Professor Alan Maynard OBE reviews the validity of the idea that teaching hospitals have higher running costs

Traditionally, it has been argued that university teaching hospitals are more costly to operate. The primary argument to support this assertion is that they have a more complex case-mix.

Is this argument tenable, or is it a case of special interests defending their inefficiency?

Economics teaches us that larger units may have lower costs due to economies of scale and economies of scope. Why are these hospitals exceptions?

In medical care, there is some evidence of economies of scale in terms of activity rates and outcome, as proxied by mortality. Thus if the comparative mortality rates of high- and low-activity hospitals in pancreatic cancer surgery, oesophageal cancer surgery and paediatric cardiac surgery are compared, the former (according to US data) have a 10-13 percent lower death rate.

About the kind of care
However for activities such as knee replacements, coronary by-pass grafting and lung cancer surgery the differential in mortality rates between the low- and the high-activity hospitals is less than 2 per cent.

Clearly, it is important not to generalise about the link between activity levels and mortality outcomes.

The limited evidence for ‘going large’
This does not inhibit some of the Royal Colleges - who continually pressure government for amalgamation of existing hospitals into large, single units in order to reap these limited and specific benefits from economies of scale.

Such blanket advocacy on the basis of partial evidence should be resisted. This resistance should be accompanied by questioning of the cost implications of having larger hospitals.

Again the literature is limited, largely American and of mixed quality. What it shows is costs in relation to bed volume falling to about 600 beds, then plateauing temporarily before rising again - i.e. bigger hospitals have higher costs.

The pursuit of big department to capture economies of scale in relation to mortality outcomes may have to be traded off against higher unit cost. Do large teaching hospitals garner better mortality outcomes in a few specialised areas of activity but at a higher cost for all their activity?

Why does having over 600 beds means higher LOS
The nice issue is why costs rise as hospital size increases above about 600 beds. Again, there is a storyline which asserts a more complex case mix in teaching hospitals which gives rise to longer lengths of stay (LOS). The evidence for this is mixed enough to fuel scepticism and concern that the longer LOS are a product of inefficiency in managing patient discharges.

The economies of scale literature is indicative and incomplete. However, it is sufficient to create scepticism about unthinking acceptance of the argument that large hospitals are “special” and that we should expect their costs to be higher.

Perhaps the mortality economies of scale could be achieved by smaller, low-cost hospitals which specialise in cancer and cardiac surgery?

Demand side economies of scope
Unlike economies of scale which affect the supply side, economies of scope affect the demand side of the market. Economies of scope lead to declining average costs of productionm as the number of goods and services produced increases.

Thus as more products are added to a firm’s portfolio, the administrative, managerial and advertising budgets can be spread over more activities.

The inference for hospitals is that economies of scope have the potential to reduce average costs as the range and volume of services offered to patients rises. Thus, for instance, as a hospital grows and its volume of activity increases, it may be able to market its services at lower cost.

The inference from this is that the large teaching hospitals should benefit from lower costs, and not have higher costs.

As the NHS totters towards mergers and hospital closures, it is timely to confront the sacred cows called university teaching hospitals. These powerful institutions with their trust funds and history tend to shake off calls for reform and sail on expensively. Their high costs deprive other institutions of resources that may produce more health gain.

In 1905, the Daily Mail noted that London hospitals were in the wrong place to serve the local population. After just 105 years, is it timely to reform sacred cows both in London and elsewhere?

With the coalition government calling for transparency and accountability, this will be a nice test of whether politicians are blathering or really sincere about protecting patients and taxpayers.