A pilot study of a cervical cancer screening service was carried out at a major teaching hospital in Perth. The service, for women inpatients aged 20 to 69 years, was staffed by a women's health nurse. The effect of providing the service (service) was compared with giving a leaflet on Pap smears to eligible women (education) and with no intervention (control). Of 517 women in the service group, 184 (36 per cent) needed a Pap smear and were well enough to be offered screening; only 29 of 184 (16 per cent) refused and 132 of 184 (72 per cent) were screened. Of those screened, 29 per cent had never had a Pap smear. Information on women in the education and control groups was obtained by mailed questionnaire. Of the eligible women in the service group, 72 per cent accepted screening in hospital, but only 24 per cent of eligible women in the education group and 20 per cent in the control group reported having a Pap smear in the four months since leaving hospital. The service group showed a very large effect relative to the control group (odds ratio (OR) 17.71, 95 per cent confidence interval (CI) 10.05 to 31.22), but there was no significant difference between the education and control groups. Other significant variables in the logistic regression model were age, marital status, and sex of the woman's general practitioner. The effect of offering the service was greater for women over 50 (OR 51.51, CI 19.01 to 139.60) A hospital‐based cervical screening service provides an important opportunity for screening women who are not being reached by other services. (Aust J Public Health 1995; 19: 288–93)
