Handling and reporting missing data in training load and injury risk research

ABSTRACT

Purpose

To map the current practice of handling missing data in the field of training load and injury risk and to determine how missing data in training load should be handled.

Methods

A systematic review of the training load and injury risk literature was performed to determine how missing data are reported and handled. We ran simulations to compare the accuracy of modelling a predetermined relationship between training load and injury risk following handling missing data with different methods. The simulations were based on a Norwegian Premier League men’s football dataset (n = 39). Internal training load was measured with the session Rating of Perceived Exertion (sRPE). External training load was the total distance covered measured by a global positioning systems (GPS) device.

Results

Only 37 (34%) of 108 studies reported whether training load had any missing observations. Multiple Imputation using Predicted Mean Matching was the best method of handling missing data across multiple scenarios.

Conclusion

Studies of training load and injury risk should report the extent of missing data, and how they are handled. Multiple Imputation with Predicted Mean Matching should be used when imputing sRPE and GPS variables.

KEYWORDS:

• Training load
• injury
• missing data
• simulation
• systematic review

Acknowledgements

We thank Torstein Dalen-Lorentsen for providing access to the male Norwegian Premier League football data. We also thank Garth Theron for high quality programming of the Norwegian Premier League football database. This research had been impossible without the collaboration of coaches and players, and we thank the participants for their contribution.

Supplemental data

Supplemental data for this article can be accessed here.

Data availability

All data relevant to the study are included in the article or are available as supplementary files. The Norwegian Premier League football data were anonymised based on requirements of the Norwegian Data Protection Agency (Datatilsynet 2017), which required the removal of the variable describing the player’s position on the football team. Any analysis which included this variable is therefore not reproducible. The review data on training load and injury risk studies are available as a machine-readable xlsx file, along with the anonymised Norwegian Premier League football data, and all statistical programming code, in a GitHub repository (Bache-Mathiesen 2021a).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Ethics Approval

The Norwegian Premier League football study was approved by the Ethical Review Board of the Norwegian School of Sport Sciences, and by the Norwegian Centre for Research Data (722773).

Twitter

@lena_kbm, @DocThorAndersen, @benclarsen, @FagerlandWang

Box 1. A summary of the simulation steps.

Additional information

Funding

The Oslo Sports Trauma Research Centre provided all funding for performing this study.