The health service is collecting more data than ever before, but it can only help patients if it knows how best to use it, say Hilary Thomas and Lizzie Tuckey

Water deluge


Harvesting and analysing data has long sat at the heart of how we understand healthcare and develop our responses.

‘We are collecting more information now than at any point in healthcare history’

However, we are still largely operating in a world of structured “small data”. Electronic health record systems, for example, are critical and certainly offer significant benefits locally, but we should remember that there are bigger questions for healthcare that local data cannot answer.

Without better population based data, we cannot reliably predict outcomes for the majority of patients and we cannot easily see whether one drug adds more value than another, beyond the boundaries of a clinical trial.

Yet “big data”, which entails unstructured information, offers a new approach to making what is invisible, visible. Globally, healthcare is now at the centre of a data boom that may prove to be one of the most significant drivers of change over the next decade.

Recent years have seen an unprecedented growth in the volume of data available, from sequencing the human genome to tracking cancer treatments. We are collecting more information now than at any point in healthcare history. Of course, collecting data is one thing, but we still need to answer questions about what we need to do to ensure we capitalise on the opportunities it provides.

Creating trust

In the retail industry, huge amounts of consumer data is already enabling personalisation and optimisation, providing massive efficiencies and revenue opportunities. While similar personalisation and early warnings present clear opportunities in healthcare, it is important to note the sector is different. Healthcare is about patient outcomes and care.

It is possible to be independent, tech savvy and data driven while maintaining compassion and putting the patient at the centre; however, there are some pressing challenges and issues.

There is a clear tension between the desire to find the signal in the noise and the fear of loss of privacy. We are not accustomed to managing data as an asset across organisations. This is a major governance issue for the public sector where there are thousands of interrelated organisations and highly sensitive data.

‘If a retailer recommends the wrong book for you, you probably don’t care. If an oncologist recommends the wrong drug, you will’

Progress is being made, though. Health secretary Jeremy Hunt and NHS England have committed to a March 2015 deadline for the introduction of the electronic referrals − part of the overall ambition to give people ownership of their own data and, even when anonymised, the opportunity for it not to be shared.

But there won’t be a data revolution until some rules get sorted. Much as we currently access sensitive personal information such as personal banking online, the same trust in security must be developed for our health data.

If we can cope with the privacy issues, then wider access to data will challenge our ability to analyse it as data insights do not emerge fully formed. After all, big data may make decision makers more certain − but not necessarily right. If a retailer recommends the wrong book for you, you probably don’t care. If an oncologist recommends the wrong drug, you will.

New rules of engagement

Of course, the risk of big data is not an explosion but an implosion, if we allow people to become paralysed by the growing tide of information without making it useful and useable.

James Hope-Falkner, creator of Medimapp − an app that maps user reviews of NHS services from its own NHS Choices data set − illustrates the problem when he says: “The data we pull in is vast and unwieldy; comes only in one lump rather than it being possible to pick and choose what is relevant; is confined to England over Wales, Northern Ireland and Scotland; and suffers from server related slowness, out of date and missing information, countless inaccurate service coordinates and poor feedback.

“A huge amount of time and cost has been spent reprocessing that data into something useable. Any other developers looking to build Choices powered locational apps will unavoidably find themselves starting from scratch in terms of database architecture.”

‘The objective will be to enable patients to take a more active role in their health’

That’s why we also need to share best practice on managing data. Examples of what is possible with coordination and transparency are visible in the life sciences industry − most notably with the 10 global pharmaceutical companies that joined forces last year to form the research group TransCelerate Biopharma.

We will need to engage with our patients differently − through different channels and continuously, on an iterative basis. Increasing innovation will see new methods of treatment and management of care, and the objective will be to enable patients to take a more active role in their health.

Further developing how we analyse and search data, enabling live processing to consider information at the point of care, means we can truly personalise treatments that benefit everyone. The NHS can − and needs to − grasp the opportunity. This means asking the right questions and ensuring we are well placed to answer them.

Professor Hilary Thomas is a partner and Lizzie Tuckey is an adviser at KPMG’s healthcare practice