1547 - Pilot Validation Study of a New Concussion Symptom Scale

Friday, February 23, 2024

5:00 PM – 6:30 PM EDT

Presenting Author(s)

ES

Eli Sepkowitz, MD

PGY3
Northwell Health
Brooklyn, New York, United States

Co-Author(s)

AK

Anshul K. Kulkarni, BS

Medical Student
Donald and Barbara Zucker School of Medicine at Hofstra/Northwell
Queens, New York, United States
JC

Joy Cambe, MD, MPH

Clinical Research Coordinator
Feinstein Institute of Medical Research at Northwell Health
Manhasset, New York, United States
FM

Fatimah Mozawalla, MD

Resident Physician
Mount Sinai Hospital
Woodbury, New York, United States
WP

Welmince Pello, MPH

Clinical Research Coordinator
Feinstein Institutes for Medical Research
New York City, New York, United States
OB

Ona Bloom, PhD

Professor
The Feinstein Institute/Zucker School of Medicine at Hofstra Northwell
Manhasset, New York, United States
RS

Rosanna C. Sabini, DO

Associate Professor
Zucker School of Medicine
Bay Shore, New York, United States

Objectives: The Post-Concussion Symptom Scale (PCSS) is a reliable self-report survey that measures the severity of 22 symptoms following a concussion. Our institution created a New Concussion Symptom Scale (NCSS) that was designed as an abbreviated alternative and also categorizes the symptoms into five domains. The symptoms of the PCSS are rated from 0-6 and the symptoms of the NCSS are rated from 0-2. In both scales, higher scores correspond to more severe symptoms. This pilot study sought to examine whether patient responses to the two scales would be meaningfully divergent.

Design: A retrospective chart review was conducted with the approval of our institutional IRB (#21-1203) of 76 patients (ages 11-72) who had undergone evaluation at the outpatient Concussion Program from 11/1/2020-5/31/2022. All subjects completed both the PCSS and NCSS during the same physician visit. Outlier data as well as incomplete or equivocal responses were removed from analysis. Following exclusion, descriptive statistics and Spearman correlations between the two scales were conducted. Symptom domains from the NCSS were extrapolated to the PCSS for comparison.

Results: A total of 44 patients were included for analysis. The mean age was 38±18 years. The number of males was 19 (43%) and females was 25 (57%). The median total scores for the PCSS and NCSS were 35.50±52.75 and 12.50±14.50 respectively; the Spearman correlation between total scores was r=0.9341 (p< 0.0001). Comparisons of all 5 domains were highly correlated: Dizziness r=0.93 (p< 0.0001), Cognitive r=0.89 (p< 0.0001), Sleep r=0.88 (p< 0.0001), Somatic r=0.84 (p< 0.0001), and Mood r=0.82 (p< 0.0001). Comparisons of matched individual symptoms within each domain were highly correlated as well.

Conclusions: Scores from the PCSS and the condensed NCSS were highly correlated. Future studies should include the time needed to complete each scale. The NCSS could reduce completion time and improve assessment efficiency without compromising integrity in clinical practice.

Scientific Paper Best Paper Award Abstract:
Objectives

After a concussion, symptom assessment tools can provide a means to objectively measure and assist in patient management. The Post-Concussion Symptom Scale (PCSS) is widely used and considered a reliable self-report survey that measures the severity of 22 symptoms following a concussion. Each symptom is measured from 0 (no report of symptom) to 6 (severe report of the symptom) on a Likert scale with a total score that can range from 0 to 132.

Our institution sought to examine whether an abbreviated questionnaire could continue to capture patient symptoms without sacrificing pertinent data. Therefore, our institution developed a New Concussion Symptom Scale (NCSS), designed to provide a more efficient assessment of concussion symptoms. The NCSS is a self-reported survey that measures the severity of 15 symptoms on a 0 (no report of symptom), 1 (sometimes report of symptoms) and 2 (always report of symptom) point Likert scale. The total score can range from 0 to a maximum score of 30. In addition, the fifteen symptoms are subdivided into 5 domains (Somatic, Vestibular, Sleep, Mood and Cognitive). This pilot study sought to examine whether patient responses to the two scales would be meaningfully divergent.

Design

A retrospective chart review was conducted with the approval of our institutional IRB (#21-1203) of 76 patients (ages 11-72) who had undergone evaluation at the outpatient Concussion Program from 11/1/2020-5/31/2022. All subjects completed both the PCSS and NCSS during the same physician visit. Thirty-two datasets were excluded from analysis based on replies that were categorized as statistical outliers (n=5) or incomplete and/or equivocal (n=27). For all patients, data collected from subsequent visits was not incorporated into the primary analysis. Following exclusion, descriptive statistics and Spearman correlations between the two scales were conducted. Symptom domains from the NCSS were extrapolated to the PCSS for comparison.

Results

A total of 44 patients were included for analysis. The mean age was 38±18 years. The number of males was 19 (43%) and females was 25 (57%). The median total scores for the PCSS and NCSS were 35.50 ± 52.75 and 12.50 ± 14.50 respectively; the Spearman correlation between total scores was r=0.9341 (p< 0.0001). Comparisons of all 5 domains were highly correlated: Dizziness r=0.93 (p< 0.0001), Cognitive r=0.89 (p< 0.0001), Sleep r=0.88 (p< 0.0001), Somatic r=0.84 (p< 0.0001), and Mood r=0.82 (p< 0.0001). Strong correlations existed between individual symptom scores between the two scales (e.g., individual NCSS cognitive symptoms versus individual PCSS cognitive symptoms). Among twelve patients whose data had been removed due to an error in completing just one of the two scales (i.e., either NCSS or PCSS, not both) errors were more frequently encountered on the PCSS than the NCSS by a tally of seven to five.

Discussion

This pilot validation study demonstrated that the scores from the PCSS and the novel, abbreviated NCSS were highly correlated. These findings provide promising evidence for the NCSS's potential as an alternative to the PCSS in routine concussion assessments and highlight its practical implications. Due to its fewer question items and response options compared to the PCSS, the NCSS can possibly reduce completion time and hence improve efficiency in clinical assessment, without losing integrity in clinical practice. Further studies should include the time to complete each scale to explore this possibility. As the field of concussion management continues to evolve, this study contributes to the ongoing efforts to improve patient experiences, optimize diagnostic accuracy, and achieve operational efficiency in the evaluation of concussions. As this study was limited to initial visits, the results cannot be extrapolated as a gauge of clinical outcomes. To this end, future endeavors examining consistency, test-retest reliability, measurement error and convergent validity are warranted.