Concussion is one of the most common injuries encountered by athletes participating in contact and collision based sports. In addition, concussion is one of the most complex injuries to manage by sports medicine professionals. Over the past 20 years, there has been significant progress made in the testing procedures used to manage the medical care for those who’ve sustained concussion during sport participation. However, this is a difficult task as many factors, other than concussion, may influence one’s performance on these testing procedures and should be accounted for in interpreting the results of these tests. Thus, understanding how various factors influence concussion test results is critical for the continued development of evidence-based concussion management guidelines.
This week we highlight the work of Mike Baum, who graduate from our masters program in Athletic Training. Under the direction of his thesis committee (Dr. Jason Mihalik, Dr. Julianne Schmidt, Dr. Benjamin Goerger, and Dr. Kevin Guskiewicz), Mike examined the influence of psychologic distress on concussion testing measures in intercollegiate athletes. Many thanks to Mike and Dr. Mihalik for contributing to this week’s EXSS Impact blog post.
My initial interest in pursuing this study was a result of the increased media attention to depression and mental health in the population, specifically the 18-22 year old college age range. At the same time, concussion was becoming more and more of health issue. My initial thought was to have two groups; an experimental group consisting of subjects diagnosed with clinical depression and a control group made up of subjects without any clinical diagnosis of depression. I wanted to see if they performed differently on concussion testing measures. After working through the logistics of performing such a research project with my research mentor team, our study morphed into something somewhat different. Instead, we examined whether “psychological distress” played a confounding role in baseline concussion testing.
We examined measures of psychological distress in freshmen student-athletes. Based on the results of the psychological distress measure, we then divided our subjects into low, moderate, and high distress groups (tertiles). We then retested subjects after 10 weeks on the same outcome measures.
We ran a repeated measures MANOVA to examine whether there was a difference in how our low, moderate, and high distress groups performed on baseline concussion testing measures (e.g. neuropsychological and balance scores, and symptom reporting). We also examined whether a change in psychological distress over the 10-week re-test period was associated with the change in performance on measures of neuropsychologic function, balance, and symptom reporting.
We found that very few variables were significantly different across the distress groups at baseline. While there were some variables that were different at baseline between the distress groups, we determined that these differences were not clinically meaningful. For example, we found that subjects in the high distress group reported significantly more somatic symptoms than the low distress group. Upon further examination, this “statistically significant difference” was truly only about ½ of a symptom score, which we determined to not be of clinical relevance.
Similar findings were observed over our 10 week re-test window, where an increase in distress level was associated with an increase in symptoms. However, we again determined this statistically significant finding to be of minimal clinical importance.
3) What is the impact of these findings on the public?
Typically, people hope to find some sort of statistical and clinically significant difference so that recommendations for change or an improvement in our practice can be made. In our case, our results suggest that our current model of baseline concussion testing is valid in terms of psychological distress. We actually validated our current method of baseline concussion testing rather than saying we need to make changes to account for a new confounding variable.