Call to measure patient outcomes and move to evidence-based health budgeting

Europe urgently needs to find ways to extract more value from healthcare budgets to put its health systems onto a sustainable footing, but in order to do this governments need detailed and actionable outcomes data to make informed decisions.

Standardisation of how data is collected and shared is essential to accelerating this shift to patient-centred care, according to experts at the first in a series of Healthy Measures events hosted by Science|Business in Paris on 8 November.

Outcomes are central to the European Commission’s main objective of supporting EU Member States through evidence-based policymaking, said Martin Seychell, Deputy Director General at DG Sante. There are “considerable gaps” in knowledge about health outcomes and this presents policymakers with the challenge of making difficult decisions in an uncertain environment, with incomplete information about what works for patients.

“There is a big shift in thinking away from a hospital-centred model towards patient-centred care,” Seychell said. “To ensure that this is more than a mantra we must measure patient-reported outcomes and record their experiences.”

Evidence-based policymaking

The European Commission can help to ensure that outcomes are measured in a scientifically-valid way and this information can be shared to help member states as they redesign their health systems. Seychell pointed to the Commission’s 2014 communication, ‘On effective, accessible and resilient health systems’, as an example the EU’s contribution to the discussion on improving health outcomes.

The EU has other instruments, including the European semester process, which can be used to support member states in optimising their expenditure. In addition, the Commission published a report on ‘Health Care and Long-Term Care Systems & Fiscal Sustainability’ in October 2016. This work includes reports on all member states’ health systems which highlight the need for well-defined outcome measures.

The diversity of Europe’s health systems is often cited as a disadvantage when it comes to replicating success stories. What works in the UK might not transfer easily to Belgium, and vice versa.

However, Seychell sees this variety as a strength. “Europe is an ideal laboratory,” he said. “It gives us a unique opportunity to determine the success factors; to see what works in different contexts.”

During the discussion that followed, some participants suggested there is a general reluctance among policymakers to embrace transparency on health outcomes. However, several speakers said transparency is the order of the day and would permeate all aspects of policy.

There is some concern about how data is used. Making superficial comparisons between hospitals or countries, without acknowledging the different contexts in which health outcomes are achieved, would discourage openness and data sharing.

Controlling variables

There are wide variations in outcomes both within and between member states, said Christina Akerman, President of the International Consortium for Health Outcomes Measurement (ICHOM) (see box).

Furthermore, current success indicators do not always capture what patients value. “We need to ask what outcomes matter most to patients, such as improvements in their symptoms, functioning and well-being,” Akerman said. “It’s interesting to see that when we scratch the surface of outcomes, we all prioritise survival, but also having as few complications as possible.”

Comparing five-year mortality following treatment for prostate cancer at the Martini Klinik to hospitals in the rest of Germany, the outcomes are almost identical. But raw survival data obscures large differences in other outcomes, such as incontinence and erectile dysfunction, that can have a profoundly negative impact on quality of life.

Such wide variations in outcomes prompted ICHOM’s global initiative to unlock the potential of value-based healthcare by defining global standard sets of outcome measures that matter to patients.

By driving the adoption of these measures, along with greater transparency in the reporting of outcomes data, ICHOM hopes to “cure inefficiencies” in healthcare.

To date, 21 standard sets have been published, covering 45 per cent of the global disease burden. By next year, ICHOM hopes to have covered around 59 per cent of disease burden. Some countries, such as Sweden, are exploring how they can use standard sets when commissioning health services. The National Health Service in Wales plans to use the standard set for cataracts when procuring technologies used by ophthalmologists.

Akerman emphasised the importance of transparency, noting the value of disease registries in improving outcomes. A Swedish registry on outcomes of patients admitted to hospital following heart attacks not only improved the performance of laggards within two years of making data public, it also raised average outcomes.

“Transparency is a very important part of accelerating improvements for all,” Akerman said. “We need to work with others to make this happen.”

Benchmarking

Patient-Reported Outcomes Measures (PROMs) are set to take centre-stage in the drive for more patient-centric healthcare delivery, according to Ian Forde, Senior Policy Analyst at the OECD Health Division, outlining the work of the OECD in this area. While collecting and sharing data is essential, governments want information that is actionable, granular and which incorporates patient input.

“We now collect data not just based on national averages but looking at differences between hospitals,” Forde said. Variation between hospitals in terms of survival rates will feature in the 2017 edition of the OECD’s Health at a Glance report.

The OECD’s High-level Reflection Group is pushing for better benchmarking of PROMs at disease level, sector level, service level and systems level. It wants improved collection of data in challenging areas such as informal care and multi-morbidity, and would also like to be better able to make international comparisons.

There is diversity in the kinds of data collected by OECD members. The most common disease groups for data collection include elective surgery, cancer and mental health, while some have moved into palliative, emergency and long-term care.

“Even in areas like knee replacement there is variability in the kinds of PROMs that are used and how the data is collected,” said Forde. “Some use paper-based surveys while others do it by phone, and a number of countries are looking at mobile apps, registries and embedding PROMs in electronic health records.”

Delegates agreed that PROMs can be used to help patients and doctors make decisions on the most appropriate treatment option. PROMs have also been used to determine the price health authorities pay for services. In one Swedish case, 10 per cent of the reimbursement of spinal surgery is dependent on whether the patient considered the outcome to be satisfactory. It also helps decision-makers to opt for products that perform best in terms of PROMs.

It was broadly accepted by contributors to the debate that while PROMs are becoming more widely used, there is too much diversity in how they are collected and used. However, speakers stressed that surveys that collect PROMs cannot simply be translated and applied universally, they must be culturally adapted and psychometrically tested in the target population.

Investing in improvement

There is evidence that PROMs can drive positive change in healthcare, said Martin Ingvar, Deputy Vice-Chancellor, Future of Healthcare at the Karolinska Institutet. “Where we have implemented PROMs we have demonstrated vast and quick improvements, meaning that it is a very good investment.”

For example, in one case in Sweden, tracking inflammation indicators in rheumatology patients reduced the need for joint replacement.

Such initiatives can inspire others to embrace the collection and sharing of patient outcome data. Ingvar also highlighted innovative approaches to data collection, for example, the use of smartphones to elicit feedback from patients receiving cognitive behavioural therapy.

There is scope for innovation too, in using outcomes to set prices. One pilot project in orthopaedic surgery offered hospitals 110 per cent of the usual price per procedure on the proviso that the hospital covers the costs of addressing anything that goes wrong.

Denmark also has experience in using PROMs to drive towards value-based healthcare. Five Danish regions have embraced PROMs to varying degrees and this approach may ultimately inform funding decisions, said Adam Wolf, CEO of Danish Regions.

By doing pilot projects it is hoped to apply what works in a national setting. “We are trying to move away from activity-based funding towards value-based healthcare,” Wolf said. “It can be very complicated but ICHOM has shown what is possible.”

Speaking the same language

If Europe is to collect and share PROMs data, it will be essential to measure the same things in the same way, said Sylvie Bove, CEO of EIT Health, one of the largest healthcare initiatives worldwide, which is supported by the European Institute for Innovation and Technology (EIT).

“It is encouraging to see progress in standardisation,” said Bove. “We must use similar approaches so that we can compare data and use it to make decisions.” As patients play an increasing role in managing their own health, access to usable data is essential, Bove said, giving an overview of EIT Health’s work in this area.

Amongst other activities, EIT Health is supporting competitions for young companies. One call for innovation attracted a start-up developing wearable sensors to monitor epilepsy patients and another which has developed a device that uses artificial intelligence to steer diabetes treatment.

The data generated using these technologies could prove to be very valuable if policymakers prioritise semantic interoperability, allowing computer systems not only to exchange information, but also to interpret it, Bove said.