Public health measures and the National Accounts: The effects of the Diabetes Prevention Programme

cube-no-animation-2

Public health measures and the National Accounts: The effects of the Diabetes Prevention Programme

By Martin Weale

Accurate measures of the output produced by public services are vital to understanding developed economies. However, measurement can be more of a challenge than in the private sector. This is because outputs are harder to define, and harder to measure, especially when they do not have a price.

An example of this is the introduction of preventative medicine. This raises issues because it reduces the risk of people becoming unwell rather than generating extra output. The conventional valuation of these non-market services at their cost of production leads to significant under-estimations of their value.

New work from Martin Weale discusses how public health measures might be shown in the National Accounts. The study first develops a methodology and then illustrates this using the Diabetes Prevention Programme, introduced in England in 2017.

Preventative medicine in the National Accounts

The appearance of new goods and services, and the disappearance of old ones poses problems for attempts to measure the output of the economy after adjusting for the effects of price changes. A number of economists have suggested that the problem can be resolved by valuating the new service at its reservation price, i.e. the price at which demand would fall to zero, in the period before its introduction. Similarly a disappearing good or service can be valued at its reservation price in the period in which it is no longer traded. However, this approach does not work unless price changes are measured using symmetric price indices – indices which treat current and previous period expenditure patterns in the same way.

The paper applies this technique, suggesting that reservation prices for new goods or services can be derived from medical information on the health benefits that they deliver. The gap between the reservation price and the actual price or cost at introduction then appears as a reduction in the price of medical services. It also appears in an output price measure for the economy as a whole, without any other changes. Given information on the monetary value of output, the impact on the volume of output can be calculated in a straightforward way. This framework is then used to explore the impact of the Diabetes Prevention Programme.

Diabetes and the Diabetes Prevention Programme

In the UK, diabetes is estimated to cost the NHS £3.2bn per year in terms of diagnosis and management. Treatment of complications (including foot problems, visual impairment and kidney disease) was estimated to cost an additional £5.6bn per year in 2021/22.
 
The Diabetes Prevention Programme was set up on a test basis in 2017 by the NHS. The programme offers classes for people at risk of developing type-2 diabetes. Patients are referred if they have blood sugar levels that are higher than normal but not yet classed as diabetic. In a series of 15 classes, advice is offered on healthy eating, weight management and exercise to help prevent the disease.

Using Quality Adjusted Life Years

The outcome of preventative health programmes can be measured using the number of extra quality-adjusted life years (QALYs) the population enjoys. QALYs consider a person’s health-related quality of life rather than just looking at their life expectancy. In this context, they can provide a straightforward way of handling reservation prices as well as quantities. NHS England suggest that one extra QALY accrues for each case of diabetes delayed or prevented. A QALY is typically valued at £60,000 but people generally value future benefits less than they value current ones. Allowing for the fact that the main benefit may occur in the distant future, we use a value of £30,000.

Estimating the effects of the programme

A 2023 study1 suggested that 3 years after course completion, 12700/100000 participants had developed diabetes, whereas the number rose to 15400/100000 for those who did not do the course. Evidence from other countries points to larger effects.

To assess the benefits of the programme and therefore its output, we need to take account of QALYs generated along with the cost saving. This allows us to compute an implied reduction in the “price” of diabetes management with the introduction of the programme. As mentioned, when working with QALYs, we need to allow for the fact that the costs saved lie largely in the future and should therefore be discounted.

Table 1 summarises the key statistics for the programme per 100,000 referrals. The target was 390,000 referrals in total, but this was not expected to be achieved in one year.  There are thought to be 5.8mn people in the UK who have elevated blood sugar levels and would therefore be eligible to participate. As the table makes clear, per 100,000 referrals we find an increase in QALYs valued at £81,000,000 and a saving in costs valued at 39,420,000.

But how might the effect of the programme be shown in measures of health service consumption, such as those found in the National Accounts?

The Impact of the Programme

Comparing the cost relative to the benefits of the programme, the introduction of the programme leads to a sharp fall in the implied price of treatment per referral. But the overall expenditure is only a small fraction of publicly-provided health service output. In 2018-19 when the programme was underway, this was valued at £151bn. Although the cost of the programme per referral was just under 14% of its benefits, the small share of expenditure on the programme limits its impact on the overall estimates of the output of medical services. Indeed, we find that the programme results in an increase in the volume of health services of only 0.02% per 100,000 referrals. Therefore, in this case, the impact of preventative treatment on the volume of health service output is small.  

To check that the methodology isn’t understating the programme, we can also look at the number of extra QALYs generated. If we look at the number of extra QALYs generated as a proportion of an estimate of the total QALYs attributed to medical treatment2, we find an even smaller percentage increase in gross output of medical services, of 0.005% per 100,000 referrals.

Why it matters

This paper shows possible ways to consider preventative medicine when producing estimates of the output of medical services in the National Accounts.  The effects in the example studied are small relative to overall health service output. While this is an important finding, it is also valuable to have a coherent framework for preventative medicine. Some preventative programmes may have very large effects, and this framework makes it possible to establish whether that is the case.

This paper is part of ESCoE work on Public Sector Productivity. For more on this topic, listen to the webinar.

ESCoE blogs are published to further debate. Any views expressed are solely those of the author(s) and so cannot be taken to represent those of the ESCoE, its partner institutions or the Office for National Statistics.


1. Ravindrararajah, R., M. Sutton, D. Reeves, S. Cotterill, E. Mcmanus, R. Meacock, W. Whittaker, C. Soiland-Reyes, S. Heller, P. Bower, and E. Kontopantelis. 2023. “Referral to the NHS Diabetes Prevention Programme and Conversion from Nondiabetic Hyperglycaemia to Type 2 Diabetes Mellitus in England: a Matched Cohort Analysis.” PLOS Medicine 20:1–18.

2. For these calculations we assume that half of the expected discounted QALYs for the population aged 65 and over are due to medical science. A larger base number of QALYs would result in a smaller percentage impact of the programme.

About the authors

Related events