Can big data mend a broken heart?
Kirstin Leslie, MRC PhD student at the Institute of Health and Wellbeing at the University of Glasgow, is the 2017 winner of our Max Perutz Science Writing Award. In her award-winning article she explains how she’s trying to find out why people stop taking drugs prescribed for preventing heart disease, and why this matters.
“When you do things right, people won’t be sure you’ve done anything at all”
That’s actually a quote from the TV show Futurama but it’s also a clear way of explaining why people are not always good at taking their medications. Imagine: you‘re taking a drug to prevent yourself from having a heart attack. But if you don’t feel any different after taking the drug, how can you know it’s even worked? Maybe you weren’t going to have a heart attack anyway? Maybe the drug you’re taking is giving you side-effects and besides, it isn’t worth it because you felt fine before. You don’t want to bother your doctor getting a new prescription and your blood pressure wasn’t that high anyway…So you stop taking your drugs and you hope for the best.
But heart disease is the leading cause of death worldwide. And it’s preventable.
This is particularly bad news if you happen to live in Scotland, where we lead the way in the number of heart attacks, strokes, and more-or-less everything that can go wrong with your heart. But why? We have access to lots of different types of drugs for your heart, and in each of these groups there are a range of specific drugs to choose from, so there should be something that works for everyone. Shouldn’t there? Unfortunately the problem isn’t that simple.
Once your doctor has identified high blood pressure, high cholesterol, or anything else that might increase your risk of a heart attack, a few steps have to happen to reduce that risk:
- Step 1 is prescribing a drug.
- Step 2 is taking the drug.
- Step 3 is the hard one.
Step 3 is taking the drug at the right time and continuing to take it for as long as you need to. And that’s hard. It’s hard because for the drugs we‘re talking about, ‘as long as you need to’ can mean ‘for the rest of your life’. And the rest of your life can be a long time.
So my research project will look at: how well people in Scotland manage to stick to Step 3, who is and who isn’t sticking to Step 3, and whether sticking to Step 3 does actually improve your chance of avoiding a heart attack or stroke.
To do this, I’m going to be using patient data from across Scotland to look back over the years and see whether or not people are picking up their medications from the pharmacy on time. This might sound straightforward before you remember that there are over 5 million people in Scotland. And, as I said before, we are not the healthiest bunch. So that’s a lot of people, with a lot of unhealthy hearts, and a lot of drugs prescribed by their doctors. To do this I’m going to have to enter a world that has always seemed distant, complicated, and honestly a little bit intimidating: the world of Big Data.
Firstly: what is Big Data? Is it Facebook working out your personality based on the number of cat pictures you like? Or apps predicting the next flu pandemic based on the number of people tweeting about a runny nose? Or targeted adverts based on your google history? The answer is, in a way, yes. Big Data is all those things and more. Big Data is what it says on the tin: data, but a lot of it.
For me, Big Data is looking at everyone in Scotland who has ever been prescribed a cardiovascular drug – or more simply, drugs for their heart – and looking to see if they picked up their next lot of drugs around about the time their first prescription should have run out. If they don’t, it means they are more likely to be skipping days, having gaps, or they might have stopped taking them altogether. By linking this to medical records I can see if people who aren’t taking their medications are statistically more likely to have a heart attack, stroke, or even die.
And by looking across the whole country I can also see if people are more likely to take their drugs if they fall into different groups: if they are older or younger, male or female, or if they are living in wealthier areas or not. By doing so, I will be able to see if certain groups of people are more likely to miss their medications, and with that information, I might be able to work out who needs help at sticking to Step 3. If we know who is at risk, we know who we can help.
And if we know who we can help, maybe we can mend a heart before it breaks.
Find out more about the competition.