Management of female prisoners with abnormal cervical cytology

BMJ 1994; 308 doi: (Published 28 May 1994) Cite this as: BMJ 1994;308:1412
  1. G P Downey,
  2. G Gabriel,
  3. A R S Deery,
  4. J Crow,
  5. P G Walker
  1. Royal Free Hospital and School of Medicine, London NW3 2QG
  1. Correspondence to: Mr. Walker.

    Women prisoners are at higher risk of developing cervical neoplasia and cervical cancer than is the general population.1,2 Her Majesty's Prison Holloway offers all women a health check on admission, which includes screening for sexually transmitted diseases and abnormal cervical cytology. Participation is voluntary. Most of the women in the prison are on remand and stay for short times; a smaller number of sentenced women remain in the prison for longer. The time for treatment and investigation is therefore limited. We studied the reasons for referral of the women with smears suggestive of moderate to severe dyskaryosis, the treatment advised, and the default rate and considered an alternative management.

    Subjects, methods, and results

    Between June 1986 and March 1991, 270 women with abnormal results on cervical screening were referred for colposcopy, which was done at the prison by a visiting gynaecologist. Treatment was carried out at a nearby hospital on an inpatient or outpatient basis. Often the women were released after court appearances before adequate assessment and treatment, and 64 women were released before attending for colposcopy.

    Each of the 206 women who attending for colposcopy had a brief gynaecological history taken, repeat screening if indicated, and a directed punch biopsy or cervical conisation arranged. The biopsy was done to determine the grade of the lesion and the treatment or follow up required. We assessed the cytological specimens to determine the prevalence of abnormal cervical smears in the screened prison population and compared the results with those from general practices in the same area. The prevalence of abnormal cervical smears in the prison population was 133 per 1000 samples, double that in the general practice population. The prevalence of high grade disease - that is, cervical intraepithelial neoplasia grade II or above - was 53.1/1000 in the prison population and 18.5/1000 in general practice (P<0.0001; 95% confidence interval 28.4 to 41.0/1000). The prison population was significantly younger (median 27 (range 17-58) years) than the general practice population (median 33.5 (range 17-58) years), which could explain the different incidence of high grade disease (P<0.0001) (table).

    Abnormal cervical smear results in both the prison and general practice population, 1986-91

    View this table:

    All 206 women were given management advice. Thirty one were advised to have follow up smear tests alone, 12 smear tests and colposcopy, 50 locally destructive treatment, 101 conisation, and one hysterectomy; 11 were pregnant and were recommended to have follow up after delivery. Only 49 women completely complied with the advice. The main reason cited for failure to attend treatment or follow up was discharge from prison, which accounted for 111 cases.

    We re-examined the data to determine the number of women who could have had large loop excision of the transformation zone as both a diagnostic and therapeutic procedure at first presentation as this would have resolved the problem of default. One hundred and fifty one (73%) women were suitable for such treatment based on presenting cytological and colposcopic assessment - that is, the lesion was assessed as grade II or above and was small enough to be removed under local anaesthesia. Forty six (30%) of these women were found to have grade I lesions or less at the final histological examination and thus would have been overtreated. However, the current policy of biopsy and subsequent ablation or conisation resulted in overtreatment of 23 women if the same criteria were used. Thus introducing a “see and treat” policy would result in a further 23 (15.2%) “unnecessary” conisations.


    People in prisons are often socially disadvantaged, and a satisfactory follow up programme is difficult to establish because they move around once released and tend to avoid contact with those in authority. Our results suggest that if a see and treat policy were introduced the overtreatment rate would be similar to that in other see and treat studies,3 but most women would be cured of their condition at first presentation. In this group of women, however, the advantages of prompt successful treatment outweigh possible overtreatment because of the high incidence of high grade disease and the high default rate from treatment and follow up.


    View Abstract