Expert Insight: How Big Data Can Help UK Cardiologists

Mamas A. Mamas, BM, BCh, MA, DPhil, FRCP


July 20, 2018

Welcome to Medscape. My name is Mamas Mamas. I'm Professor of Cardiology at Keele University, and today we're going to discuss big data.

So what do we mean by big data? Well, big data is extremely large data sets in which one can analyse and study patterns of trends, associations and clinical outcomes. And this is particularly relevant in healthcare systems because every time a patient interacts with a healthcare provider, be it primary, secondary, or tertiary care, an electronic healthcare record is produced. Big data is not new. William Farr is the father of big data in the United Kingdom and in 1800 he developed the mortality statistics in the United Kingdom. So he captured mortality data in individuals and then plotted these and studied these by age, by demographics, by marital status, and so forth.

Why Do We Need Big Data?

In the era of randomised control trials many people may ask, well, why do we need big data? You know, we have the gold standard evidence in the form of randomised control trials. Well, these are my thoughts.

I think big data is even more important in contemporary practice. In the United Kingdom, we capture routinely collected healthcare data from different interactions with the healthcare service. So for example, in procedures that we deliver, such as PCI procedures, every PCI procedure undertaken in the United Kingdom will have data captured. We then use this data to study outcomes, to study rare complications of patients, to study how we're performing these procedures, whether our outcomes are in line with the risk profile of the patient, in a way benchmarking our services. You may say, well, you can capture complications from randomised control trials, but this is real world data. Often the patients that we undertake these procedures in in clinical practice are much higher risk, much more elderly, much more common than those undertaken in randomised control trials.

Secondly, we can also look at trends. We can look at trends in practice, we can look at trends in outcomes, study what has changed these outcomes. So, for example, one of the pieces of work that my group has undertaken, looked at stroke complications, and we looked at whether stroke complications have changed over time following PCR, and whether there has been an impact by particular procedural technique or pharmacology using these devices.

Thirdly, we can use big data to study policy and its impact on healthcare. So for example, how does pay performance impact on clinical outcomes where you can't do this through randomised control trials? You can only look at this through clinical data. You can look at whether healthcare policy has been equitable, whether there's still north and south divides in healthcare provision and clinical outcomes.

We can also use routinely collected healthcare records to study the efficacy of interventions. Whilst the randomised control trial is considered to be the gold standard, there isn't a randomised control trial on every intervention. And certainly many high risk groups are excluded from randomised controlled trials. So again, studying the efficacy of the therapies that we give patients is very important. And the only way to do this is using big data.

Service Delivery

Finally, I think big data is particularly useful in understanding how our service delivery is impacted on. So, for example, looking at a service, what are the rates of re-hospitalisation? What are the implications of these re-hospitalisations? How can we structure services that reduce the rates of re-hospitalisation?

Furthermore, routinely collected healthcare data is very useful for health economic analyses. Whilst you can undertake these and randomised control trials, again, these are very specific populations undertaken in specific healthcare systems. How applicable is healthcare cost analysis in the United States and the United Kingdom, for example. So, again, routinely collected healthcare data is critical in being able to compare impact of interventions, outcomes associated with interventions, in one's own local healthcare system.

So in summary, I feel that big data has completely changed the way we practise clinical cardiology and medicine generally. I think there are challenges associated with big data and for example, there's a whole issue around governance and privacy, where we store these data, who has access to this data.

So I think I've raised a number of issues around routinely collected healthcare data in the UK and how I feel it is useful and how it can impact on clinical practice. What do you think? How do you use routinely collected healthcare data in your services?

You can follow Mamas Mamas on Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.