Study finds online symptom sites only correct half the time

Online symptom checkers are accurate only about half the time, according to a Harvard Medical School study published in the BMJ.

For the study, researchers reviewed 23 symptom checker websites, including:

• Ask MD
• DocResponse
• Isabel
• iTriage
• The Mayo Clinic
• WebMD

Researchers entered symptoms for 45 patients into the checkers, sourced from standardized vignettes used in medical student training. The patients' conditions included acute liver failure, bee stings, meningitis and mononucleosis.

The study found that one-third of the sites named the correct diagnosis as the patient's first option; 51 percent of the sites named the correct diagnosis in their top three options; and 58 percent of the sites named the correct diagnosis in their top 20 options.

The researchers found that the checkers were about as accurate as diagnoses made through primary care physician phone services, which usually offer insight on whether patients should seek urgent care. The online symptom checkers were "pretty good at effectively directing people with an (emergency) situation to seek some kind of appropriate care, and to do so quickly," said lead author Hannah Semigran, a research assistant at Harvard Medical School.

The findings show patients should use symptom checker sites with caution, said Ateev Mehrotra, another of the study authors. “People who use these tools should be aware of their inaccuracy and not see them as gospel. They shouldn't think that whatever the symptom checker says is what they have.”

The sites are not a “replacement for going to the doctor and getting a full evaluation and diagnosis. They are simply providing some information on what might be going on with you,” said Mehrotra.