Factor Structure, Validity, and Clinical Utility of a 19-Item Post-Concussion Symptom Scale

Monday, October 28, 2013: 3:00 PM
Florida Ballroom A (Hyatt Regency Orlando, formerly the Peabody)
Aisha Sophia Dharamsi, MD1, Cynthia LaBella, MD, FAAP1, Rebecca Carl, MD1, Jin-Shei Lai, Ph.D.2 and Frank Zelko, Ph.D.3, (1)Northwestern University's Feinberg School of Medicine, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, IL, (2)Northwestern University Feinberg School of Medicine, (3)Northwestern University Feinberg School of Medicine, Ann & Robert H. Lurie Children's Hospital of Chicago

Purpose:

The primary goals of this study were to (1) examine the factor structure of a self-reported 19-item post-concussion symptom scale (PCSS) in pediatric athletes following concussion and (2) investigate factor structures with respect to time. Because this 19-item PCCS utilized a Likert-type scale to assess symptom severity a secondary goal was to evaluate the utility of summation scores pertaining to each factor.

Methods:

This was a retrospective medical record review of persons between 6 and 22 years of age with a diagnosis of concussion seen at outpatient sports medicine clinics in the Chicago area between April 2008 and September 2012.  Exclusion criteria were conditions that suggested a specific cause of symptoms.

Study participants who fulfilled these criteria were identified and pertinent data abstracted included age, gender, injury date and mechanism, previous head injury, date PCSS completed, scores for each of the 19 PCSS items. Data was entered into a standardized computer-based template prior to statistical analysis.

Results:

402 concussed individuals constituted the study sample. Most (66.8%) were male and the mean (SD) age was 14.27 ± 2.67 years. 57% denied previous head injury, and the mean (SD) number of concussions sustained was 1.04 ± 0.21. The most frequently reported mechanism was football (20.7%), followed by soccer (17.7%), ice hockey (15.21%), and falls (11.97%).

Factor analyses supported a 3 factor solution for post-concussive symptoms that included 18 of the original 19 items. These consisted of (1) eight cognitive-related items, (2) six somatic-related, and (3) four emotional-related. Results supported the use of a summation score for factors (2) and (3) but not for factor (1).  Finally, it was discovered that factor (3) summation scores increased 14 days after concussion, possibly indicating the emotional impact of an injury.

Conclusion:

Concussion management emphasizes importance of self-reported symptoms as indicators of injury and post-injury recovery.

Factor analysis has been used to provide a more focused approach to the concussed athlete by grouping symptoms into common categories.  Current studies, however, remain limited by numerous methodological and analytical factors.  Additionally, although the majority of symptom checklists employ Likert-type scales to characterize symptom severity, no literature exists regarding the utility of summation scores.

Our investigation demonstrates a consistent symptom factor structure in a population of concussed individuals in the acute, sub-acute, and chronic stages of concussion.  It is the first study to date supporting the use of summation scores in 2 of the 3 factors identified (somatic, emotional).  Furthermore, the discovery of increasing severity of emotional symptoms with time has significant implications in providing a more targeted approach to concussion evaluation and management.