For healthcare professionals only


Issues in the management of prisoners infected with HIV-1: the King's College Hospital HIV prison service retrospective cohort study

BMJ 2001; 322 doi: (Published 17 February 2001) Cite this as: BMJ 2001;322:398
  1. Simon Edwards, specialist registrar (SEdwards{at},
  2. Melinda Tenant-Flowers, consultant,
  3. Joseph Buggy, health adviser,
  4. Peter Horne, health adviser,
  5. Nick Hulme, clinic manager,
  6. Philippa Easterbrook, professor,
  7. Chris Taylor, consultant
  1. Department of Genitourinary/HIV Medicine, Caldecot Centre, King's College Hospital, London SE5 9RS
  1. Correspondence to: Simon Edwards
  • Accepted 18 September 2000

Concern has been raised about the quality of health care provided to prisoners in England and Wales. 1 2 The management of prisoners infected with HIV-1 is challenging: a high proportion are injecting drug users, there are issues regarding confidentiality, and administering complex antiretroviral regimens may be difficult in prison. We reviewed our experience of providing specialist HIV care to prisoners between October 1994 and July 1999.

Participants, methods, and results

In October 1994, King's College Hospital was contracted to provide care to male prisoners with HIV-1 and sexually transmitted diseases at Wandsworth and Brixton prisons in south London. Prisoners access the service through self referral or referral by wing officers and prison healthcare workers.

Between October 1994 and July 1999 six prisoners were newly diagnosed as positive for antibodies to HIV-1 and 121 said that they had previously tested positive for antibodies to HIV-1. Of those 121, 75 were confirmed as positive for antibodies to HIV-1 and 25 tested negative for antibodies to HIV-1. Fourteen of the remaining 21 who declined to be tested gave information to support their claim, including their HIV treatment centre. In all cases this information proved to be false. Documented reasons for this subterfuge included the desire for a letter pleading mitigating circumstances in court or a request for food supplements, sedatives, or opioids.

Of 81 patients confirmed as positive for antibodies to HIV-1, 77% (62/81) were white and 16% (13/81) were black-African. The median age at first assessment in prison was 33 (range 23-65) years and the main HIV risk factor recorded was injecting drug use (59%; 48/81). The median CD4 count was 210×106/l (range 4-740×106/l) and a fifth were severely immunosuppressed (CD4<50×106/l). Twenty one (26%) had AIDS, 41 (51%) were coinfected with hepatitis C, and five (6%) also had chronic hepatitis B.

Inmates were reviewed regularly to assess clinical status and adherence to antiretroviral treatment. As expected, they were significantly more likely to keep appointments compared with our hospital outpatient cohort (88% (446/509) v 67% (1098/1645); P<0.001). Reasons for non-attendance included attendance at court or hospital or a legal or social visit (35%; 22/63), transfer to another prison (25%; 16/63), failure to locate prisoner (13%; 8/63), and lack of clinic time (6%; 4/63).

Sixteen of 34 (47%) inmates incarcerated after 1996 who were eligible for antiretroviral treatment according to national guidelines were taking it.3 In comparison, 76% (493/649) of outpatients were already taking or started antiretroviral treatment within six months of their CD4 count dropping below 350×106/l.

Seven of 18 inmates who had not been taking antiretroviral treatment started taking it in prison. The remainder were seen on only a few occasions, which was considered insufficient to initiate and monitor treatment (median=3; range 2-6).

Self reported adherence to antiretroviral treatment exceeded 90%, which compares favourably with the rate reported from a London outpatient cohort.4 Nineteen of 30 (63%) inmates reported occasions when they had not received their medication as prescribed. Reasons included confinement to cell and travel to court, hospital, or another prison. Prescription error and drug unavailability were cited infrequently.

Fifteen (19%) inmates required at least one admission to hospital for a median of seven days (range 3-84). The spectrum of clinical problems included respiratory tract infections, investigation of possible mycobacterium tuberculosis (n=12), treatment of lymphoma or Kaposi's sarcoma (n=2), meningitis (n=2), complications from hepatitis C (n=1), and neuropsychiatric problems (n=1).


Our HIV service is used by a high proportion of severely immunosuppressed prisoners, who present complicated management issues. Almost a quarter of prisoners who claimed to be positive for HIV-1 antibody were not, although the proportion may be higher because a considerable number declined confirmatory testing. We therefore recommend that HIV status be confirmed in all prisoners.

Imprisonment presents an opportunity for inmates to have closely supervised specialist HIV care.5 We identified several logistical problems that had an impact on patient monitoring and adherence to antiretroviral treatment. The provision of services to prisoners who are positive for HIV-1 antibody must be regularly audited to identify obstacles to effective healthcare delivery.


We thank Natalie Ives for assisting with the statistical analysis.

Contributors: SE designed the study, data extraction and analysis, and drafted the paper and organised its final production. MTF, PH, and JB assisted in data extraction. CT conceived the idea, guided the study design, and provided support in preparing the manuscript. NH set up the service contract. PE contributed to the data interpretation and provided support in preparing the manuscript. SE, CT, MTF, PH, and JB delivered the service for which the data are based. All authors commented on drafts of this publication. CT is the guarantor.


  • Funding None.

  • Competing interests None declared.

  • Embedded Image This article is part of the BMJ's randomised controlled trial of open peer review. Documentation relating to the editorial decision making process is available on the BMJ's website


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