Intended for healthcare professionals

Editor's Choice

Examining our failings

BMJ 2022; 377 doi: (Published 23 June 2022) Cite this as: BMJ 2022;377:o1529
  1. Tom Moberly, UK editor
  1. The BMJ
  1. tmoberly{at}
    Follow Tom on Twitter @tommoberly

Every day we fail to adequately recognise, acknowledge, and mitigate our biases. These biases affect how we think as individuals and as a society, what interests us, what questions we ask, how we ask them, and what we do with the answers.

Society’s failure to examine and tackle its biases is made painfully evident in the racism and sexism that so many doctors experience every day. This month a BMA survey heard that 57% of senior doctors from ethnic minority backgrounds had experienced verbal abuse or behaviour targeting their racial, national, or cultural heritage at least once in the past three years (doi:10.1136/bmj.o1515).1 Last week a survey found that over half of senior health leaders from ethnic minority backgrounds had considered leaving the NHS because of their experience of racism at work (doi:10.1136/bmj.o1456).2

And, this week, the BMA unveiled all male candidate lists for the posts of chair of council and treasurer (doi:10.1136/bmj.o1525).3 The shortlists led to criticism over the organisation’s record on promoting equality (doi:10.1136/bmj.o1527),4 at the same time as it sought to highlight progress made since Daphne Romney’s 2019 investigation into sexism and sexual harassment at the association (doi:10.1136/bmj.l6089).5

Failure to mitigate biases also underpins our inability to develop a sound evidence base for many interventions and to use the evidence we do have to inform policy making. For example, Allyah Abbas-Hanif and colleagues highlight how pregnant women are by default excluded from the drug development process (doi:10.1136/bmj-2022-071296).6 “This results in delayed or even absent data on risk-benefit profiles and a dangerous spiral of indecision,” they say. Feng Xie and Ting Zhou show that sponsorship biases in cost effectiveness analyses are “significant, systemic, and present across a range of diseases and study designs” (doi:10.1136/bmj-2021-069573).7

Our biases influence what we do with the answers we uncover, including how findings of studies are interpreted. Hattie Burt and colleagues point out that, even though the link between salt intake and cardiovascular disease is one of the more robust causal associations we know, bias in a handful of studies weakened the imperative for action on salt reduction targets in the UK (doi:10.1136/bmj-2022-070686).8

For the biases that prevent sound evidence from informing policy making, Steven Woolf (doi:10.1136/bmj.o1308) believes we are moving backwards, rather than making progress,9 as Haider Warraich and colleagues (doi:10.1136/bmj-2021-069308) find a widening mortality gap between US counties that vote Republican and those that vote Democrat.10 Woolf argues that the “growing influence of partisan affiliation on policies affecting health” means that it is increasingly important that research examines the impact of party politics on public health.

We have known about these issues for years, and Margaret McCartney writes about her perspective on 18 years of advocating for better ways to tackle conflicts of interest in medicine. “We need a new world order, better regulation, better testing, less pointless bureaucracy,” she says. “Our house remains conflicted as ever.”