Editor's Choice

Tricky thing, choice

BMJ 2009; 338 doi: https://doi.org/10.1136/bmj.b707 (Published 19 February 2009) Cite this as: BMJ 2009;338:b707
  1. Jane Smith, deputy editor, BMJ
  1. jsmith{at}bmj.com

    This week’s cover story is a careful dissection of the information given to women who are about to undergo breast screening with mammography (doi:10.1136/bmj.b86). Three years ago Jørgensen and Gøtzsche surveyed information given to women invited for breast screening in six countries and concluded that all of it was misleading—because it didn’t mention the major harm of misdiagnosis and overtreatment (BMJ 2006;332:538). This week they and their colleagues revisit the issue—by explaining how the leaflet used in the United Kingdom still misleads. Gøtzsche and his colleagues are known critics of breast screening programmes, having long argued that their benefits are small and their harms rarely discussed. But the authors’ point here is that women are not being enabled to make an informed choice—and they offer their own leaflet instead (available on bmj.com).

    Yet if it is hard for a woman to get a clear picture of breast screening, how much harder it is with screening for Down’s syndrome. In this week’s issue Jean Gekas and colleagues assess the cost effectiveness of three screening strategies for Down’s syndrome: an integrated test (where results from both first and second trimester tests are evaluated together to assess risk); sequential screening (first trimester tests determine which second trimester tests are offered); and contingent screening (first trimester tests are used to categorise risk, with women at high and intermediate risk undergoing different further tests). Using different cut-off levels for each strategy, they performed computer simulations with real trial data to compare 19 screening options. They conclude that contingent screening, with a cut-off value of 1 in 9 for high risk, is the optimum strategy (doi:10.1136/bmj.b138).

    Charlotte Ekelund and colleagues meanwhile describe the real world results of a combined screening strategy in Denmark in 2005-6 (doi:10.1136/bmj.a2547). Pregnant women were offered a first trimester risk assessment based on maternal age, nuchal translucency scanning, and a biochemical test; those found to be at high risk were then offered an invasive test. This strategy resulted in a halving of the numbers of infants born with Down’s syndrome and a sharp fall in the number of invasive tests.

    So what are parents to conclude from all this, asks Zarko Alfirevic in his editorial (doi:10.1136/bmj.b140). The language used to describe the evaluation of screening policies is incomprehensible, he says, and someone needs to decide the trade-off between sensitivity and specificity. In the UK the NHS has set a target for 2010 of a detection rate of over 90% with a false positive rate of less than 2%. Yet what parents want in practice are different things—to be able to choose not to be screened, to state that any risk of having an affected baby is unacceptable, or to have a risk assessment to help them decide—and they may want to change their mind at any time. “These decisions are not irrational—they are personal choices that change during pregnancy and between pregnancies.” If the NHS is genuinely to provide patients with choice, argues Alfirevic, it needs not one optimum screening test but a range of options that produce reliable results.


    Cite this as: BMJ 2009;338:b707

    View Abstract