Most agree the 18-week target is a worthy one, but how can it be measured? Lyn Whitfield asks whether current technology can keep up with the good intentions

When the government announced its 18-week 'total' waiting time target in 2005, many people asked a simple question. How would it be measured?

The NHS, after all, had no way of knowing how long people waited from GP appointment to treatment. The only information available related to points along that journey - how long people waited for a first outpatient appointment or for inpatient treatment.

But then health secretary Patricia Hewitt was adamant: 'By December 2008, the whole patient journey will be managed and measured as one,' she said. The same press release acknowledged that 'changing the way we measure waiting times is not easy'. It didn't say that it would be a major IT challenge that existing NHS systems were not set up to cope with.

Complete system

The subsequent 18 weeks implementation framework promised a 'complete measurement system' to be agreed with Connecting for Health, the agency in charge of 'digitising' the NHS.

However this tracking system, which will be delivered through the secondary uses service, has yet to emerge. The British Computer Society has argued that 18 weeks is a classic example of a mismatch between policy and the IT and other infrastructure needed to support it.

'There is an urgent requirement to align Connecting for Health implementation schedules with the NHS reform timetable' said its Way Forward for NHS Health Informatics report, published last December. 'The poor alignment between the 18-week target and IT support is a case in point. Business objectives should drive information objectives which in turn should drive IT solutions.'

This is not to say no work has gone into defining, measuring and monitoring 18 weeks. Far from it. The first additional waiting time information was released when the Department of Health started to publish data on waits for diagnostic tests.

Meanwhile, 18-week pioneer sites were instructed to look at how existing patient administration systems could be used to generate referral-to-treatment time data.

The Department of Health was then able to run a snapshot exercise to determine baseline referral-to-treatment times in a number of specialties in December 2006, to provide trusts with basic reporting tools and to start publishing times for a number of specialties this spring.

However, there is still a lot to do. One of the most frequently voiced frustrations about 18 weeks is that what counts as a starting and finishing point is still being decided.

Another is that generating referral-to-treatment information is still very difficult, since 'tactical' solutions tend to rely on new paper elements and considerable manual effort.

Data deficit

Malcolm McFrederick, project director for service delivery and development at Princess Alexandra Hospital trust in Harlow, sets out the basic problems, namely that some patient administration systems cannot generate the data required and/or cannot link different events on the patient journey.

'We know people are waiting for outpatient and inpatient appointments, but there is no way to follow them from one to another without trying to make the links manually,' he says.

'We have a tactical solution. It's basically a giant database and folk enter information into it. Getting information out is a bit of a nightmare.'

Like other organisations, the trust has created clinical outcome forms to capture decisions made at various points along the patient journey. Where there is one clock-start and one clock-stop that can be associated with one patient, the trust can be reasonably confident about its reporting.

Things become more difficult when there are a number of starts and stops - perhaps because a patient is being seen by more than one consultant or treated for more than one condition.

'When that happens, we make some assumptions about what is likely to be happening and try to check those against paper notes,' says Mr McFrederick. 'But we are still turning up patients who seem to have been waiting for more than a year but are actually clock-stopped while they get fit from another condition.'

Connecting for Health is well aware that hospital administration systems will need to be able to capture key 18-week data. It says the DoH and its 18-weeks team are working to specify data requirements through formal standards and data set change notices, and to make sure patient administration systems can implement the principle notices by December this year.

Suppliers say they are committed to this. Cerner, for example, is working on a solution that will 'link together all stages of a patient's treatment as part of the everyday administrative process, including the referral-to-treatment information that indicates clock-starts and stops.'

Flexible solutions

Managers will be watching with interest to see what suppliers can deliver. Because while referral-to-treatment reporting is an issue, the real demand is for solutions that will allow both commissioners and trusts to track patients in real time and spot target breaches before they occur.

Connecting for Health says 'modifying the secondary uses service to provide support to the NHS in tracking individual patients along their 18-week pathway' is another of its key work areas. At the moment, it is concentrating on capturing data and ensuring it is in a standard format.

'The December 2007 release of the secondary uses service [the new single repository for all data relating to care of patients except that used for direct clinical purposes], patient care will enable the loading of these new items and reporting facilities will be made available from April 2008,' it said in a statement.

However, even if the work stays on track, April is pretty close to the December deadline for meeting the target. So it is perhaps not surprising that large trusts are looking to develop their own solutions, while consultants and other commercial analysts have spotted an opportunity.

Healthcare business intelligence specialist Ardentia, for example, have launched the pathway manager system, with input from Microsoft and existing customers including Dudley Group of Hospitals trust in the Birmingham.

The solution is built on a data-warehouse. It allows information to be searched, pathways to be generated and unique identifiers to be created for patients when information like the NHS number and their date of birth suggests they are moving along a single pathway.

Internal reports and referral-to-treatment returns are generated automatically, while the trust has given its host PCT controlled access so it can monitor patients through the system.

'Eighteen weeks is a huge project. The only way we could imagine delivering it was by having good information. We could not wait for the secondary uses service,' says trust deputy director of information John Uttley. 'Some trusts are developing their own solutions, but we did not have the technical ability that Ardentia had. And [many of those solutions] involve a lot of data input. We did not want to go down that route. Being able to do all this electronically is essential.'

Policy pragmatism

Policy, of course, never stands still. Some forward-thinking commissioners are already talking about giving patients access to 18 weeks data, so they can apply their own pressure to providers to get them through the system. The DoH is keen to capture patient expectations and outcomes as well. Yet these aspirations run even further ahead of the IT available.

In the meantime, there are still more than 500 days until the NHS has to deliver on 18 weeks, so it could be argued that it is not surprising work on the IT support is still going on. Yet that work is dealing with such basic issues as what counts as 18 weeks, what data is needed to create and monitor 18-week pathways and how to adapt systems to capture it.

Given the challenges that many organisations will face once they have this information, it could also be argued that it might have been better to have more of the necessary IT infrastructure in place before declaring the target. 'I'm not saying 18 weeks is wrong - far from it,' says a frustrated IT director. 'But before you announce a policy like this, you need to have the information structure, so people can see where they are and where they need to get to. You can do that and still make the aim itself challenging.'

Complementary policies: the choose and book factor

Connecting for Health is keen to stress that the electronic choose and book system can be used in support of 18 weeks. It is now describing the system, which started life as 'airline-style booking for the NHS' as 'an enabling tool to help measure and deliver the 18-week pathway'.

In a statement, it identified 'implementing choose and book as a standard method of referral and entry into 18-week pathways' as one of its key work areas. It also said, from December this year, that data on decisions to refer and other items will be fed into the secondary uses service, to provide a 'clock-start' for the 18-week period.

The agency argues that choose and book supports 18 weeks in other ways. For example, its 'advice and guidance' functionality allows referrers to seek specialist advice on referrals and conducting specific tests before referring.

Trust information in the directory of services can be used in a similar way.

Connecting for Health added that choose and book's clinical assessment services functionality allows referrals to be triaged. 'A clinical assessment service can be based in primary or secondary care. It allows for a rapid assessment of patients and for tests and investigations to be arranged before a patient is seen.'

If a decision to refer is made, the directory of services can be used to find the most appropriate service - including the shortest outpatient waits. If referral criteria are met, patients can also be booked into two-week wait and urgent slots.

On its website, the agency gives examples of trusts that have used choose and book in support of 18 weeks in other ways, for example to reduce did-not-attends.