Can home care work be organized to promote musculoskeletal health for workers? Results from the GoldiCare cluster randomized controlled trial

Abstract

Background: Workers in home care have high sick leave rates, predominantly because of musculoskeletal pain. The Goldilocks Work Principle proposes that health should be promoted by a "just right" composition of work tasks. Weekly workloads differ substantially between home care workers, suggesting that certain workers may have workloads that are too high, impacting their musculoskeletal health. The aim of this study was to evaluate the effectiveness of a "GoldiCare" intervention redistributing weekly workloads to become more equal among the homecare workers. Outcomes were pain in the neck/shoulder and lower back, and the implementation of the intervention was also evaluated.

Methods: A 16-week cluster randomized controlled trial was conducted with 125 workers from 11 home care units, divided into six intervention units and five control units. The operation coordinators of each intervention unit were educated in the Goldilocks Work Principle and provided with a planning tool to facilitate an even distribution of high care need clients. The control group continued their usual work. Primary outcomes were pain intensity in the neck/shoulder and lower back (0 to 10). Secondary outcomes included fatigue (0 to 10), composition of physical behaviors and postures (accelerometers), adherence to the intervention (weekly usage rates of the planning tool), and performance of the intervention (percentage of workers with an even distribution of workload).

Results: The analysis showed no difference between the intervention and control groups in change in lower back pain (0.07, 95%CI[-0.29;0.43]), neck/shoulder pain (-0.06, 95%CI[-0.49;0.36]) or fatigue (0.04, 95%CI[-0.52;0.61]. No significant changes were observed in the composition of physical behaviors (p = 0.067) or postures (p = 0.080-0.131) between the two groups. The intervention was succesfully implemented in three units of the six, with adherence ranging from 82-100% across the intervention period. The remaining three units had an adherence of 0-47%. No improvement in performance was observed.

Conclusion: No significant intervention effects were observed on musculoskeletal pain, fatigue, or the composition of physical behaviors and postures. The findings suggest that the intervention was not adequately implemented within the organization. Consequently, we cannot discern whether the lack of positive results were due to poor implementation or an ineffective intervention. Results thus highlight the need for a more comprehensive understanding of organizational structures within home care to facilitate more effective implementations. The hypothetical effectiveness of a fully implemented intervention remains unknown.

Trial registration: Clinicaltrials.gov ID: NCT05487027, submitted: 03/08/2022.

Keywords: Clinical trial; Cluster randomized controlled trial; Compositional analysis; Goldilocks work; Home care; Norway; Worker health.

Fig. 1

Flowchart of participants in accordance with recommendation by the “CONSORT 2010 statement: extension to cluster randomized trials” [54]

Fig. 2

Compositional biplots illustrating the composition of physical behaviors of the intervention (blue) and control group (red) at baseline (fully drawn curves) and after the intervention (dashed curves). The center of the graph (x = 0, y = 0) represents the average composition of the groups. Individual points represent each worker’s weekly mean composition (sedentary, standing, and active). Ellipses illustrate 95% confidence regions for the variance within each group and timepoint (the tighter the ellipse the lower the variability). Arrowheads indicate the most prevalent posture in the respective direction

Fig. 3

Compositional biplots illustrating the composition of arm elevation (left panel) and trunk forward bending (right panel) of the intervention (blue) and control group (red) at baseline (fully drawn curves) and after the intervention (dashed curves). The graph centers (x = 0, y = 0) represent the average composition of the groups. Individual points represent each worker’s weekly mean composition of arm elevation (sedentary, upright ≤ 60°, upright > 60°) and trunk forward bending (sedentary, upright ≤ 30°, upright > 30°). Ellipses represent 95% confidence regions, illustrating the variability within each group and timepoint (the tighter the ellipse the lower the variability). Arrowheads indicate the most prevalent posture in the respective direction

Fig. 4

Adherence and performance of the intervention. Percentage of workers who received a balanced week (i.e., even distribution of light, heavy and medium shifts throughout the week). Breaks in lines are due to not using the tool for two weeks consecutively. Different endpoints of lines are due to different follow-up measure timepoints, as the units used the tool until the participants belonging to the respective units finished measurements. One unit excluded due to dropping out