Popular Concussion Test Inaccurate, Study Says

ImPACT software found to misidentify concussions in healthy people

All that stands between the peloton and the finish line are one last turn and 400 meters of rain-slicked road. The lead cyclist charges into the corner, leaning hard into his turn, and his tires lose their grip on the pavement. He falls to the ground, hitting his head so hard it bounces. When he manages to stand, he's wobbly and road-rashed. Later, a team trainer tests the cyclist for a brain injury using ImPACT, the world's most popular concussion evaluation system, while his teammates nervously await the results. According to a new study, though, the test may not even be all that accurate.

When researchers at the University of Texas at Arlington and Dublin City University tested healthy non-athletes with ImPACT software, it misidentified concussions in 22 to 46 percent of them. The University of Georgia also participated int the study. In case you didn't catch that, none of the student volunteers had a concussion at the time of testing.

Researchers divided participants into two groups by location—45 students in Texas and 46 in Ireland. The 91 students took the test three times, the same way athletes do before, during and after a concussion. The first time determines how a person answers when he's healthy. A second test is done to see if the athlete has a concussion. A third is conducted in order to see a return to the baseline. Again, these non-athletes had no concussions at the time of testing or within the past six months before they volunteered for the study. But after taking the test a second time, 22 percent of the American students and 37 percent of Irish students were classified as having a concussion. The third time, 28 percent of Americans and 46 percent of Irish students were identified as having concussions.

Lead researcher Jacob Resch said these results are common for any individual method of testing for a concussion. He said these tests are among a variety of methods athletic trainers and doctors use to diagnose concussions.

“It's not this one tool that needs to assess a concussion,” said Resch, assistant professor at the University of Texas-Arlington. “It's the battery of tests you need.”

Resch also said that concussions are commonly diagnosed after many different types of tests that challenge the patient’s response rate, balance and memory, in addition to a physical exam and symptoms they report. He said this combination brings the rate of misidentifying a concussion down substantially.

The results are not uncommon. A 2007 study by the University of Illinois at Urbana-Champaign using healthy students showed ImPACT misclassified about 34 percent of 110 students in the second test and 38 percent of the same students in a third test as having a concussion. Similar software showed similar misclassification rates.

Resch a said health care providers, including physicians and athletic trainers, should be aware that it may take multiple tests to confirm a concussion because an injured athlete may pass one test and fail another. He also advised parents of athletes to have their doctor do multiple tests to diagnose a concussion, warning that concussions are unique injuries.

“Every concussion is an individual injury,” he said. “Everyone is different.”

Click here for the study abstract.