Objective: Injuries from Muaythai fighting is a common outcome. Prevention strategies to reduce injury incidence and severity have focused on increased safety regulations. However, this is based on anecdotal evidence rather than empirical evidence measuring theseverity and possible predictors. The objective of this study was to determine the relationship between injury severity, age, competitive level, previous fight experience, and the fight outcome. Also, a prediction model for injury severity was built using these variables.

Methods: A total of 150 fighters were randomly sampled at sanctioned fight events and information on severity, demographics, thenumber of previous fights, professional versus amateur status, and the outcome of the bout was collected. Simple linear regression was used to assess each variable’s relationship with severity. Multiple linear regression was then used to determine a final model for the prediction of injury severity.

Results: Simple linear regression showed that competitive status (professional versus amateur) was statistically significant in predictinginjury severity, while the total number of fights and age were not. However, these models were determined to have poor fit (R2= .28, .19, .39). Multiple linear regression showed that the most appropriate final model consisted of two covariates - competitive level and fight outcome. The final model had a fit of 15%, and MSE of 590.8.

Conclusions: Two out of 4 independent variables examined appear to be related to the prediction of injury severity (the level of the fighter’s competition and the outcome of the fight).

Injury, Severity Score, Fight, Boxing

Strotmeyer SJ (2017) Predicting Injury Severity among Muaythai Fighters Using Age, Fight Experience, Competitive Status and Fight Outcome. Ann Sports Med Res 4(5): 1120.

Muaythai, often translated into English as Thai boxing, is the national sport of Thailand and is a martial art in which fighters use several parts of the body for offensive and defensive purposes. Fists, elbows, knees, and feet are used to strike the opponent or defend from the attacks of their opponent. A typical match usually consists of 3 to 5 rounds (depending on the category of fighters), 3-minutes per round, with a 1-minute break between each round. As with many combat sports, the potential for injury exists [1]. Prior research has reported similar injury rates karate and taekwondoto that in Muaythai [2-4]. The first study to investigate differentiation in rate as to the level of competition showed an association with younger, professional fighters [5]. Further, as recording injury incidence to establish rates was the study’s focus, measuring severity was not assessed [4]. Review of published data for the striking martial arts has found that younger participants and those with less experience have a higher risk of serious injury [6-12] Suggesting that among Muaythai fighters, younger participants and those with less experience would be more prone to serious injury, and these may predict the level of reported severity. In this study, our objective was to establish a relationship between injury severity, age, competition status, fight experience and fight outcome by performing multiple regression analyses.

Data collection

We randomly sampled 150 fighters out of 450 participantscompeting in the largest sanctioned event in the United States in 2015. Upon receiving consent, we emailed invitations to participate in a post-fight web survey. Nonrespondents were sent 2 addtitional email reminders. The series of emails resulted in a 72% response rate (108 returns) within a 6-week time period. For each respondent, we collected information regarding demographic characteristics,injury severity, thelevel of competition, overall fight experience and protective equipment used in the most recent competition. A sample of 20 respondents from the data is shown in Table (1).

Statistical analysis

Statistical analyses were performed using STATA v.11 software. Descriptive analysis provided in Table (2). Additionally, histograms to check the distributions of the continuous variables is shown in the following Figure (1). To check the relationships between each of the variables,

Table 1: A Sample of 20 Observations from respondents on independent variables

Figure 1: Histograms of the variables: Injury Severity Score, Age, and Number of fights.

pairwise scatterplotsare shown in Figure (2). No strong linear associations between were seen in Figure 2, and this is verified in the correlation matrix in Table (3). (NOTE: did not need to put categorical in)

We then fit simple linear regression models for the independent variables (age, fight experience, fight outcome, competitive status), and summarized the model results in Table (4). Simple linear regression results indicated the fight outcome (winning or losing) and the competitive status (professional versus amateur) weresignificant individual predictors for reported injury severity. As the coefficients of determination were low,these variables were subsequently analyzed with multiple regression models.

Model selection & diagnostics We started model selection by fitting the maximum model

Table 2: Descriptive Analysis of the variables of interest (mean, standard deviation, range).

Figure 2: Pairwise scatterplots of injury severity score (ISS), age, total fights, fight outcome and competition level.

Table 3: Pairwise correlation matrix of injury severity score (ISS), age, total fights, bout outcome and competitive status.

 Table 4: Simple linear regression models.

with injury severity as the dependent variable; and competitive status and fight outcome as independent variables. Additionally, while age was not shown to be an important predictor in the simple regression model, we opted to keep this in to evaluate possible attenuation. A total of seven models were assessed for fit as shown in Table (5).

R-squared, MSE, Fp, AIC, and BIC. Therefore, the final model indicated was: Injury severity = 16.4 x competitive status - 13.683 x fight outcome +29.10.

Residuals and linearity were checked for the final model and shown in Figure (3).

To assure collinearity was not a problem collinearity diagnostics were assessed using the variance inflation factor (VIF) shown in Table (6). Outliers and influential points were

Figure 3: Residuals versus Fitted values for Model 6.

Table 5: Multiple Linear Regression Model Selection.

noted, but not deleted due to data accuracy.

The purpose of this study was to develop a prediction model for injury severity in addition to determining if a relationship exists between injury severity, age, fight experience, competitive status and fight outcome. The analysis showed that competitive status (professional versus amateur) and the fight result (winning versus losing) was strongly related to the injury severity, but that age and combat experience was not. The final regression model can be used to predict the injury severity score using the beta coefficients for competitive status and fight outcome. The model fit was somewhat low, explaining only 15% of the variance in injury severity.

Reducing the incidence of injury among fighters presents a unique challenge within injury epidemiology. The goal of inflicting damage to ultimately knock out an opponent makes implementing safety measures incongruous. Previous research within the sport of Muaythai was merely descriptive, reporting rates of injury and simple severity classifications, but no attempts to study what

Table 6: Variance Inflation Factors (VIF) of the independent variables

predicts severity. In this study, there appears to be a relationship between self-reported injury severity, the competitive level of the fighter, and the outcome of the fight immediately before survey administration. Professional fighters losing a match would have the higher injury severity scores predictively, compared to amateur fighters winning a fight. As professionals do not fight with protective equipment, this association seems reasonable. While more experienced and better conditioned as a group, professionals may remain more vulnerable to injury since the honed fight skills may result in a greater ability to deliver strikes with considerable force and transfer more energy. Injuries occur when humans encounter energy forces that are larger than the body’s normal tolerance levels for energy absorption. The level of energy encountered exceeds a threshold.

Whether or not the fighter wins or loses (fight outcome) was an interesting finding. Losing fighters were more inclined to report a higher injury severity versus those winning. Perhaps there is some bias that losing a fight is often due to being injured and a higher likelihood to reflect that in the severity score. Another interesting finding was that while age was associated with more severe injury among previous studies involving other martial arts and Muaythai, this was not apparently the case among this Muaythai fighter sample [5,13]. Age alone was not a statistically significant predictor of injury severity, and even when trying to force it into the multiple regression models, it did not appear to attenuate the competitive status. We thought age would initially be related to the professional or amateur status as before reaching a professional career; the younger, lessexperienced fighters spend time in the amateur ranks to develop.

Further, the number of fights, often a proxy for experience, did not have a strong relationship to the injury severity score, which would further suggest that even more, seasoned and developed fighters are equally, if not more vulnerable to injurythan relative novices. This too stands to reason, as each contest is a unique instance when injury risk would likely remain the same and not be reduced by increasing exposure. The notion that seasoned or experienced may prevent injury due to conditioning and prowess suggests an area to for exploration in future studies. Stronger study designs such as longitudinal studies with repeated measures over time may help elucidate the relationships found here in thepreliminary analysis.

1. Pieter W. Martial arts injuries. Med Sport Sci. 2005; 48: 59-73.

2. Zetaruk MN, Violan MA, Zurakowski D, Micheli LJ. Injuries in martial arts: a comparison of five styles. Br J sports Med. 2005s; 39: 29-33.

3. Gartland S, Malik MH, Lovell ME. Injury and injury rates in Muay Thai kick boxing. Br J Sports Med. 2001; 35: 308-313.

4. Gartland S, Malik MH, Lovell M. A prospective study of injuries sustained during competitive Muay Thai kickboxing. Clin J Sport Med. 2005; 15: 34-36.

5. Strotmeyer S, Coben JH, Fabio A, Songer T, Brooks M. Epidemiology of Muay Thai fight-related injuries. Inj epidemiol. 2016; 3: 30.

6. Buschbacher RM, Shay T. Martial arts. Physical medicine and rehabilitation clinics of North America. 1999; 10: 35-47.

7. Wilkerson LA. Martial arts injuries. The Journal of the American Osteopathic Association. 1997; 97: 221-6.

8. Birrer RB. Trauma epidemiology in the martial arts. The results of an eighteen-year international survey. Am J Sports Med. 1996; 24: 72.

9. Stricevic MV, Patel MR, Okazaki T, Swain BK. Karate: historical perspective and injuries sustained in national and international tournament competitions. Am J Sports Med. 1983; 11: 320-324.

10.Zemper ED, Pieter W. Injury rates during the 1988 US Olympic Team Trials for taekwondo. Br J Sports Med. 1989; 23: 161-164.

11. Oler M, Tomson W, Pepe H, Yoon D, Branoff R, Branch J. Morbidity and Mortality in the Martial Arts:: A Warning. J Trauma. 1991; 31: 251- 253.

12.McLatchie GR. Analysis of karate injuries sustained in 295 contests. Injury. 1976; 8: 132-134.

13. Zazryn TR, Finch CF, McCrory P. A 16-year study of injuries to professional kickboxers in the state of Victoria, Australia. Br J Sports Med. 2003; 37: 448-451.