Diabetes: cases and costs predicted to rise

“Diabetes could ‘bankrupt’ the NHS in 20 years,” the Daily Mail has reported. The newspaper said that most spending on the condition is due to avoidable complications. Several other newspapers featured similar claims, saying that by 2035 a sixth of NHS funding will be spent on the disease.

These bold claims are based on a UK study which reported that the annual NHS cost of the direct treatment of diabetes in the UK will increase from £9.8 billion to £16.9 billion over the next 25 years. The predicted rise would equate to the NHS spending 17% of its entire budget on the condition, up from about 10% today.

The study also reported that the cost of treating diabetes complications (including kidney failure, nerve damage, stroke, blindness and amputation) is expected to almost double from £7.7 billion currently to £13.5 billion by 2035/36.

These projections are estimates based on current data. While this doesn't mean they are unreliable or unrealistic, predicting future trends involves a lot of uncertainty and assumptions, and many things can change by 2035. The study does highlight the major challenges many nations face in preventing and treating diabetes and the need to address the disease through measures such as improved education, diagnosis and management.

The newspaper headlines that suggest the costs of diabetes treatment will ‘bankrupt’ the NHS are misleading as this is not likely to happen. The study did not look at the wider spending of the NHS in detail. However, it is clear that diabetes is a major condition in the UK, and there is a need to examine the financial, personal and societal impacts of the condition to prevent people being needlessly affected by it and its complications.

Where did the story come from?

The study was carried out by researchers from the York Health Economic Consortium and was funded by Sanofi, a healthcare company working in the fields of pharmaceuticals and research.
The study was published in Diabetic Medicine, the peer-reviewed medical journal of the charity Diabetes UK.

Aside from exaggerated headlines about bankrupting the NHS, the media generally described the study findings accurately.

What kind of research was this?

This economic analysis estimated the current and future economic burdens of type 1 and type 2 diabetes in the UK.

The body uses the hormone insulin to control blood sugar (glucose) levels. Insulin is released in response to meals so the body can remove excess glucose from the bloodstream. Excess levels of blood glucose can cause several problems in the body, including damaging the eyes and internal organs, coma or even death.

There are two types of diabetes, with slightly different causes and treatments:

  • Type 1 diabetes is an autoimmune condition that occurs when the body’s immune cells destroy the insulin-producing cells in the pancreas. It usually presents at a younger age and requires life-long insulin replacement injections
  • Type 2 disease occurs when either the pancreatic cells do not produce enough insulin or the body’s cells do not respond to the actions of insulin and, therefore, do not remove glucose from the bloodstream sufficiently.

Both types of diabetes are associated with complications that involve the large blood vessels of the body, increasing the risk of cardiovascular diseases such as strokes and heart disease. They can also cause complications involving the tiny blood vessels in the body, such as those in the kidney, eye, and supplying the nerves (for example, leading to loss of feeling in the feet). In both types of diabetes, poorer control of blood glucose is associated with increased risk of these complications.

While predicting future costs of treating a certain condition is difficult, this approach can provide a useful insight into what aspects may cost the most in the future. It can also highlight areas where there may be wasteful or unexpectedly high costs that could be corrected or scrutinised.

What did the research involve?

The researchers sourced various reports on the prevalence and cost of diabetes from diabetes organisations and UK national statistics. They used them to estimate their costs for 2010/11. They then used projected prevalence and population data to predict how these costs would change up to 2035/36.

Prevalence and population data were obtained for children and adults with type 1 and type 2 diabetes from sources including the Network of Public Health Observatories (APHO) Diabetes Prevalence Model, a research study in the UK, and the Office for National Statistics (ONS) population data. Population changes, estimated using ONS projected population figures, were then combined with the diabetes prevalence data to give the projected number of people with diabetes in the UK up to 2035/36.

Direct and indirect cost data were obtained from either published literature or national data sources such as NHS reference costs. Direct treatment costs included items such as primary care consultations (visits to the GP) and prescribed drugs (insulin), consumables (such as disposable needles) and monitoring devices. Indirect costs (non-NHS costs) included social and productivity costs, such as reduced ability to work through sickness leave and working years of life lost because of death from diabetes or complications of diabetes.

Data on the frequency and cost of complications associated with diabetes were also estimated for a large and varying number of conditions, including heart disease, kidney disease, nerve damage and erectile dysfunction.

The researchers combined figures of the number of people with diabetes and the costs incurred to form an economic model that predicted the future costs of diabetes care. This was modelled on the basis that current trends and treatments would continue.

The statistical analysis of the results was appropriate. It included a "sensitivity analysis", which is a verification process that involves adjusting the inputs of a model to see how much they affect the estimates it generates.

What were the basic results?

The report found a contrast between the current costs (2010/11) of treating diabetes with those expected by 2035/36. The main findings included:

Current situation:

  • Diabetes cost approximately £23.7 billion in the UK in 2010/11, including both direct and indirect costs.
  • The current cost of direct patient care (treatment, intervention and complications) for those living with diabetes is estimated at £9.8 billion (£1 billion for type 1 diabetes and £8.8 billion for type 2 diabetes).
  • The current indirect costs associated with diabetes, such as those related to increased death and illness, work loss and the need for informal care, are estimated at £13.9 billion (£0.9 billion for type 1 diabetes and £13 billion for type 2 diabetes).
  • Deaths from diabetes in 2010/11 resulted in over 325,000 lost working years.
  • An estimated 850,000 people in the UK have diabetes that has not yet been diagnosed, and the cost of this group was estimated at about £1.5 billion.

The possible situation by 2035/36:

  • Diabetes is projected to cost £39.8 billion overall by 2035/36.
  • The cost of direct care for patients is estimated to rise to £16.9 billion (£1.8 billion for type 1 diabetes and £15.1 billion for type 2 diabetes).
  • The indirect costs associated with diabetes will increase to approximately £22.9 billion (2.4 billion for type 1 diabetes and £20.5 billion for type 2 diabetes).

The differences:

  • The annual amount the NHS spends on direct diabetes treatment in the UK will increase from £9.8 billion to £16.9 billion over the next 25 years.
  • The cost of treating diabetes complications (including kidney failure, nerve damage, stroke, blindness and amputation) is expected to almost double, from £7.7 billion currently to £13.5 billion by 2035/36).
  • Diabetes currently accounts for approximately 10% of the total NHS budget, but this is projected to rise to around 17% by 2035/36.

How did the researchers interpret the results?

The researchers concluded that type 1 and type 2 diabetes are "prominent diseases in the UK" and the costs associated with the condition represent a "significant economic burden".

They highlighted that "complications related to the diseases account for a substantial proportion of the direct health costs". They later stated that only 25% of the total cost relates to the treatment and ongoing management of diabetes, while the remaining 75% is spent on treating the complications of diabetes. Complications can result from poor control of blood sugar levels in people with diabetes.


Diabetes is one of the most prevalent and serious chronic conditions currently affecting the UK population. This new economic study has provided thought-provoking estimates on the current (2010/11) direct and indirect costs of diabetes treatment in the UK. It has also projected these costs into the future to 2035/36, providing attention-grabbing estimates that suggest costs will rise dramatically in the next few decades.

It should be noted that the projections of future spending are broad estimates based on extrapolating current estimates. This type of modelling is only as good as the information it uses and will ultimately rely on assumptions and estimates.

This is not to say the estimates are not plausible or valuable, but that a host of factors may cause future spending to deviate from this new model’s predictions. For example, the management of diabetes or the medicines available may change significantly by 2035, and this model cannot accurately account for these potential changes. Even the best health research cannot accurately predict the future.

The primary types of data used in this model were estimates on the number of people with diabetes (prevalence), and the costs associated with people with diabetes (cost impact). The authors acknowledge that the prevalence estimates varied significantly from different sources.

They also stated that "none of the sources provide an accurate indication of the numbers of people with type 1 and type 2 diabetes." Therefore, the estimates are susceptible to a certain degree of error and so are not definitive cost estimates. However, using the best available data is a practical approach to produce a broad estimate of costs and having imperfect estimates is arguably better than having no estimates at all.

The projected future costs generated by this study have attracted much media attention, with many headlines suggesting that the costs will somehow "bankrupt" or bring down the NHS. Given the uncertainties over the estimates, such claims are sensationalist and misleading. However, while we cannot say how these costs will directly affect overall NHS finances, the study suggests that diabetes currently incurs major costs, and that these will rise significantly if things stay the same.

Overall, it suggests several important issues that will need addressing in the near future to minimise the impact diabetes has on people’s lives, as well as the nation’s finances. Areas that may need to be looked at include:

  • new measures to prevent cases of diabetes, such as public health initiatives and education
  • new measures to diagnose and treat diabetes when it does occur
  • the role of education for people with diabetes to help improve their blood sugar management and minimise the complications they experience
  • further research into potential treatments and interventions that may reduce complications or provide better value without compromising the quality of treatment
  • further financial assessments of how diabetes spending might be restructured to deliver better outcomes for lower spending

There is existing guidance from the National Institute for Health and Clinical Excellence (NICE) on diabetes treatment and management. The authors of this research say that their future work will examine the cost impact of fully adopting NICE guidelines across the UK. In particular, they will look at how costs could be saved by reducing or delaying complications from diabetes.

NHS Attribution