The NHS has never lacked information, but, says Dr Foster Intelligence's Tim Kelsey, only now are managers and clinicians harnessing its power to change services. Public access is the next big challenge

When Dr Foster first began publishing analyses of mortality using routine hospital administrative data, one particularly outraged consultant wrote to me to express his opinion that our tactics were no better than those of al-Qaeda.

Although his choice of words was unusual, his sense that he was being threatened was not. Many doctors felt that control was being taken out of their hands; data sources that they had not sanctioned were being used to judge them; and information about them which they had not authorised was being provided to the public.

In the six years since then, attitudes have changed. Most people in the NHS recognise they are better off knowing what data about patients can tell them about their services. And if the data is not good enough, it is in their interests to ensure good data is available.

Probably the best way to show this is with examples - such as the orthopaedic team at United Lincolnshire Hospitals trust, led by consultant surgeon Rajiv Deshmukh.

Over the last year, he and his team have worked to improve discharge planning. They have put in place a trauma care co-ordinator and made sure that patients were supported on leaving hospital. They decided to tackle this issue because they had data showing their readmission rates were higher than would be expected for their patient population. Having implemented their changes, they have been able to track readmission rates and have seen them fall to well below similar hospitals.

Information overload

Like most senior clinicians, Mr Deshmukh had access to plenty of information about the performance of his trust. But it was the wrong sort of information. What enabled him to identify areas of potential improvement was comparative performance information, meaningful to him and his team.

Specialists need specialised information - not information about high-level government targets. They need to have access to information which is meaningful to them and their peers and which they can review as a trend over time or benchmark against others.

Such information is not an end in itself. It prompts questions: is my team performing as well as it could be? How are we performing against our peers? Why are we different? What could we do better?

At Hope Hospital in Salford, examining performance information promoted a better understanding of activity and costs in the neurosurgical directorate. Consultant neurosurgeon Kanna Gnanalingham found that it prompted a constructive dialogue with management.

Length of stay at seven to eight days was longer than the national average of six days. He was able to show that Hope's neurological stays were skewed by the presence of an in-house rehabilitation unit. Coding included the time patients spent in rehab but the trust was being compared with departments that sent patients to outside rehabilitation units.

The NHS is awash with data which can prompt these sorts of dialogues at the most important level: within clinical teams. The challenge for information providers is to increase the usage of what we already have. The quality of administrative data such as hospital episode statistics has improved in the past five years. Payment by results has driven more precise coding. Comparative studies show that HES is as accurate as clinical audit databases. The roll-out of the secondary uses service and electronic patient record will drive further improvements.

Crown the coders

Clinical record taking has been left to coders for too long. Clinical coders in the US have a graduate or postgraduate qualification and are paid accordingly. NHS coders generally work separately from clinicians and are relatively poorly paid. Some organisations, such as Hammersmith Hospitals trust, are now placing coders alongside clinicians on ward rounds. In this way the coder can verify information from both the clinician and the patient at the bedside. Data quality and confidence in that data therefore improves.

The use of administrative data offers significant benefits to the clinical community: it allows for comparison within and between organisations, according to case-mix, within near real time. HES data is also relatively cost effective and less resource intensive than other sources of data in the health service.

Although there is some way to go, clinicians in the acute sector are better off than those working in mental health, primary and community care. If the government is to meet its objective of shifting more care to local settings, we need to ensure that professionals have better access to information to allow them to do so. Why should practice-based commissioners develop new pathways of care within the community if understanding the costs and developing a business plan is resource intensive?

Commissioners and providers also need to develop their understanding of population needs by linking available data sets across public services. And we can do most of this without the sort of cross-cutting super database announced by the government this week.

Priority pressures

If clinicians and managers have information they trust about issues they care about, they can deliver changes that benefit patients. There is one small fly in the ointment. And that is the issue of competing priorities and national targets.

Mr Deshmukh knows the NHS focuses on performance information driven by the priorities of central government targets. He knows the government does it because it must provide assurance to taxpayers that investment is allocated effectively and efficiently. There is no doubt that targets have led to significant improvements in access to and the quality of services and the overall experience of patients. But they are inflexible. If your local data is telling you to focus on something different, you still have to meet the targets.

That brings us to the question: what information should the public have? In general, managers and clinicians are wary of putting too much information in the public domain. They are concerned that patients will be unduly concerned by apparently negative performance. They worry that the public is not able to understand complex statistical information.

However, a growing number of people in the public services are starting to recognise that greater transparency may be in their long-term interests. The reason is that transparency could perhaps allow a more flexible debate to take place about what the priorities for local services should be.

Currently, when people in public services complain about the impact of performance targets, the government replies: 'How else can we ensure that taxpayers' money is used to deliver real benefits?'

But this may not be the only way of ensuring public accountability. There is an alternative vision gaining increasing support in which targets are largely replaced by much increased transparency. A world in which patients and the public are able to examine data about their hospitals at almost (privacy respected) the same level of detail that a doctor or manager can.

If enough people and stakeholders such as local media and patient groups were engaged by this sort of radical transparency, it could provide a more responsive and flexible form of accountability - one that would free organisations from the tyranny of national targets and inspection. That is something doctors and clinicians could welcome with open arms.

Tim Kelsey is chair of the management board of Dr Foster Intelligence. Its report Intelligence: quality and safety is published today. It can be downloaded at