Learning Curve Analysis of a Patient Lift Assist Device

Abstract

The prevalence of musculoskeletal disorders (MSDs) among health care workers is extremely high. In response to high injury rates associated with the patient transfer tasks, one of the most physically demanding tasks in the health care industry, ergonomic interventions have been developed. These ergonomic interventions were created to lessen the risk of MSDs development by reducing the load required of the worker, however workers often complain of the added task time that using the interventions introduce. It has been shown that an increased number of replications of a task will yield an increased proficiency in the task, and that this increased productivity follows a response called a learning curve. The use of learning curve theory has not been applied to the introduction of ergonomic interventions in the workplace. The objectives of the current study were to explore the possible application of learning curve theory in the process of introducing ergonomic interventions, and to evaluate the impact of training technique (passive training vs. active training) on variables describing the learning process.

Eighteen subjects completed multiple replications of a patient transfer task after being trained in either an active or passive training method. The patient transfer task was completed with an ergonomic intervention, a mechanical patient lift assist device ('Opera' model manufactured by Arjo). The active training method took a hands-on approach to the training, while the passive method was similar to the see-one-do-one method currently prevalent in the health care industry. The task time-to-complete values were recorded for each repetition, and the dependent variable values were calculated for each subject. The dependent variable Learning Rate, a standard measure of the decrease in task performance as a function of task repetition, was 85.2% for the hands-on training and was 81.1% for the see-one-do-one training (this difference was not found to be statistically significant). However, the dependent variable Trial 0, the time-to-complete values for the first time the subject completed the patient handling task, was found to be significant (p<0.05). The average time-to-complete value for the hands-on method was 370 seconds while the average for the see-one-do-one method was 475 seconds. The dependent variable 'Delta 3' was also found to be significant. This measure quantifies the decrease in time-to-complete values for the across the first four trials. The Delta 3 values for the hands-on method was 82 seconds while the see-one-do-one method was 170 seconds, indicating a rapid 'merging' of performance in the two training methods.

The study successfully applied learning curve theory to the introduction of an ergonomic intervention, and illustrated the dominance of an active training method when training workers to use the intervention (particularly in the early stages of learning to use the device). The results emphasize the importance in considering learning curves when workers are learning to use new equipment, especially ergonomic interventions that can prevent MSDs from occurring. The results also show that the method of training given to workers can impact the initial performance of a task.