This news release is a wonderfully detailed summary of a study of a novel eye-tracking measure of brain injury, one with the potential to significantly advance the accuracy of tests to detect the severity of traumatic brain injuries, whose consequences can range from a temporary and mild concussion to permanent damage and death. The release ranks as a very good piece of explanatory journalism, going beyond the results of a single study to inform readers about the complexity of brain injury and the challenges inherent in attempts to identify precise and consistent biomarkers for concussion. Because the release makes ample use of good quotes, metaphors, and attention to numbers, journalists and other readers of this news release will come away with a good understanding of a significant clinical problem (over- and under- diagnosis of “mild” traumatic brain injury) in young people; and ample context for evaluating the significance of the eye-tracking test. The description of the eye-tracking test itself — which requires patients to watch a music or other video on a TV screen — is deft.
Physicians and other medical personnel in emergency rooms, on playing fields, and in offices struggle with diagnosing concussions and determining the severity of the underlying brain injury. While brain imaging can identify patients with fractures, bleeds, and strokes, the ability to categorize the severity of patients with head injuries without anatomic abnormalities is limited. The authors report on a new test that can potentially categorize patients with head injuries as having a concussion or not, as well as potentially to assess the severity. Though the results presented are encouraging, the value of the test remains to be determined. Only by applying this prospectively to patients presenting to clinicians with head injuries and then following outcomes over time, can we be certain whether the test will improve our prediction over usual care. Where it may be most helpful would be in patients with normal findings on the test – they can be reassured. But this has not yet been proven. Similarly, it remains to be seen whether an abnormal test result can be effectively used to improve care. A challenge the authors identify is the lack of a gold-standard definition of concussion. Because of that uncertainty, only studies that examine outcomes over time will help determine whether this test improves patient care or not.
The release, (surprisingly, given its overall thoroughness) does not supply any information about the estimated costs of traumatic brain injury, the costs (in time and money) of providing the eye tracking technology, or even the costs of CT scans. This information, even in rough numbers, would have added another layer of good reporting.
This release quantifies results of the study to a degree rarely seen in news releases, going so far as to explain how the researchers scored not only their own new measure, but also how other currently used biomarkers are scored. A bravura performance here.
Based on the nature and description of the technology and how it is used with patients, there are no direct harms from watching music videos, except perhaps acute boredom. But while the test itself may not produce harm, the question is what if the test is wrong? Say the test is positive and in fact the person does not have a concussion? May they receive additional tests or treatments that may cause harm? More importantly, what if the test is normal and the patient has a concussion? The test may falsely reassure, and the patient returns to the activity that caused the problem in the first place, risking additional serious injury. At this point, one cannot say whether or not there are harms to this test, but the potential for harm certainly exists and could have been mentioned.
The release explains pretty carefully that the eye tracking system in question — which measures how well part of the brain work to focus the eyes in tandem — is not perfect, but has a reliable specificity and sensitivity with regard to accuracy of more than 85 percent. It also explains the challenges inherent in any test for concussion, which is that it’s very difficult to establish with any method whether a concussion has, in fact, occurred. It would have been even better if the release had explained, as discussed above in our review, that it’s unclear even if an accurate test will ultimately lead to better care of patients and improved outcomes. Only a large study following many patients over time and tracking their results will be able to tell us that.
It’s hard to overstate the already severe impact of traumatic brain injury on youth and young adults worldwide, or the anxiety sustained by patients, parents, and medical personnel alike in the absence of definitive diagnostic tools for concussion. The release offers statistics and quotes that put the research in context, including quotes from a researcher at a collaborating institution.
The release notes both the sources of funding for the study and the financial interests of researchers on the study in a company that has licensed the new technology. Although we found it odd that the journal published an editorial written by the same person as the study being commented on, that’s beyond the scope of the news release.
The release goes to some lengths to describe the strengths and weaknesses of other brain injury biomarkers and tests. One way to look at alternatives is the current way doctors evaluate head trauma, which is highly variable among providers. In the future, it would be helpful to compare the test results to clinical opinion and then to see what the difference is in terms of outcomes.
Although it is implied that the technology is not currently approved or available, the article doesn’t explicitly state this. A clear statement as to availability would have been helpful for readers.
The release does a good job of explaining that the innovation is in the mathematical model that assesses eye movements. The lede could have made a bigger distinction between the means of tracking eye movements and the algorithm that assesses it, but again, in the context of the whole release, this was not a big problem.
Opinions as to the value of the new technology are properly attributed and/or qualified, although we found the headline referencing tests for heart attacks to be a stretch. There is tons of data on heart attack biomarkers, so it’s somewhat misleading to suggest that this test, based on one study, is comparable.