2 February 2023

Your shopping data could help spot cancer sooner

Big Data Clinical Research

In the brave new world of health data, a new opportunity is presenting itself: loyalty card data which can track medication purchases.  

This unusual source of data could lead to earlier detection of ovarian cancer, according to a UK study published in JMIR Public Health and Surveillance. 

Patients with ovarian cancer often present with mild symptoms in the early stages of the disease, leading them to seek relief through over-the-counter medications instead of visiting a GP. Symptoms like loss of appetite, stomach pain and bloating could be dismissed by patients as not worthy of a doctor visit, not knowing that these are common symptoms of ovarian cancer. 

And like many diseases, late diagnosis could mean more severe illness or an increased risk of death.  

But if it was possible to track changes in buying patterns for over-the-counter medication, it may become a useful diagnostic tool. 

Lancaster University’s Cancer Loyalty Card Study looked at whether there was a link between an ovarian cancer diagnosis and a history of buying pain and indigestion medications over the counter.  

The first of its kind, the study of almost 300 women found that pain and indigestion medication purchases were higher in women who were subsequently diagnosed with ovarian cancer compared to women who did not have ovarian cancer.  

The study found that the change in purchases could be seen eight months before diagnosis. However, patients with ovarian cancer only began to recognise their symptoms about four and a half months before diagnosis, on average.  

Lead author Dr James Flanagan said that if attention was brought to the fact that there was increased purchasing of pain and indigestion medication earlier, patients may be compelled to seek medical attention well before this four-and-a-half-month period.  

“Using shopping data, our study found a noticeable increase in purchases of pain and indigestion medications among women with ovarian cancer up to eight months before diagnosis, compared with women without ovarian cancer. This suggests that long before women have recognised their symptoms as alarming enough to go to the GP, they may be treating them at home,” Dr Flanagan said. 

“As we know early diagnosis of ovarian cancer is key to improving chances of survival, we hope this research can lead to ovarian cancer symptoms being picked up earlier and improve patients’ options for treatment.”   

Lancaster University’s Dr Yasemin Hirst, who led the preliminary study, said information like this expanded the possibilities of data gathered outside healthcare settings.  

“The Cancer Loyalty Card Study is one of the leading projects showing that our health behaviours can be measured beyond healthcare records using transactional data. This data is very exciting for behavioural scientists to further explore lifestyle changes, dietary behaviours and perhaps exploring other datasets that can provide more information about self-care and health outcomes,” Dr Hirst said.  

The researchers hope that these findings could lead to the development of an alert system which could warn patients when their purchasing could be reflecting a more serious health problem.  

They have received further funding from Cancer Research UK to look into whether over-the-counter purchases could also be used to detect other cancers, such a stomach, liver and bladder cancers. These can similarly have unclear symptoms in the early stages.  

JMIR Public Health and Surveillance, online 26 January.