Guest Essay

We were grateful to have Dr. Kathleen Freeman provide us with a review of a recent HbA1c bias issue in the NHS. Now the principals have responded to her article, giving us more insight into the scope and danger of the issue.

Response to ‘Lessons for Human and Veterinary Laboratories from a Recently Human Laboratory Problem’ by Dr Kathleen Freeman

April 2026
Patient Focussed Laboratory Medicine (PFLM) Group.

We, the Patient Focussed Laboratory Medicine (PFLM) Group, welcome the excellent review from Dr Kathleen Freeman on the HBA1c issue in the UK and how this is a cautionary tale for other analytes spanning both Human and Veterinary pathology. There have been repercussions within the UK following the incident and whilst that particular issue has now largely been resolved, there have been many lessons learned which we are now in the process of implementing. We would like to share what we have done and are doing as our recommendations have international relevance for both Human and Veterinary Pathology.

Background

The positive bias in the use of a particular glycated haemoglobin assay initiated a review by the NHS. Initial concerns were raised by clinicians treating patients and clinicians working in the Laboratories. Evidence showed a change in method performance (including positive bias and varying imprecision) had occurred and that there was concern that this positive bias would lead to increased diagnoses of Diabetes Mellitus and Non-Diabetic Hyperglycaemia. At Luton and Dunstable Hospital >11,000 patients were invited to be re-tested and approximately 75% of patients were eventually retested. A greater than allowable tolerance for the method from the initial result, i.e. initial result erroneous, was identified in 45% of repeat results. This changed the initial diagnosis of Diabetes Mellitus to Non-Diabetic Hyperglycaemia for 21% of patients and a further 21% of patients had their diagnosis changed from Non-Diabetic Hyperglycaemia to Normal. In early Spring 2025, the index hospital laboratory decided to call in NHS-England (NHSE) Getting it Right First Time (GIRFT) Pathology to look at the evidence to understand whether this was a method issue. A Rapid Review was undertaken which identified that the issue was method related, and NHSE GIRFT decided to undertake a deep dive into the issue to understand the root cause, understand the effect on patient management and ensure that the issue had been corrected. The final review will be completed in May 2026, but it has been reported (referenced by Dr Kathleen Freeman) that over 55,000 patients have been recalled from laboratories, and an extensive review has occurred at all laboratories using the IVD Device. 

Patient Focussed Laboratory Medicine

However, in the GIRFT Deep Dive into this incident, it became clear that there were gaps in the UK National Quality Framework, and that there is a risk that these gaps may lead to similar events for other analytes. Therefore, an Expert Advisory Group (EAG) was set up in order to understand the gaps and make recommendations. This EAG is called the Patient Focussed Laboratory Medicine (PFLM) Group. Successful system change requires peer support and co-production of recommendations. To support this principle, we organised a free Webinar to discuss the issues and to map a strategic direction for a system change in a Clinical Quality Management System (Clinical QMS). Well over 1000 people registered for the Webinar with almost 800 participants on the day. The meeting report discussed a Ten Point Plan to improve the Clinical QMS in the UK.

Recommendations

1. Each laboratory should understand the consequences to the patient on the performance of all assays / investigations and understand the impact of out of tolerance results on patients. Each laboratory needs to clearly explain and communicate consequences of the performance of all assays / investigations on the patient from recommendation 1 to clinicians and patients

1.1. All Laboratories work within the Regulatory and Accreditation Frameworks. Legislation is changing, especially on In -House Assays, modifying IVDs (such as Use Case and use of “local” mathematical correction factors, such as Adjusted Calcium, Total Iron Binding Capacity (TIBC) etc). We were concerned that some Laboratory Directors were not fully aware of the clinical effect on patients of unwarranted variation in assay performance and we recommend that Laboratories should include a Patient Focused element in their Clinical Quality Management System.

2. Fully implement Recommendation 13 of the GIRFT Pathology Report.

2.1. The GIRFT National report [https://www.gettingitrightfirsttime.co.uk/wp-content/uploads/2022/05/Pathology-4May22j.pdf] recommended that a National Quality Oversight Board with an escalation path be set up.

2.2. In response to this, a) an NHSE Diagnostic Quality and Safety Committee has been set up; b) MHRA are setting up a Post-Market Surveillance Committee to look at unwarranted variation in IVD performance; c) funding has been made available to re-instate the RCPath EQA Oversight function. These three oversight committees will ensure that a National Clinical QMS is set up allowing for identification, escalation and action on poor performing laboratories/methods.

3. Establish Analytical Performance Specifications for Decision Diagnostics

3.1. APSs have been discussed for many years yet agreement on and implementation of, have not been internationally implemented. In order to have clinical confidence in the results from the laboratory, it is critical that APSs are defined in order to ascertain whether the test performance is acceptable. These should be used when the test is initially placed and in subsequent post market surveillance of the test to ensure that performance has not drifted. APSs have to be realistic. If the APSs are set too wide then a lab may be compliant with the APS but the "acceptable" performance may be clinically unacceptable. Equally, if APSs are set too narrow, with few methods complying, then confidence in APSs will drop. A balance is required linking laboratory/method performance to clinical decisions. Birmingham Quality (a member of the UK NEQAS consortium) have implemented a TEa of 5% for HbA1c and performance can be seen on EQA reports. Weqas are doing the same. PFLM have asked the RCPath to define further APSs for key analytes, and membership of this group includes international members. To prioritise the APSs we are compiling a list of Analytes of Concern (AoC), where we think that there is a risk to patient safety due to unwarranted variation in results produced

4. Association for Laboratory Medicine (LabMed) to lead a review of criteria for the diagnosis of Diabetes Mellitus and NDH

4.1. During the Deep-Dive, it became clear that there is a mismatch between Laboratory Performance and National recommendations. Even when the performance of a test meets the new APSs, there will still be variation around the “magic numbers” used in National Guidance. Examples include the diagnosis of NDH and DM, referral for ultrasound with CA125, treatment thresholds based on bilirubin for neonatal jaundice etc. We have asked LabMed to propose a diagnostic algorithm for NDH and DM that takes into account measurement uncertainty and to include the effect of potential G6PD deficiency in ethnic groups.

5. ISO 15189:2022 accreditation should be focused on risk management of the whole patient pathway

5.1. ISO 15189:2022 embraces the preanalytical, analytical and post analytical elements of laboratory function, as well as both the Clinical and Technical elements of the service. It is therefore essential that Accreditation ensures a balance of all elements of Laboratory service.

6. MHRA introduce formal process for post-market surveillance of IVDs. Laboratories need to increase usage of the MHRA yellow card system.

6.1. The MHRA have a pre-market and post market surveillance framework to ensure that placement and use of drugs is safe. The main feedback mechanism for this is the MHRA Yellow Card system. However very few Laboratories use the Yellow Card system for IVDs, and therefore the MHRA is not getting sufficient evidence for post market surveillance of IVDs. MHRA have agreed to set up a post market surveillance team for IVDs and PFLM, with the learned Societies, will ensure that Laboratory Directors utilize the Yellow card System for IVDs to ensure that MHRA have the formal evidence of poor performance.

7. NHS Supplies to have a greater awareness of the quality infrastructure of IVDs

7.1. NHS Supplies provides excellent guidance on the procurement, financial and contractual elements of placing an IVD in the NHS. However, the Clinical element is often left to local laboratories which is variable and clinical performance is not necessarily contractual. We are working with NHS Supplies to strengthen this, supporting the Laboratory with contractual options if an assay has poor performance

8. LabMed to provide guidance/toolkits on quality processes

8.1. It became clear that there was unwarranted variation in the way Laboratories use IQC and EQA. We saw variation in IQC assignment and acceptability criteria, and with the HbA1c we even saw laboratories defining “acceptable” performance based on the Method Group means and not the results that could be traced to IFCC standards. Very few laboratories used patient means, and this would have picked up the positive bias in HbA1c months earlier. We have therefore asked LabMed to produce a Toolkit on Optimum QA.

9. LabMed to provide further training and education of consultants/laboratory directors

9.1. We have therefore recommended that Master Classes are set up for Laboratories Directors to understand their regulatory and legal responsibilities

10. Corporate members group

10.1. The relationship between the Laboratory Director and the Manufacturer is critical. Whilst this relationship is contractual, there is a risk that “when things go wrong” the actual contract may not be as useful as expected, with the potential of disagreement on what is poor performance and whether Contracts are robust enough to form a framework for mitigation. There is a risk of conflict, which we want to avoid. Having APSs which contractually support will help, but both Laboratories and Manufacturers would wish to avoid conflict. We are therefore working with BIVDA (British In-Vitro Diagnostics Association) to ensure that the Clinical Quality Management System proposed is supported by both and that there is a post market code of conduct to support any unwarranted variation in performance.

We thank Dr Kathleen Freeman for raising this issue and agree with the points raised. The HbA1c issue has focussed the mind on issues that Laboratorians throughout the world for many years. The NHS/GIRFT review will provide confidence to patients that the issue has been investigated and resolved. The work of the PFLM Group, looking at the lessons learned and applying them to other analytes, will ensure that there will be a paradigm shift in the way we interpret quality. At the moment there is a disconnect between Technical and Clinical Quality. We would argue that there is too much focus on the former and not enough on the latter; now is a time for a rebalance. it is not just about how the laboratory produces the result — it always about what the physician does with the result and the effect that the result on the patient’s health and well-being.

Whilst the GIRFT work is focussed on England and PFLM is focussed on the UK, it is clear that the process and potential problems are the same throughout the world. Solutions may vary, but the principals of the solutions remain the same. In addition the principles apply to the patient, irrespective of whether the patient is human or from other areas of the Animal Kingdom, and we agree with Dr Kathleen Freeman that the lessons learned and solutions proposed are equally applicable to Veterinary Pathology.

We are holding a second webinar on Tuesday 12th May 2026. All are welcome to join by registering at https://forms.office.com/e/sY3vd0v3q1

• Martin Myers, Co-Clinical Lead NHSE Getting it Right First Time, Pathology
• Danielle Freedman, Consultant Chemical Pathologist and Deputy Medical Director, Bedfordshire Hospitals NHS Foundation Trust
• Rachel Marrington, Consultant Clinical Scientist and Deputy Director Birmingham Quality (member of the UK NEQAS consortium), University Hospitals Birmingham NHS Foundation Trust
• Finlay MacKenzie, Consultant Clinical Scientist and Director Birmingham Quality (member of the UK NEQAS consortium), University Hospitals Birmingham NHS Foundation Trust
• David Housley, Consultant Clinical Scientist and Clinical Lead for Biochemistry and Immunology, Bedfordshire Hospitals NHS Foundation Trust
  Annette Thomas, Consultant Clinical Scientist and Director Weqas, Cardiff and Vale University Health Board
• Gareth Davies, Principal Clinical Scientist and Deputy Director Weqas, Cardiff and Vale University Health Board