The Value of Preseason Screening for Injury Prediction: The Development and Internal Validation of a Multivariable Prognostic Model to Predict Indirect Muscle Injury Risk in Elite Football (Soccer) Players

Abstract

Background: In elite football (soccer), periodic health examination (PHE) could provide prognostic factors to predict injury risk. Objective: To develop and internally validate a prognostic model to predict individualised indirect (non-contact) muscle injury (IMI) risk during a season in elite footballers, only using PHE-derived candidate prognostic factors. Methods: Routinely collected preseason PHE and injury data were used from 152 players over 5 seasons (1st July 2013 to 19th May 2018). Ten candidate prognostic factors (12 parameters) were included in model development. Multiple imputation was used to handle missing values. The outcome was any time-loss, index indirect muscle injury (I-IMI) affecting the lower extremity. A full logistic regression model was fitted, and a parsimonious model developed using backward-selection to remove factors that exceeded a threshold that was equivalent to Akaike's Information Criterion (alpha 0.157). Predictive performance was assessed through calibration, discrimination and decision-curve analysis, averaged across all imputed datasets. The model was internally validated using bootstrapping and adjusted for overfitting. Results: During 317 participant-seasons, 138 I-IMIs were recorded. The parsimonious model included only age and frequency of previous IMIs; apparent calibration was perfect, but discrimination was modest (C-index = 0.641, 95% confidence interval (CI) = 0.580 to 0.703), with clinical utility evident between risk thresholds of 37-71%. After validation and overfitting adjustment, performance deteriorated (C-index = 0.589 (95% CI = 0.528 to 0.651); calibration-in-the-large = − 0.009 (95% CI = − 0.239 to 0.239); calibration slope = 0.718 (95% CI = 0.275 to 1.161)).

Can prognostic factors for indirect muscle injuries in elite football (soccer) players be identified using data from preseason screening? An exploratory analysis using routinely-collected periodic health examination records

“…The exception was that if a hamstring IMI was sustained by a player more than 12 months (but less than 3 years) prior to PHE, their odds of sustaining a lower extremity I-IMI significantly increased 2.2-fold, which has not been previously reported. Although not directly comparable, earlier studies have also shown that a history of a previous hamstring IMI is specifically associated with an increased hazard of future hamstring IMIs in elite players.11 20 22 Nevertheless, the uncertainty in our estimates (demonstrated by wide 95% confidence intervals) and differences observed during the sensitivity analyses mean that this variable only has provisional prognostic value and needs to be established in further confirmatory studies.Indeed, age was the only variable that could be considered as an important prognostic factor, which is easily obtained even without conducting PHE; after allowing for height and weight, age retained a highly significant, modest and precise association that was robust to both Our findings confirm those of our multivariable prognostic model study, where age was retained in a model to determine individualised I-IMI risk over a season in elite players (OR = 1.10, 95% CI = 1.03 to 1.17) 17. Other studies have shown that as a multivariable prognostic factor, age is associated specifically with increased hamstring IMI risk (OR range 1.40-1.78),…”

“…28 Of the remaining eligible variables, 10 were used in our multivariable prognostic model study (represented by 12 parameters). 17 With the exception of age at PHE (which was used for adjustment purposes in this study), these candidates were therefore excluded. 28 The final number of candidate variables included for exploratory analysis was 36.…”

“…8 10 Because the predictive function of PHE remains unsubstantiated 3 11 and given that indirect (non-contact) muscle injuries (IMIs) are a significant problem observed in elite football (accounting for 30.3% to 47.9% of all injuries), [12][13][14][15][16] we have recently developed and internally validated a multivariable prognostic model to predict individualised lower extremity IMI risk in elite players using PHE data. 17 However, sample size limitations meant that only 10 candidate prognostic factors could be considered in our model and these were selected using data quality assessment, clinical reasoning, or on the basis of our related systematic review. 11 The performance of our model was modest and we concluded that its use would not be beneficial in clinical practice.…”

“…11 The performance of our model was modest and we concluded that its use would not be beneficial in clinical practice. 17 Furthermore, our previous systematic review also highlighted several methodological limitations of the current evidence, which specifically included inadequate reporting of outcomes, prognostic factor measurement and reliability. 11 Additionally, while most studies performed appropriate statistical analyses, continuous prognostic factor measurements were often categorised [18][19][20][21] and non-linear associations were not investigated, [18][19][20][21][22][23][24] which does not conform to current methodological recommendations.…”

“…[25][26][27] To assist the development of future prognostic models, there is a clear need to ascertain the existence of robust and novel prognostic factors that have an association with IMIs. 17 Therefore, using routinely collected data from a 5-season period, we conducted an exploratory cohort study to examine: 1) prognostic associations between PHE-derived variables and IMI outcomes in elite footballers, using a wider set of candidates than had previously been considered 17 and; 2) the prognostic value of these PHE-derived variables over and above standard anthropometric variables of age (which has previously confirmed prognostic value 11 17 ), height and weight. In particular, both linear and non-linear associations between candidate variables and outcomes were explored which, to the best of our knowledge, has not been conducted previously.…”