Abstract
Background
Patient-Reported Outcomes Measurement Information System (PROMIS) scales have been utilized to assess symptoms in a variety of medical conditions but have not been previously examined in patients with sports-related concussions. Our objective was to analyze PROMIS scores of University of Rochester Medical Center’s concussion patients to assess for demographic differences in symptoms and recovery.
Methods:
ICD9/10 codes retrospectively identified 214 concussion patients (51% female) less than 30 years old (mean age 15.1 +/- 2.3 years) with PROMIS metrics for physical function, pain and depression. Minimal clinically important difference for these domains is considered 4 points. Two-way ANOVA assessed the influence of gender on PROMIS domains with age as a covariate. T-tests measured the association of gender with the number of previous concussions and clinic visits. Multiple linear regression was utilized to assess the influence of the listed variables on the number of clinic visits.
Results:
Females initially present with 4.0 points lower physical function (95% Confidence interval (CI) 1.5-6.6), 5.8 points (95% CI 3.3-8.3) higher pain interference and 5.4 points (95% CI 2.7-8.2) higher depression compared to males (p <0.01). The history of previous concussions did not differ by gender (1.2 ± 0.1 for females vs. 1.2 ± 0.1 for males), however, the number of visits per episode of care was greater for females (females=8.7±0.4 vs. males=5.3±0.6, p <0.01). Multiple linear regression showed that gender, pain rating, and physical function were significant predictors of the number of clinic visits (r-value = 0.35, p <0.01).
Conclusions:
In our clinic, females and males report concussion at the same prevalence, but post-concussive females experienced statistically significant greater pain and depression and lower physical function compared to their male counterparts. Gender was not associated with the number of previous concussions, but did influence the number of visits. The combination of gender, physical function, and pain were significantly correlated with number of visits, and these variables may be considered in future efforts to determine prognosis of recovery at intake post-concussion.