The Diabetes Challenge
25 October 2016, 10:54
On the 13th June Pritesh Mistry (Director at Data Flock) and I attended the “Data Innovation Challenge – The Rising Problem of Type 2 Diabetes”on behalf of Hebe Works with Digital Catapult Centre Yorkshire at the Digital Health Enterprise Zone in Bradford. Two GP’s spent the afternoon giving us insight into Diabetes and we were given two weeks to provide a challenge entry – we decided to focus on the following challenges:
- What are the prescribing savings that could be made for CCG’s in Bradford?
- How can the patient be pointed to areas so that they can become involved in the care that they receive?
- How do we regularly capture stress, diet, excersise, blood levels, & BMI data?
We started looking through all the Open Data sources we thought would be related to the challenge (having previously worked at HSCIC for 5 years, there was perhaps an unfair advantage of knowing where to find all the data!). We used multiple datasets, but primarily focussing on prescribing data from NHS Digital.
The total cost for the initial Diabetes drugs in Bradford in 2015 was £4,570,000.08
Our idea was to identify how much money was spent on the first drugs prescribed to patients. Usually when patients are diagnosed with Type 2 Diabetes they are given advice on changes they should make to their diet and asked to come back for a review in 6 months time. If the review shows that method hasn’t been effective, they are prescribed Metformin and asked to come back for a review. If Metformin hasn’t worked, the patient is prescribed Metformin and Sulfonylureas. And if that doesn’t work they are prescribed Metformin, Sulfonylureas, and “other drugs”.
In Bradford last year CCG’s spent exactly £4,570,000.08 on these drugs – an average of £205.60 per Type 2 diabetes patient (before you count the cost of cholesterol checks, blood pressure drugs, GP visits, etc).
We started to look at differences between practices – which surgeries were prescribing significantly more than the average level. We could then create heatmaps for Wards in Bradford, and look at which high prescribing practices were close to practices who prescribed less than the average.
Our idea was to not only share prescribing knowledge between practices (one practice spent an average of £452 per patient, whereas a practice 5 minutes walk away spent £86.16 per patient), but to create an intervention before any drugs were prescribed.
What if GP’s prescribed exercise?
“On average, it takes more than two months before a new behaviour becomes automatic — 66 days to be exact.”
Diabetes UK Research states that there is “a growing body of evidence confirms that the best protection against developing Type 2 diabetes is to exercise”.
By prescribing exercise classes at Council sports centres rather than the first stage of drugs we found that there was not only a potential saving of £2,517,945.57 per year for CCG’s in Bradford, but it would also generate the council over £1 million in exercise class fees per year. CCG’s also have the power to assign budget for GP’s to prescribe in this manner, meaning practices wouldn’t have a financial loss by prescribing exercise instead of the first rounds of Type 2 Diabetes drugs.
…it would also generate the council over £1 million in exercise class fees per year
Patient’s benefit from becoming healthier, CCG’s benefit from having lower overall costs, the Council benefits from increased revenues, and GP’s are compensated for prescribing exercise rather than the initial drugs.
That feels like a win for everyone.
- Share best practices with neighbours
- Assign Practice Nurses to collect patient data after exercise to help patients form lasting positive habits
- Save over £2.5 million per year for Bradford CCG’s
- Generate over £1m for Bradford Metropolitan District Council in sport centre usage
AND we won the Digital Catapult data challenge!
View the full presentation
Contains public sector information (Mostly NHS Digital Prescribing data) licensed under the Open Government Licence v3.0.
For more information visit: http://www.dataflock.co.uk/