Factors associated with concussion during combatives activities in a military training environment

Highlights

• •This study identified specific characteristics and variables and their association with the risk of sustaining a concussion during combatives activities in a unique setting.
• •Females, elevated baseline psychological distress, worse baseline balance scores, and the endorsement of a history of headaches were each associated with greater odds of sustaining a combatives-related concussion.
• •Participating in high contact varsity sports was associated with lower odds of sustaining a combatives-related concussion.

Abstract

Background

Military service members routinely participate in combatives training (boxing, judo, martial arts, and hand-to-hand combat) to acquire and maintain mission essential skills. Despite injury mitigation strategies, high concussion incidence rates of 20.8 concussions per 100 exposures while participating in combative sports have been reported. The purpose of this study was to identify factors potentially associated with greater odds of sustaining a concussion in these combative activities in a military training environment.

Methods

A retrospective cohort study was conducted with participants enrolled at 4 military service academies participating in the concussion assessment, research, and education consortium from 2014 to 2020. Demographic information (site, varsity status, sport contact level, sex, concussion history, and headache history) and pre-injury baseline assessments (e.g., Balance Error Scoring System (BESS), Brief Symptom Inventory (BSI)) were collected at the time of enrollment. Univariate and multivariable logistic regression models were used to estimate the odds of sustaining a concussion while participating in combatives training during the follow-up period based on these pre-injury characteristics.

Results

During the study period, 17,681 participants (25% female;19.11 ± 1.45 years (mean ± SD)) completed a baseline assessment and 484 (35% female;19.88 ± 1.43 years) sustained a concussion during a combatives training. Univariate logistic regression models revealed females (odds ratio (OR) = 1.71; p < 0.001; 95% confidence interval (95%CI): 1.41–2.07), participating in high contact varsity sports (OR = 0.52; p < 0.001; 95%CI: 0.38–0.71), BSI total score (OR = 1.03; p < 0.001; 95%CI: 1.01–1.04), BESS total score (OR = 1.02; p < 0.001; 95%CI: 1.02–1.04), and headache history (OR = 1.43; p < 0.001; 95%CI: 1.18–1.73) were associated with greater odds of sustaining a combatives-related concussion. Multivariable models yielded similar results after controlling for significant covariates.

Conclusion

Females, higher BSI and BESS total scores at baseline, and participants with a history of headaches had greater odds of sustaining a combatives-related concussion during the follow-up period. Conversely, participants in high contact varsity sports had lower odds of sustaining a combatives-related concussion. These different variables should be taken into account when designing combatives training programs in a military setting.

Graphical abstract

1. Introduction

Mild traumatic brain injuries (mTBIs), also referred to as concussions, have become a major clinical concern among various populations, including athletes and military service members.¹ In the United States military, 404,450 service members have been diagnosed with an mTBI since 2000,² with 80% of documented cases in non-deployed service members.³ Additionally, approximately 800 concussions were diagnosed among U.S. Military Service Academy (MSA) cadets and midshipmen from 2014 to 2017.¹ A wealth of literature is available on the deleterious effects of concussions in different military populations.⁴ For example, Rigg and Mooney⁵ reported that military service members diagnosed with a concussion endorsed acute onset of headache, sleep disturbances, cognitive deficits, and potential psychiatric issues such as anxiety or irritability. Active duty service members and MSA cadets diagnosed with an incident concussion have also displayed greater risk of sustaining a lower extremity musculoskeletal injury within 12–15 months of the concussion.^6,7 Significant negative long-term neuropsychiatric, cognitive, and neurodegenerative sequalae have been associated with concussions as well.⁸ Additionally, concussions and subsequent injuries in service members generate a significant economic burden on the Military Health System, with traumatic brain injuries collectively costing approximately USD20 million dollars in a single year.⁹ With the evolving public awareness of the burden and potential long-term effects of concussions,^4,10 it is important to identify populations and activities associated with increased odds for sustaining these injuries to optimize injury risk mitigation strategies.

During their deployment to Iraq and Afghanistan from 2004 to 2008, 19% of soldiers reportedly engaged in hand-to-hand combat.¹¹ Combatives, tactical maneuvers, and grappling are critical skills incorporated into basic and advanced military training to prepare soldiers for these real-world scenarios.¹² These combative techniques are also incorporated into the MSA curriculum, which balances a high demand of academic, physical, and military requirements to prepare cadets for active-duty missions in the field.¹³ These types of training activities often result in exposure to repetitive head impacts.¹⁴ Increased head impact exposure may lead to an increased risk of sustaining a concussion, as cerebral tissue is unable to return to pre-impact status between strikes.¹⁵ While limited research is available on specific military combative training activities and injury rates,¹⁶ a wealth of epidemiological studies have been published assessing the injury rates of similar combatives activities in the general population and among Olympic athletes.^14,17,18 Boxing and mixed martial arts are notorious for elevated head impacts during participation and have displayed incidence rates of 20.8 and 14.7 concussions per 100 fights or athletic exposures, respectively.¹⁴ Further, a 17.5% and 25.2% concussion rate has been observed in striking sports, such as kickboxing, and in grappling sports, such as Brazilian jiu-jitsu, respectively.¹⁴

To reduce concussion rates, it is important to identify specific factors associated with elevated odds of sustaining a concussion. A paucity of research exists assessing the impact of concussion risk factors on the odds of sustaining a concussion during combative activities in military training environments and MSA settings. Previous studies have examined the impact of specific risk factors and concussion risk among amateur and professional combative athletes and found that sex, age, weight, bout type, and bout length were not associated with elevated risk of concussion.¹⁹ Among MSA members, approximately 32% of reported concussions occur during military training or physical education (PE) classes, some of which include combative activities.¹ Within this population, Van Pelt et al.¹ reported that MSA cadets with medical comorbidities such as prior concussion and a recent headache history reported within 3 months of baseline concussion testing displayed a significantly higher risk of sustaining an incident concussion. MSA members in their first year at their respective academy, females, and varsity athletes playing high-contact sports also displayed increased odds of sustaining a concussion during military training and PE classes.¹

Research is available reporting concussion incidence rates in combative sports;^14,17,18 however, risk factors for concussion during combative training in a non-deployed military environment are under-studied.²⁰ Identifying specific risk factors associated with concussion during combative training will help inform injury mitigation policies. Early risk identification is critical to decreasing the incidence of concussions to maintain military readiness and the capacity to engage in battle. Therefore, the purpose of this study was to examine the association between various risk factors and the odds of sustaining a concussion during combative activities among MSA cadets and midshipmen.

2. Methods

2.1. Design and setting

A retrospective cohort study was conducted among cadets and midshipmen enrolled at 4 MSA to examine the association between potential pre-injury baseline risk factors and the odds of sustaining a concussion during combatives training. Data were obtained from male and female MSA members who provided informed consent to participate in the Concussion Assessment, Research, and Education (CARE) consortium from 2014 to 2020. The MSA curriculum is unique and requires cadets and midshipmen to take various PE classes, including boxing and combatives (defined as hand-to-hand fighting techniques). MSA members are also required to participate in a National Collegiate Athletic Association (NCAA), competitive club, or intramural sport. Combative activities identified for this study were competitive club boxing and judo, intramural boxing, grappling, wrestling, and PE boxing and combatives classes. All procedures for this study were reviewed and approved by the Institutional Review Board at each site and by the U.S. Army Human Research Protection Office (EIRB: 965959). Each participant provided informed consent prior to data collection and their baseline concussion assessment.

2.2. Procedures

A detailed methodology of the CARE consortium study has been published previously.²¹ The following summarizes the procedures pertinent to the current analysis. Each MSA member underwent baseline concussion testing during their first semester at their respective academy. During baseline testing, pertinent demographic information (height, weight, age, site, varsity status, sport contact level, and sex) medical health history information (concussion history, headache history, and migraine headache history), and pre-injury clinical assessments (Balance Error Scoring System (BESS), Brief Symptom Inventory-18 (BSI-18), Immediate Post-concussion Assessment and Cognitive Test (ImPACT), Automated Neuropsychological Assessment Metrics (ANAM), and Brief Sensation Seeking Scale (BSSS)) were collected. Participants with a concussion diagnosed during the follow-up period underwent a standardized assessment administered by a healthcare provider (athletic trainer or primary care physician) within 48 h (initial) post-injury. During the initial intake portion of this evaluation timepoint, information regarding the activity the subject was engaged in at the time of injury was collected. The evaluating clinician noted the competition level (NCAA, club, intramural sport, PE class, military training, or outside of organized sport and military activity) and the activity during which the injury occurred (e.g., NCAA football, PE class boxing, intramural soccer). Baseline demographic information and initial post-injury assessments were documented at each respective site and recorded in a centralized computer database (QuesGen Systems, Burlingame, CA, USA).

2.3. Injury surveillance

A concussion was operationally defined as “a change in brain function following a force to the head, which may be accompanied by temporary loss of consciousness but is identified in awake individuals with measures of neurologic and cognitive dysfunction”.²² Participants diagnosed with a concussion by the primary healthcare provider at their respective site based on the results of a standardized evaluation were included for analysis. This study only analyzed incident concussion cases documented during the observation period. Additional cases were omitted if there was no injury activity (i.e., boxing, club handball, free time, etc.) reported at the initial post-injury evaluation timepoint.

2.4. Instruments

2.4.1. BESS

The BESS is used by clinicians to evaluate balance and postural control. The assessment consists of 3 stances: double-leg, single-leg (non-dominant foot on the ground), and tandem (non-dominant foot behind the dominant foot).²³ Each position is performed on a stable and unstable (i.e., foam pad) surface. The subject is instructed to close their eyes and keep their hands on their hips for 20 s.²³ The clinician tallies the number of errors that occur during the test. An error is defined as opening eyes, removing hands from hips, coming out of the stance position, lifting the forefoot or heel, abducting the hip by more than 30 degrees, or taking more than 5 s to return to position after falling out of it.²³ The BESS has previously displayed poor to moderate sensitivity (16%–60%) in diagnosing concussions within 24 h of injury.²⁴

2.4.2. BSI-18

The BSI-18 provides a total score measuring global psychological distress and is comprised of 3 sub-scores assessing anxiety, somatization, and depression.²⁵ It consists of 18 questions graded on a 0 (not at all) to 4 (extremely) Likert scale, with increased scores indicating elevated psychological distress.²⁵ BSI-18 total scores have previously displayed a high correlation with convergent measures of psychological and affective distress and moderate correlations with functional status and disability assessments.²⁶

2.4.3. ImPACT

The ImPACT is a computerized neurocognitive assessment that produces composite scores for verbal and visual memory, processing speed, and reaction time. Combined, the ImPACT composite scores have displayed a sensitivity ranging from 79.2%–81.9% for concussion diagnosis.²⁷ The reaction time component by itself has previously displayed a sensitivity of 29.3%.²⁸ However, robust reaction time deficits have been observed post-concussion despite minimal significant differences from pre-injury baseline assessments.²⁹ The ImPACT was used as the standard of care computerized neurocognitive assessment at 3 of the 4 sites in the current study.

2.4.4. ANAM

The ANAM is a computerized neurocognitive assessment using eight sub-tests (simple reaction time, code-substitution learning, procedural reaction time, mathematical processing, matching to sample, delayed code substitution, simple reaction time 2, and go/no-go) and an overall composite score to diagnose and manage concussions.³⁰ The composite score has displayed area under the receiver operating characteristic (ROC) curves starting at 0.72 within 24 h of injury and falling to 0.56 up to 45 days post-injury.³⁰ The simple reaction time component alone has displayed ROC values of 0.63 and 0.53 at those same time intervals, respectively.³⁰ The ANAM was used as the standard of care computerized neurocognitive assessment at one site in the current study.

2.4.5. BSSS-8

The BSSS-8 is a short version of a standardized assessment (Sensation Seeking Scale, Form V) that is used to determine how likely an individual is to engage in high-risk behaviors.³¹ It consists of 8 questions derived from 4 subscales of thrill and adventure seeking, experience seeking, disinhibition, and boredom susceptibility.³¹ Each question is graded on a Likert scale ranging from 0 (strongly disagree) to 4 (strongly agree) with elevated scores indicating increased sensation seeking and a general increased risk of engaging in high-risk behaviors.³¹ In a collegiate population, a one-point increase on any single item of the BSSS-8 has been associated with a 12% increased risk of sustaining a concussion.³²

2.5. Statistical analysis

Descriptive statistics were calculated for demographic variables and placed into 1 of 3 injury type groups: participants who did not sustain a concussion, participants who sustained a combatives concussion, and participants who sustained a non-combatives concussion. Combatives concussions were defined as injuries that occurred during boxing, combatives, martial arts, or grappling activities in classes, competitive club practices or games, and intramural events at the participants’ respective MSA. Based on previous research, sport contact level was defined as: (a) NCAA-contact, (b) NCAA-limited contact, or (c) NCAA non-contact and non-NCAA athlete.³³ Table 1 displays the sports that were categorized into the 3 contact levels. One-way analysis of variance (ANOVA) models were calculated to determine whether there were significant differences between mean measures of continuous variables obtained from baseline testing among participants who sustained a combatives concussion, a non-combatives concussion, or no concussion. The assumptions for analysis of variance models were assessed qualitatively and statistically using a Shapiro-Wilk test; they demonstrated normal distributions and all assumptions were adequately met. Post hoc testing using a Bonferroni correction was used on models that were statistically significant to determine differences between groups. Pearson’s Chi-squared test statistics were utilized to compare ordinal variables to expected normative values between the three injury type groups.

Table 1.

Sport contact classification.

Sport contact level	Sport
NCAA contact sports	Basketball
	Diving
	Football
	Ice hockey
	Lacrosse
	Soccer
	Water polo
	Wrestling
NCAA limited contact	Baseball
	Cross country/track
	Fencing
	Field events
	Gymnastics
	Softball
	Volleyball
NCAA non-contact	Golf
	Rifle
	Swimming
	Tennis
	All non-NCAA cadetsa

^aincludes all cadets participating in club or intramural sports.

Abbreviation: NCAA = National Collegiate Athletics Association.

Univariate logistic regression models were used to estimate the odds of sustaining a concussion while participating in a combatives activity during the follow-up period. These models compared participants who sustained a combatives concussion to participants who did not sustain a concussion. Participants who sustained a non-combatives concussion were not included in these logistic regression models. Univariate logistic regression models were also conducted to assess the impact of sex (male or female), concussion history (0 previous concussions or ≥1 previous concussion), NCAA sport contact level, endorsing a history of any type of headache within 3 months (no history of headache or ≥1 headache), history of diagnosed migraines (0 diagnosed migraines or ≥1 diagnosed migraine), BSI-18 total score, and BSSS-8 total score reported during baseline testing, as each of these covariates has demonstrated a significant association with odds of sustaining a concussion in previous research.^1,32 While baseline balance and reaction time may not be predictive of future concussions in collegiate student athletes,³⁴ univariate models were still conducted on baseline BESS total scores, ImPACT composite reaction time (RT), and ANAM composite simple RT, as both are traits considered to be important attributes in combative sports performance^35,36 and concussion management.^29,37 A multivariable logistic regression model was then used to assess the odds of sustaining a concussion in combatives activities while controlling for statistically significant covariates. The assumptions for these multivariable models were assessed qualitatively and statistically using a Pearson’s correlation coefficient test; all assumptions were met, with no multicollinearity observed among the variables analyzed.

Statistical significance for each univariate model was set at α < 0.05. Covariates were included in multivariable models if the univariate p value was <0.100. Odds ratios (ORs) and 95% confidence intervals (95%CIs) were calculated for each univariate and multivariable logistic regression model. All statistical analyses were conducted in StataSE software Version 14.2 (StataCorp, College Station, TX, USA).

3. Results

During the study period, 17,681 participants completed a baseline assessment, and 1528 (8.6% of the entire sample) participants sustained a concussion, of which 484 (31.7%) occurred during a combatives activity. A majority of concussions were sustained in non-combatives activities. Fig. 1 displays the number of combatives concussions sustained by activity type. A majority of these combative injuries occurred during boxing, followed by hand-to-hand combat, judo, and grappling. Table 2 displays descriptive statistics for baseline demographic and clinical measures along with the proportion of specific characteristics obtained during baseline testing. The one-way ANOVA and Pearson’s Chi-squared results are also displayed in Table 2, comparing the characteristics and demographics of participants who sustained a combatives concussion to participants who were not injured or who sustained a non-combatives concussion. One-way ANOVA results suggest participants who sustained a combatives concussion were older, shorter, and weighed less than participants who sustained a non-combatives concussion or who did not sustain any injury. At baseline, participants who sustained any type of concussion reported more BSI-18 symptoms and displayed balance deficits on the BESS compared to uninjured participants. No significant differences were detected between the combative and non-combative concussion groups on either measure. No other differences in continuous measures of demographic or clinical measures were observed between the two groups.

Fig. 1.

Number of concussions sustained by combative activity type.

Table 2.

Descriptive statistics for pre-injury baseline demographics and covariates.

	No injury	Combatives injury	Non-combatives injury
	(n = 16,153)	(n = 484)	(n = 1044)		p
Demographics (mean ± SD)				F
Height (cm)	176.40 ± 13.38	174.67 ± 9.67	175.14 ± 16.00	6.98	0.001
Weight (kg)	75.12 ± 13.70	72.58 ± 13.56	74.50 ± 15.59	8.66	<0.001
Age (year)	19.06 ± 1.43	19.88 ± 1.43	19.53 ± 1.53	125.94	<0.001
Clinical measures (mean ± SD)				F
BSSS-8	3.32 ± 0.68	3.31 ± 0.67	3.31 ± 0.64	0.32	0.724
BSI-18	3.08 ± 5.84	4.18 ± 6.04	4.93 ± 7.34	48.55	<0.001
BESS	14.87 ± 6.74	16.02 ± 6.45	15.63 ± 6.77	11.86	<0.001
ImPACT RT	0.59 ± 0.10	0.60 ± 0.08	0.59 ± 0.10	0.98	0.374
ANAM simple RT	250.33 ± 26.11	259.48 ± 17.74	253.12 ± 20.83	0.76	0.466
Covariates (% of specified covariate)				χ2
Sex
Male	12,320 (76%)	316 (65%)	622 (60%)	171.87	<0.001
Female	3833 (24%)	168 (35%)	422 (40%)
Concussion history
0	13,207 (82%)	372 (74%)	757 (73%)	45.00	<0.001
≥1	2946 (18%)	102 (26%)	278 (27%)
3-month headache history
No history of headaches	11,775 (73%)	307 (64%)	633 (61%)	63.01	<0.001
History of headaches	4378 (27%)	168 (20%)	392 (38%)
History of migraines
Migraine history	426 (3%)	10 (2%)	41 (4%)	7.03	0.030
No migraine history	15,727 (97%)	467 (96%)	987 (95%)
Sport contact level
Non-NCAA athlete	12,361 (77%)	408 (84%)	657 (63%)	132.67	<0.001
NCAA limited	1256 (8%)	33 (7%)	92 (9%)
NCAA contact	2526 (15%)	43 (9%)	294 (28%)

Note: percentages might add up not to 100% due to rounding.

Abbreviations: ANAM = Automated Neuropsychological Assessment Metrics; BESS = Balance Error Scoring System; BSI-18 = Brief Symptom Inventory-18; BSSS-8 = Brief Sensation Seeking Scale-8; ImPACT = Immediate Post-Concussion Assessment and Cognitive Test; NCAA = National Collegiate Athletics Association; RT = reaction time.

Pearson’s Chi-squared results indicated the frequency distribution of sex, concussion history, headache history, migraine history, and sport contact level were significantly different from expected frequencies for these predictor variables. Both combatives and non-combatives concussions were sustained by a greater number of females, participants with a headache history, participants reporting a migraine history, and participants reporting a concussion history than expected. Participants in the NCAA non-contact/non-NCAA athlete group sustained more combatives concussions and a smaller number of non-combatives concussions than expected.

Univariate logistic regression models (Table 3) revealed females experienced 71% higher odds of sustaining a concussion during combatives activities compared to males, and participants reporting a headache history of any type experienced 43% higher odds of sustaining a concussion during combatives activities than participants without a headache history. A one-point increase in baseline BSI-18 (higher behavioral symptoms) and BESS total scores (poorer balance) was associated with 3% and 2% increased odds of sustaining a concussion during combative activities respectively. Participants participating in high contact NCAA sports experienced 48% decreased odds of sustaining a combatives-activity concussion compared to non-contact NCAA and non-NCAA participants. The odds of sustaining a concussion during a combatives activity were not impacted by participating in an NCAA limited-contact sport, baseline reaction time measured by either the ImPACT or ANAM, diagnosed migraine headache history, or BSSS-8 total scores. Similar results were observed in a multivariable model (Table 3) controlling for sex, concussion history, sport contact level, headache history, BESS total score, and BSI-18 total score. Females (77% increase), participants reporting a history of headaches (26% increase), and participants reporting elevated baseline BSI-18 and BESS total scores (3% and 2% increase, respectively) displayed a significant impact on the odds of sustaining a combatives-activity concussion. Participants participating in a contact NCAA sport displayed 39% decreased odds of sustaining a combatives-activity concussion compared to non-contact NCAA and non-NCAA participants.

Table 3.

Univariate and multivariable logistic regression models to determine the odds of sustaining a combatives-activity concussion.

Variable	Univariate model			Multivariable model
	OR	p	95%CI	OR	p	95%CI
Sex
Male	1.00			1.00
Female	1.71	<0.001	1.41–2.07	1.77	<0.001	1.44–2.17
Concussion history
0	1.00			1.00
≥1	1.20	0.104	0.96–1.50	1.22	0.106	0.96–1.55
Sport contact level
Non-NCAA/non-contact NCAA athlete	1.00			1.00
NCAA limited	0.80	0.217	0.56–1.14	1.02	0.921	0.71–1.47
NCAA contact	0.52	<0.001	0.38–0.71	0.61	0.005	0.44–0.86
3-month headache history
No history of headaches	1.00			1.00
History of headaches	1.43	<0.001	1.18–1.73	1.26	0.032	1.02–1.55
Migraine history
Migraine history	1.00
No migraine history	0.78	0.441	0.41–1.47	—	—	—
BSI-18 total score	1.03	<0.001	1.01–1.04	1.03	0.001	1.01–1.04
BESS total	1.02	<0.001	1.01–1.04	1.02	0.001	1.01–1.04
ImPACT RT	1.67	0.256	0.69–4.02	—	—	—
ANAM simple RT	1.02	0.280	0.99–1.05	—	—	—
BSSS total score	0.97	0.623	0.85–1.11	—	—	—

Abbreviations: 95%CI = 95% confidence interval; ANAM = Automated Neuropsychological Assessment Metrics; BESS = Balance Error Scoring System; BSI-18 = Brief Symptom Inventory-18; BSSS-8 = Brief Sensation Seeking Scale-8; ImPACT = Immediate Post-Concussion Assessment and Cognitive Test; NCAA = National Collegiate Athletics Association; OR = odds ratio; RT = reaction time.

4. Discussion

This study aimed to determine the impact of specific variables on the odds of sustaining a concussion during a combatives activity in an MSA setting. After controlling for significant covariates, females, participants endorsing elevated baseline BSI-18 total symptom burdens, participants with higher baseline BESS total scores, and participants reporting a history of headaches during baseline testing all displayed increased odds of sustaining a concussion in a combatives activity. Participants in high contact varsity sports displayed decreased odds of sustaining a combatives concussion compared to non-contact NCAA and non-NCAA participants.

4.1. Combatives activities and concussion

Boxing was associated with the greatest number of incident concussions in the current study followed by combatives, judo, and grappling/wrestling. This has been supported by previous epidemiological studies on combative sports across different populations.³⁸ Broadly, a meta-analysis of injuries across a variety of full-contact combat sports reported the lowest odds of sustaining any injury in grappling sports (judo), with the greatest odds in striking sports (boxing).³⁸ Furthermore, the results from this meta-analysis revealed concussions comprised a greater proportion of total injuries in boxing (14% of all injuries) compared to different martial art activities (4%–6% of all injuries).³⁸ These findings also likely reflect the compulsory requirements for all cadets and midshipmen to complete boxing and combatives courses to prepare them to serve as an officer in the military upon graduation. While these activities are conducted in a controlled environment, certain activities included in the physical program are associated with increased exposure to head impacts and potential injuries. For example, the nature and objective of boxing includes a primary goal to deliver strikes to the head, while combative sports like grappling do not rely as much on head impacts to score points or win matches. It is important to note that we did not measure the cumulative exposure or concussion incidence rates during these combatives activities, which is a limitation of these data.

4.2. Impact of covariates on concussion

In the current study, a significant difference was observed in height, weight, and age between subjects who sustained a combatives concussion and those who did not. Previous research assessing both overall injury incidence and head injury incidence specifically revealed an association between weight and age and injury risk that contradicts the findings of the current study. Increased age and experience in combatives activities has been associated with decreased overall injury risk.³⁹ Comparisons between weight classes of various combatives activities, from amateur to professional levels of competition, revealed no association between weight and injury risk.^18,39 However, when focusing on head impact exposure in professional combat sport athletes, previous research has shown head injury incidence increases with progressively heavier weight classes.⁴⁰

Females displayed increased odds of sustaining a combatives-related concussion compared to males. This has been supported by previous research among a service academy cohort where females displayed an OR of 2.02 for sustaining any concussion in any activity (i.e., athletics, academy training, free time) compared to males.¹ It has been suggested that the increased rates of concussion reporting observed among females may be related to differences in neck strength musculature⁴¹ and to an increased likelihood females have displayed to endorse concussion-like symptoms and report concussions.^42,43 Participants reporting a headache history within 3 months of their baseline assessment displayed greater odds of sustaining a concussion than participants without a headache history. Reporting a history of migraine headaches specifically did not demonstrate increased odds of sustaining a concussion. Similar findings were reported previously in an MSA population.¹ The association between this increase in odds of sustaining a concussion and headache history is unclear,^1,44 though it may be explained by the symptoms reported with both migraine and non-migraine headaches. Migraines and headaches can present with symptoms such as light sensitivity and decreased cognitive function, which are symptoms frequently reported with concussions.44, 45, 46 The overlap in symptoms may increase the likelihood that a concussion is diagnosed if the subject is experiencing a headache before the injury occurs.

Incremental increases in the endorsement of psychological distress were associated with increased odds of sustaining a combatives-activity concussion in the current study. Previous research among service academy cadets also found increased odds of sustaining a concussion were associated with elevated BSI-18 scores, specifically somatization sub-scores.¹ The endorsement of elevated somatization symptoms has been previously associated with a tendency to express psychological or emotional problems as physical ailments.⁴⁷ In the management of concussions, a subject endorsing elevated somatic symptoms at baseline may tend to report exacerbated symptom burdens after sustaining a concussion and throughout the recovery process.¹

Elevated sensation seeking scores were not associated with odds of sustaining a concussion. The lack of association between odds of sustaining a concussion and elevated sensation seeking scores reported on the BSSS-8 has been observed previously among service academy cadets.¹ While contradictory results have been observed among non-MSA NCAA athletes where a one-point increase has been associated with 12% greater odds of sustaining a concussion.³² Elevated sensation seeking has been likened to elevated impulsivity, which is a trait observed in participants reporting a history of multiple concussions.³² The low BSSS-8 scores and lack of difference between the three groups in the current study suggest the cohort demonstrated a low impulsivity in these activities, potentially explaining the lack of impact observed between sensation seeking and sustaining a combatives concussion. Concussion history and reaction time composite scores were also not associated with sustaining a concussion in the current study. The lack of impact observed for reaction time on the odds of sustaining a concussion is supported by previous studies among a physically active, adolescent cohort.⁴⁸ Previous research in a service academy setting found concussion history was associated with 2.01 increased odds of sustaining a concussion during academy training.¹ Several reasons have been suggested to explain the impact of concussion history on the odds of sustaining a subsequent concussion. Increased incidence of subsequent injury may be attributed to differences in injury recall and reporting behaviors, or it may be explained by residual physiological effects from the primary concussion pertaining to deficits in neurocognition, neuromotor control, and sensorimotor control.⁴⁹

Finally, participation in a contact NCAA sport had a protective effect, as these participants displayed decreased odds of sustaining a combatives-related concussion compared to participants participating in non-contact NCAA athletics or non-NCAA athletics. This is supported by a previous study in which participants in NCAA contact sports were 44% less likely to sustain an academy training-related concussion.¹ The reason for this observation is unclear; however, it has been hypothesized that competition level and sport type may be associated with an enhanced development in multiple domains of cognitive function.⁵⁰ Elite athletes in dynamic sports (e.g., rugby, soccer) may develop a greater ability to make quick and flexible decisions and execute subsequent rapid movements compared to amateur athletes.⁵⁰ These skills may be essential to reducing concussions during combatives activities.

4.3. Clinical implications

The current study identified the percentage of concussions sustained during specific mandatory combatives activities included in the MSA curriculum. This information can be used to develop mitigation strategies to minimize the number of concussions that occur during these required activities. The most recent concussion in sport group (CISG) position statement reports on a wealth of emerging research regarding concussion prevention, including policy and rule changes, personal protective equipment, and different training strategies.³⁷ Using mouthguards in collision sports was associated with a 26% decrease in concussion rates, with no difference observed between dentist-fit mouthguards compared to off-the-shelf types.⁵¹ Rules minimizing intentional contact with the head or neck led to a significant reduction in concussion rates in contact sports during both practices (incidence rate ratio (IRR) = 0.29) and games (IRR = 0.51).⁵¹ Utilizing a neuromuscular warm-up strategy consisting of different balance, whole body resistance, static neck contraction, and plyometric maneuvers at least 3 times weekly was associated with a significant reduction (59%) in concussion rates in adolescent rugby.⁵¹ These are all strategies that could prove beneficial in decreasing the odds of sustaining a combatives concussion in the populations identified in the current study. While limiting intentional head contact can be the most challenging prevention measure to implement based on the nature of certain combatives activities, other strategies including mouthguard use, limiting high-impact events, and implementing a neuromuscular warm-up may help reduce the odds of sustaining a combatives-activity concussion.

4.4. Limitations/future research

This study was not without limitations. The subject population consisted of healthy MSA cadets and midshipmen and so may not be directly applicable to the general population, different age groups, or people with various comorbidities. The methodology of the current study was also unable to account for cumulative head impact exposure across the various activities. Emerging research has reported an association between the use of specific psychostimulant medications on different aspects of post-concussion recovery; however, medication use was not accounted for in the current study due to the lack of a statistically significant association between specific medications and the risk of sustaining a concussion.⁵² Furthermore, the use of psychostimulants would be medically disqualifying for most cadets and midshipmen, preventing them from entering military service; as a result, medication use was not evaluated in the current study. Future research should focus on methods to decrease the odds of sustaining a concussion among participants endorsing the factors associated with increased odds for sustaining a concussion in the current study. Additional studies should also examine the protective effect observed in athletes participating in different sports and competition levels.

5. Conclusion

The current study identified specific factors associated with increased odds of sustaining a concussion during combatives activities in an MSA setting. Females, reporting a history of headaches, endorsing elevated baseline psychological distress, and demonstrating balance deficits during baseline were all associated with increased odds of sustaining a concussion in combatives activities. Engaging in combatives activities in this controlled environment is critical to preparing service members for the demands of the military. The variables identified in this study should be taken into consideration when designing policies or guidelines for combatives activities to reduce the odds of sustaining a concussion to optimize and maintain troop and soldier readiness.

Authors’ contributions

MJA was responsible for data collection, data analysis, and manuscript writing; SRM, JDR, and AS was responsible for data collection; GM and JJ were responsible for data collection and conceptualization and design of the study; JR and RMB were both responsible for data collection; KLC was responsible for data analysis, manuscript writing, and conceptualization and design of the study; MHR was responsible for manuscript writing and conceptualization and design of the study; KW was responsible for manuscript writing; SJS, SPB, MAM, TM, and PFP were responsible for conceptualization and design of the study. All authors have read and approved the final version of the manuscript, and agree with the order of presentation of the authors.

Competing interests

The authors declare that they have no competing interests.