Getting the most out of intensive longitudinal data: a methodological review of workload-injury studies

A novel approach for sports injury risk prediction: based on time-series image encoding and deep learning

The rapid development of big data technology and artificial intelligence has provided a new perspective on sports injury prevention. Although data-driven algorithms have achieved some valuable results in the field of sports injury risk assessment, the lack of sufficient generalization of models and the inability to automate feature extraction have made it challenging to deploy research results in the real world. Therefore, this study attempts to build an injury risk prediction model using a combination of time-series image encoding and deep learning algorithms to address this issue better. This study used the time-series image encoding approach for feature construction to represent relationships between values at different moments, including Gramian Angular Summation Field (GASF), Gramian Angular Difference Field (GADF), Markov Transition Field (MTF), and Recurrence Plot (RP). Deep Convolutional Auto-Encoder (DCAE) learned the image-encoded data for representation to obtain features with good discrimination, and the classifier was performed using Deep Neural Network (DNN). The results from five repeated experiments show that the GASF-DCAE-DNN model is overall better in the training (AUC: 0.985 ± 0.001, Gmean: 0.930 ± 0.007, Sensitivity: 0.997 ± 0.003, Specificity: 0.868 ± 0.013) and test sets (AUC: 0.891 ± 0.026, Gmean: 0.830 ± 0.027, Sensitivity: 0.816 ± 0.039, Specificity: 0.845 ± 0.022), with good discriminative power, robustness, and generalization ability. Compared with the best model reported in the literature, the AUC, Gmean, Sensitivity, and Specificity of the GASF-DCAE-DNN model were higher by 23.9%, 27.5%, 39.7%, and 16.2%, respectively, which confirmed the validity and practicability of the model in injury risk prediction. In addition, differences in injury risk patterns between the training and test sets were identified through shapley additivity interpretation. It was also found that the training volume was an essential factor that affected injury risk prediction. The model proposed in this study provides a powerful injury risk prediction tool for future sports injury prevention practice.

Intraindividual variability in sleep among athletes: A systematic review of definitions, operationalizations, and key correlates

Via systematic review with narrative synthesis of findings, we aimed to document the ways by which researchers have defined, operationalized, and examined sleep variability among athletes. We identified studies in which scholars examined intraperson variability in sleep among athletes via a search of six databases (Web of Science, Embase, Medline, PsycINFO, CINHAL Plus, and ProQuest Dissertations and Theses Global) using a protocol that included keywords for the target outcome (sleep*), population (athlet* OR sport*), and outcome operationalization (variability OR variation OR "standard deviation" OR fluctuate OR fluctuation OR stability OR instability OR reactivity OR IIV OR intraindividual). We complemented this primary search with citation searching of eligible articles. Assessments of study quality captured eight core elements, namely aims/hypotheses, sample size justification, sample representativeness, number of days sleep assessed, measures of sleep and its correlates, missing data, and inferences and conclusions. From a total of 1209 potentially relevant papers, we identified 16 studies as meeting our eligibility criteria. Concept definitions of variability were notably absent from this work and where available were vague. Quantitative deviations from one's typical level of target sleep metrics reflected the essence by which all but one of the research teams operationalized sleep variability. We assessed the overall quality of empirical work as moderate in nature. We propose a working definition of sleep variability that can inform knowledge generation on the temporal, day-to-day dynamics of sleep functioning that is required for personalized interventions for optimizing sleep health.

Injury incidence in male elite youth football players is associated with preceding levels and changes in training load

Objectives

Elite youth football players miss out on a large part of seasonal training due to injury. Limited research suggests an association between external and internal training load (TL) and injury incidence in elite youth football. This study analysed external and internal TL variables and their association with injury incidence in a group of male elite youth football players over four seasons.

Methods

Measures of external and internal TL and injury incidence of 56 male elite youth football players (age 17-19 years) were collected throughout four seasons. Heart rate, session rating of perceived exertion andGlobal Positioning System (GPS) variables were analysed. Individual players' TL during the 30 days leading up to injury was compared with 30-day injury-free control periods. Change in TL through the periods was also analysed.

Results

Eighty-five injuries were included for analysis, showing that for most TL variables, the average levels were significantly lower during the period leading up to injury. Significant increases for the majority of TL variables were also found during the periods leading up to injury, while the control periods did not show any significant change.

Conclusion

A lower and/or increasing average TL volume over 30 days might increase the risk of injury in male elite youth football players. Avoiding long-term drops in TL and balance increases in TL might be beneficial to reduce injury risk.

Relationship Between Posterior Shoulder Tightness and Lower-Limb Flexibility in College Baseball Players

CONTEXT

Posterior shoulder tightness evaluated as limited glenohumeral internal rotation and horizontal adduction is a risk factor for throwing-related shoulder and elbow injuries. Given that the throwing motion uses the entire body kinematics, limited lower-limb flexibility might be associated with posterior shoulder tightness. Therefore, we aimed to investigate the relationships between posterior shoulder tightness and lower-limb flexibility in college baseball players.

DESIGN

Cross-sectional study.

SETTING

University laboratory.

PARTICIPANTS

Twenty-two college baseball players (20 right-handed and 2 left-handed).

METHODS

To investigate the relationships between glenohumeral range of motion and lower-limb flexibility using simple linear regression analysis, we measured passive range of motion of glenohumeral internal rotation and horizontal adduction, hip internal/external rotation in the prone/sitting position, ankle dorsiflexion, and quadriceps and hamstrings flexibility from both shoulders and legs.

RESULTS

Our analysis indicated that decreases in the lead leg hip external rotation in the prone position were moderately associated with limitations in glenohumeral internal rotation (R2 = .250, β [95% confidence interval, CI] = 0.500 [0.149 to 1.392], P = .018) and horizontal adduction (R2 = .200, β [95% CI] = 0.447 [0.051 to 1.499], P = .019) on the throwing shoulder. Furthermore, there were significant moderate relationships between decreases in glenohumeral internal rotation and limited lead leg quadriceps flexibility (R2 = .189, β [95% CI] = 0.435 [0.019 to 1.137], P = .022), and between decreases in glenohumeral horizontal adduction and limited stance leg ankle dorsiflexion (R2 = .243, β [95% CI] = 0.493 [0.139 to 1.438], P = .010).

CONCLUSION

College baseball players with limited lower-limb flexibilities including the lead leg hip external rotation in the prone position, the lead leg quadriceps flexibility, and the stance leg ankle dorsiflexion showed excessive posterior shoulder tightness. The current results support the hypothesis that lower-limb flexibility is associated with posterior shoulder tightness in college baseball players.

Forecasting neuromuscular recovery after anterior cruciate ligament injury: Athlete recovery profiles with generalized additive modeling

A retrospective analysis of longitudinally collected athlete monitoring data was conducted to generate a model of neuromuscular recovery after anterior cruciate ligament (ACL) injury and reconstruction (ACLR). Neuromuscular testing data including countermovement jump (CMJ) force-time asymmetries and knee extensor strength (maximum voluntary contraction_ext ) asymmetries (between-limb asymmetry index-AI) were obtained from athletes with ACLR using semitendinosus (ST) autograft (n = 29; AI measurements: n = 494), bone patellar tendon bone autograft (n = 5; AI measurements: n = 88) and noninjured controls (n = 178; AI measurements: n = 3188). Explosive strength measured as the rate of torque development was also calculated. CMJ force-time asymmetries were measured over discrete movement phases (eccentric deceleration phase, concentric phase). Separate additive mixed effects models (additive mixed effects model [AMM]) were fit for each AI with a main effect for the surgical technique and a smooth term for the time since surgery (days). The models explained between 43% and 91% of the deviance in neuromuscular recovery after ACLR. The mean time course was generated from the AMM. Comparative neuromuscular recovery profiles of an athlete with an accelerated progression and an athlete with a delayed progression after a serious multiligament injury were generated. Clinical Significance: This paper provides a new perspective on the utility of longitudinal athlete monitoring including routine testing to develop models of neuromuscular recovery after ACLR that can be used to characterize individual progression throughout rehabilitation.

Evidence for stratified conflicts of interest policies in research contexts: a methodological review

OBJECTIVES

The purpose of this study was to conduct a methodological review of research on the effects of conflicts of interest (COIs) in research contexts.

DESIGN

Methodological review.

DATA SOURCES

Ovid.

ELIGIBILITY CRITERIA

Studies published between 1986 and 2021 conducting quantitative assessments of relationships between industry funding or COI and four target outcomes: positive study results, methodological biases, reporting quality and results-conclusions concordance.

DATA EXTRACTION AND SYNTHESIS

We assessed key facets of study design: our primary analysis identified whether studies stratified industry funding or COI variables by magnitude (ie, number of COI or disbursement amount), type (employment, travel fees, speaking fees) or if they assessed dichotomous variables (ie, conflict present or absent). Secondary analyses focused on target outcomes and available effects measures.

RESULTS

Of the 167 articles included in this study, a substantial majority (98.2%) evaluated the effects of industry sponsorship. None evaluated associations between funding magnitude and outcomes of interest. Seven studies (4.3%) stratified industry funding based on the mechanism of disbursement or funder relationship to product (manufacturer or competitor). A fifth of the articles (19.8%) assessed the effects of author COI on target outcomes. None evaluated COI magnitude, and three studies (9.1%) stratified COI by disbursement type and/or reporting practices. Participation of an industry-employed author showed the most consistent effect on favourability of results across studies.

CONCLUSIONS

Substantial evidence demonstrates that industry funding and COI can bias biomedical research. Evidence-based policies are essential for mitigating the risks associated with COI. Although most policies stratify guidelines for managing COI, differentiating COIs based on the type of relationship or monetary value, this review shows that the available research has generally not been designed to assess the differential risks of COI types or magnitudes. Targeted research is necessary to establish an evidence base that can effectively inform policy to manage COI.

Statistical analysis considerations within longitudinal studies of physical qualities in youth athletes: A qualitative systematic methodological review

Background

The evaluation of physical qualities in talent identification and development systems is vital and commonplace in supporting youth athletes towards elite sport. However, the complex and dynamic development of physical qualities in addition to temporal challenges associated with the research design, such as unstructured data collection and missing data, requires appropriate statistical methods to be applied in research to optimise the understanding and knowledge of long-term physical development.

Aim

To collate and evaluate the application of methodological and statistical methods used in studies investigating the development of physical qualities within youth athletes.

Methods

Electronic databases were systematically searched form the earliest record to June 2021 and reference lists were hand searched in accordance with the PRISMA guidelines. Studies were included if they tested physical qualities over a minimum of 3 timepoints, were observational in nature and used youth sporting populations.

Results

Forty articles met the inclusion criteria. The statistical analysis methods applied were qualitatively assessed against the theoretical underpinnings (i.e. multidimensional development, non-linear change and between and within athlete change) and temporal challenges (i.e. time variant and invariant variables, missing data, treatment of time and repeated measures) encountered with longitudinal physical testing research. Multilevel models were implemented most frequently (50%) and the most appropriately used statistical analysis method when qualitatively compared against the longitudinal challenges. Independent groups ANOVA, MANOVA and X2 were also used, yet failed to address any of the challenges posed within longitudinal physical testing research.

Conclusions

This methodological review identified the statistical methods currently employed within longitudinal physical testing research and addressed the theoretical and temporal challenges faced in longitudinal physical testing research with varying success. The findings can be used to support the selection of statistical methods when evaluating the development of youth athletes through the consideration of the challenges presented.

Attenuated Lower Limb Stretch-Shorten-Cycle Capacity in ACL Injured vs. Non-Injured Female Alpine Ski Racers: Not Just a Matter of Between-Limb Asymmetry

A retrospective analysis of routine countermovement jump (CMJ) testing, a coupled eccentric-concentric (stretch-shorten-cycle: SSC) movement, was performed in female elite alpine skiers with anterior cruciate ligament (ACL) reconstruction (ACLR) and without ACLR. A total of 567 tests obtained from the daily training environment were analyzed in 41 elite female athletes (non-injured control: n = 30, ACLR: n = 17), including n = 6 athletes with pre-injury data, between 16 and 32 years of age from alpine ski racing (n = 32) and skier cross (n = 9). Bilateral CMJ testing was conducted on a dual force plate system, and the limb-specific vertical ground reaction force (Fz) was analyzed to obtain the net eccentric deceleration impulse (Ecc), lower limb stiffness (Stiff), maximal vertical jump height (JH), peak external mechanical power (PP) exerted on the body center of mass (BCM), modified-reactive-strength-index (RSImod), and the loss in BCM velocity during the final phase of the takeoff Δ(Vmax-Vtakeoff). Eccentric and concentric phase-specific between-limb asymmetry indexes (AIs) were also calculated. Additive mixed effects models (AMMs) were used to compare the age-dependent and post-injury time course change between groups. The mean values for non-injured controls >25 years of age were used as a comparative benchmark for recovery given the absence of pre-injury data. Net eccentric deceleration impulse increased and Δ(Vmax-Vtakeoff) decreased with age for the non-injured control group (p < 0.001) while between-limb AI (mean ± SD) fell between 1 ± 5% for the concentric phase and 3 ± 7% for the eccentric deceleration phase. Between-limb asymmetry became smaller in ACLR skiers with time-from-surgery to reach non-injured control values by 2 years, but SSC function, such as JH and PP, remained depressed up to 5 years post-surgery (p < 0.01), indicating impairments in SSC function. This highlights the importance of evaluating SSC performance capacity alongside vertical jump force-time asymmetries in female ACLR alpine skiers.

Synthetic playing surfaces increase the incidence of match injuries in an elite Rugby Union team

OBJECTIVES

To examine differences in match injury incidence between three playing surfaces in elite Rugby Union.

DESIGN

Prospective cohort.

METHODS

Match injury incidence was assessed in 89 elite Rugby Union players over two-seasons of professional competition (44 matches, 1014 h player exposure). Match injury incidence was assessed on three different playing surfaces; natural grass, hybrid (natural grass combined with approximately 3% synthetic fibres) and fully synthetic (sand and rubber infill). Overall injury incidence, contact and non-contact injury incidence, and the incidence of minor (≤7 d lost) and major (≥8 d lost) injuries were considered using mixed effect models.

RESULTS

Overall match injury incidence doubled on hybrid and synthetic surfaces compared to natural grass (hybrid: OR = 2.58 [95% CI 1.65-4.03], p < 0.001; synthetic: OR = 2.16 [95% CI 1.07-4.37], p = 0.033). Furthermore, the odds of sustaining a contact injury on a pitch containing any synthetic content also increased compared to natural grass (hybrid: OR = 2.31 [95% CI 1.41-3.78], p = 0.001; synthetic: OR = 2.19 [95% CI 1.00-4.77], p = 0.049). The hybrid surface elicited a four times greater likelihood of non-contact injury incidence compared to natural grass (OR = 4.18 [95% CI 1.16-15.04], p = 0.028). However, the playing surface did not affect the severity of match injuries (all p > 0.05).

CONCLUSIONS

The present study suggests that even a small percentage (3%) of synthetic content in the playing surface significantly increases match injury incidence, with an effect seen on both contact and non-contact injury incidence. These findings are important to enable practitioners to be aware of the injury implications of playing matches on hybrid and synthetic pitches.

Not straightforward: modelling non-linearity in training load and injury research

OBJECTIVES

To determine whether the relationship between training load and injury risk is non-linear and investigate ways of handling non-linearity.

METHODS

We analysed daily training load and injury data from three cohorts: Norwegian elite U-19 football (n=81, 55% male, mean age 17 years (SD 1)), Norwegian Premier League football (n=36, 100% male, mean age 26 years (SD 4)) and elite youth handball (n=205, 36% male, mean age 17 years (SD 1)). The relationship between session rating of perceived exertion (sRPE) and probability of injury was estimated with restricted cubic splines in mixed-effects logistic regression models. Simulations were carried out to compare the ability of seven methods to model non-linear relationships, using visualisations, root-mean-squared error and coverage of prediction intervals as performance metrics.

RESULTS

No relationships were identified in the football cohorts; however, a J-shaped relationship was found between sRPE and the probability of injury on the same day for elite youth handball players (p<0.001). In the simulations, the only methods capable of non-linear modelling relationships were the quadratic model, fractional polynomials and restricted cubic splines.

CONCLUSION

The relationship between training load and injury risk should be assumed to be non-linear. Future research should apply appropriate methods to account for non-linearity, such as fractional polynomials or restricted cubic splines. We propose a guide for which method(s) to use in a range of different situations.

Training load monitoring in team sports: a practical approach to addressing missing data

Training load (TL) is a modifiable risk factor that may provide practitioners with opportunities to mitigate injury risk and increase sports performance. A regular problem encountered by practitioners, however, is the issue of missing TL data. The purpose of this study was to examine the impact of missing TL data in team sports and to offer a practical and effective method of missing value imputation (MVI) to address this. Session rating of perceived exertion (sRPE) data from 10 male professional soccer players (age, 24.8 ± 5.0 years; height, 181.2 ± 5.1 cm; mass, 78.7 ± 6.4 kg) were collected over a 32-week season. Data were randomly removed at a range of 5-50% in increments of 5% and data were imputed using 12 MVI methods. Performance was measured using the normalized root-mean-square error and mean of absolute deviations. The best-fitting MVI method across all levels of missingness was Daily Team Mean (DTMean). Not addressing missing sRPE data may lead to more inaccurate calculations of other TL metrics (e.g., acute chronic workload ratio, training monotony, training strain). The DTMean MVI method may provide practitioners with a practical and effective approach to addressing the negative consequences of missing TL data.

Is the Acute: Chronic Workload Ratio (ACWR) Associated with Risk of Time-Loss Injury in Professional Team Sports? A Systematic Review of Methodology, Variables and Injury Risk in Practical Situations

BACKGROUND

The acute: chronic workload ratio (ACWR) is an index of the acute workload relative to the cumulative chronic workloads. The monitoring of physical workloads using the ACWR has emerged and been hypothesized as a useful tool for coaches and athletes to optimize performance while aiming to reduce the risk of potentially preventable load-driven injuries.

OBJECTIVES

Our goal was to describe characteristics of the ACWR and investigate the association of the ACWR with the risk of time-loss injuries in adult elite team sport athletes.

DATA SOURCES

PubMed, EMBASE and grey literature databases; inception to May 2019.

ELIGIBILITY CRITERIA

Longitudinal studies that assess the relationship of the ACWR and time-loss injury risk in adult professional or elite team sports.

METHODS

We summarized the population characteristics, workload metrics and ACWR calculation methods. For each workload metric, we plotted the risk estimates for the ACWR in isolation, or when combined with chronic workloads. Methodological quality was assessed using a modified version of the Downs and Black scale.

RESULTS

Twenty studies comprising 2375 injuries from 1234 athletes (all males and mean age of 24 years) from different sports were included. Internal (65%) and external loads (70%) were collected in more than half of the studies and the session-rating of perceived exertion and total distance were the most commonly collected metrics. The ACWR was commonly calculated using the coupled method (95%), 1:4 weekly blocks (95%) and subsequent week injury lag (80%). There were 14 different binning methods with almost none of the studies using the same binning categories.

CONCLUSION

The majority of studies suggest that athletes are at greater risk of sustaining a time-loss injury when the ACWR is higher relative to a lower or moderate ACWR. The heterogenous methodological approaches not only reflect the wide range of sports studied and the differing demands of these activities, but also limit the strength of recommendations.

PROSPERO REGISTRATION NUMBER

CRD42017067585.

The Use of the Intensive Longitudinal Methods to Study Financial Well-Being: A Scoping Review and Future Research Agenda

Financial well-being is a positive financial condition that has an objective (e.g., income) and a subjective (e.g., financial satisfaction) side. Much research has examined financial well-being using cross-sectional and classic longitudinal designs. More recently, researchers have begun to examine financial well-being using intensive longitudinal designs, collecting data in a repeated (at least five measurements) and intensive (short time interval between measurements) way. The goal of the current study was to systematically review all published research on financial well-being using intensive longitudinal methods, summarize themes from this work, and suggest future research directions. Searching three databases (Scopus, PsycINFO, Econpapers), we found nine articles that respected inclusion and exclusion criteria. From each selected article, we extracted information about (1) research field diffusion, (2) data collection methods, (3) financial well-being's definition and operationalization, (4) research questions addressed and (5) data analysis. Findings showed that most of the studies adopted an interval-contingent research design, collecting data once a day; that both the objective and subjective sides of the construct were assessed, and that, most of the time, the construct was conceptualized as financial stress (lack of financial well-being). Different kinds of research questions were addressed across studies and these were often analyzed using multilevel analysis. In the discussion section, future research directions are suggested.

A Cherry, Ripe for Picking: The Relationship Between the Acute-Chronic Workload Ratio and Health Problems

OBJECTIVE

To investigate whether the relationship between the acute-chronic workload ratio (ACWR) and health problems varies when different methodological approaches are used to quantify it.

DESIGN

Prospective cohort study.

METHODS

An online questionnaire was used to collect daily health and training information from 86 elite youth footballers for 105 days. The relationship between players' training load and health was analyzed using a range of different definitions of ACWR and health problems. We used 21-day and 28-day chronic periods, coupled and uncoupled calculations, and the exponentially weighted moving average and rolling average. Acute-chronic workload ratio data were categorized as low, medium, or high, using predefined categories and z scores. We compared medium to high, medium to low, and low to high categories. The outcome was defined in 3 ways: "all health problems," "all injuries," and "new noncontact injuries." We performed random-effects logistic regression analyses of all combinations, for a total of 108 analyses.

RESULTS

We recorded 6250 athlete-days and 196 health problems. Of the 108 analyses performed, 23 (21%) identified a statistically significant (P<.05) association between the ACWR and health problems. A greater proportion of significant associations were identified when using an exponentially weighted moving average (44% of analyses), when comparing low to high categories (33%), and when using the "all health problems" definition (33%).

CONCLUSION

The relationship between the ACWR and health problems was dependent on methodological approach. J Orthop Sports Phys Ther 2021;51(4):162-173. Epub 20 Jan 2021. doi:10.2519/jospt.2021.9893.

Evaluating Methods for Imputing Missing Data from Longitudinal Monitoring of Athlete Workload

Missing data can influence calculations of accumulated athlete workload. The objectives were to identify the best single imputation methods and examine workload trends using multiple imputation. External (jumps per hour) and internal (rating of perceived exertion; RPE) workload were recorded for 93 (45 females, 48 males) high school basketball players throughout a season. Recorded data were simulated as missing and imputed using ten imputation methods based on the context of the individual, team and session. Both single imputation and machine learning methods were used to impute the simulated missing data. The difference between the imputed data and the actual workload values was computed as root mean squared error (RMSE). A generalized estimating equation determined the effect of imputation method on RMSE. Multiple imputation of the original dataset, with all known and actual missing workload data, was used to examine trends in longitudinal workload data. Following multiple imputation, a Pearson correlation evaluated the longitudinal association between jump count and sRPE over the season. A single imputation method based on the specific context of the session for which data are missing (team mean) was only outperformed by methods that combine information about the session and the individual (machine learning models). There was a significant and strong association between jump count and sRPE in the original data and imputed datasets using multiple imputation. The amount and nature of the missing data should be considered when choosing a method for single imputation of workload data in youth basketball. Multiple imputation using several predictor variables in a regression model can be used for analyses where workload is accumulated across an entire season.

"To Tech or Not to Tech?" A Critical Decision-Making Framework for Implementing Technology in Sport

The current technological age has created exponential growth in the availability of technology and data in every industry, including sport. It is tempting to get caught up in the excitement of purchasing and implementing technology, but technology has a potential dark side that warrants consideration. Before investing in technology, it is imperative to consider the potential roadblocks, including its limitations and the contextual challenges that compromise implementation in a specific environment. A thoughtful approach is therefore necessary when deciding whether to implement any given technology into practice. In this article, we review the vision and pitfalls behind technology's potential in sport science and medicine applications and then present a critical decision-making framework of 4 simple questions to help practitioners decide whether to purchase and implement a given technology.

What Role Do Chronic Workloads Play in the Acute to Chronic Workload Ratio? Time to Dismiss ACWR and Its Underlying Theory

AIM

The aim of this study was to examine the associations between the injury risk and the acute (AL) to chronic (CL) workload ratio (ACWR) by substituting the original CL with contrived values to assess the role of CL (i.e., the presence and implications of statistical artefacts).

METHODS

Using previously published data, we generated a contrived ACWR by dividing the AL by fixed and randomly generated CLs, and we compared these results to real data. We also reproduced previously reported subgroup analyses, including dichotomising players' data above and below the median CL. Our analyses follow the same, previously published modelling approach.

RESULTS

The analyses with original data showed effects compatible with higher injury risk for ACWR only (odd ratios, OR: 2.45, 95% CI 1.28-4.71). However, we observed similar effects by dividing AL by the "contrived" fixed and randomly generated CLs: OR 1.95 (1.18-3.52) dividing by 1510 (average CL); and OR ranging from 1.16 to 2.07, using random CL 1.53 (mean). Random ACWRs reduced the variance relative to the original AL and further inflated the ORs (mean OR 1.89, from 1.42 to 2.70). ACWR causes artificial reclassification of players compared to AL alone. Finally, neither ACWR nor AL alone confer a meaningful predictive advantage to an intercept-only model, even within the training sample (c-statistic 0.574/0.544 vs. 0.5 in both ACWR/AL and intercept-only models, respectively).

DISCUSSION

ACWR is a rescaling of the explanatory variable (AL, numerator), in turn magnifying its effect estimates and decreasing its variance despite conferring no predictive advantage. Other ratio-related transformations (e.g., reducing the variance of the explanatory variable and unjustified reclassifications) further inflate the OR of AL alone with injury risk. These results also disprove the etiological theory behind this ratio and its components. We suggest ACWR be dismissed as a framework and model, and in line with this, injury frameworks, recommendations, and consensus be updated to reflect the lack of predictive value of and statistical artefacts inherent in ACWR models.

More than a Metric: How Training Load is Used in Elite Sport for Athlete Management

Training load monitoring is a core aspect of modern-day sport science practice. Collecting, cleaning, analysing, interpreting, and disseminating load data is usually undertaken with a view to improve player performance and/or manage injury risk. To target these outcomes, practitioners attempt to optimise load at different stages throughout the training process, like adjusting individual sessions, planning day-to-day, periodising the season, and managing athletes with a long-term view. With greater investment in training load monitoring comes greater expectations, as stakeholders count on practitioners to transform data into informed, meaningful decisions. In this editorial we highlight how training load monitoring has many potential applications and cannot be simply reduced to one metric and/or calculation. With experience across a variety of sporting backgrounds, this editorial details the challenges and contextual factors that must be considered when interpreting such data. It further demonstrates the need for those working with athletes to develop strong communication channels with all stakeholders in the decision-making process. Importantly, this editorial highlights the complexity associated with using training load for managing injury risk and explores the potential for framing training load with a performance and training progression mindset.

Training Load and Injury Part 2: Questionable Research Practices Hijack the Truth and Mislead Well-Intentioned Clinicians

BACKGROUND

In this clinical commentary, we highlight issues related to conceptual foundations and methods used in training load and injury research. We focus on sources of degrees of freedom that can favor questionable research practices such as P hacking and hypothesizing after the results are known, which can undermine the trustworthiness of research findings.

CLINICAL QUESTION

Is the methodological rigor of studies in the training load and injury field sufficient to inform training-related decisions in clinical practice?

KEY RESULTS

The absence of a clear conceptual framework, causal structure, and reliable methods can promote questionable research practices, selective reporting, and confirmation bias. The fact that well-accepted training principles (eg, overload progression) are in line with some study findings may simply be a consequence of confirmation bias, resulting from cherry picking and emphasizing results that align with popular beliefs. Identifying evidence-based practical applications, grounded in high-quality research, is not currently possible. The strongest recommendation we can make for the clinician is grounded in common sense: "Do not train too much, too soon"-not because it has been confirmed by studies, but because it reflects accepted generic training principles.

CLINICAL APPLICATION

The training load and injury research field has fundamental conceptual and methodological weaknesses. Therefore, making decisions about planning and modifying training programs for injury reduction in clinical practice, based on available studies, is premature. Clinicians should continue to rely on best practice, experience, and well-known training principles, and consider the potential influence of contextual factors when planning and monitoring training loads. J Orthop Sports Phys Ther 2020;50(10):577-584. Epub 1 Aug 2020. doi:10.2519/jospt.2020.9211.

Limited Support for Trunk and Hip Deficits as Risk Factors for Athletic Knee Injuries: A Systematic Review With Meta-analysis and Best-Evidence Synthesis

OBJECTIVE

To determine whether neuromuscular deficits in trunk and hip-related function are risk factors for athletic knee injuries.

DESIGN

Etiology systematic review with meta-analysis.

LITERATURE SEARCH

Six online databases (MEDLINE, Web of Science, Embase, CINAHL, Scopus, and SPORTDiscus) were searched up to April 2019.

STUDY SELECTION CRITERIA

Studies assessing trunk and hip neuromuscular function as risk factors for knee injuries in healthy athletic populations were included.

DATA SYNTHESIS

Outcomes were synthesized quantitatively using meta-analysis of odds ratios, and qualitatively using best-evidence synthesis.

RESULTS

Twenty-one studies met the inclusion criteria. There was very low-certainty evidence that greater hip external rotation strength protected against knee injuries (odds ratio = 0.78; 95% confidence interval: 0.70, 0.87; P<.05). There was limited evidence that deficits in trunk proprioception and neuromuscular control, and the combination of excessive knee valgus and ipsilateral trunk angle when landing unilaterally from a jump, may be risk factors for knee injuries.

CONCLUSION

Most variables of trunk and hip function were not risk factors for injuries. Further research is required to confirm whether hip external rotation strength, trunk proprioception and neuromuscular control, and the combination of knee valgus angle and ipsilateral trunk control are risk factors for future knee injuries. J Orthop Sports Phys Ther 2020;50(9):476-489. Epub 1 Aug 2020. doi:10.2519/jospt.2020.9705.

Training Load and Its Role in Injury Prevention, Part 2: Conceptual and Methodologic Pitfalls

In part 2 of this clinical commentary, we highlight the conceptual and methodologic pitfalls evident in current training-load-injury research. These limitations make these studies unsuitable for determining how to use new metrics such as acute workload, chronic workload, and their ratio for reducing injury risk. The main overarching concerns are the lack of a conceptual framework and reference models that do not allow for appropriate interpretation of the results to define a causal structure. The lack of any conceptual framework also gives investigators too many degrees of freedom, which can dramatically increase the risk of false discoveries and confirmation bias by forcing the interpretation of results toward common beliefs and accepted training principles. Specifically, we underline methodologic concerns relating to (1) measure of exposures, (2) pitfalls of using ratios, (3) training-load measures, (4) time windows, (5) discretization and reference category, (6) injury definitions, (7) unclear analyses, (8) sample size and generalizability, (9) missing data, and (10) standards and quality of reporting. Given the pitfalls of previous studies, we need to return to our practices before this research influx began, when practitioners relied on traditional training principles (eg, overload progression) and adjusted training loads based on athletes' responses. Training-load measures cannot tell us whether the variations are increasing or decreasing the injury risk; we recommend that practitioners still rely on their expert knowledge and experience.

Time series analysis of intensive longitudinal data in psychosomatic research: A methodological overview

Time series analysis of intensive longitudinal data provides the psychological literature with a powerful tool for assessing how psychological processes evolve through time. Recent applications in the field of psychosomatic research have provided insights into the dynamical nature of the relationship between somatic symptoms, physiological measures, and emotional states. These promising results highlight the intrinsic value of employing time series analysis, although application comes with some important challenges. This paper aims to present an approachable, non-technical overview of the state of the art on these challenges and the solutions that have been proposed, with emphasis on application towards psychosomatic hypotheses. Specifically, we elaborate on issues related to measurement intervals, the number and nature of the variables used in the analysis, modeling stable and changing processes, concurrent relationships, and extending time series analysis to incorporate the data of multiple individuals. We also briefly discuss some general modeling issues, such as lag-specification, sample size and time series length, and the role of measurement errors. We hope to arm applied researchers with an overview from which to select appropriate techniques from the ever growing variety of time series analysis approaches.

Utilizing time series data embedded in electronic health records to develop continuous mortality risk prediction models using hidden Markov models: A sepsis case study

Continuous mortality risk monitoring is instrumental to manage a patient's care and to efficiently utilize the limited hospital resources. Due to incompleteness and irregularities of electronic health records (EHR), developing continuous mortality risk prediction using EHR data is a challenge. In this study, we propose a framework to continuously monitor mortality risk, and apply it to the real-world EHR data. The proposed method employs hidden Markov models (temporal technique) that take account of both the previous state of patient's health and the current value of clinical signs. Following the Sepsis-3 definition, we selected 3898 encounters of patients with suspected infection to compare the performance of temporal and non-temporal methods (Decision Tree (DT), Logistic Regression (LR), Naive Bayes (NB), Random Forest (RF), and Support Vector Machine (SVM)). The area under receiver operating characteristics (AUROC) curve, sensitivity, specificity and G-mean were used as performance measures. On the selected data, the AUROC of the proposed temporal framework (0.87) is 9-12% greater than the nontemporal methods (DT: 0.78, NB: 0.79, SVM: 0.79, LR: 0.80 and RF: 0.80). The results also show that our model (G-mean=0.78) provides a better balance between sensitivity and specificity compared to clinically acceptable bed-side criteria (G-mean=0.71). The proposed framework leverages the longitudinal data available in EHR and performs better than the non-temporal methods. The proposed method facilitates information related to the time of change of the patient's health that may help practitioners to plan early and develop effective treatment strategies.

Statement on methods in sport injury research from the 1st METHODS MATTER Meeting, Copenhagen, 2019

High quality sports injury research can facilitate sports injury prevention and treatment. There is scope to improve how our field applies best practice methods—methods matter (greatly!). The 1st METHODS MATTER Meeting, held in January 2019 in Copenhagen, Denmark, was the forum for an international group of researchers with expertise in research methods to discuss sports injury methods. We discussed important epidemiological and statistical topics within the field of sports injury research. With this opinion document, we provide the main take-home messages that emerged from the meeting.

Statement on Methods in Sport Injury Research From the First METHODS MATTER Meeting, Copenhagen, 2019

High-quality sports injury research can facilitate sports injury prevention and treatment. There is scope to improve how our field applies best-practice methods-methods matter (greatly!). The first METHODS MATTER meeting, held in January 2019 in Copenhagen, Denmark, was the forum for an international group of researchers with expertise in research methods to discuss sports injury methods. We discussed important epidemiological and statistical topics within the field of sports injury research. With this opinion document, we provide the main take-home messages that emerged from the meeting. Meeting participants agreed that the definition of sport injury depends on the research question and context. It was considered essential to be explicit about the goal of the research effort and to use frameworks to illustrate the assumptions that underpin measurement and the analytical strategy. Complex systems were discussed to illustrate how potential risk factors can interact in a nonlinear way. This approach is often a useful alternative to identifying single risk factors. Investigating changes in exposure status over time is important when analyzing sport injury etiology, and analyzing recurrent injury, subsequent injury, or injury exacerbation remains challenging. The choice of statistical model should consider the research question, injury measure (eg, prevalence, incidence), type and granularity of injury data (categorical or continuous), and study design. Multidisciplinary collaboration will be a cornerstone for future high-quality sport injury research. Working outside professional silos in a diverse, multidisciplinary team benefits the research process, from the formulation of research questions and designs to the statistical analyses and dissemination of study results in implementation contexts. This article has been copublished in the British Journal of Sports Medicine and the Journal of Orthopaedic & Sports Physical Therapy. J Orthop Sports Phys Ther 2020;50(5):226-233. doi:10.2519/jospt.2020.9876.

Absolute and Relative Load and Injury in Elite Junior Australian Football Players Over 1 Season

PURPOSE

To investigate the association between training and match loads and injury in elite junior Australian football players over 1 competitive season.

METHODS

Elite junior Australian football players (n = 290, age 17.7 [0.3] y, range 16-18 y) were recruited from the under-18 state league competition in Victoria to report load and injury information. One-week load (session rating of perceived exertion multiplied by duration) and all time-loss injuries were reported using an online sport-injury surveillance system. Absolute load measures (weekly sums) enabled the calculation of relative measures such as the acute:chronic workload ratio. Load measures were modeled against injury outcome (yes/no) using a generalized estimating equation approach, with a 1-wk lag for injury.

RESULTS

Low (<300 arbitrary units [au]) and high (>4650 au) 1-wk loads were associated with significantly higher risk of injury. Furthermore, low (<100 au) and high (>850 au) session loads were associated with a higher risk of injury. High strain values (>13,000) were associated with up to a 5-fold increase in the odds of injury. There was a relatively flat-line association between the acute:chronic workload ratio and injury.

CONCLUSIONS

This study is the first investigation of elite junior athletes demonstrating linear and nonlinear relationships between absolute and relative load measures and injury. Coaches should focus player loads on, or at least close to, the point at which injury risk starts to increase again (2214 au for 1-wk load and 458 au for session load) and use evidence-based strategies across the week and month to help reduce the risk of injury.

Correction to: An individualised versus a conventional pneumoperitoneum pressure strategy during colorectal laparoscopic surgery: rationale and study protocol for a multicentre randomised clinical study

After publication of our article [1] the authors have notified us that there are changes in the primary outcome and the statistical analysis plan of the study.

Measurement of Dynamical Resilience Indicators Improves the Prediction of Recovery Following Hospitalization in Older Adults

OBJECTIVES

Acute illnesses and subsequent hospital admissions present large health stressors to older adults, after which their recovery is variable. The concept of physical resilience offers opportunities to develop dynamical tools to predict an individual's recovery potential. This study aimed to investigate if dynamical resilience indicators based on repeated physical and mental measurements in acutely hospitalized geriatric patients have added value over single baseline measurements in predicting favorable recovery.

DESIGN

Intensive longitudinal study.

SETTING AND PARTICIPANTS

121 patients (aged 84.3 ± 6.2 years, 60% female) admitted to the geriatric ward for acute illness.

MEASUREMENTS

In addition to preadmission characteristics (frailty, multimorbidity), in-hospital heart rate and physical activity were continuously monitored with a wearable sensor. Momentary well-being (life satisfaction, anxiety, discomfort) was measured by experience sampling 4 times per day. The added value of dynamical indicators of resilience was investigated for predicting recovery at hospital discharge and 3 months later.

RESULTS

31% of participants satisfied the criteria of good recovery at hospital discharge and 50% after 3 months. A combination of a frailty index, multimorbidity, Clinical Frailty Scale, and or gait speed predicted good recovery reasonably well on the short term [area under the receiver operating characteristic curve (AUC) = 0.79], but only moderately after 3 months (AUC = 0.70). On addition of dynamical resilience indicators, the AUC for predicting good 3-month recovery increased to 0.79 (P = .03). Variability in life satisfaction and anxiety during the hospital stay were independent predictors of good 3-month recovery [odds ratio (OR) = 0.24, P = .01, and OR = 0.54, P = .04, respectively].

CONCLUSIONS AND IMPLICATIONS

These results highlight that measurements capturing the dynamic functioning of multiple physiological systems have added value in assessing physical resilience in clinical practice, especially those monitoring mental responses. Improved monitoring and prediction of physical resilience could help target intensive treatment options and subsequent geriatric rehabilitation to patients who will most likely benefit from them.

Match and Training Load Exposure and Time-Loss Incidence in Elite Rugby Union Players

Objective

To investigate the impact of match and training load on time-loss incidence in elite, professional Rugby Union players.

Materials and Methods

Eighty-nine Rugby Union players were monitored over two seasons of training and competition. Load was measured for all training sessions and matches using subjective [session ratings of perceived exertion (sRPE) load; RPE × session duration] and objective [global positioning systems (GPS); distance and high-speed running distance] methods and quantified using multiple approaches; absolute match and training load, acute:chronic workload ratio (ACWR), exponentially weighted moving average (EWMA) and cumulative 7, 14, 21, and 28 d sums. Mixed effect models were used to assess the effect of each variable on time-loss incidence.

Results

Of the 474 time-loss incidences that occurred across the two seasons, 50.0% were contact injuries (86.5% occurred during matches and 13.5% during training), 34.8% were non-contact injuries (31.5% occurred during matches and 68.5% during training) and 15.2% were cases of illness. The absolute match and training load variables provided the best explanation of the variance in time-loss incidence occurrence [sRPE load: p < 0.001, Akaike information criterion (AIC) = 2936; distance: p < 0.001, AIC = 3004; high-speed running distance: p < 0.001, AIC = 3025]. The EWMA approach (EWMA sRPE load: p < 0.001, AIC = 2980; EWMA distance: p < 0.001, AIC = 2980; EWMA high-speed running distance: p = 0.002, AIC = 2987) also explained more of the variance in time-loss incidence occurrence than the ACWR approach (ACWR sRPE load: p = 0.091, AIC = 2993; ACWR distance: p = 0.008, AIC = 2990; ACWR high-speed running distance: p = 0.153, AIC = 2994).

Conclusion

Overall, the absolute sRPE load variable best explained the variance in time-loss incidence, followed by absolute distance and absolute high-speed running distance. Whilst the model fit using the EWMA approach was not as good as the absolute load variables, it was better than when the same variables were calculated using the ACWR method. Overall, these findings suggest that the absolute match and training load variables provide the best predictors of time-loss incidence rates, with sRPE load likely to be the optimal variant of those examined here.

Impact of free-living pattern of sedentary behaviour on intra-day glucose regulation in type 2 diabetes

Purpose

To investigate how the pattern of sedentary behaviour affects intra-day glucose regulation in type 2 diabetes.

Methods

This intensive longitudinal study was conducted in 37 participants with type 2 diabetes (age, 62.8 ± 10.5 years). Glucose and sedentary behaviour/physical activity were assessed with a continuous glucose monitoring (Abbott FreeStyle Libre) and an activity monitor (activPAL3) for 14 days. Multiple regression models with generalised estimating equations (GEEs) approach were used to assess the associations of sedentary time and breaks in sedentary time with pre-breakfast glucose, pre-lunch glucose, pre-dinner glucose, post-breakfast glucose, post-lunch glucose, post-dinner glucose, bedtime glucose, the dawn phenomenon, time in target glucose range (TIR, glucose 3.9–10 mmol/L) and time above target glucose range (TAR, glucose > 10 mmol/L).

Results

Sedentary time was associated with higher pre-breakfast glucose (p = 0.001), pre-dinner glucose (p < 0.001), post-lunch glucose (p = 0.005), post-dinner glucose (p = 0.013) and the dawn phenomenon (p < 0.001). Breaks in sedentary time were associated with lower pre-breakfast glucose (p = 0.023), pre-dinner glucose (p = 0.023), post-breakfast glucose (p < 0.001) and the dawn phenomenon (p = 0.004). The association between sedentary time and less TIR (p = 0.022) and the association between breaks in sedentary time and more TIR (p = 0.001) were also observed.

Conclusions

Reducing sedentary time and promoting breaks in sedentary time could be clinically relevant to improve intra-day glucose regulation in type 2 diabetes.

Electronic supplementary material

The online version of this article (10.1007/s00421-019-04261-z) contains supplementary material, which is available to authorized users.

Does Mathematical Coupling Matter to the Acute to Chronic Workload Ratio? A Case Study From Elite Sport

Purpose: Criticisms of the acute to chronic workload ratio (ACWR) have been that the mathematical coupling inherent in the traditional calculation of the ACWR results in a spurious correlation. The purposes of this commentary are (1) to examine how mathematical coupling causes spurious correlations and (2) to use a case study from actual monitoring data to determine how mathematical coupling affects the ACWR. Methods: Training and competition workload (TL) data were obtained from international-level open-skill (basketball) and closed-skill (weightlifting) athletes before their respective qualifying tournaments for the 2016 Olympic Games. Correlations between acute TL, chronic TL, and the ACWR as coupled/uncoupled variations were examined. These variables were also compared using both rolling averages and exponentially weighted moving averages to account for any potential benefits of one calculation method over another. Results: Although there were some significant differences between coupled and uncoupled chronic TL and ACWR data, the effect sizes of these differences were almost all trivial (g = 0.04–0.21). Correlations ranged from r = .55 to .76, .17 to .53, and .88 to .99 for acute to chronic TL, acute to uncoupled chronic TL, and ACWR to uncoupled ACWR, respectively. Conclusions: There may be low risk of mathematical coupling causing spurious correlations in the TL–injury-risk relationship. Varying levels of correlation seem to exist naturally between acute and chronic TL variables regardless of coupling. The trivial to small effect sizes and large to nearly perfect correlations between coupled and uncoupled AWCRs also imply that mathematical coupling may have little effect on either calculation method, if practitioners choose to apply the ACWR for TL monitoring purposes.

No association between perceived exertion and session duration with hamstring injury occurrence in professional football

Training and competition loads have emerged as modifiable composite risk factors of non-contact injury. Hamstring strains are the most common injuries in football with substantial burden on the individual player and club. Nevertheless, robust evidence of a consistent load-hamstring injury relationship in professional football is lacking. Using available data from the Qatar Stars League over three competitive seasons, this study investigated the separate and combined effects of perceived exertion and session duration on hamstring injury occurrence in a sample of 30 outfield football players. Load variables were calculated into 7-day, 14-day, 21-day, 28-day periods of data, and week-to-week changes for average ratings of perceived exertion (RPE; au) score and session-RPE (s-RPE; session-duration  ×  score), plus the cumulative training and match minutes and s-RPE, respectively. Conditional logistic regression models estimated load-injury relationships per 2-within-subject standard deviation increments in each candidate variable. Associations were declared practically important based on the location of the confidence interval in relation to thresholds of 0.90 and 1.11 defining small beneficial and harmful effects, respectively. The uncertainty for the corrected odds ratios show that typically high within-subject increments in each candidate variable were not practically important for training- and match-related hamstring injury (95% confidence intervals range: 0.85 to 1.16). We found limited exploratory evidence regarding the value of perceived exertion and session duration as etiological factors of hamstring injury in Middle-East professional football. Monitoring remains valuable to inform player load management strategies, but our exploratory findings suggest its role for type-specific injury risk determination appears empirically unsupported.

Robust Exponential Decreasing Index (REDI): adaptive and robust method for computing cumulated workload

Objective

The purpose of this study was to define a new index the Robust Exponential Decreasing Index (REDI), which is capable of an improved analysis of the cumulative workload. This allows for precise control of the decreasing influence of load over time. Additionally, REDI is robust to missing data that are frequently present in sport.

Methods

200 cumulative workloads were simulated in two ways (Gaussian and uniform distributions) to test the robustness and flexibility of the REDI, as compared with classical methods (acute:chronic workload ratio and exponentially weighted moving average). Theoretical properties have been highlighted especially around the decreasing parameter.

Results

The REDI allows practitioners to consistently monitor load with missing data as it remains consistent even when a significant portion of the dataset is absent. Adjusting the decreasing parameter allows practitioners to choose the weight given to each daily workload.

Discussion

Computation of cumulative workload is not easy due to many factors (weekends, international training sessions, national selections and injuries). Several practical and theoretical drawbacks of the existing indices are discussed in the paper, especially in the context of missing data; the REDI aims to settle some of them. The decreasing parameter may be modified according to the studied sport. Further research should focus on methodology around setting this parameter.

Conclusion

The robust and adaptable nature of the REDI is a credible alternative for computing a cumulative workload with decreasing weight over time.

Time-to-event analysis for sports injury research part 1: time-varying exposures

BACKGROUND

'How much change in training load is too much before injury is sustained, among different athletes?' is a key question in sports medicine and sports science. To address this question the investigator/practitioner must analyse exposure variables that change over time, such as change in training load. Very few studies have included time-varying exposures (eg, training load) and time-varying effect-measure modifiers (eg, previous injury, biomechanics, sleep/stress) when studying sports injury aetiology.

AIM

To discuss advanced statistical methods suitable for the complex analysis of time-varying exposures such as changes in training load and injury-related outcomes.

CONTENT

Time-varying exposures and time-varying effect-measure modifiers can be used in time-to-event models to investigate sport injury aetiology. We address four key-questions (i) Does time-to-event modelling allow change in training load to be included as a time-varying exposure for sport injury development? (ii) Why is time-to-event analysis superior to other analytical concepts when analysing training-load related data that changes status over time? (iii) How can researchers include change in training load in a time-to-event analysis? and, (iv) Are researchers able to include other time-varying variables into time-to-event analyses? We emphasise that cleaning datasets, setting up the data, performing analyses with time-varying variables and interpreting the results is time-consuming, and requires dedication. It may need you to ask for assistance from methodological peers as the analytical approaches presented this paper require specialist knowledge and well-honed statistical skills.

CONCLUSION

To increase knowledge about the association between changes in training load and injury, we encourage sports injury researchers to collaborate with statisticians and/or methodological epidemiologists to carefully consider applying time-to-event models to prospective sports injury data. This will ensure appropriate interpretation of time-to-event data.