07 Nov 2016   |   Viewpoint

Countries need to dig deeper on patient outcomes data to improve healthcare systems

The head of the health division at the OECD, Francesca Colombo, says members are making poor use of health data. She is calling for a broader collection of personal health information, to reflect the outcomes that matter to patients

Countries need to cast a wider net and collect a broader range of personal health information, in order to have a better idea of how well they are performing, according to Francesca Colombo, head of the OECD’s health division.

“What we’re critically missing is patient perspectives of care,” Colombo told a conference on improving health system performance at the European Policy Centre in Brussels last week.

Colombo called on governments and health systems to change what they measure, going beyond traditional patient indicators such as mortality or life expectancy, towards a more relevant and important set of statistics on care quality.

“We’d like to capture quality of life, capacity to manage an illness, ability to function and the ability to live without care. We’re trying to move towards having patient-reported outcomes – but it’s going to take some time,” Colombo said.

Since 2003, the OECD has published comparisons of how the health systems in its 35 member countries perform in its ‘Health at a Glance’ series.

The data have been powerful in informing policy in some countries. For example, when South Korea saw its lowly ranking against other OECD members in 30-day mortality after admission to hospital for a heart attack, it put a huge amount of effort into improvements.

Similarly, comparative data on unplanned admissions to hospital for people with diabetes show huge variations and drove policy change in France, which was doing badly.

Prevention not cure

Better patient data will help healthcare providers prevent illness—and not merely treat it, Colombo said. Knowing more about how patients live after treatment could prove useful as doctors look for clues to poor health and tailor interventions to address patients’ needs, potentially preventing illness and saving money.

The big obstacle is not availability of data, of which there is an astounding amount. “It’s a paradox – the health sector produces the most data, but somehow it’s not being leveraged to improve performance. I believe health systems use data poorly,” said Colombo.

Teasing out the appropriate information from the reams of data is tough, despite advances in big data analytics.

“We hear a lot about big data and the effect it can have on health systems,” said Colombo. “The reality is more complex. Regular data is not collected or aggregated in a good way. There’s a missed opportunity to use big data to cut out waste and improve surveillance and research.”

Only half of OECD countries have comprehensive electronic health records in hospitals and primary care centres, according to Colombo. Hospitals have many different datasets and they are rarely linked together.

The European Commission is encouraging national authorities to develop Health Systems Performance Assessments (HSPAs) in order to identify where improvements are needed and to measure progress. It wants to see stronger country assessments and inter-country comparison tools, to help member states modernise their healthcare systems.

The OECD is working with the Commission on this, developing a new generation of health statistics, which will include better measures of patient outcomes. Such globally standardised datasets have the power to spur quality improvements both within and between countries.

Colombo said, “We’re trying to become more granular and are starting to collect data directly from hospitals, rather than just relying on national data.”

This is to circumvent the problem that national health averages hide huge diversity in performance inside countries. “It’s extraordinary – for the share of hip replacement operations in a country for example, you can see a difference of up to six times between regions,” Colombo said.

OECD health reports aim to collect data that is at the most two years old. But this should be improved on. “How can we support policy if we’re working on minus two years?” Colombo asked.

Never miss an update from Science|Business:   Newsletter sign-up