Feature / Added value
At the HFMA’s first international symposium on value, clinicians and managers agreed on the importance of good data and a common language as the core foundations for value-based decision making, Steve Brown reports
Data quality and the need for a common language between clinicians and finance staff. These were mentioned time and time again as the key ingredients for moving to value-based healthcare at the Healthcare Costing for Value Institute’s first ever international symposium on value.
A mixed audience of clinicians, finance and other managers were brought together in London in October to hear speakers from across the world and the UK talk about progress towards using the delivery of value – taking account of quality and cost – as the primary goal for healthcare services.
Setting the scene, Dr Paul Buss (above centre and right), medical director at Aneurin Bevan University Health Board, said there was growing evidence that better outcomes cost less and that changing clinical behaviour was key to closing the growing ‘value gap’.
He stressed that an evidence-based approach – understanding cost-benefit ratios and measuring clinical outcomes – was fundamental to unlocking this change in behaviour. In terms of the use of patient-level and activity-based cost data, he said clinicians had to start demanding ‘why aren’t we doing this?’ rather than ‘why are we doing this?’. Four questions needed answering, he said – ‘Can we develop costing systems that influence clinical behaviour? Can we develop systems where the cost element is used as an early marker of poor performance? Can we use costing to close the value gap? And can we develop costing mechanisms that accurately portray our clinical activity?’
Jason Neil-Dwyer, a plastic surgeon at Nottingham University Hospitals NHS Trust, is one clinician who has seen the potential of cost data. The trust was an early implementer of patient-level information and costing systems (PLICS). While clinicians initially responded negatively to a first cut of the granular cost data for his division, pointing out errors, he also saw how more robust data could support improvement in the way services were delivered. But first, finance practitioners and clinicians needed to develop a common language.
‘There is an inevitable clash of worlds,’ he added. ‘It can be a case of P-values [measures of statistical significance] versus £-values.’
Finance tended to be ‘linear, hierarchical and deal in absolute values’, said Mr Neil-Dwyer, while clinical care was ‘networked, interdependent with inherent variability’. A core example was the healthcare resource groups used as the currency to show clinicians their costs for various activities. Describing the HRGs for plastic surgery as ‘not obvious’, Mr Neil-Dwyer said they were ‘our work as imagined by someone else’.
The first task was to create a shared mental model of work done. This involved clinicians acknowledging the starting point was that there was a £3m deficit on the departmental budget. For finance it meant presenting costs in something other than HRGs. ‘We needed to express the financial concerns in a clinical “work as done” context,’ he told the conference.
Refocusing exercise
For plastic surgery at Nottingham, this meant initially examining procedure codes grouped in a way that made more sense to the clinicians and then focusing in on the areas where most difference could be made. Examining the 30 codes where the service made the biggest absolute losses and where it exhibited the biggest unit cost deficits helped the team to identify trauma and skin cancer as the agreed areas to target for improvement.
‘Now my colleagues could see the cost in terms they recognised and finance understood the clinical pathways generating the costs,’ he said.
Using quality improvement methodologies and with clinicians now welcoming the PLICS detail, the department looked at areas including the costs involved in single prolonged stay cases, complex breast reconstruction and skin cancer pathways.
For example, in breast reconstruction, it identified that more than 80% of costs were in theatre, with an operation time (7.5 hours) significantly above the international gold standard. This has led to improvements, with an extra consultant on each case helping to bring down theatre time – two cases can now be undertaken in a lengthened 9.5-hour theatre session. It has also helped mount an evidence-based challenge to the current tariff rate set for the procedure.
Duncan Orme, deputy director of finance at the trust, said that with the trust still making a significant underlying loss, the work in plastic surgery and other areas had to be expanded to other parts of the trust. ‘The theory is beginning to set,’ he said. ‘We have half a dozen examples of senior clinicians leading motivated teams in using good information [to drive value]. We now have to make that the norm in our hospital, not the exception.’
There was support for ensuring a focus on quality as a means of engaging clinicians. However, Dr Buss said the trick was to ‘bring money into the room and discuss it in a clinical context’. Mr Orme added that making sure data was published regularly was also important. ‘If you produce cost data once a year, clinicians will probably have forgotten the patients concerned, but if it is six weeks after discharge then they are more likely to be able to remember specific details.’
There was agreement that focusing on value in acute and community services was a good starting point – with a focus on improving safety, flow and starting to build links with wider quality and outcome data. But there was also recognition that in future the focus should broaden to include primary care, enabling clinicians and managers to target best value across the whole patient pathway.
The importance of ‘using the same language’ was raised
again in a workshop led Dr Jean MacLeod and Stuart Burney (right) from North Tees and Hartlepool NHS Foundation Trust. They said the integrated trust was still trying to improve communication across its sites, almost a decade after taking on community services, which itself had followed an acute merger. ‘So clinicians talking to finance is just another layer,’ said Dr MacLeod, a consultant physician and the GP liaison officer at the trust.
However, clinical teams and the finance department have worked well together to put in place and build a business case for a completely revised chronic obstructive pulmonary disease pathway. The traditional model saw patients with exacerbations of COPD being frequently admitted to hospital – often at some distance from home and with poor transport links. Oversight was often by generalist teams rather than specialists and, as part of the admission, the patients would typically have a chest X-ray and numerous blood tests. Average length of stay was just over six days.
Hospital at home
The aim under a new hospital at home model was for patients to be cared for in their own homes, with guaranteed specialist input and no travelling for them or their relatives.
With the commissioner only committing to a one-year contract for the revised service (as part of a local block contract), the challenge has been to demonstrate the new model works both for patients and financially for the commissioner and provider – and that has meant closer working between clinicians and finance.
Dr MacLeod said the original business case was based on a number of assumptions and predictions for the impact of the revised service. ‘There were no concrete examples we were copying and nowhere with a comparable set-up to the trust,’ she said, adding that the trust had targeted an ambitious 50% reduction in admissions over two years.
With five months’ real data since starting the service in April, over the summer the trust started to build its business case for continuing with the new model. By the end of August, it had counted 176 avoided admissions along with more than 50 early discharges made possible
by the enhanced home support. In both cases the numbers are trending upwards month by month.
‘There is no doubt this is the right clinical thing to do, but now we have to make the business case as well,’ said Dr MacLeod.
Even on the early numbers while the service was still taking hold, projections suggest there will be a 26% reduction in admissions (compared with 1100 previously) over the full year. This translates to a reduction of nearly 1,700 bed days, and reductions in both the number of blood tests ordered (£3 per patient) and chest X-rays (£35 per patient). There are also unanticipated savings from reduced use of tiotropium inhalers. While new inhalers often had to be handed out in hospital settings, seeing patients at home means they can access their existing inhalers (and get them checked). Clinicians were largely unaware that these inhalers cost around £32 per patient. In total, the trust is confident it will save at least £450,000 over a full year.
Multiple benefits
The biggest benefits are for patients – avoiding unnecessary admissions and greater control over their condition. But the financial case is also stacking up. However, the trust is keen to examine other metrics that it should be collecting to help reinforce the case and to improve the service.
For example, Dr MacLeod said the trust wanted to build in patient experience and link test results with the data set, in part to verify exactly how many of the ordered blood tests are looked at.
Head of contracting income and costing Stuart Burney accepted that the trust would lose out in net terms from losing the income tied to admissions, even with the new community service being separately funded. However, he said the trust needed additional acute capacity, and the changes would support this. ‘So although we may lose income, we will be more cost-effective,’ he said.
He also encouraged finance managers to investigate the data currently held by clinical audit teams. ‘We submit over 300 sets of audit data,’ he said. ‘Everything we do [through our PLICS system] is based on data that is already there. Go and talk to your clinical audit and effectiveness teams and present this data alongside your PLICS data,’ he said.
One thing was clear, all international health systems face similar challenges. And many recognise a focus on value – rather than simply cost or quality – as the best way to approach them. The HFMA’s symposium is likely to be the first of many.
• For details about the HFMA Healthcare Costing for Value Institute, see ‘our networks’ at www.hfma.org.uk. Institute members can also access videos from the symposium
Data processing
Prescription has played a big
part in creating a
foundation of meaningful patient cost data compiled using a consistent approach in Germany. INeK (the Institut für das Entgeltsystem im Krankenhaus) is the body that oversees the hospital remuneration system. Its head of economics Michael Rabenschlag (above) gave symposium delegates an insight into the country’s 15 year journey to improve costing, creating what is now a leading example of a large scale diagnosis-related group costing/pricing system.
The organisation is in fact responsible for more than just costing – collecting inpatient activity data, maintaining the currency, setting coding rules and producing grouping software as well as overseeing the patient costing and tariff setting process.
While the hospital inpatient activity collection is mandatory for all hospitals, cost data is submitted on a voluntary basis. Nearly 250 hospitals are part of this costing sample and they are paid for their efforts, covered along with InEK’s running costs by a €1.15 surcharge on all invoices submitted by hospitals. For larger hospitals, Dr Rabenschlag said the funding could be enough to cover three people working on producing the costing data. And with InEK-focused work only taking about half their time, this left hospitals with a paid resource to get value locally from the cost data.
Echoing comments from other speakers, he said that data quality was vital – with data of insufficient quality rejected from the sample either partially or wholly. Some basic data checks are built into the submission system, such as missing costs in cost centres and comparisons with previous years, while more sophisticated checks look at coherency and variation, for example, ensuring appropriate prosthesis costs are included for a hip replacement.
Cost data is collected in a matrix covering 11 cost centres (such as wards, intensive care and theatres) and 10 cost types (arranged in three broad groups covering staff, consumables and overheads) and the allocation methods for each cost type in each cost centre are mandated.
The Australian healthcare system has long collected patient-level cost data, using the data to inform an activity-based funding system for hospitals. But in the last few years, the state of New South Wales has stepped up a gear and made firm moves towards using the data to support real activity-based management. This has been underpinned by a portal that enables clinicians and managers to analyse and compare activity and cost data across health districts.
Deputy director of the state’s Activity-based Funding Taskforce Alfa D’Amato (above left) told the symposium that the state was taking the use of patient data ‘very seriously’ now and that the portal had helped to embed patient-level costing across the system, with single submissions of data now being put to multiple uses. He said there were four key stages in the journey: improve the data; improve the process; have an impact in terms of better care; and then deliver improved patient outcomes.
Mr D’Amato (right) described the portal in detail in
Healthcare Finance (July/August 2016 p23), but he told the symposium that getting data fit for purpose was key. Without this, it would be difficult to get the all-important clinical engagement. There is extensive data validation, with the data submission platform open for four weeks and hospitals encouraged to submit improved data in response to any issues flagged by the system. A ‘reasonableness and quality’ app scores the data quality to support practitioners with these submissions. The data is also subject to local audit as part of the internal audit programme.
New South Wales has already started to see a reduction in the rate of increase of average costs – and believes the portal has played a part in this improvement. It recognises it still has a way to go before it has fully embedded value-based decision making in local hospitals – however it is clear that system leaders see huge benefits in the approach and the portal and approach has now been expanded to the whole of Australia.
Capitation and outcomes
With the NHS now exploring accountable care models, Santiago Delgado Izquierdo, described how the model was delivering benefits in Valencia. The Ribera Salud Hospital
System has been on a journey towards more integrated care for more than 15 years and is now seen as a pioneer of capitated payment approaches in healthcare.
Mr Izquiero said the hospital was no longer the centre of the system and the culture was now one of financing health, not healthcare. ‘Capitated payment gives predictability for the government and makes us accountable for the healthcare outcomes of the population,’ he said. ‘And it incentivises providers to take on greater financial risks and invest in preventative care and treatment in the most appropriate settings.’
Thomas Kelley, a vice-president at the International Consortium for Health Outcomes Measurement (ICHOM), told the symposium that health systems needed to get better at measuring what really mattered to patients for specific conditions. Structural indicators (is there a stroke unit?), process measures (was an intervention performed in a certain timescale?) and clinical measures are important – but what the patient wanted to achieve is also vital. For prostate cancer patients, this might mean continence or sexual function, not just survival. ICHOM now has 21 agreed standard outcome sets and is overseeing international benchmarking pilots in two service areas – cataracts and hip and knee osteoarthritis.
Erasmus Medical Center in Rotterdam was an early adopter of ICHOM outcome standards and is a good example of an organisation putting value-based healthcare theory into practice.
By mid-2017 it plans to have outcomes in place for 36 service areas, some adopted from ICHOM standard sets, some self-developed. Chief medical information officer Jan Hazelzet said that value-based healthcare was written into the hospital’s strategy. Outcomes – including before and after questionnaire responses – are now part of normal practice and always discussed in consultations.
Related content
The Institute’s annual costing conference provides the NHS with the latest developments and guidance in NHS costing.
The value masterclass shares examples of organisations and systems that have pursued a value-driven approach and the results they have achieved.