Background:

A longstanding hypothesis in physical therapy and occupational research is that workers repeating a task very stereotypically will be more prone to develop musculoskeletal disorders than workers that manage to vary postures and loads. Increased movement variability (MV), presumably, modify tissue loads, distribute stresses more equally, and thus reduce the cumulative load on any particular tissue. A handful of studies of MV have indicated less overuse injuries and faster recovery from musculoskeletal pain disorders. Even when repeating strictly controlled tasks individuals may differ in motor consistency, some showing higher levels of MV than others. However, whether the extent of MV is indeed a consistent individual trait across different tasks and different days is not known.

Purpose:

To investigate whether individual profiles of MV is stabile between days, the consistency of MV patterns from kinematic recordings, repeated across three days, was studied when performing repetitive upper-extremity precision work.

Methods:

A laboratory-based simulation of precision work; a 'pipetting' task paradigm, was developed in which liquid was repeatedly transferred from one tube to another, with a cycle time of 2.8s. Fourteen healthy female subjects, aged 20-45 years, right-handed and with experience in pipetting participated on 3 different days under identical conditions. Kinematic data were obtained using an electromagnetic motion capture system (FASTRAK). MV in shoulder elevation, elbow flexion and shoulder-elbow coordination were operationalized using cycle-to-cycle standard deviations across 20 pipetting cycles of kinematics parameters including joint range of motion, average and peak velocities, time to peak velocities, average angle and phase. Multivariate analysis was conducted using principal component analysis (PCA) (SIMCA+P, 12.0) to analyze relationships among variables and individual patterns in the data matrix of the recordings from day1. Thereafter, in order to confirm the observed structure of inter-individual MV patterns, classification of the data from day2 and day3 was performed using the parameters of the model from day1.

Results:

Four PCA components (Eigenvalues>1) accounted for 80 percent of the total variance in the model for day1. In the subsequent prediction model where data from day2 and 3 were projected into the model of day1, all subject observations except one could be predicted with 95% confidence (Hotelling T2). And individual data scores from all three days were clustered in relative proximity to each other, indicating consistency in MV between days.

Conclusion(s):

The findings indicate, even in this small and homogenous sample of young healthy females, that there may indeed be consistent individual traits in motor variability. A next step would be to answer whether these traits remain consistent if work factors such as work pace or precision are altered, and whether individual profiles of MV are associated with physiological responses related to risk for developing musculoskeletal disorders.

Implications:

Consistency of individual MV patterns substantiate previous notions that some people appear prone to repeat themselves while others tend to vary their motor behavior when performing the same task. Assessment of MV by physical therapists in research and practice could be valuable to further explore and address the relation of MV and musculoskeletal health.