Grocery bills can predict diabetes rates by neighborhood
Dietary habits are notoriously difficult to monitor. Now data scientists have analyzed sales figures from London’s biggest grocer to link eating patterns with local rates of high blood pressure, high cholesterol, and high blood sugar.
Food is an important factor in this state of affairs. So researchers desperately want to know more about dietary habits and how they relate to health.
Enter Luca Maria Aiello at Nokia Bell Labs in Cambridge, UK, and colleagues, who have studied diet by mining the data from grocery bills and then comparing it with the population’s health, as determined by medical prescriptions issued in the area.
They show, as expected, that increased consumption of carbohydrates, fat, and sugar is positively correlated with metabolic syndrome, while increased fiber intake is negatively correlated. They go on to show that item weight—which is a proxy for calorie consumption—is also positively correlated with metabolic illness, while greater diversity of nutrients is negatively correlated.
The team go on to predict the levels of high blood pressure, high cholesterol, and high blood sugar in a neighborhood merely by looking at the local calorie and nutrient consumption. Indeed, they say their classifier can identify unhealthy areas with an accuracy of 91%.