The Identification and Comparison of Factors Affecting Musculoskeletal and Concussion Injury Disclosure

For this cross-sectional study, we recruited a convenience sample from a single National Collegiate Athletic Association (NCAA) Division I institution. Participating student-athletes were administered self-report surveys asking about their history of concussion and MSK injury. Specifically, they were asked how many injuries they perceived sustaining as well as how many of those perceived injuries they reported and to whom. Moreover, the athletes were asked about their knowledge of signs and symptoms for each injury and their personal beliefs and attitudes toward disclosure of injury. The study was approved by the Institutional Review Board at the University of Texas at San Antonio and all individuals provided written, informed consent before participation. Collegiate student-athletes were invited to join the study during the fall preparticipation physical examination before the academic year. All individuals who were receiving fall physicals (146) were approached; 115 (78%) agreed to participate, and 107 (73%) athletes completed all components of the survey used for this assessment and were therefore included in the final analysis. The average completion time for the survey was 16 minutes.

The questionnaire was administered during the preparticipation physical examination using an online survey platform (Qualtrics). Recruits who consented to be involved in the study were invited to sit in a quiet room where they were given a tablet on which to complete the online questionnaire. The online questionnaire consisted of 3 parts: (1) comprehensive demographic information, including lifetime concussion and MSK injury history; (2) concussion disclosure behaviors and potential determinants of intention to disclose concussions; and (3) MSK injury disclosure behaviors and potential determinants of intention to disclose MSK injuries. The concussion disclosure portion of the survey contained previously validated items to explore disclosure behaviors and intention to disclose concussions.^28–30  Using existing concussion disclosure literature, we created the MSK disclosure portion of the survey based on a similar framework to examine potentially influential factors. Given the novelty of MSK injury disclosure concepts, a panel of 5 experts (certified athletic trainers, including those with doctoral degrees), with more than 100 combined publications related to the diagnosis and treatment of MSK injury, reviewed the survey for face and content validity. As part of the validation process, experts were asked to grade questions for relevance and clarity (1 = not clear or relevant to 5 = very clear and relevant) and if anything should be included that was not originally considered. This process identified 4 additional signs and symptoms of an MSK injury (compensating to complete a task, pain at night, headache, and difficulty falling asleep). Lastly, the survey was piloted with 5 student-athletes. They were given an opportunity to comment on any items that were confusing. Based on the feedback, we revised the instrument. Additionally, internal consistency between individual questionnaire domains was assessed. Specific determinants adapted for the MSK injury portion of the survey are explained in detail in the subsequent section.

Knowledge of MSK injury was evaluated using 21 yes-or-no items concerning symptom recognition, potential signs and symptoms, potential long-term side effects, effects of premature return to play, and consequences of incurring multiple MSK injuries. Correct answers received 1 point, and incorrect answers received 0 points, with a maximum composite score of 21 (higher scores meant better MSK injury knowledge).

Questions about attitudes toward disclosing MSK injuries were adapted from previously validated questions about disclosing concussions.^28–30  The same 8 validated 7-point scale items were adapted and used to determine attitudes toward and perceptions of MSK injury disclosure. Values for the answers were summed, resulting in a composite attitude score ranging from 8 to 56 (higher scores meant more favorable attitudes toward MSK injury disclosure). For example, the concussion attitude question read: “Disclosing possible concussive symptoms to a medical professional or someone in authority during physical activity is…,” while the MSK attitude question replaced “concussive symptoms” with “possible MSK symptoms.” The same eight 7-point items were then presented: cowardly : brave, shameful : prideful, harmful : beneficial, extremely difficult : extremely easy, bad : good, unimportant : important, waste of time : worthwhile, and weak : strong.

The items on MSK injury perceived social norms were adapted from the concussion perceived social norm questions and consisted of the same 7 validated 7-point scale items adapted to reflect the perceptions of the organization, social referent expectations, and actions concerning MSK injury instead of concussion. Scores for the answers were summed, resulting in an MSK composite social norm score ranging from 7 to 49 (higher scores meant more favorable perceived social norms surrounding MSK injury).

This section consisted of a single 7-point scale question regarding how much control student-athletes believed they had over disclosing MSK injury. A perceived control score was categorized as high (score 6 or 7), reflecting agreement or low (scores 1–5), reflecting some agreement to strong disagreement with having control over MSK injury disclosure.

This section consisted of a single 7-point scale question regarding a student-athlete’s intention to disclose symptoms of an MSK injury. Intention scores were categorized as high (score 6 or 7), reflecting agreement or strong agreement with disclosing symptoms of an MSK injury, or low (scores 1–5), reflecting some agreement to strong disagreement with the intention to disclose symptoms of an MSK injury.

Musculoskeletal injury disclosure behavior was measured by asking the question “How many of these musculoskeletal injuries did you report/disclose to a medical professional or someone in authority at the time of injury?” to anyone who reported having had an MSK injury.

We calculated descriptive statistics for all variables. Medians and interquartile ranges (IQRs) were computed for all continuous variables and frequencies and percentages for all categorical variables. The a priori α level was set to .05 for each analysis. Before running models, we examined the variables for multicollinearity. Consistent with previous concussion research, we assessed multicollinearity using condition indices and corresponding variance inflation factors (VIFs). If a condition index was >30, individual assessments of multicollinearity were performed using VIFs. Variables were determined to be collinear if the VIF was >2.5. The disclosure percentage was defined by the number of previous injuries disclosed to a medical professional by each athlete divided by the total number of injuries the athlete recalled having sustained. Nondisclosure was determined by athletes recalling ≥1 injury that was not disclosed at any point in time. Lastly, the intention to disclose a concussion and intention to disclose an MSK injury were converted into categorical variables and dichotomized as high (6 or 7) or low (0–5), which allowed for exploration of differences between those who selected high versus low intention to report injury in accordance with previous literature.³⁰ 

We initially assessed the association between high intention to disclose concussion and high intention to disclose MSK injury using a χ² test. Spearman rank correlations were conducted to compare the relationships between (1) concussion disclosure intention and previous concussion disclosure percentage, (2) MSK disclosure intention and previous MSK injury disclosure percentage, and (3) previous concussion disclosure percentage and previous MSK injury disclosure percentage. Cronbach α tests were performed to evaluate the internal consistency between explored determinants (previous injury, knowledge of injury symptoms and consequences, attitudes toward disclosure, perceived social norms, and perceived behavioral control of disclosure) of concussion and MSK disclosure. The strength of the Spearman rank correlations was determined using the magnitude classifications developed by Schober et al,³¹  while χ² tests were interpreted following suggestions from McHugh.³² 

We then generated univariable logistic regression models to examine each individual determinant and how it affected high disclosure intentionality for both concussion and MSK injury. Two multivariate binomial regression models were then used to identify which of the hypothesized determinants influenced high intentionality for disclosure of each concussion and MSK injury. Perceived behavioral control and gender were not included in the multivariate binomial model exploring concussion disclosure intentionality due to collinearity and reduced sample size. Perceived behavioral control was not included in the MSK injury disclosure intentionality model due to reduced sample size. Odds ratios (ORs) with corresponding 95% CIs were computed to model high intention to disclose.

Overall, 107 student-athletes completed the survey and were included in the analyses (Table 1). Details pertaining to each participant’s primary NCAA sport and high school sport involvement are shown in the Figure. Most respondents were female (n = 74, 69.2%) and White (n = 53, 49.5%) and participated in collision or contact sports (n = 59, 55.1%). Additionally, 24.3% (n = 26) reported a history of concussion, while 68.2% (n = 73) reported a history of an MSK injury. Strong internal consistency (α = 0.918–0.968) was present for each domain in the questionnaire. Student-athletes’ MSK and concussion injury knowledge, attitudes, perceived norms, and descriptive statistics are described in Tables 2 and 3.

High intention to disclose a concussion and high intention to disclose an MSK injury were moderately associated (χ² = 19.276, P < .001, Cramer V = 0.482). The Spearman rank order test revealed a correlation between intention to disclose MSK injury and a history of disclosing MSK injury (ρ = 0.299, P = .019), whereas the intention to disclose a concussion was not correlated with previous concussion disclosure behaviors (ρ = −0.005, P = .982). Despite a trend toward significance in the correlation between actual disclosure in individuals who reported a history of both a concussion and an MSK injury, no relationship was found (ρ = 0.314, P = .061).

Our sample’s knowledge, attitudes, perceived norms, and behaviors pertaining to concussion disclosure were comparable with previous values.^28,29  The median total concussion knowledge score was 39.0/50.0 (IQR = 32–41), while the median total MSK knowledge score was 29.5/40.0 (IQR = 25.5–33.0). In general, higher concussion knowledge scores were positively correlated with higher MSK injury knowledge (ρ = 0.586, P < .001). Pertaining to attitudes, a strong positive relationship was noted between the median concussion attitude score (median = 49.5, IQR = 44.3–56.0) and median attitude score toward MSK injury (median = 47.0, IQR = 37.0–53.0; ρ = 0.788, P < .001). Lastly, when exploring the relationship between median perceived social norm scores for concussion (median = 40.0, IQR = 34.0–42.0) and MSK injury (median = 38.0, IQR = 33.0–41.0), we observed a strong positive relationship (ρ = 0.741, P < .001). Regarding disclosure behavior, participants reported having failed to disclose 25% of perceived concussions compared with 52% of perceived MSK injuries.

The results of the individual univariate regression models to identify individual predictors of high intention to disclose a concussion as well as the multivariate binomial regression model to explore the shared variance between predictors of high intention to disclose a concussion are provided in Table 4. The strongest significant individual predictors of high intention to disclose a concussion were concussion knowledge (β = 1.133, P = .0041, 95% CI = 1.041, 1.233), attitudes toward concussion disclosure (β = 1.059, P = .014, 95% CI = 1.012, 1.109), and perceived norms toward concussion disclosure (β = 1.102, P = .004, 95% CI = 1.032, 1.178). Further, the multivariate regression model explained 68% of the observed variance in high intention to disclose a concussion. Perceived social norms for concussion was the strongest determinant (β = 1.365, P = .002, 95% CI = 1.121, 1.663) of high intention to disclose concussion after accounting for the shared variance. Due to the limited number of females (n = 1) with low intention to disclose a concussion, gender was not incorporated into this model.

The results of the individual univariate regression models to identify individual predictors of high intention to disclose an MSK injury as well as the multivariate binomial regression to explore the shared variance between predictors of disclosing MSK injuries are available in Table 5. Musculoskeletal injury knowledge (β = 1.121, P = .012, 95% CI = 1.025, 1.224), attitudes toward MSK injury disclosure (β = 1.065, P = .002, 95% CI = 1.024, 1.193), and perceived social norms surrounding MSK injury (β = 1.102, P = .004, 95% CI = 1.032, 1.178) were the strongest significant individual predictors of high intention to disclose MSK injury. The multivariate regression was able to account for 43% of the observed variance in high intention to disclose MSK injury with 78% sensitivity and 87% specificity. The strongest determinants of high intention to disclose an MSK injury after accounting for shared variance were attitudes toward MSK injury (β = 1.067, P = .005, 95% CI = 1.020, 1.117) and perceived social norms toward MSK injury (β = 1.099, P = .013, 95% CI = 1.020, 1.185).

The purpose of our study was to explore the association between intention to disclose concussion and intention to disclose MSK injury. Furthermore, we sought to compare the association between previously identified determinants of high intention to disclose a concussion and proposed determinants of high intention to disclose MSK injury using the same domains. A moderate association between high intention to disclose concussion and high intention to disclose MSK injury was demonstrated.

In assessing the hypothesized relationship between known determinants of high intention to disclose a concussion and potential determinants of MSK injury disclosure, we found strong relationships between attitudes and perceived social norms in the same domain. Moreover, correlations identified a moderate relationship between concussion and MSK injury knowledge. Perceived social norms was a strong predictor of intention to disclose both concussion and MSK injury, with attitudes further affecting the intention to disclose MSK injury. Of our participants with a concussion history, 25.0% described having purposefully failed to disclose a concussion symptom, a value that is lower than the previously observed 35.0%–50.0% rate of concussion nondisclosure.^20,21  In contrast, our participants with a history of MSK injury indicated having not disclosed 52.0% of their prior perceived MSK injuries.

To our knowledge, this is the first study to explore determinants of MSK injury disclosure in any population. Although a moderate association existed between intention to disclose a concussion and intention to disclose MSK injury, MSK injury was twice as likely to go undisclosed as a concussive injury (52.0% versus 25.0%, respectively). This finding may have been influenced by participants’ attitudes toward reporting MSK injuries. Musculoskeletal injury represents a broad spectrum, with some injuries leading to significant disability, whereas others present with minimal discomfort that resolves quickly. Therefore, some athletes may perceive some MSK injuries as minor, with no need to be reported. In contrast, widespread media attention and education programs have stressed the dangers of concussion to student-athletes, possibly influencing them to perceive a potential concussion as a more reportable event. Despite the lack of statistical significance in the difference between intention to disclose MSK injury and intention to disclose concussion, MSK injuries and concussion may not be viewed as equally necessary to report based on each athlete’s individual knowledge and perceptions or other potential determinants. Future researchers should strive to identify differences in reporting behaviors between these distinct injuries to improve the behavior of early disclosure. However, a stronger correlation was evident between high intention to report MSK injury and the behavior of reporting a concussion. This finding suggests that MSK injuries are more common than concussions, and significant knowledge gaps pertaining to what does and does not constitute a reportable MSK injury may exist. Moreover, although we found associations between injury disclosure determinants, strong individualized factors, such as attitudes and perceived social norms, may affect the behavior of true disclosure as it relates to MSK injury and concussions differently.

Knowledge of symptoms and potential consequences, attitudes, and perceived social norms toward concussions have been identified as strong determinants of high intention to disclose a concussive injury as well as true disclosure of concussive injury.^{22,23,25,28,30,33}  Authors of recent studies^33–36  have shown decreases in the rate of concussion symptom nondisclosure, highlighting the significance of interventions designed to improve knowledge of the injury itself while enhancing the social environment and context of sport participation to make disclosure more socially accepted. Local ecosystem-level factors such as university policies, support from coaches and administrators, as well as freedom for athletic trainers to do their job appropriately without fear of external pressure have increased concussion disclosure.^33,37  As high as 52% of perceived MSK injuries in our study population went undisclosed at the time of injury. Although various logical explanations could be offered (ie, lack of knowledge about potential consequences, perceived severity, disclosure not being socially acceptable, or fear of letting friends or family or coaches down), future research is necessary to explore and identify determinants of MSK injury disclosure in athletes. Previous results regarding the disclosure of concussion symptoms should be leveraged as a starting point for investigating interventions to improve MSK injury disclosure and, in turn, better understand the role disclosure behavior plays in the observed increased risk of MSK injury after a concussion.

In considering future educational programs on MSK injury disclosure, it is crucial to note the large observed deficit in knowledge of MSK injury symptoms in our sample. The median concussion symptom knowledge score was 18.7/24 versus a median MSK injury symptom score of 12.2/21. The most prevalent deficits in MSK injury knowledge pertained to the perception of pain as well as the role of emotional symptoms in identifying an MSK injury. Nearly all participants falsely dismissed symptoms such as headache, trouble falling asleep, [becoming] more emotional, irritability, and sadness as potential symptoms of MSK injury. Moreover, pain that goes away during activity, pain after activity, and pain at night were also frequently falsely dismissed as symptoms of MSK injury by most respondents. This is perhaps perpetuated by the stigma of “no pain, no gain” or the social context surrounding pain in sport.³⁸  Because these are key early-stage indicators of MSK injury, if they remain undisclosed, the condition may worsen.^38,39  These outcomes suggest that health care professionals seeking to improve MSK injury reporting could educate athletes on the signs and symptoms of MSK injury to help them better understand symptoms that may necessitate reporting or further evaluation.

As indicated earlier, MSK injuries in the United States are associated with significant consequences.^14,15  Given the high percentage of MSK injuries that went undisclosed in our sample, stakeholders who desire to further understand the increased risk of MSK injury after a concussion should start by focusing on community-based participatory research projects to improve MSK injury knowledge and help athletes better understand MSK injury. Previous interventional research targeted at improving concussion symptom knowledge, perceived social norms, attitudes, and perceived behavioral control offers an evidence-based starting point for an interventional strategy to improve MSK injury disclosure.^21,24  As our respondents revealed a 52% nondisclosure rate of MSK injury, perhaps the correlation between concussion and MSK injury is even more robust than presently understood. Creating an environment in which student-athletes feel support from coaches and other personnel in their athletics department to report MSK injury may help to mitigate some of the economic, social, and cultural burden of MSK injury by permitting detection of injuries in the early stages. These initiatives should seek to facilitate interpersonal communication among and between members of the health care team, student-athletes, and other stakeholder groups at the local level to influence the attributes of groups that drive safety-related behavior and improve shared values pertaining to the disclosure of MSK injury.^34–36 

This study was cross-sectional and from a single institution. Therefore, we could not measure serial changes in disclosure perceptions or behavior or address generalizability to different populations. The cross-sectional nature of the study also inhibited our ability to establish a causal relationship between disclosure behavior and increased likelihood of MSK injury after a disclosed concussion. Due to the paucity of research, little is known about which criteria must be met pertaining to the severity of an injury for an athlete to report the injury or what other determinants (eg, perceived norms, socioeconomic class, gender, coaches’ pressure) may alter that person’s decision, further limiting the interpretability of the present findings. Moreover, very few women disclosed a history of concussion, so we could not consider gender as a predictor variable. As with any survey, the potential also exists for response, recall, and social desirability bias. We attempted to reduce response bias by approaching all student-athletes who were undergoing the preseason screening at the time of the survey, achieving a 78% response rate. Additionally, we tried to address social desirability bias by ensuring the survey results were anonymous and by reminding the student-athletes that neither the research nor medical staff would be able to identify their responses.

Although we observed strong relationships between high intention to disclose concussion and high intention to disclose MSK injury, concussions were disclosed more frequently than MSK injuries. Some similarities between intention to report concussion and intention to disclose MSK injury were seen in the explored determinants, namely, the importance athletes’ attitudes toward disclosure played in both concussion and MSK injury disclosure. To better determine the role of injury disclosure in the known increased risk of sustaining an MSK injury after a concussion, a clearer understanding of the factors influencing MSK disclosure is necessary. Characterizing determinants of MSK injury disclosure will help stakeholders identify targets to include in interventional steps aimed at improving MSK injury disclosure in the early stages to help reduce the social, emotional, and economic costs of these conditions.