Is data the key to curing the health service?

The problems besetting the Irish health service are well documented and seemingly resistant to all the solutions proposed thus far. However, according to KPMG head of healthcare Frank O'Donnell, significant improvements in quality of care and overall efficiency could be realised through better use of the data generated within the system.

He believes this data could be used to support better decision-making and deliver better outcomes for patients. The problem is that, in a very complex system, the data cannot easily be collected and accessed.

“The challenge with anything around healthcare is that the whole area is so complicated and complex,” he says. “This is due to the human dimension. There is a natural element of unpredictability around when people want to use the service. But there is a lot of data collected in various forms throughout the health sector and there is great potential to collect it, use it and share it in ways that help make the service better and more efficient.”

This applies to all levels of the system. “At its simplest, it gives the clinician better-quality information about the patient, and that has to be good,” O’Donnell says. “When you move upwards, it can be at an organisational level where you can use it for rostering staff to better meet demand patterns and help make more efficient use of health funding.”

Population health

A higher level again is population health. Measuring this takes into account the various social, economic, physical and environmental factors that affect health, as well as personal practices and lifestyles. Data focuses on the conditions and factors that influence the health of populations, identifies variations in their patterns of occurrence, and uses the resulting knowledge to develop and implement policies and actions in response.

“Data is a key requirement of a population health approach,” says O’Donnell. “We need more and better data if we are to realise those benefits.

“Getting the data is the challenge. Some areas are more difficult than others. Take primary, social and home care, for example: there is not much collection of data around what’s happening there. It is not well defined and there is no technology to support it. Even where there is a lot of data collected, there is still a question of what use it is put to.”

He points out that Health Service Executive and Department of Health reports contain a wealth of potentially useful data, which needs to be turned into knowledge and insight. One of the barriers to progress thus far has been the culture.

“When we look at the issues, they can be simple things but difficult to change just the same,” he says. “Culture is one of those. We need to create a culture where everyone values data, where they understand that little pieces of data which might appear to be minor and unimportant are actually very valuable when used appropriately and combined with other data.

“Some corporates actually value data on their balance sheets. That’s the sort of thinking we need to encourage in the health service.”

Better technology

Technology has a role to play as well, of course. “If you have better technology at the point of care, you can get better data in a more useful form,” says O’Donnell. “This will help with the development of electronic patient records (EPRs) in future. These records will mean that if you live in Dublin and you are on holiday in Kerry and visit a GP or a hospital, they will have access to the full record of your medical history, your allergies, previous prescriptions, and so on.

“That will lead to better quality of care for patients, speed things up and reduce the capacity for mistakes. Other countries have already gone down this road and are quite advanced on it. We need to do that as well.”

The actual gathering of large volumes of data is less important than what’s done with it afterwards, according to O’Donnell. “Sharing the information is where the value is delivered. At GP level, for example, they all have some sort of system to record patient visits and what happens during them. The question is how that information gets shared. The technology exists there to do that, but is not being utilised.”

From a macro perspective, data collection and analysis can also be used to plan and design health services. “To begin with, you can use it to analyse population groups, such as people with diabetes,” O’Donnell says. “We can use that information to help keep those people out of the health service. In reality, it’s a sickness service, not a health service. If you are healthy, you’re not using the health service. We need to focus on keeping people well. That’s the best way to improve people’s overall health.”

Data can also be used in deciding where to put new healthcare facilities, based on demand patterns and the behaviour of populations, he adds.

Greatest benefit

Perhaps the greatest benefit that better use of data can offer is an understanding of exactly why things happen in the system. “Very often, what we think are the reasons for something happening turns to be incorrect,” O’Donnell says. “You need to look at a number of data points to see the whole story. People talk a lot about health outcomes. The challenge is to relate them to what is actually happening.

“Cancer is one area where outcomes have improved, but causality is difficult to establish. It’s not impossible, and that’s where data analysis comes in. That’s how we can achieve the good outcomes we are all looking for in terms of improved health and quality of care for the whole population.”

Achieving this will require a combination of knowledge and skills. “Data will only take us so far,” O’Donnell says. “We need experienced analysts as well, data scientists who understand how the health system works.

“We need to combine that knowledge of the health system with expertise in data to be able to read the signals and draw the right conclusions. If you have the right data, you will be able to look back and gain insights and see the patterns which only become visible over time. This will allow the factors that deliver good outcomes to be more readily identified and replicated across the system, and decisions can be made based on evidence and good data.”