Prediction of work-related musculoskeletal discomfort in the meat processing industry using statistical models

Abstract

Musculoskeletal disorders are one of the most common occupational disorders in the manufacturing industry, and cause pain, suffering, disability and a decrease in productivity. The objective of this study was the development of statistical models for the prediction of work-related musculoskeletal discomfort. A sample of 174 workers of the meat processing industry was taken. Diverse ergonomic evaluation methods were applied on data collected by means of direct observation and surveys. Later, pattern recognition techniques were used to identify the relevant predictor variables from an initial set of 20 variables. A prevalence of musculoskeletal discomfort of 77% was found. The most suitable classification models to predict the discomfort were the models based on logistic regression and decision trees. Statistical models were obtained to predict discomfort in shoulders, back, hands/wrists and neck with a precision between 83.3% and 90.2%. The findings can be useful to guide improvement initiatives according to the specific characteristics of the job and the profile of the worker.