The General Medical Council's own data confirm what Black, Asian and minority ethnic (BAME) doctors and international medical graduates (IMGs) have long alleged: employers refer ethnic minority doctors to the GMC at roughly twice the rate of white doctors, and non-UK graduates at close to three times the rate of UK graduates. Thirty-three years after Aneez Esmail's first BMJ paper documenting racial discrimination in UK medicine, the regulator concedes the problem is “persistent and pernicious.” For candidates sitting the MLA and AKT in 2026, this is not an abstract policy dispute. It sits at the intersection of Good Medical Practice 2024, the Equality Act 2010, the Public Sector Equality Duty, and a regulatory apparatus that has been associated with doctor suicides. Understanding it is a test of professional literacy, not political posture.
The numbers behind the claim of disproportionality
The foundational peer-reviewed evidence is Humphrey, Hickman and Gulliford's BMJ cohort study (2011), analysing 7,526 GMC fitness-to-practise inquiries. Adjusted for sex, years since qualification, specialty and inquiry source, non-UK EU graduates had 2.14 times the odds (95% CI 1.46–3.16) of being referred for adjudication compared with UK graduates, and non-EU graduates had odds of 1.68 (1.31–2.16). Odds of erasure or suspension were 2.16 for EU and 1.48 for non-EU doctors. At every stage - triage, adjudication, sanction - IMGs moved further through the funnel.
That pattern persisted through the decade. The GMC's own Fair to Refer? headline - ethnic minority doctors ~2× more likely to be referred by employers, non-UK graduates ~2.5–3× more likely - has been reiterated in every annual fitness-to-practise report since 2019. BAPIO's widely cited claim that BME doctors are 3–5 times more likely to receive public warnings or restrictions draws on raw GMC sanctions data and reflects cumulative disadvantage across the regulatory funnel rather than adjudication bias alone. Crucially, employer complaints comprise only ~4% of total GMC referrals, but 77% result in full investigation, compared with roughly 9% of patient complaints. Employer referrals are the choke point - which is why their demographic skew matters so much.
The most recent GMC data (the 2024 EDI progress report, and the 2025 update) show movement. The proportion of designated bodies with disproportionate referrals fell from 5.6% in 2016–2020 to 2.9% by 2025 — a ~48% reduction. The absolute referral-rate gap between ethnic minority and white licensed doctors narrowed from 0.28 percentage points (2016–2020) to 0.13 pp (2019–2023). The GMC judges its 2026 target for FtP parity reachable. Its 2031 education target, however, is off-track: the Foundation Year 1 preparedness gap between white and ethnic-minority graduates has widened from 7.8 pp (2019) to 11.7 pp (2023), and undergraduate indices have barely shifted.
Esmail's thirty-year evidence trail
The UK has known about racialised outcomes in medicine longer than most doctors sitting AKT in 2026 have been alive. Professor Aneez Esmail's 1993 BMJ "CV study" with Sam Everington - paired fictitious job applications differing only in surname - found white-named candidates twice as likely to be shortlisted for hospital posts. The pair were briefly arrested by the fraud squad; charges were dropped. Esmail and colleagues then traced differential outcomes through medical school admissions (1995), clinical exam failure rates at Manchester (1998), distinction awards (1998, 2003), and the MRCGP (2000).
His GMC-commissioned 2013 BMJ paper with Chris Roberts analysed the MRCGP Clinical Skills Assessment (CSA) and reported that UK BME graduates had an odds ratio of 3.54 (95% CI 2.70–4.63) of failing the CSA at first attempt versus white UK graduates, while IMGs failed at roughly 15 times the rate. Esmail concluded that the CSA's design was not the primary cause but added: "We cannot exclude subjective bias owing to racial discrimination in the marking of the clinical skills assessment." That finding triggered BAPIO's 2014 judicial review against the RCGP and GMC, heard by Mr Justice Mitting. The court dismissed all three legal claims, ruling the CSA a "proportionate means" of protecting the public, but ordered the College to strengthen equality monitoring - a “moral victory” that cut BAPIO's costs from £288,000 to £50,000.
Esmail's 30-year anniversary commentary in the BMJ (2022, co-authored with Everington) was blunt: “The GMC has been failing doctors and patients for 30 years.” Readers who take the claim seriously have to reckon with a cumulative evidence base spanning recruitment, training, exams, awards, and regulation - one actor, many stages, one consistent pattern.
What Fair to Refer actually found
The 2019 GMC-commissioned report Fair to Refer? Reducing disproportionality in fitness to practise concerns reported to the GMC, by Doyin Atewologun (Cranfield) and Roger Kline (Middlesex), interviewed 262 participants across primary and secondary care. Its power lies in explaining the mechanism, not just quantifying the gap.
The authors identified five structural drivers of disproportionate referrals:
- Doctors treated as "outsiders" rarely receive honest, timely feedback; managers avoid difficult conversations across ethnic difference, so concerns escalate from correctable to career-ending.
- Inadequate induction and ongoing support for doctors new to the UK or NHS leaves IMGs unprepared for NHS-specific norms around consent, safeguarding, and consultation style.
- Isolated or segregated roles - SAS grades, locums, single-handed GPs - lack the mentorship and peer scaffolding that constitute informal "protective factors."
- "In-group/out-group" dynamics allocate sponsorship, informal learning, and benefit of the doubt to insiders; outsiders face stereotyping and heightened scrutiny.
Remote leadership and blame-focused cultures replace learning with formal process, including defensive GMC referrals made partly to avoid accusations of favouritism or racism.
Crucially, Kline and Atewologun rejected "unconscious bias" as an adequate framing, preferring structural analysis of how "risk factors pile on some groups while protective factors accrue to others." The recommendations - comprehensive induction, learning-not-blame cultures, inclusive leadership, a UK-wide delivery mechanism - remain only partially implemented seven years on.
The programme formerly known as TDMR
What the blog brief calls the "Tackling Disadvantage in Medical Regulation" programme is more accurately the umbrella of three GMC workstreams launched in February 2021: the Fairer Employer Referrals (FER) programme, the Regulatory Fairness Review (RFR), and Tackling Disadvantage in Medical Education (TDME). The GMC committed to eliminating disproportionality in employer FtP referrals by 2026 and eradicating disadvantage in education and training by 2031.
Progress is uneven. FER metrics have improved materially and the GMC's headline numbers (the 43% and then 48% reductions in KPI1) look real. The IMG-to-UK specialty exam attainment gap has narrowed from 30.8 percentage points to roughly 22.1. But the Foundation preparedness gap has widened, senior management ethnic-minority representation remains below target, and Charlie Massey himself admitted in 2024 that discrimination in medicine is “persistent and pernicious” and that the system is "moving at two speeds" — visible momentum on employer referrals, stalled progress on education. The 2031 target is, by the regulator's own assessment, off-track.
Differential attainment and the uncomfortable question it poses
The second front is the attainment gap itself. GMC data show postgraduate exam pass rates of roughly 76% for white UK-trained doctors, 63% for UK-trained BME doctors, and 41–43% for non-European IMGs. Katherine Woolf's landmark BMJ 2011 meta-analysis (with Potts and McManus) established that ethnic-minority UK medical graduates had roughly 2.5 times the odds of failing exams compared with white peers - a gap not explained by prior attainment, socioeconomic background, school type, first language, personality, or learning habits.
The intellectually honest question is whether differential attainment explains FtP disproportionality or is another manifestation of it. Tiffin and colleagues (BMC Medicine, 2017) tracked 27,330 IMGs and found PLAB and IELTS scores genuinely predicted later sanctions - i.e. the regulator captures some real risk signal. McManus and Wakeford (BMJ, 2014) argued the PLAB pass mark is set below UK F1-equivalent standards, producing a cohort of IMGs who enter the register with genuinely lower measured performance. Wakeford et al. (2015) showed CSA and MRCP(UK) PACES performance track each other closely by ethnicity, suggesting differences reflect underlying competencies rather than assessment-specific bias. McManus and colleagues (BMC Medical Education, 2013) found no evidence of examiner "own-kind favouritism" across 2,000+ PACES examiners.
Those findings form the steelman counter-argument: disproportionality may reflect real, legitimately regulated variation in performance, mediated through structural factors (specialty concentration in psychiatry, geriatrics and emergency medicine; over-representation in SAS and locum roles; deployment in deprived practices with higher-complexity caseloads) that are confounders rather than proof of institutional bias. Independent audits of the GMC's own adjudication stage have not demonstrated racial bias in decision-making once cases enter the process - a point the GMC repeats in every defence.
The rebuttal, advanced by BAPIO, Kline, and Esmail, is that structural sorting into unsupported roles is itself a product of upstream discrimination - in hiring, sponsorship, training allocation - so confounders are mediators, not exculpatory explanations. Fair to Refer documented that managers bypass informal resolution with BME and IMG doctors precisely to avoid accusations of racism, inflating formal referrals independently of risk. After every known confounder is adjusted, a residual unexplained excess remains. The truth is that both accounts are partially correct, and both sides know it.
The WRES and BAPIO picture
The NHS Workforce Race Equality Standard reinforces the picture across non-medical staff. Indicator 3 - relative likelihood of BME versus white staff entering formal disciplinary processes - improved from 1.56 in 2016 to 1.03 in 2023 before rising back to 1.09 in 2024. More tellingly, 51% of NHS trusts now report BME staff are at least 1.25 times more likely to face discipline, up from 46% in 2023: national averages can narrow while local disparities widen. Career progression, experience of discrimination from managers, and shortlisting outcomes remain stubbornly skewed.
BAPIO, under successive leadership (Ramesh Mehta, JS Bamrah, and from 2024 Geeta Menon - the organisation's first woman Chair), has been the most consistent advocacy voice. It won a 2007 House of Lords challenge against Department of Health guidance that preferred EU over non-EU IMGs; secured interested-party status in the Bawa-Garba Court of Appeal (2018); campaigned successfully for IMGs to receive CCT rather than the inferior CESR-CP certification; and, in its March 2026 Sushruta editorial drawing on FOI data, documented that PLAB Part 2 passers rose 47% between 2022 (8,775) and 2024 (12,902), yet only 2.4% secured training posts by 2024, with 79.9% not connected to a designated body - a stark unemployment proxy. A striking BAPIO analysis of Medical Protection Society FOI data found 93% of overseas-qualified GPs at MPTS hearings without legal representation were suspended or erased, versus 52% of those represented - a procedural inequity with life-altering consequences.
The human cost and the Horsfall legacy
The Horsfall Review (2014) remains the most important document in this debate. Commissioned after a FOI request revealed 96 doctors had died while under GMC investigation between 2004 and 2012, independent consultant Sarndrah Horsfall identified 28 suicides or suspected suicides between 2005 and 2013. Key demographics cut against naive assumptions: 79% held UK primary medical qualifications; 71% were male; 71% had health concerns alongside conduct or performance issues. In 30% of cases where health concerns existed, suicide risk was known to GMC staff but no formal risk assessment occurred. One doctor's suicide note read: "I am extremely stressed and cannot carry on like this. I hold the G.M.C. responsible for making my condition worse with no offer of help."
Horsfall's nine recommendations included case conferencing for health cases, a dedicated senior medical officer to oversee such cases, and treating doctors under investigation as innocent until proven guilty. Louis Appleby's subsequent work for the GMC introduced single points of contact and pause provisions for unwell doctors. Yet the BMA passed a formal no-confidence motion in the GMC and MPTS in July 2020, and the 2018 case of Dr Suresh - a consultant anaesthetist who died by suicide within hours of receiving a GMC letter - prompted a letter-before-action alleging foreseeable breach.
Ongoing data are devastating. A 2023 Medical Protection Society survey of ~200 doctors investigated in the preceding five years found 91% reported stress and anxiety, 78% mental-health damage, and 31% experienced suicidal thoughts during investigation. Brooks and colleagues' 2014 qualitative work in BMJ Open ("You feel you've been bad, not ill") documented how GMC correspondence reads as accusatory and confusing to already-unwell doctors. The Dr Daksha Emson tragedy (2000), in which a psychiatry registrar with bipolar disorder killed her infant daughter and herself after concealing her illness from supervisors owing to stigma, catalysed the NHS Practitioner Health Programme - but the underlying stigma around unwell doctors, and the culture of fear around regulators, has proved far harder to dismantle.
The Bawa-Garba case sharpened a parallel fear. After the GMC successfully appealed the MPTS's 12-month suspension of Dr Hadiza Bawa-Garba to secure erasure in January 2018 - later overturned by the Court of Appeal in August 2018 - surveys found 70% of GPs felt reflective writing was “unsafe” and 81% of junior doctors in the Williams Review sample had changed their reflective style. The Williams Review (June 2018) responded by removing the GMC's right to appeal MPTS decisions and publishing the joint Reflective Practitioner guidance. Thorpe et al. (BMJ Open, 2021) found attitudes towards the GMC became significantly more negative during the erasure period and only partially recovered. The chilling effect on candour, the core engine of patient safety, is well documented and disproportionately affects trainees and IMGs.
Progress, reform, and a controversial new Act
The past two years have produced genuine movement alongside genuine backsliding. Good Medical Practice 2024, in force from 30 January 2024, introduced explicit duties around sexual harassment, bystander intervention, speaking up against discrimination, and doctor wellbeing - the first edition to treat cultures of mistreatment as patient-safety issues. The GMC began regulating Physician Associates and Anaesthesia Associates on 13 December 2024, a decision that the Leng Review (July 2025) subsequently qualified with 18 recommendations (all accepted by government) constraining PA scope and renaming the roles. The BMA's judicial review was dismissed in April 2025, and the Anaesthetists United / Emily Chesterton family challenge was also dismissed; the GMC's unified FtP framework was upheld as lawful.
The most consequential development for IMGs is the Medical Training (Prioritisation) Act 2026 (c.7), which received Royal Assent on 5 March 2026 and commenced the next day. Fast-tracked by Secretary of State Wes Streeting in response to 47,000+ applicants chasing ~9,500 specialty training posts in 2026, the Act prioritises at offer stage: UK medical graduates, then graduates from Ireland, Norway, Iceland, Liechtenstein and Switzerland, then those on qualifying UK training programmes, then specified immigration statuses (British/Irish citizens, those with ILR, EU Settlement Scheme). From 2027, the immigration tier is replaced by a "significant NHS experience" test to be defined by regulations. The DHSC's own equality impact statement concedes the Act disproportionately affects certain ethnic, national and religious groups - an explicit acknowledgement of differential impact written into the founding document.
Reactions split the profession. Royal Colleges, NHS Confederation and HCSA welcomed it. BAPIO, led on this issue by Foundation President Joydeep Grover, opposed it and is convening stakeholders through the British Institute of Human Rights. DAUK flagged lack of clarity on IMGs already in-country. The BMA welcomed it cautiously but deferred its own policy to July 2026 to engage on the definitions. For IMGs currently on Skilled Worker visas whose right to remain is tied to NHS employment, the combination of training-place prioritisation, tightened visa routes, and the 26% year-on-year rise in IMGs relinquishing their GMC licence in 2024 (from 3,869 to 4,880) amounts to a second structural disadvantage layered on top of the FtP issue — one with serious moral and operational implications for NHS workforce planning.
What MLA and AKT candidates actually need to know
For exam purposes the debate crystallises into a small number of anchor concepts you should be able to deploy fluently:
- The GMC's statutory overarching objective under s.1(1A) of the Medical Act 1983 is tripartite: protect public health and safety; promote public confidence in the profession; maintain professional standards. Disputes about FtP disproportionality are disputes about how these three limbs are balanced and whether procedural fairness to registrants falls inside or outside the "public interest."
- The Public Sector Equality Duty (s.149 Equality Act 2010) applies to the GMC as a Schedule 19 body and requires it to have "due regard" to eliminating discrimination and advancing equality of opportunity. Independent evidence (Fair to Refer 2019; the Arora Independent Learning Review by Iqbal Singh and Martin Forde KC, November 2022) documents patterns that raise serious questions about PSED compliance in practice.
- The professional duty of candour (GMC/NMC joint guidance, 2015) differs from the statutory duty (Regulation 20, Health and Social Care Act 2008 Regulations 2014, enforced by CQC) in scope: professional duty covers all harm and applies to individuals; statutory duty applies to organisations and triggers at specified harm thresholds. Post-Francis, both sit alongside reflective practice guidance that the profession still regards as legally unprotected despite GMC/BMA lobbying.
The Medical Practitioners Tribunal Service, a statutory committee of the GMC since 2012, is operationally independent; since the Williams Review 2018 the GMC can no longer appeal its decisions — only the Professional Standards Authority can refer under s.29 of the 2002 Act. Doctors retain 28-day appeal rights to the High Court under s.40 of the Medical Act.
The Arora case is the single most important recent case for exam-oriented candidates to understand. Dr Manjula Arora, a Manchester GP of Indian origin, was suspended for one month by an MPTS panel in May 2022 over whether a Medical Director had "promised" her a laptop or merely "noted her interest." The High Court quashed the determination unopposed. The Singh/Forde review concluded the dishonesty allegation "should not have been taken forward", identified "a regulatory process lacking in fairness; a system in which the stakes seem much higher if you are a black and minority ethnic doctor", and made 29 recommendations - all accepted. It is the test case for what the intersection of low-level dishonesty, race, and regulatory proportionality looks like in practice.
What changes when you hold all of this together
The honest conclusion is that both the disproportionality and the counter-arguments are real, and holding both is the only intellectually defensible position. The GMC does capture genuine risk signal - PLAB scores predict later sanctions, exam performance tracks across independently designed assessments, and adjudication at MPTS has not been shown to be racially biased once cases enter the system. But the entry to the system is skewed at source: employers refer BAME and IMG doctors at roughly twice the rate of comparators, they do so partly because informal resolution is avoided, and the doctors most exposed are those placed into unsupported roles by structural sorting that begins at medical school.
The consequence is a regulatory apparatus that identifies real problems through a filter that systematically over-samples certain demographics - with mental health consequences that have killed dozens of doctors we know about and probably more we do not. The GMC's 2021 targets are approaching partial success on the referral-rate metric and clear failure on the education metric. The 2026 Medical Training (Prioritisation) Act adds a new legislative axis of differential impact at exactly the moment the regulatory one is beginning to yield.
For doctors sitting MLA or AKT in 2026, the learnable content is procedural and ethical: duty of candour, Good Medical Practice 2024, Equality Act 2010, PSED, MPTS structure, reflective practice post-Bawa-Garba, Arora's lessons on cultural curiosity and low-level dishonesty. The deeper professional content is harder and more permanent. The question is not whether to trust the regulator - you are bound to it - but whether the profession will continue to accept statistical disparities of this magnitude as the price of public safety, or whether it will insist, as Esmail and Kline have for three decades, that a system can be both rigorous and fair without trading one for the other. The evidence now sits on the profession's side of that argument. What remains is whether it will act on it before another cohort of IMG and BAME colleagues pays the full price.