The call to the government to prioritise COVID-19 vaccinations for all people with learning disability (LD) has featured in the mainstream press recently. This campaign was in response to evidence published by Public Health England (PHE) over 8 months into the pandemic, that highlighted the disproportionately high COVID-19 mortality rate within the population of people living with LD.1 The advice that followed was that GPs should use their clinical discretion when deciding to prioritise vaccinations for patients with LD on their lists.2 Some clinical commissioning groups committed to prioritising all patients with LD known to them3 before the Joint Committee on Vaccination and Immunisation decided to recommend this approach, recognising that the severity of LD may not be documented in GP records.4 Alongside this press coverage, the lack of fully comprehensive GP registers for patients with LD slipped briefly into the limelight. The challenges of this have been discussed within BJGP before.5,6 Maintaining an LD register is incentivised by the Quality Outcomes Framework (QOF).7 In 2017/2018, NHS England (NHSE) increased the financial incentive for undertaking annual health checks for those on a practice’s LD register.8 The number of patients on this register acts as a target when completing annual health checks and now weighs into identifying patients with LD for priority COVID-19 vaccination.
DO THE NUMBERS ADD UP ON LEARNING DISABILITY REGISTERS?
The QOF recorded national prevalence rates for LD were at 0.5% in the data provided by NHS Digital for 2018/2019 and 2019/2020.9 This percentage corresponds, more or less, to the 244 184 patients on practice-based LD registers published by NHS digital.10 Of note, this figure only covers patients aged ≥14 years on practice-based registers who are eligible for an annual health check, and not all practices reported the number of eligible patients on their register, therefore this figure underestimates the number of people with known LD using primary care services. However, research commissioned by the Department of Health estimated that in 2004, 828 000 adults in England were likely to have LDs, and in 2015 this estimate increased to 1 087 100 people.11 Based on 2019 population data from the Office for National Statistics (ONS), Mencap estimates that 1.2 million people live with LD in England.12 The QOF prevalence rate/LD health check data is less than 25% of this and therefore 75% of people with LD are not on a GP register. Why is our data on the number of patients with LD so poorly representative of the actual prevalence rate?
We know the reason for this may lie partly in issues with terminology and coding. Confusingly, some guidance indicates there may be patients found to be eligible for LD annual health checks (identified via the local authority) who do not qualify for the LD QOF register; highlighting that even at a high level capturing all the patients with LD in one place is problematic.8 Other guidance for practices lists hundreds of SNOMED CT codes that could indicate a patient has LD and thus the likelihood of some of these slipping through the net when trying to collate a comprehensive register is high.13 Then there are the known unknowns — patients who have a known LD that is not coded in an appropriate way — the transition from Lloyd George records and Read codes will factor into this. Changing and inconsistent terminology used across the sector complicates the issue further. Finally, and most worryingly of all, there are the unknown unknowns — those patients who have an LD that has not been recognised.
KNOWN AND UNKNOWN PATIENTS LIVING WITH LEARNING DISABILITY
National Institute for Health and Care Excellence (NICE) Clinical Knowledge Summary (CKS) on LD states that in practice most patients seen with LD in primary care will have already received a specialist diagnosis.14 Unfortunately, it is difficult to evidence this, since there is no comprehensive record of patients with a confirmed specialist diagnosis of LD. Previous estimates have been based on extrapolations from numbers of children with certain types of special educational need (SEN), though this is not interchangeable with a diagnosis of LD. There is a notable decrease in the number of children with these types of SEN versus the number of adults with LD receiving services from health and social care. PHE lists possible reasons for this including rationing of specialist service provision, higher thresholds to access these services, and the impact of LDs being more hidden in non-educational settings.11 Overstretched community LD teams no longer have the resources to routinely engage with practices to validate their registers, and consistent communication between health care, social care, and educational settings is lacking. Patients not known to services are more likely to have mild LD, with less effect on their functioning, though they may still have significant healthcare needs. Sadly, patients with more ‘hidden’ disability might not present for reasons around stigma. Even beyond the primary care setting, there are known unknowns and unknown unknowns when it comes to identifying patients with LD.
Inviting patients with LD to have the COVID-19 vaccine is the tip of the iceberg. They have a diverse range of healthcare needs, and while practices are incentivised to meet those needs, the ‘denominator’ of how many patients to target is out of range by several-fold. The worst-case scenario is that we do not know who these additional patients with LD are, and are oblivious to the additional healthcare needs they may have. A marginally better scenario is that they are known to specialist services but not to their practice teams. Better still is the scenario that they are known to their own practice teams, but if this is the case, the absence of national-level data, let alone a central record for easy identification by the practice, is significant. The reason for a shortfall in data quality is multifactorial and extends beyond general practice, though part of the solution lies in better record keeping. Once again, the pandemic has highlighted the importance of good record-keeping in primary care, and our shortfalls despite aligned financial incentives. SNOMED CT coding has the potential to muddy the waters and impede this task unless we improve our understanding and implementation of it. Hopefully, the media attention on LD registers will encourage stakeholders to recognise the importance of improving data quality in this area,7 which will be the first step in improving health outcomes for this vulnerable population.
Notes
Provenance
Commissioned; externally peer reviewed.
Competing interests
The authors have declared no competing interests.
- © British Journal of General Practice 2021