The additional healthcare needs of people with learning disabilities have been long established and the findings of a confidential Inquiry suggested that people with learning disabilities are two and a half times more likely to die before the age of 50.
Directed Enhanced Scheme
In 2008, the Learning Disability Directed Enhanced Scheme (LD-DES) was introduced by NHS England to incentivise Primary care teams to carry out annual health checks for people with learning disabilities.
Early reviews of health checks suggested that they were effective in identifying unidentified conditions and led to targeted actions to address health needs (Robertson et al 2011)
The researchers in this study were interested in looking at the effects of the first 3 years of the scheme by comparing the numbers of health checks, assessments and investigations and diagnoses of common health conditions recorded in those practices that had opted into the scheme, (incentivised) as opposed to in practices that had not (non- incentivised)
What they did was to carry out a longitudinal cohort study (a study that involves repeated observations of the same variables over long periods of time.) They used data from ‘The Health Improvement Network’ database (THIN) a clinical primary-care database of general practices that have opted in. The data have been found to be representative of the UK population. Data come from 10 million patients registered with 578 practices since 1988
They developed a set of inclusion criteria for practices and patients to identify the study population, which enabled them to identify GP practices who had been incentivised through the Directed Enhanced Service (DES) – Learning Disabilities (introduced in England in 2008/9) and those that had not. They were then able to compare data from incentivised and non-incentivised practices.
There are a number of issues with the identification of patients with learning disabilities in General Practice. For example, specific genetic conditions associated with learning disabilities may be coded as that condition, and therefore not identifying the person has having a learning disability.
Local authority thresholds for support also may exclude some people from their learning disability registers, meaning the person may not be identified on GP lists as having a learning disability.
Working with the lists, they carried out a logistic regression analysis to look at the association between practice-level characteristics and opting into the incentivised scheme for all active patients.
The practice characteristics they looked at included
- number of patients
- proportion of men
- patients older than 60
- Townsend deprivation score (measure of material deprivation within a population)
- patients with learning disability
- former strategic health authority
They then assessed data for three years (2009-11) looking at data for patients from incentivised practices with at least one health check in the 3 years, looking at the recording of health assessments completed, interventions offered, and newly identified common health conditions and compared the same data with patients in the non-incentivised practices.
They also separated data from patients with a learning disability into attendees or non-attendees and did a multivariate logistic regression analysis to test the association between attending a health check and strategic health authority, sex, age, Townsend deprivation quintile, carer type, and comorbidity.
Finally they compared data for patients with identified disorders associated with learning disability, but who had not been given a QOF-ID code with patients with relevant QOF-ID codes to look at the proportions offered health checks as they wanted to test whether all learning disability-related activities were more frequent in practices opting into the Learning Disability- Directed Enhanced Service (LD-DES).
- There were 222 incentivised practices with 8,692 registered patients with a learning disability and 48 non-incentivised practices with 918 learning disabled patients identified in the study.
- In terms of demographic data, they found that 10% of patients from incentivised practices were in the London strategic health authority, compared with only 3% from non-incentivised practices.
- They found that the total number of patients with learning disabilities in individual practices was the only variable significantly associated with opting into the LD-DES scheme.
- 88.4% of patients registered with incentivised practices had received at least one health check during the 3-year study period
- They found some tests were more likely to be carried out in incentivised practices, for example blood tests to measure haemoglobin and total cholesterol concentrations in patients with ID.
- Also, if someone was a patient of an incentivised practice they were more likely to be offered general health status checks, specific health assessments for hearing or vision, medication reviews, recorded health action plans, and secondary referrals.
- The rate of newly identified health conditions were higher for patients in incentivised practices, including rates of gastrointestinal and thyroid disorders, constipation, and being underweight or obese.
- Interestingly, the rate of new cases of epilepsy and diabetes was slightly higher in non-incentivised practices, but the differences were not statistically significant.
- Patients not attending health checks were found to be younger and to live in more deprived neighbourhoods.
The researchers had identified over 2000 patients of incentivised practices with disorders associated with learning disabilities, (e.g Down syndrome or autism). 60% had QOF-ID codes recorded, but 800 (40%) did not. If they did not have a QOF-ID code, they were not offered an annual health check in either incentivised or non-incentivised practices.
Conclusion and Comment
This is a large scale study of the Directed Enhanced Service which aims to ensure that every adult with a learning disability is offered an annual health check. The data used came from a reliable source and has been shown in studies elsewhere to be representative of patients throughout England.
The findings suggest that those practices opted into the scheme had increased rates of general and specific health assessments of their patients with learning disabilities and so an increased likelihood of identifying new co-morbidities.
One of the clear policy outcomes associated with the scheme is to ensure that people with learning disabilities have health action plans to respond to their additional healthcare needs, and the findings suggest that the incentivised practices were more likely to offer such action plans and to refer their patients with learning disabilities to secondary care.
Previous reviews of the impact of health checks have suggested that they were effective in identifying unidentified conditions and led to targeted actions to address health needs and this study adds to that evidence base. The authors point to the importance of this finding in the face of evidence from the recently published Confidential Inquiry report which showed that many premature deaths of people with learning disabilities were from potentially preventable causes, amenable to good quality health care interventions.
One of the continuing concerns about the scheme, and supported by the findings of this study is the variability of uptake between English regions, with the Learning Disability Public Health Observatory suggesting that only 52% of patients eligible for incentivised health checks receive them
A concerning finding was also that adults with disorders known to cause learning disabilities were excluded from the health checks because no QOF-ID codes had been recorded. This would require a cross referencing protocol for primary care doctors in the scheme to ensure eligible patients were included.
The authors call for this to be a national-level criterion for participation in the LD-DES.
The authors themselves point out some of the limitation of the study, including
The need to infer which practices had opted in to the LD-DES from data collected on reimbursement
All data were assessed together as it was not possible to look at the quality of health checks within individual practices.
It is likely that in those practices where there are high numbers of patients with a learning disability, there will be doctors with experience and confidence in working with this population. The authors suggest that it may be possible to design a scheme for additional training for those practices where there are fewer learning disabled patients or create a model of support from experienced doctors working across several practices.
Attempts to reduce confounding variables was limited to known confounders and only 5,256 (55%) of 9,610 patients had complete data for the 3-year period.
- Primary care health checks for people with learning disabilities were associated with increases in health related activities, identification of important co-morbidities and referrals to secondary care.
- 40% (811 of 2,034) of patients with specific syndromes did not have QOF-ID codes recorded were not offered a health check.
- Provision and uptake of health checks continues to be variable and to reduce this variability, more work needs to be done on understanding the contributing factors
Assessment of an incentivised scheme to provide annual health checks in primary care for adults with intellectual disability: a longitudinal cohort study, Marta Buszewicz, Catherine Welch, Laura Horsfall, Irwin Nazareth, David Osborn, Angela Hassiotis, Gyles Glover , Umesh Chauhan , Matthew Hoghton, Sally-Ann Cooper, Gwen Moulster, Rosalyn Hithersay, Rachael Hunter, Pauline Heslop, Ken Courtenay, André Strydom in The Lancet Psychiatry, 1, 7, 522 – 530, [abstract]
The impact of health checks for people with intellectual disabilities: a systematic review of evidence, Robertson J et al., in Journal of Intellectual Disability Research, 55: 1009–1019 [abstract]
Conﬁdential Inquiry into premature deaths of people with learning disabilities (CIPOLD): Final report (PDF). Norah Fry Research Centre, University of Bristol, March 2013puf
6 out of 10 GP surgeries are signed to DES but 40% of pwld get a health check – England http://t.co/WswkjpMz6f
RT @LearningDisElf: BUT 40% of patients with learning disabilities did not get a health check http://t.co/ZfUXaE7ov5 (2/2)
RT @LearningDisElf: .@mhoghton @ld_CI @heslop_pauline @ihal_talk Pls RT and comment on our DES blog today http://t.co/ZfUXaE7ov5
RT @LearningDisElf: Six out of ten GP surgeries are signed up to the Directed Enhanced Scheme in England http://t.co/ZfUXaE7ov5 (1/2)
Six out of ten GP signed Directed Enhanced but 40% of patients learning disabilities did not get health check http://t.co/pczCb7GNHk
Don’t miss: @LancetPsych study finds that 40% of patients with learning disabilities did not get a health check http://t.co/ZfUXaE7ov5
These findings are in line with the recently released total uptake of annual health checks in England in 2013-4. 94,647 people received a learning disability health check in 2013 to 2014, an increase of 2,318 (3%) on the 2011 to 2012 figure. The number on GP learning disabilities (QOF) registers was 214,352, a rise of 4%. The coverage (defined as the number receiving checks as a percentage of the number on GP learning disabilities (QOF) registers) fell by 1 percentage point, from 45% to 44%. http://www.improvinghealthandlives.org.uk/securefiles/150110_1159//Learning%20disability%20health%20checks%20report%202014.pdf
Since the start of the DES in England the median age at death of people identifiable as having learning disabilities in death certificates has risen steadily from 2008 to 2012 from 54 to 58 years http://www.improvinghealthandlives.org.uk/securefiles/150110_1202//People%20with%20learning%20disabilities%20in%20England%202013.pdf
Six out of ten GP surgeries are signed up to the Directed Enhanced Scheme in England, but 40% of patients with… http://t.co/ok5D2tuLdw
The continuing variability between regions is shocking. I know NHSE are running a workshop on health checks in the spring. Could one action be how to tackle those areas/practices that are not incentivised?
I think a better understanding of the reasons why there is such variability could really help to focus that discussion and identify some actions that are likely to be successful