Intended for healthcare professionals


Screening for prostate cancer in the UK

BMJ 2001; 323 doi: (Published 06 October 2001) Cite this as: BMJ 2001;323:763

Seems to be creeping in by the back door

  1. Jenny L Donovan, professor of social medicine (jenny.donovan{at},
  2. Stephen J Frankel, professor of epidemiology and public health,
  3. David E Neal, professor of surgery,
  4. Freddie C Hamdy, professor of urology
  1. Department of Social Medicine, University of Bristol, Bristol BS8 2PR
  2. School of Surgical Sciences, University of Newcastle upon Tyne, Newcastle upon Tyne NE2 4HH
  3. Division of Clinical Sciences, University of Sheffield, Sheffield S10 2JF

    Screening for prostate cancer is controversial. Findings from systematic and other reviews consistently conclude that there is insufficient evidence to recommend its introduction because of concerns that it may not improve survival or quality of life and may thus cause more harm than good.13 Current government policy in the United Kingdom, expressed in the NHS prostate cancer programme, confirms this view, but adds that “any man considering a PSA [prostate specific antigen] test will be given detailed information to enable him to make an informed choice about whether to proceed with a test or not.”4 This implies that asymptomatic men may have the test if they want, so there is now ambiguity about whether screening is supported and confusion about what this policy means in practice.

    The assumption may be that most men will not want to be tested once they are informed of the uncertainties. In the United States several studies have shown that informed choice can reduce prostate specific antigen testing in some groups by up to one half.57 But this may not apply in the United Kingdom. A systematic review of the use of decision aids has shown that though such aids result in higher levels of knowledge, they have variable effects on the decisions themselves, with reduced preferences for prostate specific antigen testing found in two studies but no effect in two others.8 Further, close inspection of the landmark study5 shows that though prostate specific antigen testing was reduced by half among scheduled clinic attenders who viewed a video, a parallel (rarely quoted) trial found that only 3 out of 206 men attending free prostate specific antigen clinics refused the test after viewing the video.8 Prostate specific antigen testing in the NHS will be free and available to those anxious about prostate cancer. Moreover, in the feasibility study for the ProtecT (Prostate testing for cancer and Treatment) trial, around 90% of men given detailed information about the implications of prostate specific antigen testing and the lack of evidence about treatment consented to a test.

    Thus prostate cancer screening may creep in, even though it does not satisfy the basic criteria for screening.9 Though prostate cancer is clearly an important public health problem (17 210 new cases registered in England and Wales 1993 and 8570 deaths10), many uncertainties surround the acceptability and efficiency of screening tests, the natural history of the disease, and the effectiveness of treatments.

    Serum prostate specific antigen testing is generally considered to be acceptable, but confirmation of cancer requires transrectal ultrasound and biopsy, an uncomfortable procedure requiring prophylactic antibiotics. In terms of efficiency, about 10% of men aged 50-69 will have raised prostate specific antigen levels, but only about a quarter will be confirmed to have prostate cancer, and some tumours will be missed.

    Clinically apparent prostate cancer is primarily a disease of older men, and postmortem studies confirm the aphorism that “more men die with prostate cancer than of it.” Tumours are mostly slow growing and it is not possible to distinguish with certainty between non-fatal lesions, which probably require no treatment, and fast growing tumours that metastasise quickly. Although high grade tumours tend to behave aggressively, there are uncertainties about the effectiveness of early treatment.

    Each of the main treatments—radical prostatectomy, radiation therapy, and monitoring—has risks. Radical treatments offer the potential for cure, but can have serious side effects, including pain, hospitalisation, varying levels of incontinence and impotence, and, rarely, death. With monitoring, men have to live with the knowledge that they have untreated cancer and the risk of progression that in a few cases may be fatal. 1 2

    The new policy is likely to lead to increased detection of prostate cancer among asymptomatic men, who will then be referred to specialist services. NHS directives indicate that they must be seen within two weeks, a speed incompatible with the degree of uncertainty and which will put additional pressure on stretched resources. What treatment should they then receive? Studies have shown that specialists in the United States and United Kingdom overwhelmingly recommend the treatment they themselves deliver—urologists radical prostatectomy and radiation oncologists radical radiotherapy. 11 12 Many of these men will therefore be advised to receive radical treatment, even though this is not based on sound evidence.

    These dilemmas illustrate the difficulties inherent in devising evidence based policy where evidence is weak and the options so disparate.13 The policy outlined in the NHS programme4 is a reflection of the confusion caused by fundamental uncertainties about the effectiveness of early treatment for prostate cancer in the context of consumer led demand for testing. Clear and simple policy in this area will not be possible until evidence shows whether or not screening can improve survival and quality of life. In the meantime the Department of Health should ensure that the information to go with the new policy is accurate, understandable, and acceptable to men who express an interest in prostate specific antigen testing. In our experience in the ProtecT study at least 20 minutes is required to cover the main issues. In addition, it would be helpful for men and general practitioners if strong statements were issued explaining the uncertainties, particularly that early detection and treatment of localised prostate cancer is of unproved benefit and may be harmful.

    It is encouraging that there has been considerable investment in prostate cancer research in the United Kingdom, including the funding by the NHS Health Technology Assessment Programme of the ProtecT study, a randomised trial of radical prostatectomy, radical radiotherapy, and active monitoring for localised prostate cancer. The results of this and other studies should eventually determine the appropriateness of prostate cancer screening, but until then the Department of Health needs to reduce the ambiguity of its policy to ensure that men concerned about prostate cancer can make truly informed choices.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract