PSA screening leads to overdiagnosis, study says

BMJ 2002; 325 doi: (Published 13 July 2002) Cite this as: BMJ 2002;325:61
  1. Deborah Josefson
  1. Nebraska

    Routine screening for prostate cancer using the prostate specific antigen (PSA) test, is responsible for the observed increased epidemiological incidence of the disease, says a new report (Journal of the National Cancer Institute 2002;94:981-90).

    Moreover, PSA screening causes overdiagnosis rates of the disease in about 29% of white men and 44% of black men, so many of those whose cancers were diagnosed through PSA screening would likely have died of something other than prostate cancer.

    As prostate cancer treatment itself causes significant morbidity, patients with abnormal PSA results should carefully weigh their options before proceeding to radical treatment, the study suggests.

    Moreover, most of the prostate cancers detected via PSA testing in recent years would have presented clinically, the investigators assert.

    Dr Ruth Etzioni and colleagues from the Fred Hutchinson Cancer Research Center in Seattle, Washington, along with Eric Feuer and others from the National Cancer Institute in Bethesda, Maryland, reached these conclusions by comparing incidence rates of prostate cancer since the introduction of PSA screening with a computer based model that predicted incidence of prostate cancer in the absence of such screening.

    The computer model followed a hypothetical cohort of two million men for 10 years who were aged 60-84 in 1988 and calculated all cause mortality for this cohort.

    Rates of screen detected prostate cancer were defined as the number of men diagnosed with prostate cancer within three months of having had a PSA test divided by the number of men who had had at least one PSA screen in that year. For purposes of the study, overdiagnosis was defined as the detection of prostate cancer via PSA testing that would not otherwise have been detected in the patient's lifetime.

    Before the advent of PSA screening the lifetime probability of a prostate cancer diagnosis was 9% for both black and white men. Historical postmortem data show that prostate cancer is detected in 36% of white and 28% of black men.

    Applying their estimates of screen detected overdiagnosis rates, the researchers concluded that, at most, the PSA screen detects 15% (in white men) or 37% (in black men) of latent tumours that would otherwise have shown up only at postmortem examination. Thus, most of the tumours picked up by PSA screening would have eventually presented clinically, (85% of those in whites and 63% in blacks), the researchers concluded.

    Embedded Image

    General Norman Schwarzkopf, who had prostate cancer, is a keen proponent of PSA testing


    View Abstract

    Log in

    Log in through your institution


    * For online subscription