Challenging the funding model of the National Health Service

When the NHS was founded on 5 July 1948, it was done out of an ideal that good healthcare should be available to all, regardless of wealth. More than 65 years on, few would argue that this commitment to healthcare for all has had anything less than a profound impact upon life in Britain. But the assumption that it is meeting the needs of everyone is being challenged by more than a decade’s worth of research from a multidisciplinary team from Plymouth University.

“Most people believe we have a universal NHS, one in which everyone has equal access to care,” said Sheena Asthana, Professor of Health Policy in the School of Law, Criminology and Government. “But actually, by prioritising healthcare inequalities, economists within the Department of Health have diverted funding for healthcare away from those who need it most - the elderly.”
The origin of the University’s research in this area dates back to 1999 when Sheena secured Economic and Social Research Council funding to produce estimates of coronary heart disease in English Primary Care Trusts (PCTs), and then to compare those figures with actual rates of surgical intervention (e.g. coronary artery bypass operations) obtained from hospital data and other sources.

The results revealed that although the burden of coronary heart disease tended to be highest in areas with older populations - even if those populations were relatively affluent - hospitalisation and surgical intervention rates were highest in areas with socially disadvantaged, and usually younger, populations.

“The underlying driver was that those areas suffering from urban deprivation were assumed to have a greater claim to NHS resources,” said Sheena. “The legitimate healthcare demands of, in particular, ageing populations in rural and coastal areas were not being given appropriate weight within the allocation formulae used to distribute NHS resources.”
Sheena, with her colleague Dr Alex Gibson, began to scrutinise in more detail the formula that was being used to allocate budgets to different PCTs. They discovered that, in 2004-05, just 7 per cent of trusts in deprived urban areas were running a deficit, compared with 70 per cent in affluent rural areas. This, they argued, implied an underlying flaw in resource allocation and, at least in part, explained the huge variations in per capita expenditure on critical healthcare. For instance, in 2010, about £4,000 was spent on each cancer patient in Dorset, compared to £15,000 in some areas of London.

On scrutinising funding formulae for public health services and questioning their philosophical, technical and empirical bases

What we have been saying is controversial. It is even politically difficult to question the idea that deprivation is synonymous with public service need. But our research has proven to be methodologically pioneering, and our results intuitively coherent.

Professor Sheena Asthana.

Having questioned the current ‘utilisation-based’ approach to resource allocation (whereby the healthcare needs of populations are predicted by modelling existing patterns of healthcare use), in 2007 Sheena led an NHS-funded feasibility study to examine whether a very different approach could be developed. This exploited existing large-scale health surveys to estimate the prevalence of various illnesses across different population cohorts (defined by factors such as age, sex, ethnicity, educational status and, importantly, location). Applying these estimates to local areas using census data, and drawing upon national cost data, these ‘direct’ estimates could then be used to allocate resources in a way that reflected the actual healthcare needs of populations.

This work was methodologically as well as conceptually innovative. Dr Paul Hewson, Associate Professor in Statistics, applied Bayesian statistical methods to ensure that the team were able to capture and describe all uncertainty around the resulting ‘risk-adjusted resource need’ estimates. This allowed policy makers to understand, in particular, potential problems associated with allocating fixed budgets to very small populations, such as those served by individual practices.

This research supported their claims that the existing funding formula was not fit for purpose, and resulted in the team being asked to develop a practice-level allocation formula for distributing the £8 billion mental health budget. This they did using a ‘casemix-based modelling approach’ - once again using factors such as age, sex, ethnicity, tenure and employment status to predict the likelihood that different ‘person types’ would fall into each of the casemix categories. The team then combined these likelihood estimates with NHS administrative and census data to predict the number of people in each practice falling into each casemix category. Using treatment costs attached to all patients included in the original casemix study, the final step was to apply casemix cost distributions to patient counts to generate an overall estimate of the resource needed to meet the mental healthcare needs of each practice population.

Sheena said: “We found a complex relationship between mental health needs, age and deprivation, and this gave rise to a more nuanced understanding of the geographical pattern of prevalence. Mental health needs were generally very much higher in northern England, especially in the major cities. However, there was also a notable coastal fringe of ‘high-need’ practices which, by and large, reflected the high proportion of elderly people in retirement hot-spots.”

The Department of Health described the model, which was used for funding decisions from 2009 to 2011 as “a step-change improvement in the way we model mental health need.”

A new approach to funding has been introduced with the shift to Clinical Commissioning Groups, but the research continues to impact on the public sector. The team has written ministerial briefings, provided evidence to a wide range of stakeholders from the parliamentary Health Committee to the National Audit Office and NICE, and been quoted in parliamentary debates and health journals.