The importance of data collection in measuring effectiveness of diabetes care
Aligning clinical practice, policy and research highlighted in Banting Memorial Lecture at conference for diabetes healthcare professionals.
Claire Levy, PR and communications at DRWF, reporting from the Banting Memorial Lecture at the Diabetes UK Professional Conference held recently at the SEC in Glasgow.
Professor Jonathan Valabhji
Clinical Chair in Medicine
Honorary Consultant Diabetologist, Chelsea and Westminster Hospital Campus, Imperial College London
From April 2013 until September 2023, National Clinical Director for Diabetes and Obesity at NHS England
Presenting the Banting Memorial Lecture, Professor Valabhji recognised the importance of granular data collection to support evaluation of everything. He said if you are getting taxpayers money to invest, you must prove that it is effective and cost effective.
- If you have a good policy idea and get the funding for it and implement it, follow the evidence.
- Implement evidence-based policy.
- There were often political pressures to do something in a space for which there was very little evidence, and in which case the principle was pilot and evaluate.
- A term bandied about from around a decade ago was accelerated access to the pipeline from inception of research idea to implementation at scale.
Professor Valabhji highlighted that whatever we need to do and want to do nationally, we must in this day and age, consider the availability of workforce capacity, resilience and protecting workloads of our healthcare professionals across the NHS.
He saw the importance of a multidisciplinary approach, for example, all of the relevant clinicians in the same place at the same time, reduced amputation rates considerably.
As National Clinical Director, he wanted to improve access across England to such multidisciplinary settings. Over a period of eight years, he achieved an additional £50 million of investment across England to improve access to multidisciplinary foot care.
The major amputation incidence in people with diabetes across England directly standardised rate per 10,000 people with diabetes has reduced. The additional investment in the better access to care has contributed to that.
Over this period, teams were getting better at saving limbs, but not necessarily saving lives. Mortality is high of those presenting with foot ulcers 20% at two years, 40% at five years.
In 2014, Bob Young and William Jeffcoat had the foresight to implement the Diabetic Foot Care Module of the National Diabetes Audit. Data collection, supporting evaluation of everything. At the same time, they set up a quality improvement initiative to try and address the high mortality to suggest a 12 lead ECG (electrocardiogram) in the routine care pathway.
Ten centres across England tried to implement that and via the audit data collection, they could compare mortality in those ten centres versus the other centres in the country.
They achieved five-year mortality numbers. There was a statistically significant 13% lower mortality rate at the five-year mark.
Professor Valabhji said this granular data collection showed the effectiveness of the work being done.
He went on to give the example of helping people with prediabetes towards intensive lifestyle interventions to delay or prevent the onset of type 2 diabetes.
He made the case, got the funding and this was the first national programme of work that he undertook. He put together an expert reference group that he chaired over the next nine years. The group digested best evidence internationally and produced a service specification based on that best evidence.
This service specification was used to run a national procurement exercise, through which they appointed four providers to a national framework. The intervention, based on best evidence at that time, was group-based face to face delivery. Nine-month intervention. At least 16 hours of contact time with 13 different sessions focusing on behavioural support for losing weight, exercising more, and eating more healthily.
The rollout started in financial year 2017. They achieved 51% geographical coverage across England in the first financial year, 75% in the second year, and by the summer of 2018, they became the first country internationally to achieve universal geographical coverage across the whole nation.
Going back to granular data collection – Professor Valabhji said their providers did not get paid unless they populate a minimum data set on every participant every month through NHS linked to the National Diabetes Audit and to ONS (Office for National Statistics) mortality and hospital episode statistics. They could see who had an MRI (magnetic resonance imaging) scan. They developed an infrastructure for evaluating the programme well into the future.
During the Covid-19 pandemic the programme went digital, with remote delivery. There have been 1.7 million people with prediabetes referred into the programme. It grows and grows, said Professor Valabhji, suggesting it probably assessed around a third of the entire population with prediabetes or non-diabetic hyperglycaemia in England.
It was reported that someone who attends all the sessions had approximately half the subsequent rate of conversion to type 2 diabetes compared to someone who attended none of the sessions. That was reflected in similar trajectories of weight change and HbA1c reduction, related to the number of sessions attended.
Professor Valabhji also talked about the work of Roy Taylor and Mike Lean, who were able to demonstrate that we could put type 2 diabetes into reverse and achieve 46% remission at 12 months. It was delivered by healthcare professionals. However, there was not the capacity to implement a national programme with delivery by healthcare professionals.
In the same year, published in the BMJ by the Oxford Group, Paul Aveyard and Susan Jebb, used the same intervention. To achieve that profound weight loss meant three months of soups and shakes diet, followed by food reintroduction, followed by ongoing support to complete 12 months of behavioural support to maintain the weight loss.
A similar study from Oxford, while not in a diabetes population but crucially, delivery was by lay people trained up as health coaches. They combined these two approaches. They pitched in the long-term plan, which was funded to do a pilot and then, if successful, to roll out nationally.
With full national coverage as of March of last year, so when people come off programme, they are an average of ten kilograms lighter. This was published in the Lancet Diabetes and Endocrinology in the summer 2024. The remission rate is 30%, in “the real world”.
Policymakers, decision makers need to know what they are going to get from their investment in the real world, which is ongoing work.
Routine diabetes care and clinical outcomes – Bob Young, who was the father of the National Diabetes Audit – made it all happen, according to Professor Valabhji. In 2003 he established the National Diabetes Audit, and it has continued since. It is now a fully comprehensive register of everyone with a coded diagnosis of type 2 diabetes.
In the early years, it was developed to support the measurement of performance against routine diabetes care.
Following NICE guidelines, it focused on delivering nine care processes, for every person with diabetes and looking to achieve treatment targets for glucose and blood pressure and cholesterol. The audit measured the NHS performance against that.
Professor Valabhji said: “We see lower mortality associated with better care delivery of those processes using the national diabetes audit, lower hospital admissions, lower amputations, lower GI complications. So routine diabetes care works. Better things happen to people, patients, with diabetes if we deliver routine care.
“In 2020/ 2021 we were improving and then Covid hit, and we had a major reduction in routine care delivery over that first pandemic year. We could link ONS mortality data to whether people got care processes in the early years. And we showed a very clear association with that fall off in routine care delivery during Covid, and the higher mortality that was being subsequently realised that provided a very powerful lever for additional investment £37 million to support routine care recovery. The data points show you that two years on we have achieved a recovery from that.”
Professor Valabhji went on to focus on some home truths around epidemiology showing the proportional contribution to the mortality burden of leading cause specific diseases in those with diabetes for the two decades leading up to the pandemic.
At the turn of the millennium, the commonest cause of death was cardiovascular causes. Looking at the amazing diminution in cardiovascular causes of death over the ensuing two decades, they did a similar analysis for causes of hospitalisation, showing pretty much exactly the same – this reduction in cardiovascular causes.
What does that relate to? Well, almost certainly this has something to do with reduced smoking rates at the population level from the 1990s, Professor Valabhji suggested, but also, this focus on routine care delivery through treatment targets have contributed to some of this. This has resulted in increased longevity of our diabetes population.
Professor Valabhji said: “What we are seeing, therefore, with the diminution in cardiovascular disease is a diversification of the causes of both morbidity hospitalisation and mortality.
“In parallel, we have expanding waistlines over the last few decades across our whole population. And that is playing out as younger and younger age of onset of type 2 diabetes. We're seeing younger age of onset increase longevity, greater exposure to that metabolic milieu.
“We're seeing a diversification of causes of morbidity and mortality because of the reduced cardiovascular disease. And that's all culminating in much higher prevalence of what is now called multiple long-term conditions. Our clinics are populated by people who don't just have diabetes. Some of them have many other conditions, some of the complications of diabetes. Some are not. The complications of diabetes, multiple long-term conditions are the clinical reality in specialist clinics, in GP surgeries coming through the doors of A&E.”
When Professor Valabhji stood down as National Clinical Director, he had formed a data analytical workstream around multiple long-term conditions, he is still retaining a leadership role in the multiple long term conditions space for this reason. The next transition in the global diabetes epidemic is multiple long-term conditions.
Professor Valabhji said: “Early reports in 2020 coming out of Wuhan, before we had our first cases, suggested, rather worryingly, an excess of people with diabetes in those dying from Covid. Now, older people were also more likely to die from Covid and prevalence of diabetes goes up with age. Was this a real signal or was it not? We realised that we were very well placed to disentangle some of those early unknowns.
“We linked the National Diabetes Audit to something called the Master Patient Index, which is everyone registered with a GP in England to something called the British to Health Segmentation Data set, which draws together many other data sets and national audits and gives us essentially registers of 35 different long term conditions, including diabetes, and finally to ONS mortality in the Covid patient notification system. We were the first internationally to give an accurate estimation of the increased absolute and relative risk of Covid related mortality in people with diabetes by type. And that paved the way for prioritisation to vaccination for people with diabetes when vaccines became available.
“The diabetes prevention program is 1.7 million. We have got 35,000 in the remission programme. I have not talked about the digital weight management program, which is another lifestyle intervention that we have rolled out nationally that has had half a million people referred in. But we took the diabetes prevention programme cohorts. We matched every individual who passed all the way through and completed the programme to someone who looked very similar in terms of BMC level and age and sex and ethnicity and deprivation and both points of the GP surgery.
“Now, we already knew this, that the programme works to reduce incidence of type 2 diabetes. But we were interested, we hypothesise. Surely a nine-month lifestyle intervention. Will it not do more? Will it not reduce incidence of other conditions? And so, this was a retrospective look using a matched cohort to try and first address that question.”
Professor Valabhji continues to research whether lifestyle intervention really does reduce incidence of other long-term conditions.
In conclusion, he hoped to have given examples of how each of these granular data collections are fundamental. If you can follow the evidence in policy implementation, the diabetes prevention program being a cardinal example in terms of pilot and evaluate you see results.
I would like to make a regular donation of
I would like to make a single donation of
There are lots of ways to raise money to support
people living with all forms of diabetes.
Bake, Swim, Cycle, Fly ... Do It For DRWF!
Fundraise with us
Recent News
