Advanced Computing in the Age of AI | Friday, March 29, 2024

Boston Children’s Combats Strep Throat from Home with Big Data 

A new health assessment that combines big data with patient symptoms, called a “home score,” may keep patients suffering from strep throat from having to take a trip to their doctors’ offices, according to a new paper in the Annals of Internal Medicine. The study, conducted by doctors Andrew Fine and Kenneth Mandl of Boston Children’s Hospital found that by combining public health data with patient symptoms, they could help patients avoid unnecessary doctor visits.

The score is calculated by combining patient data such as symptoms and age with a geographic health statistic developed by Mandl and Fine that captures the number of strep throat instances in a patient’s area. From this, the doctors hope that patients will be able to determine whether or not they should get a strep test without needing to leave their home.

“Using the home score could empower patients to make informed decisions,” says Fine, who works alongside Mandl in Boston Children’s’ Division of Emergency Medicine and Informatics Program. “Integrating local epidemiologic context with the symptom information permits calculation of personal, local risk of strep throat.”

Helping to gather this data was MinuteClinic, CVS Caremark’s retail health clinic. Mandl and Fine estimate that broad use of the data to determine home scores could eliminate 230,000 unnecessary doctor visits annually.

“The basic math here is that if group A strep is present in patients around you then you are mork likely to have strep,” Mandl says. “The local epidemiology is so informative that when combined with just a few additional facts from an individual we can arrive at a reasonable initial diagnosis, without a health care visit.”

“Because sore throat is so common, reducing these visits could alleviate strain on the health system while saving significant opportunity costs for patients,” adds Fine.

Beyond strep throat, however, Mandl and Fine are hoping that their work will bolster the larger, ongoing effort to make disease risk assessment more accessible to the population at large. But by combining surveillance data with patient-reported personal data, Mandl and Fine say that it will also help clinicians to make better, more informed decisions about patient treatment.

EnterpriseAI