Study: Statistics Aid Accurate Diagnoses

Providing physicians with live information on current rates of infection can help them make more accurate diagnoses of illnesses, according to a study published by Harvard researchers earlier this week.

By disseminating real-time statistics on local infections, the medical system could potentially reduce the number of doctor visits and the unnecessary use of antibiotics for uninfected individuals, while preventing complications for those who actually are sick.

The study, published in the Annals of Internal Medicine, was conducted by Harvard Medical School Professor of Pediatrics Dr. Kenneth D. Mandl, along with HMS Assistant Professor of Pediatrics Dr. Andrew M. Fine ’91 and Dr. Victor Nizet, a professor at the University of California, San Diego.

The researchers examined 82,000 patient cases gathered from CVS MinuteClinics—small medical clinics located within CVS stores—where all patients’ symptoms are tested and recorded.

The study found that monitoring rates of infection allowed physicians to use local incidence of strep throat in order to better understand the probability that an individual was actually sick. If there was a high incidence rate in the area, then the chances that an individual had the infection increased.

If there was a low rate of incidence, then the probability of infection dropped.

“We identified this application and put together this data set in a unique way that allowed us to see how data from a population can be used for individual clinical cases,” said Mandl.

Mandl explained that if such data were used to increase the likelihood of a correct diagnosis, then it could potentially also be used to improve the efficiency of medical treatment.

“Getting the right care at the right time is going to benefit individuals and society,” Mandl explained.

Mandl has been working in the field of biosurveillance—the transmission of information about disease outbreaks from hospitals to the government—for over a decade.

According to Mandl, it is currently difficult for physicians to quickly get access to information collected about communicable diseases’ incidence.

“We want to bring a degree of situational awareness to the doctor,” Fine said.

Mandl currently serves as the co-director of the Substitutable Medical Apps Reusable Technologies project at Harvard, a federally funded program which seeks to develop infrastructure to allow real-time information to be delivered directly to health care providers.

The SMART project is seeking to develop this infrastructure by creating an “iPhone-like” application, through which doctors could access information and statistics about regional infection rates.