of 1,139), and since the non-London addicts were spread over a large area, it was decided to restrict sampling to those patients attending the London clinics only. The criteria for the selection of an individual patient were that he should be attending a London clinic on an out-patient basis and be prescribed heroin at the time of selection. During the period of the survey (March-November 1969) the 15 London clinics were between them prescribing heroin to a steadily decreasing number of people and, at any one time, the total was no more than 450.

The consultants in charge of 14 out of the 15 clinics agreed to allow us access to their patients. At one clinic there was only one patient and this clinic was not included in the study. From each of the remaining 13 clinics we obtained a list of the names of patients currently attending and receiving heroin. A one-in-three sample of names was drawn from each list using a table of random numbers. A supplementary list of names was also drawn up at the same time, and names were picked from this list to replace any patients who were ineligible for interview at the time of our first attempt to contact them. A person was considered ineligible if, at the time of our first attempt to interview him, he was no longer receiving heroin on prescription or if he had been admitted to hospital, sent to prison, or had left the clinic. No replacements were selected for people who were eligible for the study, yet whom we were unable to interview (see below).

The interview

The interview schedule was built up on the basis of extensive pilot work and concentrated on aspects of the addicts' current life situation. It covered work record, pattern of drug use, conditions of drug use, ability to care for the self, involvement in criminal activities, involvement with other addicts, various measures of attitudes, and indices of behaviour prior to drug use. The majority of subjects were contacted personally when they arrived at the clinic. If the clinic had space and the subject had time the interview was conducted, or at least started, at the time of first contact. If this was not possible, a suitable time and place for the interview were arranged. A subject selected for interview was told on first contact that the interview was confidential and that no part of it would be disclosed to the clinic or to any other authority. The research was presented as attempting to "understand a bit more about what is happening to people who are coming to these clinics". No monetary incentive was offered, except to those subjects attending one clinic where for a variety of reasons, considerable difficulty was experienced in contacting them. A letter was sent to these subjects offering £1 for the interview. This incentive did not, however, make contacting the subjects easier.


The sampling procedure yielded a list of 128 subjects to be interviewed. Of these 111 were successfully interviewed, giving a response rate of 86.7%. (The response rate for males was 91.3% and for females 75%.) Non-response was due to refusal [ 9] , failure to make contact with the subject after five or more attempts (6) or because the subject was too disturbed to be able to participate in the interview [ 2] . The mean length of the interview was 100 minutes with a range from 55 to 235 minutes.

Basic information on non-interviewed subjects was abstracted from clinic records in all cases but one. Comparing the non-interviewed with the interviewed subjects, the non-interviewed were significantly older (mean age 27.8 years, as compared to 24.6 years, P<.05) and there were proportionately more females (8 out of 17 as against 27 out of 111, x 2 = 3.97, P<.05). In the following sections data are presented for the interview sample only (N = 111), unless otherwise stated.


Age. The mean age for the total sample was 25.05 years (s.d. = 6.94) with a range from 17 to 52 years. Forty per cent of the sample were aged 21 years and under, and 69% were aged 25 and under.

Place of birth. Eighty-four (76%) of the interviewed subjects were born in England or Wales, four (4%) came from Scotland, thirteen (12%) from Northern or Southern Ireland and ten (9%) were born outside the British Isles, including three Americans and four Canadians.

Marital status. Of the total sample eighty (62%) were single, thirty-six (28%) were married, six (5%) were divorced and three (2%) were widowed, (three not known).

Socio-Economic Group. The Socio-Economic group distribution for the total sample, based on the present or last job is shown in table I.


Socio-economic group of subjects

Subject's occupation



Socio-economic groups *



Population statistics **

Professional workers
3,4 2 1.6 5.6
Employers and managers
1,2,13 10 7.8 12.0
Foremen, skilled manual workers and own account workers (other than professional)
8,9,12,14 12 9.4 35.0
Non-manual workers
5,6 43 33.6 23.5
Personal service workers, semi-skilled manual workers and agricultural workers
7,10,15 33 25.8 14.5
Unskilled manual workers
11 23 18.0 8.0
Members of armed forces and persons with inadequately described occupations
16,17 1 0.8 1.3
Not known
4 3.1

* As classified by the Registrar General in census reports of the United Kingdom.

** Economically active males, Greater London area, Sample Census 1966

As the average age of the sample is low, it is difficult to interpret differences between our sample and the general population. Table II compares the socio-economic group for the fathers of our subjects with the socio-economic group distribution for England and Wales.


Current work status. In the total sample, 52 (40%) were currently employed full-time, 11 (9%) were employed part-time or casually, 50 (39%) were unemployed, 11 (9%) were housewives, one was a student and one was of indeterminate employment status, (two not known). Table III shows the relationship between age and current employment status. The proportion of unemployed in the age group 20-24 was significantly larger than that for any other age group ( x 2 = 4.23, P<.05).


Father's occupation



Socio-economic groups *



Population statistics **

Professional workers
3,4 10 7.8 4.5
Employers and managers
1,2,13 23 18.0 10.6
Foremen, skilled manual workers and own account workers (other than professional)
8,9,12,14 36 28.1 39.4
Non-manual workers
5,6 20 15.6 17.1
Personal service workers, semi-skilled manual workers and agricultural workers
7,10,15 10 7.8 18.0
Unskilled manual workers
11 15 11.7 8.3
Members of armed forces and persons with inadequately described occupations
16,17 3 2.4 3.0
Not known
11 8.6

* As classified by the Registrar General in census reports of the United Kingdom.

** Economically active males, Great Britain, Sample Census 1966.


Age and employment status


Employment Status



Part-time or casual


Housewives and others

Not known


Under 20
16 3 12 1 0 32
15 1 23 6 0 45
11 4 5 4 1 25
Over 29
10 3 10 2 1 26
52 11 50 13 2 128

Work stability. Of the 43 (39%) interviewed subjects reporting full-time employment, three had not worked at all in the week prior to interview and a further 11 had worked during the week but had taken some time off. Five of the 11 took time off to visit their clinic or chemist, six took time off for other reasons including sickness due to withdrawal, buying drugs and lack of interest in the job; (two cases not known). Of all interviewed subjects then, 27 (24%) were known to have worked a full week in the week prior to interview.

Table IV shows the relationship between an individual's current work status and the amount of time he has spent working in the three months prior to interview. As can be seen, an individual's current state of employment is not necessarily transitory; those currently employed are very likely to have worked for 12 or 13 of the previous 13 weeks, and those currently unemployed are unlikely to have worked at all during this period, (P<.01).

Furthermore, these subjects currently in full-time employment compared to all others had worked for a greater proportion of the time since they began using heroin ( x 2 = 10.00, P<.01) and they less frequently report having lost a job through their use of heroin ( x 2 = 4.54, P<.05).


Current work status and number of weeks worked in the eight months prior to interview


Employment status

Weeks worked


Part-time or casual


33 2 1
9 7 7
1 *
0 36

Employed, but off sick all 13 weeks.

Means of support

Means of support during the month preceding interview are given in table V, (sources are not mutually exclusive). Twenty-four (22%) reported that their own earnings were their sole means of support during that period and forty-two (38%) reported that at least one of their means of support was illicit, i.e. shoplifting, stealing, selling drugs


Means of support in last month



Per cent

Own earnings
54 49
Student grant
1 1
Income from rent, maintenance, pension
5 5
Social Security
31 28
Unemployment Benefit
4 4
Sickness Benefit
10 9
Money from family
28 25
25 23
Selling things
23 21
Money from friends
24 22
5 5
13 12
8 7
Petty stealing (including shoplifting)
17 16
Selling drugs
25 23

Subjects can be classified into the groups shown in Table VI according to the types of support reported. These types are "Own income," (items 1 - 3) in table V above, "Social Security" (items 4 - 6), "Hustling" (items 7-13) and "Illicit" (items 14-16) and each type is mutually exclusive.


Types of support in last month




Own income only
29 26
Own income plus Social Security only
17 *
Social Security only
8 7
Own income plus Social Security plus hustling
and/or illicit
12 11
Social Security plus hustling and/or illicit
Hustling only
7 6
Hustling and illicit only
12 11

Own income and Social Security were not necessarily concurrent for these people but both sources were used during the month prior to interview.

Drug use

Everyone in the sample was, by definition, receiving heroin on prescription. The mean dose prescribed was 134.9 mg per day with a range from 10 mg to 1140 mg. Fifty-eight per cent of subjects received 120 mg or less and only nine subjects received 360 mg or more.


Drug use (N = 111)


In last month




Ever used








91 82 95 86 107 97
Other opiates, e.g. morphine, opium, pethidine
0 0 25 22    
Amphetamine, amphetamine/barbiturate mixtures and other stimulants
14 13 49 44 109 98
Sedatives and hypnotics
51 46 83 75 103
95 *
5 4 19 17 90 81
14 13 32 29 104 94
0 0 68 61 111 100
0 0 13 12 72 65

* N = 108.

In addition to heroin, ninety-one were prescribed methadone and fifty-one a sedative or hypnotic drug. Table VII shows the numbers receiving each class of drug on prescription, the numbers reporting having used the drug in the last month, and the numbers reporting having ever used the drug.

The majority of subjects reported some illicit use of drugs in the month prior to interview, in fact all but 18 subjects (16%) were using one or more drugs illicitly during that time and an additional five (4%) were using cannabis only as well as prescribed drugs. Forty-one subjects (37%) reported that on occasion they sold, exchanged, or loaned some of the drugs that they were prescribed.

Onset of drug use. The mean age for first use of any drug was 16.8 years with a range from 11 to 40 years. Ninety-three per cent of subjects had used their first drug before they were aged 21. The mean age for first use of heroin was 19.0 years with a range from 14 to 48 years. Ninety-one per cent had used heroin before they were aged 25. In nine (8%) cases, heroin was the first drug ever used. In 21 (19%) cases the first drug injected was methylamphetamine, in 80 (72%) cases heroin was the first drug injected, and in ten (9%) cases it was other drugs (morphine two, methadone two, cocaine two, sodium pentobarbitone two, hydromorphone hydrochloride one, dimethyltriptamine one).

The mean length of time from first use of heroin to the time of interviewing was 5.25 years (s.d. 5.06) with a range from 1 to 30 years.

Conditions of drug use

All subjects were using heroin by injection. In the week prior to interview 28 (25%) subjects had injected themselves at home only, a further 17 (15%) had injected themselves at their clinic or at a day centre as well as at their home, 46 (41%) had injected themselves at some time during the week in a public toilet as well as at home, clinic or day centre, and 20 (18%) had injected themselves in a "public place", e.g. shop doorways, in the street, or in telephone kiosks, in addition to any of the above places. Sixty-two (56%) had injected themselves whilst other addicts were present and also injecting themselves.

The majority of subjects (101, (91%)) used a disposable syringe for their injections and of these 67 (66%) had used their syringe more than once (50, (45%) using it more than twice) during the previous week. Seventy-two subjects (68%) did not clean their arms prior to injection.

Although the most frequent method for preparing an injection was to dissolve heroin in sterile water, 54 (49%) had at some time during the week used ordinary unboiled tap water and 12 (11%) had used water from a lavatory basin. Nine subjects reported sharing a syringe with another addict during the week.

If injection practices are defined as sterile when normal medical procedure is followed (i.e. using a new disposable syringe for each injection or using an adequately sterilised glass syringe, not sharing a syringe, cleaning the arm prior to injection, and making up the injection with sterile water) then in the total sample 12 subjects (11%) are using sterile injection practices. In the week prior to interview all others had engaged in some non-sterile practice.

Involvement with other addicts

Subjects were asked how many other addicts they knew well enough to say hello to, and the proportion of addicts amongst their close friends. Forty-nine (49%) claimed to know 40 or fewer addicts. These were less likely to claim addicts amongst their close friends than those who knew 41 or more addicts. Twenty-five subjects (24%) claimed to have no friends who were not using drugs of some sort. Of the 59 subjects (53%) who had an ongoing heterosexual relationship, in 22 cases (20%) the partner was using heroin and for a further 18 (16%) the partner was using other drugs but not heroin.


Forty-seven subjects (43%) reported that they had been convicted of an offence under one of the Drugs Acts. Fifty-six (51%) reported a conviction for a non-drug offence before their first use of heroin and 40 subjects (36%) reported such a conviction since using heroin. In all eighty-seven subjects (79%) admitted to a conviction of some sort. Thirty-one subjects (28%) stated that they had been sentenced to prison and 43 (39%) that they had been to a borstal, remand home, approved school, or detention centre.

Current criminal activities

Twenty-two (20%) subjects reported no criminal activity in the three months prior to interview. Forty (36%) reported only activities that would be covered by the Drugs Acts-selling drugs, being in illegal possession of drugs, altering a prescription. Thirty-eight (34%) admitted to activities of an acquisitive type-stealing things from shops, receiving stolen goods, breaking into a house, breaking into a meter or vending machine, obtaining things by false pretences, taking and driving away, forging cheques, in addition to reporting a drug activity. Three (3%) reported stealing only, (8 not known).

Sleeping and eating

In the week prior to interview, 29 subjects (27%) reported that they had not slept during one or more nights, and 47 subjects (45%) reported that they slept abnormal hours, i.e. that they slept at irregular hours during the day and night and did not take all their sleep in one period.

Thirty-five subjects (33%) reported that they usually ate two meals or more each day, 39 (37%) that they usually had one meal each day, and 32 (30%) that they usually ate snacks only.


Forty-four subjects (41%) reported that they were living with their parents or in a house or flat which they themselves owned. Thirty-one (29%) lived in furnished or unfurnished flats, 21 (19%) in bedsitters and 12 (11%) were of no fixed abode. Twenty-five subjects (22%) had been at their current address for three months or less, 29 (26%) for between four months and one year, 16 (14%) up to three years and 29 (26%) for four years or more, (12, (11%) N.F.A.).


Subjects were asked about periods of complete abstinence from heroin since first becoming aware of their addiction. Thirty-four (32%) had never been abstinent from heroin, 29 (27%) had been abstinent once only, and the remainder (44, (41%)) had been abstinent two or more times. For all those who had been abstinent at some time, the mean number of times that this had occurred was 2.0. Periods of abstinence ranged from one week or less (10 cases) up to five years (2 cases), with 34 (47%) of those ever abstinent off for nine weeks or less. Twenty-eight subjects (25%) reported having been abstinent in prison and 43 (39%) in hospital. Forty-six subjects (41%) had at some time remained abstinent from heroin whilst living in the community.


Hospitalization for reasons connected with drug use, e.g. physical complications, overdose or withdrawal, had been experienced by 72 subjects, (67%), at some time, and in 14 cases (13%) this had been in the three months prior to interview. Mean number of admissions for the sample was 2.26. Between them the 72 subjects who had been admitted to hospital in connexion with their drug use had a total of 242 admissions.

In-patient withdrawal treatment of some kind or another had been given to 55 subjects (51%) and inpatient treatment for septicaemia, hepatitis, abscesses or overdose, had been given to 39 subjects (36%).

Complications reported by subjects were septicaemia (18 cases, (16%)), hepatitis (44 cases, (40%)), overdose (44 cases, (40%)) and abscesses (51 cases, (46%)). Twenty-two (20%) had abscesses in the three months prior to interview.


Separation from both parents for a period of a year or more prior to the age of sixteen was reported by 24 subjects (22%). A further 28 subjects (25%) had been separated from one parent, only, 20 from the father and 8 from the mother.

Sixty-six subjects (62%) left school at the age of 15 years or younger and eight (7%) stayed on until the age of 18 years or older. Type of last school attended is given in table VIII.

Minor truancy whilst at school was reported by 46 subjects (44%), 15 (14%) reported excessive periods of absence through truancy, and 14 (13%) reported that they were known as "troublemakers" whilst at school. The rest (29, (28%)) reported no trouble at school.


Type of last school attended



Per cent

Secondary Modern
56 50
22 20
4 4
Comprehensive "A"
3 3
Comprehensive other
3 3
Approved school
5 4
Schools in other countries
7 6
5 4
Not known
6 5

Further education had been undertaken by 45 subjects (42%) and this included evening classes and correspondence courses (11, (10%)), day release and apprenticeship (6, (5%)), university (5, (4%)), and other full-time or part-time college (18, (16%)). Of those who did enter into some kind of further education, 27 did not complete their course.


In the introduction it was noted that there is a need in the field of addiction research to go beyond simple descriptions of drug-using populations, and to examine the relationships between variables. As a prelude to further analysis along these lines, the relationship between two variables - current work status and involvement with addicts - and other variables were examined. The following results are not an exhaustive list of significant relationships, but have been chosen from different areas of behaviour to highlight the differences within the population sampled.

For subjects in this sample, current work status is significantly related to previous work record (see above) indicating some consistency over time. Comparing those subjects who were in full-time employment at the time of the study with those who were unemployed reveals further significant differences with respect to variables relating to current and previous behaviour.

Those addicts currently employed were more likely than those currently unemployed (P<.05) to :

Not run out of heroin before the next prescription is due;

Be prescribed smaller amounts of heroin;

Not sell, exchange or lend drugs that are prescribed;

Not misuse the drugs prescribed, e.g. not to inject drugs not designed to be injected;

Use fewer drugs illicitly;

Have used fewer drugs ever;

Not have visited Piccadilly in the week prior to interview;

Not have had septicaemia, abscesses or jaundice in the past three months;

Be rated as more conventional in appearance by the interviewer.

As one measure of involvement in the drug sub-culture, subjects were asked the number of other addicts that they knew. The sample was divided into two approximately equal groups by taking those who knew 40 or fewer addicts and those who knew 41 or more. No significant differences in employment status was found between these two groups. The following differences between the two groups in other areas of behaviour were found.

Those who said they knew 41 or more addicts were more likely than those who knew 40 or less (P<.05) to:

Run out of heroin before the next prescription is due;

Have used "Chinese" heroin in the past three months, (i.e. illegally imported heroin);

Use more drugs illicitly;

Misuse the drugs prescribed;

Inject themselves in a "public" place;

Inject themselves in the company of other addicts;

Have had more than two convictions;

Have been convicted of a drug offence;

Have sold drugs in the last three months;

Have visited Piccadilly in the week prior to interview;

Have had septicaemia, abscesses or jaundice in the past three months;

Be rated by the interviewer as unconventional in dress.

The interpretation of these relationships needs some care, for they do not necessarily indicate cause and effect, and it is probable that some third variable or combination of variables is operating that can explain the relationships. The relationships cannot be explained in terms of length of addiction.


Representativeness of the sample

The sample selected represents 34.4% of all patients prescribed heroin on an Out-Patient basis at the 13 London clinics included in the study. The sample was selected by a random procedure and is representative of the population of addicts receiving heroin at these clinics. Some bias in the results may have occurred due to non-response, as we know that those subjects who we were unable to interview differ from the interviewed sample in terms of age and sex.

An attempt to predict the characteristics of the total London clinic population receiving heroin must take into account the fact that interviewing did not take place at two of the 15 clinics. At the first clinic the caseload was too small (varying from one to two) to choose a sample. At the second there were approximately 75 patients, of whom an unknown proportion were prescribed heroin. This clinic appears to be similar to other clinics included in the study in terms of patient population and so the bias caused by exclusion of this clinic is likely to be small. We are, therefore, confident that the sample described in this paper can give us information about the total population from which it is drawn.

Less than 50% of patients attending clinics were being prescribed heroin during the time of the study. Other patients "on the books" were either not at the time attending the clinic on an out-patient basis, or were being prescribed substitute drugs (mainly physeptone), but not heroin. How those who are prescribed heroin differ from those who are prescribed substitute drugs and why one patient will be prescribed heroin and another physeptone, is not known. It would appear that those who are prescribed heroin include ( a) those with a long history of addiction and of being prescribed for their addiction, ( b) those who the clinic doctor considers are able to function on heroin, i.e. those who are employed and who do not become arrested or hospitalised for reasons connected with drug use, ( c) those who the doctor considers are unable to function without heroin, and ( d) those who are prescribed heroin on a steadily reducing basis.


A problem that faces any social investigation concerns the validity of the data obtained. It could be argued that heroin addicts are more likely than other groups to distort information either because the nature of addiction means that it is both difficult to recall past events and the "reality" of truth disappears, or they are motivated to give certain types of answers in order to influence the treatment they are receiving or to justify their present condition. Respondent bias can be reduced by using skilled interviewers and by using well piloted instruments. In this study the interviewers had considerable previous experience with addicts and the interview was developed by the use of extensive piloting. The interviewers also stressed to the subject that the interview was in no way connected with the clinic, would not be seen by the clinic, and could in no way affect the treatment they were receiving.

As one external measure of our interview we abstracted information from clinic case-notes for comparison with our interview. Twenty-five pieces of information were abstracted on every other subject. Table IX gives the items, the percentage coverage of these items in clinic case-notes, and the agreement between our interview and the case-note abstraction for each item as a percentage of the possible agreements.

The large disagreement on socio-economic group of subject and father is due in part to lack of information in case-notes or interview, which in either case made detailed classification difficult. Disagreement on some items (e.g. parents' marital status and father's occupation) may be partly due to a time lag between the time when the information in the case-notes was collected and the time of the abstraction. The disagreement in the two items "age at first drug use" and "age at first heroin use" is interesting in that there is a tendency for the interviewer to have obtained an earlier age than the case-notes. We have no simple explanation for this.

In general there is then a high agreement between case-note and interview material. It may be, however, that both methods are getting the same "wrong" information. The internal consistency of the interview, the subject's behaviour during the interview, and information on subjects from outside the interview, suggest that this is not the case. It is well known among clinic staff that many addicts are not reluctant to talk about their activities, including illicit drug use and crime. The data here are also supported by the findings of Ball [ 1] into the reliability and validity of data obtained from Puerto Rican addicts. It would seem a prerequisite of future studies of addiction in this country to further pursue the problem of validity.



Percentage coverage in case-notes

Percentage agreement between case-notes and interview

100 91
Current work status
93 92
Socio-economic group of subject
95 68
Amount of heroin prescribed (to within 10 mg
100 93
Methadone prescribed
100 100
Other drugs prescribed
98 99
Change in prescription on last month
89 92
Age at first drug use (to within one year)
70 80
Age at first heroin use (to within one year) .
84 79
Date of first attendance at clinic
100 86
Frequency of attendance at clinic
96 88
Ever convicted
82 100
Parents' marital status
90 88
Socio-economic group of father
66 86
Number of brothers and sisters
86 87
Age left school
79 82
Marital status
98 98
Current residence
93 100

Drug use

Although all the subjects in the sample were by definition heroin addicts, few subjects confined their drug use to heroin or other opiates. The high incidence of barbiturate use amongst heroin addicts had been recorded elsewhere [ 11] , and although the way in which barbiturates were used is not known for our sample, the similarity between Mitcheson and co-workers' sample and our own suggests that the purposes they are put to may be similar, i.e. a majority of subjects using them for reasons other than sleeping. A wide range of other drugs were also being used by subjects in the sample, especially amphetamines and cannabis, and only 18 (16%) had used no other drugs than the ones prescribed during the month prior to interview. Poly-drug use is now a familiar phenomenon and it would appear that the clinics are not dealing with addicts to a particular drug, i.e. heroin, but with people who are willing to use a wide range of substances by injection or otherwise.

Few of the subjects in the sample used injection practices that would be medically defined as sterile. Bewley [ 4] and Sapira et al [ 12] have pointed to the connexion between unsterile injection practices and physical complications such as liver damage, septicaemia and abcesses. The high incidence of these complications and the general lack of use of sterile injection techniques in the sample reported here seems to confirm this. The incidence of abscesses reported in the three months prior to interview may be due to the injection of sedatives which, when injected sub-cutaneously, are highly irritant and may produce sterile abscesses. Most clinics do provide patients with the necessary equipment for sterile self-injection. However, whether the supplies given to addicts are sufficient and whether clinics give addicts advice and training in self-injection procedures is not known.

Stable addicts

The literature suggests that there may be one pattern of adjustement to heroin that can be described as "stable". The first Brain report describes such addicts as using a relatively small, non-progressive quantity of the drug of addiction, and leading a "relatively normal life". Schur [ 13] described the stable addict as one who is likely to be working, who is uninvolved with other addicts and who is uninvolved in criminal activities.

In the present sample, the high incidence of illicit drug use means that the number of addicts who are stable in terms of the drugs they use is small. Taking these subjects who report no illicit use of drugs in the previous month and no form of criminal involvement in the previous three months, then there are 14 cases. Of these, nine were in full-time employment, three were housewives and two were unemployed. None of the 14 regularly used sterile injection procedures, although none had shared a syringe or had made up an injection with lavatory water in the previous week. Three reported abscesses in the previous three months. All 14 had somewhere to live. None had visited Piccadilly in the previous week. All claimed to have friends who were not drug users. Eleven had ongoing heterosexual relationships and in three cases the partner was also using heroin. In all other cases the partner was using no drugs at all.

The criteria used above may not be the most important to consider when assessing stability; they are relatively crude criteria. Others in the sample are not necessarily leading chaotic lives, and for many illicit drug use, although occurring in the previous month, was a rarity and may have been the only criminal activity they had entered into.

It is expected that other syndromes of addict behaviour will be found. The factors that contribute to addicts adopting particular patterns of adjustment-the part played by the subjects' past experiences, the clinic policy, the length of use, and the quantity of drug used-need further investigation.


We would like to express our thanks to the following consultants who so kindly allowed us access to patients and made the study possible: Drs.: H. Dale Beckett, T. H. Bewley, P. H. Connell, P. J. Dally, J. Denham, F. P. Haldane, G. B. Oppenheim, J. L. Reed, R. F. Tredgold, J. G. Weir, J. Willis, and Professor J. Hinton, and to their colleagues and staff at the clinics who made our task so much easier. Our thanks too to the subjects who participated in the study, to Carl Emmons who helped with the interviewing, and to Christine Guest and Joyce Hedge for secretarial assistance.



J. C. Ball, Reliability and Validity of Interview Data Obtained from 59 Narcotic Drug Addicts. Amer. J. Sociol ., 1967, 72, 6, 650-654.


T. H. Bewley, O. Ben-Arie and I. P. James, Survey of Heroin Addicts Known to the Home Office. Brit. Med. J . 1968, 1, 725-726.


T. H. Bewley and O. Ben-Arie, Study of 100 Consecutive In-Patients. Brit. Med. J ., 1968, 1, 727-730.


T. H. Bewley, O. Ben-Arie and V. Marks, Relation of Hepatitis to Self-Injection Techniques. Brit. Med. J . 1968, 1, 730-732.


R. de Alarcon, The Spread of Heroin Abuse in a Community. Bull. Narcot ., 1969, XXI, 17-22.


D. V. Hawks, The Epidemiology of Drug dependence: Priorities for Future Research. Paper delivered to the Royal Society of Medicine. 13 Jan. 1970. Published in Bull. Narcot . 1970, XXII, 3, p. 15-24.


H. Heweston and R. Ollendorf, Preliminary Survey of 100 London Heroin and Cocaine Addicts, Brit. J. Addict. Aug. 1964, 109-114.


Interdepartmental Committee on Drug Addiction, 1961. H.M.S.O.


Adele Kosviner, M. Mitcheson, K. Myers, A. Ogborne, G.V. Stimson, J. Zacune and G. Edwards, Heroin Use in a Provincial Town. Lancet, 1968, 1, 1189-1192.


A.S. Meyer, R. Brothan and A.M. Freedman, Continuities and Discontinuities in the Process of Patient Care for Narcotic Addicts. Final Report to Health Research Council, N.Y. Medical College. 1965, (Mimeograph).


M. Mitcheson, J. Davidson, D.V. Hawks, L. Hitcheis and S. Malone, Sedative abuse by Heroin Addicts, Lancet, 1970, 1, 606-607.


J. D. Sapira, D.R. Jasinsky and C.W. Goredetsky, Liver Disease in Narcotic Addicts: II The Role of the Needle. Clin. Pharmacol. and Therap. 1968, 9, 725-739


E. M. Schur, Narcotic Addiction in Britain and America. Tavistock, 1963.


H. B. Spear, The Growth of Heroin Addiction in the United Kingdom, Brit. J. Addict, 1969, 64, 245-255.