If you’re stranded without a weather forecast and you want to know if there’s rain ahead, you might be better off looking at creatures than clouds. Before our own bodies pick up the rising humidity and falling pressure, honeybees get busier, birds change how they nest and frogs start singing love songs.

These behaviours aren’t based on predictions. Birds, bees and frogs respond to the subtle atmospheric changes that come before rain. Similarly, when sharks sense changes in water pressure associated with approaching hurricanes, they dive deeper to stay safe. Elephants can even hear storms up to 150 miles away.

We can’t hear tomorrow’s weather, but we do shift our behaviour according to slight changes that happen deep inside us – even if we don’t fully recognise them. So, sometimes, the best way to understand how someone’s feeling is to pay attention to the way they act.

That got scientists at Imperial College London thinking. With collaborators across the UK, they’re investigating whether we can monitor patterns in what people buy to spot the signs of cancer earlier.

The Cancer Loyalty Card Study

Today, with our funding, the team are launching the second part of their Cancer Loyalty Card study (CLOCS-2).

In part one, they found that changes in over-the-counter medicine purchases could indicate early signs of ovarian cancer up to eight months before people were formally diagnosed. Now, they’re building on what they’ve learned with a bigger study across ovarian cancer and nine other types: bladder, bowel, endometrial, liver, oesophageal, pancreatic, stomach, uterine, and vulval.

People with these cancers often have vague or non-specific symptoms, such as bloating, indigestion, or fatigue – the kind of things we tend to manage with over-the-counter medicines before going to the doctor. 

CLOCS-2 aims to uncover subtle trends in the way people buy these medicines and identify some human equivalents to animal’s pre-rain behaviours. The better we understand these trends, the more likely it is we’ll be able to spot early signs of cancer and start treatment sooner, when it’s more likely to be successful.

“This study ultimately has the potential to revolutionise how we can use everyday data to understand and improve people’s health,” said Professor James Flanagan from Imperial’s Department of Surgery and Cancer, who’s leading the research. “We’re excited to build on the success of our first CLOCS study and to see if looking back at people’s shopping history can give us clues to the way different conditions start to appear.”

How the CLOCS go forward

CLOCS-2 is being made possible thanks to a collaboration with Boots and Tesco.

The research team will recruit 1,450 cancer patients and 1,450 healthy volunteers through GPs and the Be Part of Research platform, which is run by the National Institute for Health and Care Research (NIHR). They’ll then use up to six years of participants’ Tesco and Boots loyalty card data to identify differences in shopping patterns between the two groups.

The goal is to define a purchasing ‘threshold’ that sets people with early-stage cancer apart from other groups, as well as to identify what products are purchased for each cancer type.

Finding out this information will help researchers understand when to encourage people with suspected cancer to speak to their doctor.

Flanagan noted that while some of the products people buy to manage early cancer symptoms line up with what we might expect – like laxatives in people with very early bowel cancer – there could also be some surprising connections. “In our previous study, indigestion medicines turned out to be linked to shoppers with ovarian cancer. So, part of this study is very much about finding out which products are in play,” he explained.

An early warning system

If CLOCS-2 is successful, the researchers hope to launch another study to monitor the purchases of healthy individuals for specific ‘triggers’. This could eventually pave the way for a digital alert system that prompts shoppers to seek medical advice based on their purchasing trends.

Of course, in most cases, the sort of non-specific symptoms people buy medicines for don’t have anything to do with cancer. The power of CLOCS-2 comes from its potential to pick out wider trends in behaviour that could point to something more. A lone frog might start croaking for any number of personal reasons, but a chorus of croaks all around the pond suggests there’s something bigger happening.

In our world, where people use loyalty cards as naturally as frogs jump between lily pads, CLOCS-2 might just help turn a trip to the shops into a new way of keeping us safe.