Intended for healthcare professionals


Outcomes of screening to prevent cancer: analysis of cumulative incidence of cervical abnormality and modelling of cases and deaths prevented

BMJ 2003; 326 doi: (Published 26 April 2003) Cite this as: BMJ 2003;326:901

This article has a correction. Please see:

  1. A E Raffle, consultant in public health medicine (angela.raffle{at},
  2. B Alden, systems analysta,
  3. M Quinn, directorb,
  4. P J Babb, senior cancer epidemiologistb,
  5. M T Brett, consultant pathologistc
  1. a Avon Health Authority, King Square House, Bristol BS2 8EE
  2. b National Cancer Intelligence Centre, Office for National Statistics, London SW1V 2QQ
  3. c Department of Cellular Pathology, Southmead Hospital, Bristol BS10 5NB
  1. Correspondence to: A E Raffle
  • Accepted 27 February 2003


Objective: To determine the frequency of different outcomes in women participating in cervical screening.

Design: Analysis of screening records from 348 419 women, and modelling of cases of cervical cancer and deaths with and without screening.

Setting: Cervical screening programme in Bristol.

Results: For every 10 000 women screened from 1976 to 1996, 1564 had abnormal cytology, 818 were investigated, and 543 had abnormal histology. One hundred and seventy six had persistent abnormality for two years or more. In the absence of screening 80 women would be expected to develop cancer of the cervix by 2011, of whom 25 would die. With screening 10 of these deaths would be avoided. Comparison of cumulative abnormality rates with numbers expected to develop cancer in the absence of screening suggests that at least 80% of high grade dyskaryosis and of high grade dysplasia would not progress to cancer. The lifetime risk of having abnormal cytology detected could be as high as 40% for women born since 1960.

Conclusions: Screening is labour and resource intensive. It involves treatment for many women not destined to develop invasive cancer. The increased intervention rate for cervical abnormality in England is due to change in practice, not a cohort effect, and is probably the reason for the marked fall in incidence and mortality during the 1990s. For other cancers there is scope for major iatrogenic harm from screening because of invasive tests and treatments.

What is already known on this topic

What is already known on this topic Since the mid-1980s incidence of and mortality from cervical cancer in women born since the 1930s in England and Wales has fallen; screening is the most likely explanation

For each death prevented many women have to be screened and many are treated who would not have developed a problem

What this study adds

What this study adds In the NHS cervical screening programme around 1000 women need to be screened for 35 years to prevent one death

Over 80% of women with high grade cervical intraepithelial neoplasia will not develop invasive cancer, but all need to be treated

For each death prevented, over 150 women have an abnormal result, over 80 are referred for investigation, and over 50 have treatment

Before the 1988 relaunch of screening with strict quality standards, for each death prevented there were 57 000 tests and 1955 women had abnormal results


  • Funding BA received funding from NHS Research and Development Directorate. The guarantor accepts full responsibility for the conduct of the study, had access to the data, and controlled the decision to publish.

  • Competing interests None declared.

  • Accepted 27 February 2003
View Full Text