Harm reduction measures and injecting inside prison versus mandatory drugs testing: results of a cross sectional anonymous questionnaire survey
BMJ 1997; 315 doi: https://doi.org/10.1136/bmj.315.7099.21 (Published 05 July 1997) Cite this as: BMJ 1997;315:21
- A Graham Bird, consultant immunologista,
- Sheila M Gore, senior statisticianb (gore{at}mrc-bsu.cam.ac.uk),
- Sharon J Hutchinson, statisticianc,
- Stephanie C Lewis, statisticianc,
- Sheila Cameron, clinical virologistd,
- Sheila Burns on behalf of the European Commission Network on HIV infection and hepatitis in prison, clinical virologiste
- a Churchill Hospital, Oxford OX3 7LJ
- b MRC Biostatistics Unit, Cambridge CB2 2SR
- c MRC-BIAS, Edinburgh EH9 3JN
- d Regional Virus Laboratory, Glasgow G20 9NB
- e Regional Virus Laboratory, Edinburgh EH10 5SB
- Correspondence to: Dr Gore sheila.
- Accepted 25 April 1997
Abstract
Objectives: (a) To determine both the frequency of injecting inside prison and use of sterilising tablets to clean needles in the previous four weeks; (b) to assess the efficiency of random mandatory drugs testing at detecting prisoners who inject heroin inside prison; (c) to determine the percentage of prisoners who had been offered vaccination against hepatitis B.
Design: Cross sectional willing anonymous salivary HIV surveillance linked to a self completion risk factor questionnaire.
Setting: Lowmoss prison, Glasgow, and Aberdeen prison on 11 and 30 October 1996.
Subjects: 293 (94%) of all 312 inmates at Lowmoss and 146 (93%) of all 157 at Aberdeen, resulting in 286 and 143 valid questionnaires.
Main outcome measures: Frequency of injecting inside prison in the previous four weeks by injector inmates who had been in prison for at least four weeks.
Results: 116 (41%) Lowmoss and 53 (37%) Aberdeen prisoners had a history of injecting drug use but only 4% of inmates (17/395; 95% confidence interval 2% to 6%) had ever been offered vaccination against hepatitis B. 42 Lowmoss prisoners (estimated 207 injections and 258 uses of sterilising tablets) and 31 Aberdeen prisoners (229 injections, 221 uses) had injected inside prison in the previous four weeks. The prisons together held 112 injector inmates who had been in prison for more than four weeks, of whom 57 (51%; 42% to 60%) had injected in prison in the past four weeks; their estimated mean number of injections was 6.0 (SD 5.7). Prisoners injecting heroin six times in four weeks will test positive in random mandatory drugs testing on at most 18 days out of 28.
Conclusions: Sterilising tablets and hepatitis B vaccination should be offered to all prisoners. Random mandatory drugs testing seriously underestimates injector inmates' harm reduction needs.
Key messages
Half of injector inmates of two Scottish prisons who had been in prison for more than four weeks had injected in the previous four weeks—an average of six times
Injector inmates used sterilising tablets to clean injecting equipment as often as they injected
Only 4% of inmates had ever been offered vaccination against hepatitis B
Vaccination against hepatitis B and sterilising tablets are prisoners' rights
Random mandatory drugs testing is likely to detect only one third to two thirds of heroin injectors in prison and so seriously underestimates injector inmates' drug reduction needs
Introduction
Willing anonymous salivary HIV (WASH) surveillance in Scottish prisons provided consistent estimates in 1991-5 of the prevalence of risk behaviours inside and outside prisons, of their geographic and sentencing correlates, and of the prevalence of HIV among injector inmates.1 2 3 4 Two studies in 1996 were important for different reasons5: the study in Lowmoss prison, Glasgow, monitored changes since willing anonymous salivary HIV surveillance in 1994 at Barlinnie prison, also in Glasgow; the study in Aberdeen prison gave the first insight to injector and HIV prevalence in north east Scotland at a time of concern about the availability of cheap heroin.6
Questions were added in 1996 to evaluate prisoners' access to harm reduction measures, including hepatitis B vaccination. Sterilising tablets have been available to all Scottish prisoners since December 1993 for purposes including the cleaning of needles and works. But are they being used for the intended harm reduction purpose? Random mandatory drugs testing has been challenged as “a means of gathering information.”7 8 We used volunteered information on the frequency of injecting inside prison in the previous four weeks to estimate the likely efficiency of random mandatory drugs testing at detecting inmates who inject class A drugs such as heroin inside prison. So far as we know this is the first such study.
Methods
Willing anonymous salivary HIV surveillance was conducted with ethical approval and as described1 2 3 9 by external teams of volunteers at Lowmoss prison on 11 October and at Aberdeen prison on 30 October 1996. Saliva samples were tested for HIV antibodies at the regional virus laboratories in Glasgow and Edinburgh respectively. On the surveillance days AGB briefed prisoners about why the survey was being performed, and explained the linkage of questionnaire and saliva sample by sealed number pair (chosen at random by the prisoners)2 and that the survey and research team were unconnected with random mandatory drugs testing (due to be introduced to both prisons in 1997).
Results
The participation rate was 94% (293 of 312 prisoners) at Lowmoss and 93% (146/157) at Aberdeen (table 1). Two Lowmoss prisoners (both injectors) tested positive for HIV antibody. HIV prevalence was 0.7% overall and 1.7% (2/116) for injector inmates. At Aberdeen one saliva sample was insufficient and two prisoners (both non-injectors and heterosexual, one known to the prison's medical service) were HIV positive; HIV prevalence was 1.4% (2/145) overall but nil for injector inmates.
Composition of study groups in HM prisons Lowmoss and Aberdeen
Table 2 gives prisoners' answers to questions about injecting, use of sterilising tablets, and uptake of hepatitis B vaccination for all 286 Lowmoss and 143 Aberdeen participants whose questionnaires passed logical checks and, separately, for the 116 and 53 injecting drug users (that is, people with a history of injecting drug use).
Results of questionnaire survey at Lowmoss and Aberdeen. Figures are numbers (percentages) of prisoners
Injecting drug users and injecting inside prison
Forty five per cent (52/116) of Lowmoss's injecting drug users began their injecting career in 1989 or later compared with 77% (41/53) of Aberdeen prison's injector inmates, 19 of whom had started injecting very recently (in 1995 or later). Fifty four per cent (61/113) of Lowmoss's injecting drug users had injected on the day of entering prison or the day before, as had 52% (27/52) of injecting drug users at Aberdeen. Fifty seven per cent (66/116) of Lowmoss's injecting drug users had ever injected inside prison but only five (4%) had started to inject inside; at Aberdeen these figures were 74% (39/53; P<0.03) and 19% (10/52; P<0.01).
Forty two Lowmoss prisoners (15% of all inmates; 37% of injector inmates) and 31 Aberdeen prisoners (22% of all inmates; 58% of injector inmates) had injected in prison in the previous four weeks. The prisons together held 112 injector inmates who had been in prison for more than four weeks, of whom 57 (51%) had injected in prison in the previous four weeks as follows: once (10 inmates), two to five times (24), 6-10 times (17), 11-20 times (5), 21-50 times (1); their estimated mean number of injections in the four weeks was 6.0 (SD 5.7).
Sterilising tablets and hepatitis B vaccination
Injector inmates' answers to questions about both injecting in prison and the use of sterilising tablets to clean injecting equipment in the past four weeks were broadly concordant (Lowmoss: estimated 207 injections and 258 uses of sterilising tablets to clean needles or works; Aberdeen: estimated 229 injections and 221 uses of sterilising tablets) (table 3). In both prisons local arrangements satisfactorily allowed prisoners to access sterilising tablets for the harm reduction purpose intended.
Frequencies of injecting and of using sterilising tablets among injector inmates at Lowmoss and Aberdeen
Only 4% of Lowmoss's inmates (11/262) and 5% of inmates of Aberdeen prison (6/133) had ever been offered vaccination against hepatitis B. The low reported offer rate is the more surprising because 43% of convicted prisoners in Lowmoss and Aberdeen prisons (169/392) had been sentenced for more than six months and is disappointing in view of the high rates of clinical hepatitis reported by injector inmates (20%; 34/167; table 2).
Discussion
Results at Aberdeen prison indicated that local injecting drug use was as prevalent but more recent than in Glasgow. Injecting behaviour in Aberdeen prison mirrored that in nearby Perth prison10: 74% of Aberdeen's injector inmates had ever injected inside prison (Perth 85%) and 19% had started to inject inside prison (Perth 31%). Three fifths of injecting drug users in Aberdeen prison first injected in 1992 or later and 43% were under 26 years of age.
Use of sterilising tablets
In contrast with England and Wales,11 12 the Scottish Prison Service has provided sterilising tablets in accordance with the World Health Organisation's recommendations13 since December 1993.14 Prisoners regularly used these tablets to clean injecting equipment.
Hepatitis B immunisation
In contrast with almost universal acceptance by prison officers, at most 5% of the inmates studied had ever been offered immunisation against hepatitis B. Similarly low rates were discovered in audit studies in Oxford and Anglia prisons (J Cassidy, personal communication) and in over 3500 prisoners in Victoria, Australia,15 where only 5% of non-immune prisoners had been immunised. Other studies suggest that low awareness16 and unwillingness (M Rotliy, personal communication) contribute to low immunisation uptake by injecting drug users. Specific resources need to be allocated to prison medical services, general practitioners, and drug treatment centres for universal offering of hepatitis B immunisation to prisoners and former prisoners as routine.17 18
Frequency of injecting inside prison: implications for random drugs testing
The combined data showed that 51% (57/112) of injectors who had been in prison for more than four weeks had injected in the past four weeks while inside. Their mean number of injections was 6.0 (SD 5.7). If we assume that the substance injected remained in the urine for three days (as occurs with heroin), then these prisoners would be liable to have a positive result in random mandatory drugs tests on a maximum of 18 days out of 28. If, however, random mandatory drugs testing did not operate at weekends, as in England and Wales, and prisoners could organise their injecting accordingly (for Friday evenings and one Tuesday and one Wednesday evening, say), then they may test positive on many fewer days—for example, on (4 Mondays + (Wednesday + Thursday + Friday) + (Thursday + Friday))=9 days out of 28. On these assumptions we would expect only two thirds to one third of prisoners who are injecting heroin inside prison to test positive in random mandatory drugs tests.
Random mandatory drugs testing is therefore likely seriously to underestimate prisoners' injection related drug use problems. Underestimation will entail underresourcing of these and other prisons in respect of the healthcare and drug reduction needs of their injector inmates.
Obligations to injector inmates
The current extent of injecting among inmates of our prisons demands that we think again about society's obligation to these addicted people. As they continue to take class A drugs by injection inside prison, we must enable them to do so more safely. This requires that ready access to sterilising tablets should be extended to all prisoners in Britain, not just to those in Scotland. It also requires that all prisoners who could benefit from substitute prescribing should receive it, the primary goal being to help them stop injecting; drug reduction is a secondary objective. If sterilising tablets are used suboptimally they may not protect absolutely against bloodborne virus transmission; hence prisoners, like other citizens in Britain, need access to various harm reduction measures. Some are denied methadone, all are denied needle exchange. Prison needle exchange has been pioneered in Switzerland,19 is being evaluated in Germany, and is under consideration in Canada.20
If the current limited access to harm reduction measures is perpetuated it represents a serious gulf between the standards of health care and public health available to the same individuals in prison and outside. Prison medical service policy promotes equality but is short on delivery.21
Acknowledgments
We thank HM prison governors Mr Bill Middleton (Lowmoss) and Mr John Bywalec (Aberdeen) and their officers for welcoming WASH surveillance, for the efficiency of the escort, and for domestic arrangements for the volunteer team; Mrs Linda McDonald and Mrs Karen Wilson, and Mr Tom Shaw, for overseeing saliva samples at the regional virus laboratories in Glasgow and Edinburgh; and the prisoners at Lowmoss and Aberdeen for their participation. We also thank our volunteer teams, both local and from the European Commission Network for HIV Studies in Prisons: Dr Gwen Allardice, Ms Margaret Beveridge, Dr Graham Bird, Mrs Pat Bolam, Mr Donald Cameron, Ms Amanda Cardy, Ms Jan Cassidy, Ms Leonie Craig, Mrs Shona Donald, Ms Laura Ewen, Dr Sheila Gore, Ms Anne Greenhill, Mr Harry van Haastrecht, Mrs Vivien Herring, Dr Richard Herriot, Ms Sharon Hutchinson, Ms Jillian Ireland, Dr Emma Jandolo, Dr Kerstin Kall, Ms Stephanie Lewis, Ms Helena Liddell, Mr George Macdonald, Mrs Joy Macdonald, Mr Bryce MacGregor, Dr Pam Molyneux, Ms Sarah Morrison, Mr Dougal Quin, Mr Tony O'Reilly, Dr Michel Rotily, Ms Sarah Sutherland, Ms Jenny Wallin, Dr Karen Weilandt, Mr Andrew Weild. Neither the European Commission nor any person acting on behalf of the European Commission is responsible for the use which might be made of the data presented.
Funding: Medical Research Council (grant G9102632) and the European Commission (grant DGV, SOC-95-202181-05F02).
Conflict of interest: None.