An
article in the British Medical Journal argues that screening for cancer has
never been shown to save lives. The
authors focus on the difference between overall mortality (the death rate from
any cause) and disease specific mortality (the death rate caused specifically by
a disease). They say that while there is
evidence to suggest that screenings reduce disease specific mortality, overall
mortality is either unchanged or negatively affected.
The authors point to, among others, a
review of meta-analyses of cancer screening trials in the International
Journal of Epidemiology, which found that while some trials showed reductions
in disease specific mortality, none showed reductions in overall mortality.
There are two main reasons mentioned for the discrepancy between
disease specific mortality and overall mortality. The first is that studies might be
underpowered to detect a small overall mortality benefit. (Underpowered studies are those that have a
low probability of detecting an effect of practical importance.) The second is that disease specific mortality
reductions may be offset by deaths due to the downstream effects of screening.
Given that the overall goal of a person considering a cancer
screening is to reduce their risk of dying, the possible harmful effects of
screening need to be considered in addition to potential benefits. Screenings can cause harm if they result in false
positives (abnormal results that turn out to be normal) or overdiagnosis (harmless
cancers that never cause symptoms). For
instance, false positive results from prostate cancer screenings contribute to more
than one million prostate biopsies a year, which are linked with serious harm,
including admission to hospital and death.
False positive results affect between 12-13% of all men
who have undergone three or four screening rounds with PSA, and over 60%
of women undergoing screening mammography for a decade or more.
Do you plan to get screened for cancers? Are you worried about false positives,
overdiagnosis of non-harmful cancers, or detection of incidental findings
leading to treatment that you don't need and could cause you harm?
|
Comments rated to be "almost" Good Answers: