Health of cohort of heroin addicts from London clinics: 22 year follow up

BMJ 1996; 312 doi: (Published 08 June 1996) Cite this as: BMJ 1996;312:1458
  1. Clive Tobutt, research nursea,
  2. Edna Oppenheimer, senior lecturera,
  3. Ronaldo Laranjeira, research psychiatrista
  1. a Addiction Research Unit, National Addiction Centre, Institute of Psychiatry, London SE5 8AF
  1. Correspondence to: Mr C Tobutt, Centre for Sexual Health and HIV Studies, Wolfson School of Health Sciences, Thames Valley University (London), London W5 2SU.
  • Accepted 12 January 1996

Data are limited on the long term health consequences of drug misuse in the United Kingdom. In the United States results from longitudinal studies differ over whether heroin addicts misuse alcohol after they stop misusing opiates.1 2 We recently followed up a cohort of heroin addicts after 22 years and found that 43 (34%) of the original sample of 128 had died3; in this report we describe the health of the remainder.

Subjects, methods, and results

The original sample in 1969 was a random one in three sample of (prescribed) heroin injectors attending 13 of the 15 London drug clinics. This group was followed up in 1976-7.4 The tracing methods for the 22 year follow up, begun in 1991, have been described.3 The personal interviews were partly structured and covered prescribed and illicit drug use, social and medical aspects, the severity of alcohol dependence questionnaire, and the general health questionnaire. Of the original 128 in the sample, 63 were interviewed in person, providing the current data. Of the remaining 22, 15 were traced but not interviewed (four of them used opiates and 11 did not) and seven (none of them opiate users) were not traced but there was no evidence that they had died. Statistical methods used were Student's t test and Fisher's exact test.

Table 1 shows reported physical symptoms during the 12 months before the interview and drinking in the 28 days before according to subjects' use of opiates. Of the 23 opiate users, 15 reported physical symptoms of ill health related to drug use. Generally, subjects who did not use opiates reported no acute medical health problems as a result of drug misuse. In both groups alcohol consumption was generally moderate and alcohol dependence was low—11 opiate users and 14 non-users scored 0 on the severity of alcohol dependence questionnaire. Only one subject in each group had high scores (51 and 41); these two subjects also reported heavy drinking at the 1969 index interview and at the 1976 follow up. Both also had high scores on the general health questionnaire (130 and 157). One had a chaotic drug career and the other a history of mental illness throughout the follow up.

Table 1

Characteristics at 22 year follow up of heroin addicts according to current opiate use. Values are numbers (percentages) of subjects unless stated otherwise

View this table:


This follow up indicates that subjects who continued to use opiates have large health problems compared with those who have abstained. Like the drug abuse reporting programme study,5 our results show that heroin addicts who stop misusing opiates do not then misuse alcohol. Only 4% of opiate users and 5% of non-users drank moderately nearly every day (more than 41 g during the previous 22-28 days, there being 10 g of alcohol in one pint (570 ml) of medium strength beer) compared with 32% in the drug abuse reporting programme.5 Only 4% of opiate users drank excessively nearly every day (more than 100 g during the previous 22-28 days) compared with 30% in the drug abuse reporting programme.5 Only two subjects in the cohort were dependent on alcohol, and they both previously drank excessively.

Most of our cohort smoked tobacco every day. We previously found that seven of the 43 subjects who died during follow up had died from natural causes and were not using opiates at the time of death.3 One of these seven died from lung cancer and five from bronchopneumonia.

Despite the problems of retrospective data and of the validity of self reports from addicts, we believe that the data from these face to face interviews are accurate, given confirmatory information from the Home Office and general practice and hospital records. That this cohort drank alcohol only moderately but continued to smoke tobacco is of concern for drug treatment clinics and general practitioners. We might expect to find tobacco related disease in a future follow up.

We thank Colin Taylor and Paul Griffiths for their help.


  • Funding Medical Research Council programme grant held by Griffith Edwards, National Addiction Centre.

  • Conflict of interest None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract