Changes in Coping Skills Throughout Concussion Recovery in College-Aged Individuals

Participants

A prospective, repeated-measures–design study of college-aged individuals was conducted. Participants with a medically diagnosed concussion by a medical doctor, doctor of osteopathy, nurse practitioner, or physician assistant and matched controls were recruited from varsity, club, and recreational sports teams at a Division I university. Referrals were made by athletic trainers (ATs), coaches, and sports medicine physicians. Participants aged 18 to 30 years were included in the study if they had sustained their diagnosed concussion within 5 days of enrollment and either currently or recently participated in a varsity, club, or recreational sport. Healthy controls were closely matched to concussed participants based on similar sport and demographic information. Participants were excluded if they presented with a complex case, including hospital admission for over 24 hours, loss of consciousness for over 20 minutes, abnormal neuroimaging, or history of neurological disease. Healthy controls were excluded if they had sustained a concussion within the past 6 months.

Operational Definitions and Measures

Procedures

This study received approval from the Michigan State University institutional review board of record, and all participants completed informed consent before beginning study procedures. Participants with a concussion were tested within 5 days of injury (acute visit) and at FMC (within 3 days of clearance). Healthy controls followed the same schedule as their matched participant with concussion. At the acute visit, participants completed demographic and medical history information, injury characteristics, the symptom evaluation from the Sport Concussion Assessment Tool-5, and the ACSI-28.^22,24 At the FMC visit, participants reported recovery information, including days to symptom resolution, days to FMC, and the ACSI-28. Participant data were collected and managed through the online database Research Electronic Data Capture.²⁵

Statistical Analyses

Descriptive statistics were used to calculate both groups’ demographic and medical history variables. Continuous variables were calculated as means with standard deviations, and categorical variables were calculated as frequencies with percentages. Independent-samples t tests were used to compare continuous data between groups, and χ² tests were used to compare categorical data with Fisher exact tests when expected cell counts were less than 5.

To assess our primary purpose, a 2 × 2 mixed analysis of variance (ANOVA) was used to investigate differences in coping skills for each group throughout recovery. The within-participants variables were time (acute visit and FMC), and between-participants variables were group (concussion and healthy control). All assumptions of a 2 × 2 ANOVA were met. Effect size estimates were determined by partial η squared (η_p²) and interpreted as small (0.01–0.08), medium (0.09–0.24), and large (>0.25).²⁶ Follow-up pairwise comparisons using Bonferroni-corrected paired t tests were conducted to examine differences across time points within the same individual. Additionally, post hoc Bonferroni-corrected independent-samples t tests were conducted to examine differences between groups (ie, concussion and control) at each time point. The α level was set at .05.

To assess our secondary purpose, 2 separate multiple linear regressions were run to determine the association between ACSI-28 total score and recovery times (days to symptom resolution and days to FMC) for college-aged individuals with concussion. All assumptions of the multiple regression were met. To determine which covariates were entered into the model, separate univariate linear regressions were performed with known variables that affect recovery outcomes. However, none of these variables were significant. It was determined that sex, race, history of depression and anxiety, and Sport Concussion Assessment Tool-5 symptom severity score at the acute visit would be added as covariates in the multiple regression models, as previous evidence has indicated that these variables are related to recovery outcomes.⁴ For the multiple regression models, the overall percentage of explained variance of the model (R²), unstandardized regression coefficient (B), standardized coefficient (β), 95% CI, and P values were calculated with statistical significance set a priori at P < .05. All statistical analyses were conducted in SPSS (version 28; SPSS Inc).

Power Analysis

To achieve an acceptable power of 0.80 in a 2 × 2 mixed ANOVA using a small effect size of 0.15 and α set at .05, an estimated 90 participants (ie, 45 participants per group) were needed. A separate a priori power analysis was conducted to determine the minimum power needed for a multiple regression analysis on the concussion group. To achieve an acceptable power of 0.80 using a small effect size of 0.15 and α set at .05, an estimated 55 participants were needed. To ensure that the estimated sample size was met for both purposes, we enrolled a total of 96 participants, with 64 in the concussion group and 32 in the control group.

Demographic and Medical History Information

A total of 96 college-aged individuals (concussion n = 64, 53.1% female, age = 21.28 ± 2.36 years; control n = 32, 46.9%, female, age = 21.94 ± 2.84 years) were included in this study.

Table 1 provides demographic and medical history information. A significantly greater number of participants in our concussion group than in our control group reported having a history of concussion (P = .04); however, no other significant differences were noted between groups. For the concussion group, the average time from injury to acute visit was 3.38 ± 1.6 days, and time from acute visit to FMC visit was 14.92 ± 12.4 days. For the control group, the average time from acute visit to FMC visit was 13.41 ± 7.9 days. There was no difference in time between visits for the groups (t₉₅ = −0.63, P = .53, d = −0.14).

Table 1.Descriptive Data for Participants With Concussion and Controls

View Fullscreen

Changes in Coping Skills

Statistical outcomes of the mixed ANOVA are summarized in Table 2, with means and SDs for the ACSI-28 listed in Table 3. The group × time interaction for Concentration (F_1,94 = 7.28, P = .01, η_p² = 0.07) was significant. Post hoc comparisons with Bonferroni-corrected independent-samples t tests revealed that Concentration scores were significantly lower for the concussion group than the control group at the acute visit (mean = −1.41, SE = 0.53, P = .01), but were not different at the FMC visit (mean = −0.33, SE = 0.54, P = .54). Post hoc comparisons with Bonferroni-corrected paired-samples t tests revealed that Concentration scores improved from the acute to FMC visit for the concussion group (mean = 0.95, SE = 0.23, P < .001), but Concentration scores were not different between visits for the control group (mean = −0.13, SE = 0.33, P = .70). There were no additional significant group × time interactions for any other ACSI-28 scores (P > .05). The mixed ANOVA revealed significant main effects for time for the ACSI-28 total score (F_1,94 = 22.08, P < .001, η_p² = 0.19), Coping With Adversity (F_1,94 = 4.53, P = .04, η_p² = 0.05), Confidence and Achievement Motivation (F_1,94 = 8.71, P = .004, η_p² = 0.09), Goal Setting and Mental Preparation (F_1,94 = 6.53, P = .01, η_p² = 0.07), and Freedom From Worry (F_1,94 = 10.25, P = .002, η_p² = 0.10). Finally, we did not find any significant main effects of group for any ACSI-28 scores (P > .05).

Table 2.Results of the 2 × 2 Mixed ANOVAs Used to Analyze Variables From the ACSI-28

View Fullscreen

Table 3.ACSI-28 Total and Subscale Scores at the Acute and FMC Visits for the Concussion and Control Groups

View Fullscreen

Relationship Between Coping Skills and Recovery Outcomes

The multiple regression model results, examining the relationship between the ACSI-28 total score and days to symptom resolution while controlling for sex, race, history of depression/anxiety, and acute visit symptom severity, was not significant (F_5,63 = 1.39, P = .24, adjusted R² = 0.03), and the ACSI-28 total score did not significantly add to the model (B = 0.06, P = .62). The results for this multiple regression are presented in Table 4. Additionally, the multiple regression model results examining the relationship between ACSI-28 total score and days to FMC while controlling for sex, race, history of depression/anxiety, and acute visit symptom severity was not significant (F_5,63 = 1.54, P = .19, adjusted R² = 0.04), and the ACSI-28 total score did not significantly add to the model (B = 0.089, P = .46). The results of this multiple regression are presented in Table 5. Additionally, exploratory regressions were run with and without covariates between our independent variable (ACSI-28 total score) and dependent variables (days to symptom resolution and days to FMC). These resulting models were still insignificant.

Table 4.Multiple Regression Results for Days to Symptom Resolution With the ACSI-28 Total Score as the Independent Variable^a

View Fullscreen

Table 5.Multiple Regression Results for Days to FMC With the ACSI-28 Total Score as the Independent Variable^a

View Fullscreen

Changes in Coping Skills

Previous researchers have investigated the influence of coping strategies on postinjury and neurocognitive outcomes after concussion,¹⁶ and differences in coping among athletes with concussion, athletes with orthopaedic injury, and healthy controls.¹⁷ Furthermore, researchers examining coping skills using the ACSI-28 have predominantly focused on recovery time and return to activity after long-term musculoskeletal injuries.²¹ This study expands on existing literature by observing changes in existing coping skills in college-aged individuals after concussion while including a healthy control group for comparison. No significant interaction was observed between group and time for the ACSI-28 total score and most subscales, except Concentration. The lack of interaction between groups at both time points suggests that existing coping skills may not be affected by concussion and do not play a role in concussion recovery.

Most ACSI-28 scores were not significantly different at either time point between groups, except for one group × time interaction. The Concentration subscale was significantly lower for the concussion group than the healthy control group at the acute visit. Experiencing difficulty with concentration is common after concussion and can be exacerbated by the severity of concurrent symptoms (eg, headache, dizziness) and impairments (eg, migraine, vestibular).^27,28 Therefore, it is logical that individuals with concussion would exhibit reduced concentration skills and face more significant challenges with concentration compared with the control group. Notably, the Concentration score for the concussion group was not different than the control group at the FMC visit. This suggests that, although individuals with concussion might struggle initially with concentrating on tasks, these issues should be less prominent and return to preinjury levels by the time they are cleared for full activity. This likely coincides with the resolution of other concussion-related symptoms and impairments with rehabilitation.

The lack of interaction between groups for the ACSI-28 total score and additional subscales could be due to several factors. It is important to note that most participants in our study were collegiate athletes. These athletes were likely placed into a stepwise progression protocol after concussion, as is recommended by several consensus statements for concussion in sport.^2,23,29 Consequently, many athletes in our sample could have been recovering with their ATs or physicians before their return-to-play visits and possibly even before their initial visit. Because many ATs use PST during injury recovery, athletes working with ATs may use these skills throughout recovery.¹³ This exposure could mitigate the negative psychological effects of a concussion during recovery. Additionally, an active plan for recovery might have improved their cognitive appraisal of their injury, leading to better sport-related coping skills.³⁰ However, interventions and treatments were not noted for participants, so we cannot make conclusions from this information. Moreover, individuals dealing with shorter injuries like concussion may not have sufficient time to process their injury within the context of their sport before returning to play. This shorter recovery period might not be enough time to negatively affect their preexisting coping skills. Other studies whose authors found significant changes in total scores over time using the ACSI-28 involved long-term (>6 months) musculoskeletal injuries (eg, ACL reconstruction surgeries).²¹ The average time to recovery for individuals in our sample was 16.69 ± 12.19 days, which is much shorter than the year after surgery in the study by Ellis et al.²¹

The main effect of time showed similar increases for both groups in the ACSI-28 total score and subscales including Coping With Adversity, Confidence and Achievement Motivation, Goal Setting and Mental Preparation, and Freedom From Worry. Finding increases in total ACSI-28 scores from the acute to the FMC visit was expected for participants with concussion; however, it was not expected that matched controls would also increase in total ACSI-28 scores over time. This finding could be attributed to the large number of athletes in the current study. Previous researchers have noted that many collegiate athletes rely on the social support of their teammates to help them cope with stressors related to sport.¹⁸ This existing social support and reliance on teammates may lead to a preexisting ability to cope with general stressors that all college students face independent of injury, like school, which may extend to injury-related stressors like concussion.¹⁸ Approximately 90% of controls were athletes, often referred from the same team or by the injured athlete. This shared social support may have been used by uninjured teammates during injury recovery, leading to similar coping skills between groups. Regardless of injury, reliance on one’s teammates and shared support of injured teammates might help improve coping skills between visits for injured and noninjured athletes.

Relationship Between Coping Skills and Recovery Outcomes

Further analyses were conducted on the concussion group to determine if ACSI-28 scores were associated with recovery outcomes. The ACSI-28 was not significantly associated with any recovery outcomes (ie, average days to symptom resolution and average days to FMC). Previous researchers using the ACSI-28 to predict recovery outcomes for athletes with ACL injuries found that lower scores preoperatively on the ACSI-28 predicted a significantly greater number of days to recovery.²¹ This finding, and other findings noting the importance of coping to injury recovery outcomes, contradicts the current study, which did not find a significant association in average days to recovery, specifically days to symptom resolution and days to FMC.^16,31 Covassin and colleagues assessed their populations at 2 specific time points: 3 and 8 days postinjury.¹⁶ In contrast, we waited until full recovery to assess the groups, with the mean days to recovery being 16.69 ± 12.19. Our second visit occurred more than 8 days later than the second visit in the study by Covassin et al.¹⁶ Future researchers might consider adding an additional time point between the acute phase and full recovery to determine if there are any associations between coping mechanisms and recovery outcomes. Furthermore, the study by Covassin et al used the Brief COPE, which evaluates general coping strategies throughout injury.¹⁶ This study used the ACSI-28, which focuses on how existing coping skills, such as peaking under pressure and confidence and achievement motivation, affect recovery outcomes.

Limitations

This study is not without limitations. First, the ACSI-28 is a self-report survey, which may be subject to recall or response bias by participants. The concussion and control group sizes were also unequal, which may have introduced error in our statistical analyses. Our control group sample size was also below the sample estimate to reach minimum statistical power, which may have contributed to insignificant results between groups. This limitation adds several constraints to the generalizability of our findings. In this study, we did not find any significant differences between groups for total score or any subscales of the ACSI-28, except for the Concentration subscale, at either initial or recovery visits. A more balanced sample size for each group might have increased the likelihood of detecting a significant difference between groups for the total scores and subscales. Future researchers should focus on recruiting larger and more balanced sample sizes for concussion and control groups to further assess the effect of coping on recovery outcomes. Additionally, our sample consisted of athletes and nonathletes sustaining sport-related and non–sport-related injuries. Due to the nature of some of the athlete-focused questions in the ACSI-28, not being an athlete might limit the ability to provide accurate and meaningful responses to specific questions relating to coachability or feeling pressure to return to sport. Additionally, some athletes likely followed a structured protocol per the consensus statement guidelines. No interventions or protocols that the concussion group participated in throughout recovery were noted, which could have affected their coping skills related to activity; however, most participants in our sample with concussion were athletes (n = 58, 90.6%), so this likely had no meaningful effect on our results. Lastly, our sample consisted of college-aged individuals, mostly managed by team ATs in a university setting, which limits the generalizability of our findings to other populations and different clinical settings.

Conclusions

This study examined differences in coping skills assessed via the ACSI-28 from acute to FMC visits between participants with concussion and healthy matched controls. A significant interaction between groups from the acute to FMC visit was observed for the Concentration subscale, whereas results for all other subscales and the total ACSI-28 score were insignificant. The Concentration score for participants with concussion was significantly worse acutely but normalized to the level of controls by the FMC visit. The ACSI-28 total score, Coping With Adversity, Confidence and Achievement Motivation, Goal Setting and Mental Preparation, and Freedom From Worry significantly increased from the acute to FMC visit. For the concussion group, the ACSI-28 total score was not significantly associated with recovery outcomes, including average days to symptom resolution and days to FMC. These results suggest that coping skills may not be an important factor for clinicians to consider in short-term injury recovery like concussion. Future researchers should examine relationships between coping skills and concussion in a more diverse sample, including a greater variety of age, race, and sociocultural backgrounds. Previous researchers examining coping throughout injury found that coping strategies affect athletes’ neurocognitive performance postconcussion,¹⁶ so future researchers should expand on specific coping strategies and styles and their role in concussion recovery.