Pathology, the present and the future

On July 8 2010, E-Health Insider convened a round table debate on clinical pathology, sponsored by iSOFT. The event gathered together some of the leading players in pathology in the UK, with industry, the NHS and academia represented.

The event was chaired by Ian Barnes, who is currently national clinical director for pathology for the Department of Health, on secondment from Leeds Teaching Hospital, although he was speaking in a personal capacity.

He started by setting out what he hoped the debate would cover, arguing that technology had moved faster than pathology had been able to respond.

“We are interested today in hearing from industry and from a user perspective what we think needs to be done from a national and systems perspective to move towards a service that is more patient focussed, better integrated and more interoperable.”

With that in mind, he asked the participants to introduce themselves, and set out where their interests lay:

Pathology: a complex environment 

The discussion started by outlining the complex scenario that pathologists now face. New technologies and new ways of working hold out the possibility of making efficiencies and increasing productivity, as well as adding to quality of care and patient safety.

Digital pathology, in which images of slides are digitised and can be shared, is one example of a new technology that could make a real impact on service delivery. Integrating pathology into patient pathways so that pathology could terminate a pathway based on a lab result as well as sending a result to a GP is another.

Meanwhile, the Carter review of pathology services has called for labs to consolidate to create economies of scale. In some places this has occurred – and delivered large savings.

Leeds and Bradford, for example, has created one central lab and, in the process, moved from 12 stand-alone Laboratory Information Management Systems to just one.

Yet with 70% of diagnoses dependent on pathology, managers view labs as cost centres, rather than as services that add value. Creating the business case for increased investment in LIMS, pathology systems integration, decision-support tools and new technologies has proved hard.

“The business cases I have seen have been very unstable,” said Dr Jones. “We need to convince managers that they get pay back from investment in pathology.”

This is not a problem that is simply caused by the current financial situation, either. Professor Batstone said: “Where we have been more successful is where we have had other clinicians, such as kidney or liver specialists, saying ‘we need this new technology or investment’.

“Where we can have sensible conversations with commissioners about how pathology can take loops out of pathways, then we have found that people start pressing on our behalf.”

It is, said Ian Barnes, a case of “getting away from a lab focus”. Nick Harte agreed but felt that examples of commissioners doing this are few and far between. 

“Until now, lab results have been pitched into a relative void,” he said. “In existing electronic care record systems, a combination of incomplete patient results histories and a lack of semantic context means that the support that can be offered to aid clinical decision making is sub optimal.

“There is an opportunity once such intelligent systems are in place to see radical changes in the way that care is delivered.”

Getting take-up of good practice

The round table agreed that this is beginning to happen in places, but felt the proven advances made by some cutting-edge departments have not transferred widely across the NHS.

Professor Murphy discussed work at Oxford Radcliffe NHS Trust to improve blood transfusion services; a well-known source of error in the NHS and one that hospitals have been asked to address.

The trust had looked at the transfusion process end-to-end, he explained, and asked whether technology could help re-organise it.  Clinicians had trialled and tested a bar code system that linked the patient, the blood to be transfused and the lab, using hand-held bar code readers and a wireless network.

Professor Murphy said: “It was piloted. We persuaded the trust to implement it throughout the hospital at some considerable expense by demonstrating savings, reduced blood usage and wastage. We developed a national spec for the system.

“Since then, we have been disappointed that there has not been take up by many hospitals. Now we are looking to build on what we have done in terms of patient safety, developing decision support systems to reduce inappropriate blood use.

“We are looking for other clinical applications of the system we have developed, for example in drug administration.”

George Philp was as baffled as anyone about the lack of uptake. “Why isn’t everyone using it?” he asked. “It’s very hard to answer that,” said Professor Murphy.

“Colleagues have difficulty persuading trust chief executives and boards to provide the start up funding to deliver it. Maybe they are not convinced by the business cases. Maybe we have to move to smarter business cases.”

This reinforced Professor Batstone’s earlier point. “If you try to get RFID up and running it will help all areas,” he said. “We need to corral all these areas to make the argument more convincing.”

Virtual microscopy or digital pathology was another good example of a proven – although still novel – technology that has not been taken up, said Professor Hamilton.

It offers real benefits, not just in terms of allowing remote working and sharing of slides with multi-disciplinary teams, but in improving diagnosis and selecting patients for appropriate treatments, he said. “Yet it has not been adopted.”

He went on: “There are labs beginning to adopt it and places that have championed it, places where pathologists have the ear of the chief executive.”

Professor Kay agreed that technology and process transfer was a real stumbling block. He cited the Leicester Thyroid Registry as another example of good practice that had failed to transfer.

“These problems are part of the wider NHS problems,” he said, and stemmed from the focus on computerising hospitals rather than using technology to support patient care.

 “We need to look at the clinical shape of the problem,” he went on. “There is no explicit commissioning of pathology and no one person responsible for delivering it… If no-one is responsible for commissioning or providing it, then it is no great surprise if you do not find solutions.”

Making the case for change

George Philp took the discussion back to the labs using a factory analogy. “Heinz makes tins of beans. Labs used to produce paper reports. This is now virtually all electronic, certainly to data repositories.  LIMS do all this.

“The pain points are outside that with the sharing of information. There is now a universal language for exchanging information – HL7v3 with a SNOMED dialect – and we seriously need to adopt these standards.”

Dr Jones agreed. “Hospitals buy LIMS then have a massive problem around interoperability and integration.” However, Nick Harte said this was changing. “Interoperability is now a main driver,” he said. “We cannot create something that’s excellent and sits on its own.”

There are other drivers too, pointed out Ian Barnes, referring to the coalition government’s health white paper – unpublished at the time of the debate – and only affecting England.

“I do not think pathology is in the right place for the direction that’s being set out,” he said. “The big issue here is about how the professions of pathology and industry work with clinical users to make the case for change.”

 Who, he asked, were the major players at this point? The Department of Health? The professional bodies? Industry? The general agreement was that all of these had a part to play.

Professor Hamilton referred back to digital pathology. “The professional bodies have a very strong role to play. We have been trying to co-ordinate with the Pathological Society, trying to engage and develop strategic studies and explore how it might be developed.”

And again, there was general agreement that technology was not the main issue, and the focus needed to shift to issues such as patient safety.

“We need a whole systems approach demonstrating improvements in patient safety and treatment, and how pathology can help get the right drugs to the right people at the right time, said Professor Batstone.

Participants also suggested that a lot could be learned from PACS and RIS about the benefits of digital imaging, its impact on service design, patient safety and patient trust, and the use of standards for sharing information.

Going direct to patients 

Ian Barnes raised another driver: the shift of services into the community and empowerment of patients. What, he asked, are the challenges? Professor Kay was upbeat as he discussed the monitoring of chronic diseases.

GPs, he said, have targets and monitoring systems under the Quality and Outcomes Framework. They receive the results of blood tests automatically, which they then have to view and report back to patients.

This is ripe for automation, he argued, creating a safe system in which GPs are alerted only to those results that are outside the norm and require clinical intervention. “GPs will then say ‘that was a success. What next?’.”

Professor Batstone took the idea further. “Results could be sent directly to patients. Technology can really help people to understand what their tests mean and empower them.”

He pointed out that the US healthcare organisation Kaiser Permanente is already sending blood work to patients, raising their involvement in their care and reducing transaction costs by 3%.

He also pointed out that Lab Tests Online, a not-for-profit site sponsored by industry and developed by professional bodies, already exists to help healthcare consumers understand why they have tests and what their results mean.

It was developed in the US and has now been adapted for the UK; Leeds laboratory service already has automated links to the UK version. Ian Barnes agreed that Lab Tests Online is a “fantastic product” but wondered where the drive was from pathology to get people to use it.

“I am not sure it is very widespread,” said Dr Jones. “One presumes that doctors are viewing it with the patients and then giving them the link. Whether patients are aware of how to use it is another question.”

Labs could even start sending data to iPhones or other patient-held devices, although there would be technical challenges around how to present the data and what to include.

George Philp said this was already happening in Greater Glasgow, where patients with HIV receive a text message about their HIV load. “It’s not done by pathology but by the sexual health service,” he said. “If you can send sexual health messages over the ether you can send almost anything.

“The healthcare professionals of tomorrow are the YouTube and MySpace instant messaging generation. Ask those people to go to a PC and log on and look up information and they will say: ‘no, give me an app to look it up or send me an alert’.”

Another question facing labs is the extent to which they continue to carry out tests in the face of safe, “idiot-proof” kits available over the counter. This is not without precedent, said George Philp.

Gone are the days when it was only labs that did pregnancy tests – home testing kits are quicker, as accurate and liked by consumers. Blood glucose testing kits are now available as is home testing for patients who are prescribed the anticoagulant warfarin.

With more near-patient testing, a new role may emerge for pathology in quality assurance, visiting high street providers such as Boots and Lloyds the Chemist. Pathology should be driving this change, said Ian Barnes, and developing the expertise to do it.

The importance of standards

He went on to raise a question of patient safety. GP systems, he said, are not designed the same way as LIMS. A number of problems have cropped up as a result: such as GP systems batching results from different labs when the results may not be compatible, or delinking results from the comments that contextualise and explain them. “How do we stop that?” he asked.

This was something that the DH was addressing, said Professor Batstone. “We need to have some pragmatism about what helps patients. The DH has funded a piece of work about how to handle this data, what should be combined and so on. It’s a start.”

Nick Harte pointed out that this had brought the discussion back to interoperability and standards. “We need definitions,” he said.

Dr Jones agreed that standards were important. “In other industries, standards are viewed as cost saving and agreed standards are preferred. That does not seem to have penetrated the health system.”

He added that it was a matter of getting knowledge into the systems. “There is a role for pathology to get well engaged with system suppliers to get that information out to suppliers,” he said. The Common User Interface also addressed some of this, he added.

So, with professional bodies pushing for innovation, the DH issuing guidance and developing standards, what was the contribution of industry in taking up these standards?

Nick Harte said: “There have been standards on healthcare for many years and we are moving in the right direction. But some of the implications of adopting certain standards are quite dramatic on the underlying technology of existing systems. It can be difficult to implement new standards as a result.”

Gareth Dellenty added: “I am concerned that pathology is going down a route with digital pathology where it does not adopt standards.”

Professor Kay threw this back. “Without supplier commitment, you do not have usable standards,” he said. “There is no alternative to them coming through suppliers but the question is what do we have to do to make the suppliers adopt the standards?”

George Philp pointed out the complexity of making systems interoperable. Blood banks, for example, need to talk to several systems such as the PAS, laboratory analysers - which have to be 100% accurate - GP reporting systems and the NBS for electronic dispatch notes.

“The amount of interoperability is quite astonishing. There is no common language adopted for LIMS to talk to other systems.”

Professor Kay urged pathology to adopt universal generic standards. “IP is everywhere,” he said. “The web works. It’s all about clinical meaning. It is not that people argue about standards, it is that we do not know how to present information to clinicians. The core of what we do has virtually no published literature on this.”

Better data for research 

Interoperability also came up as an issue when the round table turned to discussing the electronic care record and using data for research purposes.

Professor Batstone said: “Interoperability is needed to make sure that far more data is available to more clinicians wherever those clinicians may be. The question then is how that stuff is stored and how people access it.”

If researchers could access diagnostic information, there could be real benefits for future healthcare, he suggested. The vast amount of data in LIMS could potentially be tapped to answer questions about what range of drugs is therapeutic in different parts of the population, for example by age or ethnic group.

Nick Harte agreed. “Think about the number of patients in the iSOFT repository,” he said. “It is probably the world’s largest medical database. We are increasingly interested in becoming a knowledge-based company that adds value to the commercial offering.”

Funding and ‘disease shaped’ business cases

Finally, the debate turned back to finance as being the big, thorny issue to grasp. Ian Barnes said: “In today’s world, everybody has to justify their existence. There is a really important issue here about the whole knowledge surrounding tests and results. Not everybody believes that a test result should have anything but a number.”

Dr Jones said this was a global issue. Pathologists, he suggested, were experts in interpreting the clinical meaning of raw data, looking at the numbers alongside the clinical presentation to find meaning. “One of the roles of pathology is to write the algorithms for this and send them to the outside world.”

Professor Batstone agreed. “We need to move from being a results service to an interpretation service and then to a transactional service.”

Again, this would mean pathologists getting out of the labs and into the wider clinical setting. Dr Jones talked about “unshackling pathology from the yolk of acute care”, while Professor Batstone argued: “Until labs become something other than trust-based cost centres, you stand no chance.”

Pathology could show its value – rather than its cost – by looking at services around specific conditions, such as liver disease or diabetes, suggested Ian Barnes. Professor Kay called for “disease-shaped business cases” that showed the economic benefits that could be delivered through utilising diagnostic services to the full.

These sorts of conversations are now underway with the ‘czars’ (cancer, renal, heart) at national level, said Dr Jones.  “I think we are potentially at a threshold,” he said.

At the end of the debate, that was the general feeling around the table. Yes, there are cultural barriers to integrating pathology better into the whole system. Yes, there are technical barriers around interoperability and sharing data.

Yes, there are financial barriers to making sound business cases for investment in diagnostics. Yes, there are barriers to sharing innovation. But there is a willingness to tackle these barriers; one that is motivated by the need to do the best for patients.

George Philp summed up: “There is a high degree of unanimity about where we are now and where we need to go. I think we are gathering momentum and if we had others here round this table I am sure they would be saying the same thing.”

Go to top of this page | Go to E-Health Insider homepage