The link between healthcare spending and health outcomes

Executive summary

This report presents preliminary results from research undertaken as part of the Quest for Quality and Improved Performance (QQUIP), an initiative of The Health Foundation.

For three years, each primary care trust (PCT) in the English National Health Service (NHS) has prepared data on expenditure on healthcare across 23 ‘programmes’ of care, based on the International Classification of Diseases (ICD) Version 10 disease categories. These programme budgeting data seek to allocate exhaustively to disease categories all items of NHS expenditure, including expenditure on inpatient care, outpatient care, community care, primary care and pharmaceuticals. In 2004/05 the average size of the programmes varied considerably, with the three largest being mental health (£145 per head per year), circulatory disease (£122) and cancer (£75).

The programme budgeting data offer immense opportunities for examining the link between healthcare expenditure and health outcomes across PCTs. There is extensive international literature on this topic, but very little solid empirical evidence on the magnitude of the link. Indeed, many authors claim that – at the margin – extra healthcare spending has little impact on health.

The main reason for the lack of evidence is the difficulty of disentangling cause and effect. Areas with high health needs and poor outcomes tend to attract high levels of healthcare spending. This phenomenon is confirmed by examining the link between programme budgeting expenditure and health outcomes (standardised mortality rates or SMRs) among the 303 PCTs (the figure relevant for the study period 2004/05). For example, there is a strong positive correlation between expenditure and under-75 SMRs in cancer and circulatory disease.

However, the question for policy-makers is whether – after adjusting for need – extra spending gives rise to better health outcomes. Addressing this question requires substantial additional data (in order to model needs) and advanced statistical methods. This report examines the link between expenditure and outcomes in two programmes: cancer and circulatory disease. It models both programme expenditure as a function of needs, and then outcomes as a function of expenditure.

These preliminary results are encouraging. For both cancer and circulatory disease programmes, it proved possible to develop robust and well-specified statistical models in line with expectations. They demonstrate a strong positive link between expenditure and better health outcomes (lower SMR) in the two disease categories, and that the link is stronger in circulatory disease than in cancer.

Using a measure of ‘years of life lost’ instead of SMR as the measure of health outcome, it is also possible to estimate the expenditure required to ‘save’ a year of life in each disease category. Our estimate is that, for a PCT with average needs and expenditure, the marginal cost of a life year saved in cancer is (on average) about £13,100, and in circulatory disease about £8,000. It must be emphasised that these results have quite large confidence intervals, will vary between PCTs and should be treated with caution. Very importantly, they are not adjusted for quality of life. However, it is noteworthy that they do appear to compare favourably with the threshold of £30,000 per quality adjusted life year (QALY) often attributed to the National Institute for Health and Clinical Excellence (NICE).

These results are useful from a number of perspectives. Scientifically, they challenge the widely held view that healthcare has little marginal impact on health. From a policy perspective, they can help set priorities by informing resource allocation across programmes. They can also help NICE decide whether its current QALY threshold is at the right level.

• Email to a colleague
• Add to briefcase
• Print page