Thirty Times Less Likely: NHS Data Exposes Racial Discrimination in Medical Training Recruitment

Thirty Times Less Likely: NHS Data Exposes Racial Discrimination in Medical Training Recruitment

Black doctors in the United Kingdom stand almost no chance. In certain medical specialties, they are up to thirty times less likely to be offered a training post than white candidates — a disparity so extreme it demands more than policy language. It demands accountability.

The figures, published exclusively by The BMJ and drawn from 2024 NHS England data obtained through a Freedom of Information request, lay bare a structural pattern of racial exclusion embedded in medical training recruitment. Across all specialties combined, Black doctors are four times less likely than white applicants to be offered a training place. The gap widens dramatically in individual disciplines, and the data — compiled by senior human resources professional and independent researcher Sheila Cunliffe — suggests that NHS England may not be complying with its legal obligations to eliminate discrimination in selection processes on the basis of protected characteristics.

The most severe disparity appears in core training 1 for anaesthetics. In 2024, only 10 of 1,158 Black applicants received an offer — a success rate of less than one percent. White applicants, by contrast, received offers at a rate of roughly one in three: 556 of 1,668 candidates. Asian applicants fared better than Black applicants but remained severely disadvantaged, with 111 of 1,696 receiving offers, a rate of approximately 7 percent. The ratio between Black and white offer rates in this single specialty reaches thirty to one.

The pattern holds across multiple disciplines. In general practice, shortlisting rates were broadly similar across ethnicities — a detail that makes the subsequent divergence all the more telling. At the offer stage, 64 percent of white applicants succeeded, against 23 percent of Asian applicants and 20 percent of Black applicants. The selection process, in other words, does not filter by ethnicity at the door; it does so at the threshold that matters most. In core psychiatry, just 5 percent of Black applicants and 9 percent of Asian applicants received offers, compared with 41 percent of white applicants. In obstetrics and gynaecology, white applicants were nearly eleven times more likely than Black candidates to be offered a specialty training place. In acute care common stem emergency medicine, 48 percent of white applicants were offered posts against 7 percent of Black applicants.

Anton Emmanuel, consultant gastroenterologist and head of the Workforce Race Equality Standard for Wales — and a former WRES director at NHS England — describes the data as revealing a pattern that has been “hiding in plain sight” for years, one that the existing WRES monitoring framework has not been equipped to expose. That framing is significant. The WRES was introduced precisely to track racial disparities within NHS institutions. Its failure to surface findings of this magnitude is itself a governance failure, not merely a measurement gap.

Cunliffe’s report also examines the qualitative dimensions of disadvantage. A Black trauma and orthopaedics consultant, speaking to The BMJ on condition of anonymity, described what they called “an unwritten curriculum” — a set of informal advantages accruing to candidates whose social backgrounds, networks, and cultural reference points align with those of assessors. Mentoring relationships, insider knowledge of institutional expectations, and the capacity to perform fluency in a particular professional register can all shape outcomes in ways that formal scoring rubrics do not capture. These dynamics are difficult to quantify and easy to dismiss. They are also, the evidence suggests, consequential.

The dataset carries acknowledged limitations. It does not permit adjustment for whether a doctor received their medical training in the United Kingdom or qualified abroad as an International Medical Graduate, nor does it disaggregate by gender or disability status. Those variables matter, and their absence means the figures cannot be read as a precise causal account. What they do establish, with considerable force, is a consistent and statistically significant pattern of racial disparity at the point of offer — a pattern that cannot be explained away by shortlisting rates alone, since those rates are often comparable across ethnic groups. The discrimination, such as it is, concentrates at the moment of decision.

NHS England has not, as of publication, indicated what corrective measures it intends to take. The legal framework is not ambiguous: public bodies in England bear a statutory duty under the Equality Act 2010 to have due regard to the need to eliminate unlawful discrimination. Cunliffe’s analysis argues that the current recruitment architecture does not meet that standard. The data, now public, makes continued inaction a political and legal choice — not an administrative oversight.