Concussion Outcomes 3 months After an Intervention for Young Adults with Co-Occurring Anxiety: A Randomized Controlled Trial

Introduction

Persistent concussion symptoms (lasting ≥3 months) often have sustained impacts on physical, emotional, and cognitive functioning.¹ Anxiety is a strong predictor of persistent post-concussion symptoms.² Young adults experience disproportionate rates of concussion³ and anxiety,⁴ rendering them susceptible to persistent post-concussion symptoms. Nevertheless, no evidence-based psychosocial interventions exist to prevent persistent concussion symptoms in this population. To address this, we developed the Toolkit for Optimal Recovery after Concussion (TOR-C), a 4-session mind-body program adapted from the Toolkit for Optimal Recovery after Orthopedic Injury.⁵ In a recent feasibility randomized controlled trial of TOR-C and an active control (Health Enhancement after Concussion; HE-C), both interventions demonstrated excellent feasibility and preliminary improvements between baseline and post-intervention in outcomes including post-concussion symptoms, physical function, and anxiety.⁶ Given TOR-C’s ultimate aim of preventing persistent concussion symptoms, integration of 3-month follow-up data is important. Consistently, the current study assesses these outcomes 3 months post-intervention.

Materials and Methods

This single-blind study was approved by the Massachusetts General Brigham-IRB (P#2022P001521) and registered on ClinicalTrials.gov (NCT05524402). Recruitment launched in October 2022 and concluded in March 2025. Fifty young adults (18–35 years old) with a recent concussion (3–10 weeks prior) and co-occurring anxiety (Generalized Anxiety Disorder-7 [GAD-7] score≥5), provided written informed consent to participate and were randomized (1:1 via permuted blocks sequence developed by the study’s biostatistician) to TOR-C (n=25) or HE-C (n=25) (see Supplemental Data: Demographic characteristics of the sample, Supplemental Data 2: CONSORT Flow Diagram, and Supplemental Data 3: CONSORT 2025 checklist). TOR-C teaches mind-body techniques, cognitive-behavioral skills, acceptance and commitment skills, return to activity skills, and psychoeducation about recovery. HE-C, dose and time-matched to TOR-C, provides education about healthy living post-concussion (sleep, nutrition, activity), without skills. Participants self-administered measures of concussion symptoms (Post-Concussion Symptom Scale [PCSS]; Minimal Clinically Important Difference [MCID]=26.5),⁷ anxiety (GAD-7; MCID=4),⁸ and physical function (World Health Organization Disability Assessment Scale [WHODAS]-2.0; MCID=3.5)⁹ at baseline (pre-randomization), post-intervention, and 3 months post-intervention. Participant adverse events were monitored until their 3-month follow-up. All event monitoring ceased in June 2024. We used mixed-model ANOVA with a shared baseline to adjust for baseline differences and assessed within-group changes in outcomes from baseline to 3 months. Concordant with recommendations for pilot trials, the study was not powered for efficacy,¹⁰ and analyses were limited to within-group comparisons.

Results

From baseline to the 3-month follow-up, those randomized TOR-C experienced clinically and statistically significant improvements in post-concussion symptoms (−32.75 [95% CI −38.66 to −26.60], p<0.001), physical function (−10.40 [95% CI −13.88 to −6.93], p<0.001), and anxiety (−7.72 [95% CI −10.21 to −5.23], p<0.001).

Participants in HE-C statistically and clinically improved in physical function (−7.62 [95% CI −10.91 to −4.33], p<0.001) and anxiety (−5.27 [95% CI −7.64 to −2.90], p<0.001) at 3 months. Post-concussion symptoms significantly improved (−25.06 [95% CI −35.16 to −14.96], p<0.001) but did not meet the MCID threshold (Figure 1).

Figure 1.

Improvements in outcomes*

*Horizontal line indicates clinical significance (i.e., exceeding minimal clinically important difference; MCID)

**Primary outcome in proposed trial

Discussion

In sum, preliminary improvements in concussion symptoms 3 months post-intervention were statistically significant in both groups, but clinically significant only for TOR-C. Baseline versus 3-month follow-up improvements in physical function and anxiety were statistically and clinically significant for both groups. This suggests both interventions are associated with improvements in specific post-concussion outcomes 3 months later. However, improvements in HE-C may be due to natural recovery and/or nonspecific intervention factors (clinician attention, emotional support, education). While TOR-C improvements showed greater magnitude, potentially due to the dissemination of specific skills, a larger RCT is necessary to detect between-group differences.

The trial’s primary limitation is the modest sample size, which, although typical of feasibility studies, limits statistical power. Additionally, natural recovery from an acute/subacute concussion is expected regardless of intervention assignment. Overall, findings suggest TOR-C may improve post-concussion recovery, but lack statistical power, supporting the need for a subsequent fully-powered trial to establish the efficacy and sustained effects of TOR-C versus HE-C.