Proper accountability and the importance of data post-Francis

Data should play a fundamental role in assessing the performance of healthcare organisations. Yet NHS ‘culture’ – a theme so central to the Francis report – fundamentally mistrusts, undervalues and misuses performance information.

This is not a new development, either in the UK or worldwide. Indeed one could invoke the defeat of Florence Nightingale’s valiant efforts to instigate rigorous and systematic comparison as an early example of government indifference and medical hostility to performance comparison.

We like to think we live in more enlightened times. However, even with the unbelievably enhanced capacity to capture data that we now enjoy, it seems an enormous task to make material advances in performance comparison – every step forward seems to be defeated.

There was a brief flourishing of comparative hospital performance data distributed in England electronically in the 1980s – including risk adjusted hospital mortality rates – but that initiative fizzled out (bizarrely) with the advent of the internal market, just when you might have thought it was most needed by purchasers.

Scotland experimented with similar comparative data in the 1990s, but so timidly that few took any notice of the apparently huge variations in mortality and other important comparisons. It took the efforts of Dr Foster to start the English ball rolling again, but the reaction of the NHS was largely hostile and defensive.

Since then we have experimented with Quality Accounts – a good idea, but not implemented with enough vigour or commitment. Of course Performance Ratings secured gains in the area of waiting times, but had little to say about clinical quality.

To an outside observer, the NHS continues to be – at best – a grudging user of performance data, especially those relating to clinical quality. Of course, there will always be legitimate debates about quality of data and analytic methods. But the NHS needs to engage with those debates creatively, and to seek to explain the cause of variations, rather than, as sometimes seems to be the case, to question the legitimacy of quantitative comparison.

What is needed to change this state of affairs? In a word: proper accountability. Where the focus of that accountability should be is a secondary issue, but there are several potential sources of accountability, such as:

• markets (properly informed patient choice)
• the professions (offering assurance that local services are satisfactory)
• local electoral process (holding local institutions and their boards properly to account)
• commissioners (taking decisions on the basis of quality rather than convenience)
• or a decently functioning system of quality inspection.

It is likely that a mix of all these mechanisms will be needed, with different processes being more appropriate for different functions. What matters however is that, at every turn, the relevant people should be properly held to account for the performance of the services for which they are responsible, and that appropriate rewards and sanctions arise as a result. Data should be central to all these mechanisms, and no proper accountability can be secured in the absence of comparative performance data.

As things stand, we have been relying on semi-informed patient choice, bland professional assurance and poorly focused regulation to secure quality assurance. This state of affairs cannot be allowed to continue after Francis. I think we need the following:

• Relevant, comparable, timely and accurate data should become central to the functioning of the health system.
• All providers of NHS services, whatever their ownership, should be required to provide performance information if they are to treat NHS patients.
• Provision of incomplete, inaccurate or delayed data should result in proportionate financial penalties, increased inspection and withdrawal of NHS business.
• Boards of directors should be required to secure assurance of the quality of the services provided by the institution on the basis of proper analysis and scrutiny of comparative information.
• Aberrant organisational performance should trigger regulatory interventions that are at first precautionary, but which become increasingly severe as unexplained adverse performance either becomes more extreme or persists over a long time period.

My own view is that all such scrutiny and interventions should be made public, with a statement of what corrective actions are being taken, or a reasoned explanation for why any sustained apparently adverse performance is being tolerated.

Comparative performance information should be the foundation on which the NHS runs its affairs, not an inconvenient irritant, to be ignored, discredited or dismissed. The scandal in Mid-Staffordshire was as much about the failure to respond properly to adverse data as it was about the failures in care standards.

Peter is Professor of Health Policy at Imperial College Business School & Centre for Health Policy.