Next Article in Journal
Analysis of the Development Status of eLoran Time Service System in China
Previous Article in Journal
Analysis of the Effect of Ultra-Fine Cement on the Microscopic Pore Structure of Cement Soil in a Peat Soil Environment
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 13
Issue 23
10.3390/app132312699
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Visualization of Caregiving Posture and Risk Evaluation of Discomfort and Injury

by
Xin Han
1,
Norihiro Nishida
2,
Minoru Morita
1,
Mao Mitsuda
2 and
Zhongwei Jiang
1,*
1
Faculty of Engineering, Yamaguchi University Graduate School of Sciences and Technology for Innovation, 2-16-1 Tokiwadai, Ube City, Yamaguchi Prefecture 755-0097, Japan
2
Department of Orthopedic Surgery, Yamaguchi University Graduate School of Medicine, 1-1-1 Minamikogushi, Ube City, Yamaguchi Prefecture 755-8505, Japan
*
Author to whom correspondence should be addressed.
Appl. Sci. 2023, 13(23), 12699; https://doi.org/10.3390/app132312699
Submission received: 25 October 2023 / Revised: 18 November 2023 / Accepted: 23 November 2023 / Published: 27 November 2023
(This article belongs to the Special Issue New Trends in the Biomechanical Assessment of Posture and Gait)
Browse Figures
Versions Notes

Abstract

:
There is a high risk of musculoskeletal discomfort and injury due to the lack of professional guidance and training in caregiving postures. This study aimed to develop a risk assessment and visualization method by analyzing caregiving postures. Participants with (n = 8) and without (n = 10) caregiving experience were recruited to simulate patient transfer from bed to wheelchair. The Rapid Entire Body Assessment (REBA) method lacked sensitivity in distinguishing the experienced and inexperienced groups. We found that the visualization of the center of gravity (COG) trajectory could represent distinct posture differences between the two groups. Based on this finding, we considered a modified REBA method combining the COG trajectory, load-bearing time, and asymmetric load parameters, named the Caregiving-REBA (C-REBA) method. Our results demonstrated that C-REBA could effectively distinguish experienced and inexperienced caregivers, especially in caregiving task Stages 2–4. In conclusion, the present work explored adjusting to the parameters of the REBA method. The proposed C-REBA method could be easily imbedded into the Internet of Things (IoT) device to assess the caregiving posture for providing visual guidance and warning of the risk of discomfort or injury.

1. Introduction

Caregiving work poses great challenges to some developed countries as the burden of the aging population increases. It involves providing daily life support and medical care services to individuals who require specialized care or supervision. These individuals may be unable to care for themselves due to age, illness, physical disabilities, or other reasons [1]. However, the capacity of the caregiving industry is limited, and a number of patients can only access care offered by their family members at home. Patient transfer and lifting tasks are the most common tasks in caregiving work and the most likely causes of musculoskeletal discomfort and injury. For inexperienced home caregivers, learning caregiving posture through video or verbal guidance is far from enough, and the learning effects are limited [2,3]. Incorrect caregiving posture carries a critical risk of musculoskeletal discomfort and injury, such as lower back pain and lumbar disc herniation [4,5]. Home caregivers are about three times more likely to suffer from lumbar spine discomfort than hospital caregivers [6]. However, there is limited research on home caregiving posture analysis and guidance currently. It is hard to obtain posture-guiding parameters from experienced caregivers as well. Therefore, inexperienced caregivers urgently need a targeted posture ergonomics assessment method for guidance, to help them avoid the risks of discomfort and injury in the caregiving process.

1.1. Literature Review

Comprehensive ergonomics assessment methods have a great effect in avoiding musculoskeletal discomfort and injury [7]. Compared with other ergonomics assessment methods [8,9], the Rapid Entire Body Assessment (REBA) [10] is a postural load assessment method that covers the entire body, and it is based on the workers’ posture, joint angles, and muscle loads, mapping these factors to a specific scoring table to determine the risk level of working postures and provide improvement suggestions for reducing potential health problems. Also, REBA is generally applicable for posture risk assessment in a wide range of occupations, such as forestry timber harvesting [11,12], mining industries [13], quarry industries [14], and especially in nursing [15]. REBA is more suitable for the assessment of hospital nursing activities [16]. Utilizing the REBA method to assess the postures of nurses in the operating room contributes to the mitigation of musculoskeletal disorders [17,18]. REBA also showed good guidance with respect to musculoskeletal risks in a sample of emergency medical technicians [19]. Some researchers believed that REBA has a good reference significance in the assessment of inappropriate posture by dentists [20,21]. The REBA method was also reported to evaluate caregiving tasks with assistive devices, where the risk of musculoskeletal disorders could be assessed [22]. Furthermore, the reliability of REBA has been confirmed compared with other traditional methods [23]. In our research, REBA was selected because it considers all body parts as well as all well-known awkward postures, loads, and types of activity. However, some studies have pointed out that there is an overassessment of risk by REBA in certain scenarios [24]. To make REBA more suitable in our caregiving scenario, we explored the differences from several objective standpoints, and we adjusted REBA’s rules based on these differences.
It is common knowledge in daily life that if you support the center of gravity (COG) you can support the gravity of each part. The COG is crucial for maintaining balance and stability during physical activities like walking, running, and sports [25,26]. It also plays a vital role in optimizing design for safety and efficiency in fields such as ergonomics, industrial design, and rehabilitation [27]. During lifting tasks, the body needs to complete a transfer from a lower position to a higher position. To maintain the body’s balance, the COG of the standing posture will be systematically lowered [28]. Simultaneously, maintaining an upright posture in the upper body allows for minimal muscle activation and enhances postural stability [29]. This helps alleviate pressure on the waist and reduces the risk of musculoskeletal disorders.
Experienced caregivers, through extensive training, have acquired familiarity with optimal postures and positions. This expertise allows them to adopt power positions that help prevent musculoskeletal injuries. Power position refers to the optimal physical position and body pose to maintain balance and maximize strength output during lifting activities, thus preventing muscular injury [30]. Experienced caregivers often change the COG’s height to achieve power positions with the minimal physical strain. There is a certain correlation between the height of the trunk bending and the posture instability, and bearing loads in a tilted manner would cause further body instability [31]. The power position is very important for caregiving tasks. Most of the previous studies have only applied the REBA rules in the assessment of caregiving posture, ignoring the relationship between COG changes and power positions. Therefore, incorporating COG and power positions into the REBA score holds the potential to make it more applicable to the caregiving process.

1.2. Contributions

The aim of this study is to develop a risk assessment for discomfort and injury associated with caregiving postures, and to provide a visualization method that could provide easy understanding and guidance. Since the original REBA method could not distinguish the caregiving postures of the experienced and inexperienced groups, we attempted to find the most relevant factors affecting caregiving posture, and to modify the REBA rule by adding these parameters to enhance the applicability of the REBA method. The following are the main contributions of this study:
(1) A modified REBA (C-REBA) method for caregiving is proposed by adding new parameters of COG trajectory, asymmetric load, and load-bearing time, which are more suitable for the risk assessment of caregiving tasks.
(2) Our results illustrate that the COG is a vital factor in the evaluation of the caregiving process, and the visualization of COG trajectories directly indicates the caregiving power position.
(3) The developed analysis method could be extended to other scenarios if more relevant data are collected and the values of the adjusting parameters are determined.

2. Materials and Methods

2.1. Study Design

The daily work in the caregiving industry is individualized and involves many complex working postures. Patient transfer, one of the most common postures in the caregiving, was chosen for our research. Subjects were recruited by convenience sampling, and the subjects were selected based on the following criteria:
Eight professional nurses from the Department of Rehabilitation Therapy at the co-authors’ hospital were invited and served as the experienced group (Exp group). The inclusion criteria for the Exp group were as follows: (1) aged between 25 and 35 years, (2) at least 5 years of professional caregiving experience, and (3) no history of back injury or pain in the past year [32].
Ten inexperienced volunteers were recruited. The inclusion criteria for the inexperienced group (Inexp group) were as follows: (1) aged between 20 and 35 years, (2) without any caregiving experience, and (3) no history of back injury or pain in the past year.
One nurse at the co-authors’ hospital acted as the patient (age: 30 years, height: 168 cm, weight: 62 kg). Another 8 nurses (Exp group) and 10 volunteers (Inexp group) were requested to transfer the patient from bed to a wheelchair. The sitting posture and movements of the patient were kept consistent. Their age, height, and weight are shown in Table 1.
The Intel RealSense Depth Camera D435 was used to capture the caregiving posture, and the kinematic analysis system based on the OPENPOSE algorithm was used to obtain skeletal joint data. OPENPOSE-based ergonomic assessments were robust to non-ideal task conditions [33]. The COG trajectory calculation method employed in this study was adapted from the works of González et al. and Cotton et al. [34,35].
A previous study proved that the Wii Balance Board is a valid tool for assessing standing balance [36]. Two Wii Balance Boards (Nintendo Co., Ltd., Kamitoba Hokodatecho, Minami-ku, Kyoto, Japan) were used to measure the ground reaction forces of each foot and then calculate the trajectories of the center of pressure (COP). Each participant (Exp group, n = 8; Inexp group, n = 10) was asked to stand on the Wii Balance Board and adjust their suitable caregiving postures. The distance between the feet of each participant was measured before starting. To analyze and refine the entire patient transfer process, the patient transfer caregiving process was divided into the following 5 stages (Figure 1):
Stage 1: The caregiver placed their hands in a hugging position around the patient’s waist, adjusted their posture, and prepared to start.
Stage 2: The caregiver began to lift the patient and prepared to take them from the support of the bed.
Stage 3: The caregiver lifted the patient to the proper point and prepared to rotate the patient to the side of the wheelchair.
Stage 4: The caregiver rotated the patient to the side of wheelchair and prepared to put them down.
Stage 5: The caregiver placed the patient on the wheelchair.

2.2. REBA

REBA is designed for quickly assessing the posture load to identify which working postures need further attention and improvement, in an attempt to reduce the risk of work-related physical discomfort and injuries. Assessing the angle changes of major joints is a part of REBA. The body is divided into two independent assessment parts. The neck, trunk, and legs are assessed in Part A, and the upper arms, lower arms, and wrists are assessed in Part B (Figure 2a). Individual scores for the trunk, legs, and neck are checked as Table A, adding to the Part A score. Score A is the sum of the Part A and Extra A scores. The Extra A score is determined by the strain and load associated with the work. Similarly, the scoring rules for Part B, Extra B, and Score B are calculated in the same way; the Extra B score is determined based on the coupling ability of the hands. Finally, Scores A and B are integrated into Table C to generate Score C, and final REBA score is the sum of Score C and Extra C (Figure 2b), where the Extra C score is determined by the activity’s difficulty. The REBA score ranges from 1 to 12. These scores are correlated with the risk of musculoskeletal disorders, and the higher the REBA score, the greater musculoskeletal discomfort and injury. Detailed scoring rules can be found in the literature [10].

2.3. Caregiving-REBA (C-REBA)

The REBA method showed limitations in assessing the risk of caregiving postures, as it failed to differentiate experienced and inexperienced caregivers, and it could not provide meaningful postural guidance for inexperienced caregivers. We firstly found that the original rules for the Extra A and Extra C scores are limited in their applicability to caregiving work. The initial rule for assessing Extra A primarily focuses on load considerations. In the realm of caregiving tasks, where the load consistently exceeds 10 kg, this corresponded to an additional 2 points within the original Extra A scoring rule. Consequently, this inconsistency invalidated the applicability of the scoring rule to caregiving tasks. Asymmetric load was a notable contributor to muscle injuries [29], and this evaluative parameter is frequently disregarded during caregiving procedures. Thus, we adjusted Extra A by emphasizing the aspect of asymmetric loads.
The initial iteration of Extra C predominantly emphasized the rapid and large range changes in postures, the duration of posture maintenance, and the action frequency as evaluative metrics. These aspects are more tailored towards evaluating postures among factory laborers. Conversely, for caregiving work, factors such as the load-bearing time, the COG trajectory of the power position, and the pose stability related to COG changes seem to be more pertinent as assessment criteria. Consequently, we adjusted the Extra C rules accordingly. The specific parameter adjustments for adjusting Extra A and Extra C were determined based on statistical variations observed between the inexperienced and experienced groups. Elaborate details regarding these improvement methodologies are expounded upon in Section 3.3. Our C-REBA method optimized the Extra A and Extra C parameters of the original REBA, amalgamating critical factors that are pertinent to caregiving tasks, notably including asymmetric loads, load-bearing time, and COG changes. These adjustments rendered C-REBA more precise and applicable in the assessment of caregiving.

2.4. Data Collection and Statistical Analysis

The experiments were conducted and the data were collected in 2022, with each subject recording three sets of data. The average values were calculated to obtain the final results. Following the completion of the caregiving task, the kinematic analysis system automatically generated posture skeleton joint information in pixel coordinates, which was synchronized with the PC terminal. Additionally, COP and COG trajectory information was also synchronized. To obtain the REBA assessment result, we developed a system that utilized a neural network model to evaluate joint angles from the experimental videos.
SPSS 16.0 software (SPSS Inc.) and GraphPad Prism 9 (GraphPad Inc.) were used for statistical analysis. The Shapiro–Wilk test was used to check whether the mean differences of all variables were normally distributed, since the sample size was less than 20. Student’s t-test was used for continuous data from two groups that met normal distribution, and the 1-way ANOVA method was used for continuous data from two groups that did not meet normal distribution. For some statistical tests, the mean value and standard deviation were reported, and p-values less than 0.05 were considered to be statistically significant.

3. Results

3.1. REBA

We employed the REBA method to assess the caregiving postures in both the Exp and Inexp groups. The results (Figure 3) revealed that, in the Inexp group, the average REBA scores were in the high-risk range (above 8 points) across all stages. Similarly, the average REBA scores for the Exp group in Stages 1 to 4 were also in the high-risk range. These findings suggest an overestimation of risk levels in caregiving postures by the REBA method. Furthermore, there were no significant differences between the Exp and Inexp groups in the assessment of caregiving postures at each individual stage (Stage 1: p = 0.319; Stage 2: p = 0.343; Stage 3: p = 0.183; Stage 4: p = 0.0596; Stage 5: p = 0.113). This indicated that the REBA method could not distinguish the Exp and Inexp groups in caregiving postures, and that it lacks sensitivity and specificity in assessing postural loads. Therefore, the REBA method is not suitable for providing guidance and reference for inexperienced caregivers, as it does not adequately assess the ergonomic demands of caregiving postures.

3.2. The Difference in COG

We used a camera to capture the entire process of patient transfer and implemented COG trajectory visualization (Figure 4a). The Exp group showed a trajectory of the COG first dropping, then rising, and then dropping again during patient transfer (Figure 4b). They tended to lower their COG to keep their upper body upright for reducing the torque in the waist, and to keep their trunk as straight as possible. Conversely, the COG trajectory of the Inexp group showed an initial rise before directly dropping (Figure 4c). Inexp group caregivers, unfamiliar with power positions, tended to bend their trunk as much as possible to lift the patient.
The height changes of the COG were visualized, where notable differences between the Exp and Inexp groups were found during patient transfer (Figure 5a). The Exp group showed a downward trend in COG changes during Stage 2 of patient transfer, but the Inexp group showed an opposite upward trend. During Stage 2, the Exp group used a backward body movement to lower the COG and maintained an upright trunk to increase their body stability. The Inexp group lifted the patient by bending over directly, which increased the burden on their waist. The differences in Stages 2 (p < 0.001), 3 (p = 0.002), and 4 (p = 0.003) were statistically significant, suggesting that Stages 2, 3, and 4 might be meaningful intervention stages for caregiving posture (Figure 5b). The visualization results provided a reference for the Inexp group, which reminded them to adjust their caregiving posture and reduce the risk of musculoskeletal discomfort and injury.
In addition, the COG heights from Stage 1 to the highest lifting position (Stage 3 or Stage 4) were also compared between the Exp and Inexp groups. The Exp group utilized power positions to control the COG height in Stage 1, minimizing the fluctuation and reducing the work performed by gravity and the waist burden. The Inexp group showed an overall tendency of higher COG heights from Stage 1 to Stage 3 and Stage 4, indicating that the Inexp group held the patient at a higher height during the caregiving task. This also increased the waist burden and the risk of injury.

3.3. Caregiving-REBA (C-REBA)

3.3.1. C-Extra A

The asymmetric load in caregiving work is an important factor of musculoskeletal discomfort and injury. We redefined the scoring rules of Extra A by adding the factor of the asymmetric load according to the changing trend of the COG. With two Wii Balance Board sensors, we measured the ground reaction forces on the left foot (COP1) and right foot (COP2) for all of the participants during the transfer operation. Figure 6a shows the results of statistical analysis of the load difference between the left and right feet. It was evident that the inter-foot load difference quartile in the Inexp group was larger than that in the Exp group. Experienced caregivers had smaller fluctuations in the load difference, while inexperienced caregivers had large fluctuations. We attempted to use the quartile value of the Exp group as an evaluation criterion to assess the difference in asymmetric loading of the Inexp groups. The inter-foot load quartile value of the Exp group showed that the values of the first quartile and the third quartile were 9.99 kg and 24.54 kg, respectively (Figure 6a). We added new scores in the C-Extra A rule by defining the load difference range as provided in Figure 6b.
To achieve the automatic calculation of C-REBA from the image level, we should convert the load difference range (Figure 6b) into the distance range between the left and right feet. As shown in Figure 7a, COP1 and COP2 were measured by the two Wii Balance Boards, and the fused COP was calculated from COP1 and COP2. We assumed that the ground reaction forces were F1 and F2 corresponding to COP1 and COP2, respectively, Fc corresponded to the fused COP, the load difference force between the left and right feet was ∆F, and the distance between the feet was L. Then, the location of Fc on the x-axis could be calculated as follows:
Fc = F1 + F2,
∆F = F2 − F1 = (2Xc − (X1 + X2))/(X2 − X1) Fc,
Xc/L = (∆F (X2 − X1))/(2Fc L) + ((X1 + X2))/2L,
After calculating Xc along the range of ∆F by Equation (3) and taking its average, the load difference score range could be converted into the distance score range of COP on the x-axis, which was given by ∆F < 9.99 kg corresponding to 0.38 < Xc/L < 0.75, 9.99 < ∆F < 24.54 kg corresponding to 0.19 < Xc/L < 0.38 or 0.75 < Xc/L < 0.86, and ∆F > 24.54 kg corresponding to Xc/L < 0.19 or Xc/L > 0.86. Figure 7b shows the load difference values (∆F < 9.99 kg, green; 9.99–24.54 kg, blue; >24.54 kg, red) on each point corresponding to the COP. Considering the quasi-static state, the COG trajectory calculated by the OPENPOSE algorithm was approximated to the COP trajectory obtained by the Wii Balance Board [34], and the center coordinate of the COG on the x-axis (Figure 7c) could be assumed to be the same as the Xc of the COP (Figure 7a). In doing so, instead of using the Wii Balance Board, Xc could be calculated from the camera image. Therefore, the scores in Figure 7c could be automatically determined by calculating the COG trajectory in the x-coordinate.

3.3.2. C-Extra C

We collected data on the loading time in caregiving work and found a significant difference (p < 0.001) in load-bearing time between the two groups. The median load-bearing time of the Inexp group was as much as 1.65 times higher than that of the Exp group (Figure 8a). Therefore, for the Extra C score in C-REBA, we set the evaluation standard for loading time at 3.87 s, which was 1.5 times the median loading time of the Exp group. Similarly, based on the fact that the less the COG height changes, the less work the waist does and the easier it is to maintain the power position, we determined the differences in the COG changes from the initial stage to the highest point of the COG trajectory. We found that the median COG height change of the Inexp group was significantly larger than that of the Exp group (Figure 8b, p < 0.001). Furthermore, we set the evaluation standard for Extra C score in C-REBA at 7.54 cm, which was two times the 75% quantile COG height of the Exp group. Therefore, the adapted C-Extra C score was redefined (Table 2).

3.4. Comparison of REBA and C-REBA

When the Exp group and Inexp group caregivers were evaluated by our C-REBA method, there were significant differences between the Exp and Inexp groups at each individual stage (Stage 1: p < 0.01; Stage 2: p < 0.001; Stage 3: p < 0.001; Stage 4: p < 0.001; Stage 5: p < 0.05). This indicates that the C-REBA method could provide good differentiation in each stage of the caregiving task (Figure 9), as well as being suitable for providing guidance and reference for inexperienced caregivers. More specifically, the mean C-REBA scores for the Exp group in Stages 1–5 were all below eight points, at a medium risk level. For the Inexp group, on the other hand, the mean C-REBA scores were above eight points except for Stages 1 and 5. This indicated that the caregiving posture in Stages 2, 3, and 4 needed to be corrected to reduce musculoskeletal discomfort and injury risks. It also suggested that Stages 2, 3, and 4 might be the most meaningful caregiving posture intervention stages.
To verify the improvement of C-Extra A and C-Extra C in C-REBA, we set up ablation experiments to verify the REBA scoring rules, REBA with C-Extra A scoring rules, and C-REBA (REBA with C-Extra A and C-Extra C) scoring rules. The scoring results of REBA and REBA with C-Extra A were significantly different (Table 3; all groups and all stages p1 < 0.05). The evaluation rules for C-Extra C focused on the caregiving power position and the load-bearing time. For the Exp group, they were familiar with the caregiving power position, and their load-bearing time was relatively short. Therefore, the C-Extra C rules had little effect on the Exp group. However, for the Inexp group, there were significant differences in Stages 2 (p < 0.001), 3 (p < 0.001), and 4 (p = 0.035) in the scoring results of REBA with C-Extra A and C-REBA, indicating that the Inexp group had a greater change in the COG and a longer load-bearing time in these stages. Thus, urgent intervention was needed in these stages. In addition, we calculated the proportion of high-risk (scores above eight points) frames in the caregiving process to assess the risk adaptability of these methods.
There was a significant difference (Table 4; all groups and all stages p1 < 0.05) in the results of the proportion of high-risk frames (scores above eight) between the REBA and C-REBA methods. For the Inexp group, the proportion of high-risk frames in Stages 2, 3, and 4 was high, indicating that the caregiving postures needed to be corrected in these stages. These results suggest that C-REBA avoided the problem of overassessment of caregiver posture. Also, the C-REBA method evaluated the COG changes in the caregiving power position and load-bearing time, and it provided reliable posture risk feedback for the caregivers.

4. Discussion

In this study, we identified the incomplete applicability of the REBA rules in caregiving scenarios. By investigating the differences between the Exp and Inexp groups, we preliminarily explored parameter adjustments for the REBA rules and proposed the C-REBA method. The C-REBA method incorporates crucial factors, including asymmetric load assessment, changes in COG height, and duration of load-bearing. C-REBA effectively differentiated the experienced and inexperienced caregivers, offering posture assessment references and guidance for inexperienced caregivers.
Although the REBA method is widely used in the nursing industry, its risk assessment criteria need to be adjusted according to different scenarios. Raman et al. considered REBA to be an easy-to-apply and fairly reliable tool for alerting clinical dental nurses to ergonomic risks [37]. Law et al. used the existing REBA assessment system to assess the risk of musculoskeletal disorders in transferring patients [22]. However, our study found that the REBA method overestimated the artificial caregiving scenarios and could not distinguish the experienced and inexperienced caregivers. As Yazdanirad et al. found in the risk-adaptation test, the REBA method overestimated the risk level of musculoskeletal disorders [24], which is consistent with our findings. Hence, we explored the key factors involved in caregiving tasks and made adjustments to the additional scoring rules of the REBA method. Our C-REBA method offers caregivers a more accurate and targeted evaluation of the risk levels associated with caregiving postures.
The original Extra A scoring rule is related to the load on the body, ignoring the injury to the body caused by the asymmetric load. Asymmetric loads tend to cause muscle injury on one side of the body, making a great impact. We found notable differences in caregiving power position and COG change trajectory between experienced and inexperienced caregivers, and then we reconstructed the scoring rules of Extra A from the perspective of the COG trajectory. The asymmetric load could be directly reflected in the position change of the COG between the feet. Keeping the COG position in the middle of the feet largely avoided the impact of the asymmetric load. It was also closely related to the caregiving power position. We converted asymmetric loads into the COG position trajectory between the feet to monitor the trend of asymmetric loads and incorporate it into the evaluation rule of C-Extra A. This ensured that C-REBA achieved comprehensive evaluation of caregiving tasks from the level of the asymmetric load.
The original Extra C rule focused on maintaining fixed positions and repeating small ranges of action. There were differences in the load-bearing time and the range of caregiving movements between the Inexp and Exp groups. The load-bearing time was directly proportional to the risk of musculoskeletal discomfort and injury. The greater the change in the COG height, the more work the waist did, which was related to the maintenance of the power position [38,39]. The effects of the load-bearing time and COG height changes were ignored by the original REBA method. We took the Exp group as the standard for load-bearing time and COG height variation, and we integrated this into the evaluation rules of the Extra C rules. The evaluation results of C-REBA indicated that the role of C-Extra C was to distinguish the Exp group from the Inexp group. Our research also found Stages 2, 3, 4 to be the most meaningful postural intervention phases for the inexperienced caregivers. Therefore, we integrated the load-bearing time and COG height change factors into the C-Extra C rules.
Research findings have consistently demonstrated a heightened susceptibility to musculoskeletal injuries among inexperienced caregivers when compared to their experienced counterparts [40,41]. This susceptibility is especially evident in physically demanding tasks, including transferring, lifting, and repositioning patients, and inexperienced caregivers are more prone to experiencing muscle strains and contusions [42]. Unfortunately, little information has been reported in the literature about the physical demands placed on inexperienced caregivers and any association with musculoskeletal injury [43]. Furthermore, the lack of research scrutinizing posture disparities between experienced and inexperienced caregivers, particularly regarding caregiving power positions, has resulted in a lack of clarity among inexperienced caregivers regarding appropriate caregiving power positions. While sporadic face-to-face training sessions on caregiving postures might exist, they are not wildly accessible. Consequently, the regular and timely correction of incorrect postures through repetitive practice presents significant challenges for inexperienced caregivers [44]. Importantly, the power position has demonstrated a certain correlation with the COG [29], and our study also demonstrated significant differences in COG trajectories between the experienced and inexperienced groups. Consequently, visualizing the COG trajectories provided valuable feedback on power positions, allowing inexperienced caregivers to make necessary adjustments based on the trajectory of the COG and the C-REBA scores. This facilitated the development of enhanced proficiency in executing power positions during caregiving.
The risk of musculoskeletal discomfort and injury in nursing can be reduced by reasonable ergonomic posture interventions [45]. Posture feedback is the most effective method of posture intervention [46]. The diversified posture feedback effectively reduces the ergonomic risk of nurses [47]. Posture intervention by merely teaching and training has been proven to be ineffective in reducing the risk of musculoskeletal discomfort and injury, while the postural interventions with biofeedback are more effective [48]. Therefore, visualizing the trajectory of the COG and calculating the C-REBA score for the caregivers in a timely manner could improve the effect of ergonomic posture intervention. Inexperienced caregivers could observe the trajectory of the COG to explore their optimal power position, and they can be made aware of the risk of their posture through the C-REBA score. The proposed C-REBA method could be easily embedded into IoT devices with cameras to infer pose information through neural network models and image processing techniques, as well as providing risk assessment and visual guidance concerning discomfort and injury in caregiving postures. The integration of these devices and technologies aims to build an intelligent caregiving posture assessment system.
Given that caregiving postures depend on the specific needs of patients and encompass a wide range of postures, the means of adjusting the parameters of the REBA rules differs for different postures. Therefore, it is crucial to seek appropriate parameter adjustments to the REBA rules in different caregiving scenarios. Our C-REBA method represents preliminary exploratory research results, providing reference and inspiration for other caregiving posture assessments and parameter adjustments in the REBA rules. In the development of this method, its applicability could be expanded to diverse domains by using the collected data and adjusting specific parameters. For the general population, C-REBA exhibits potential for evaluating the susceptibility to musculoskeletal disorders associated with routine activities, such as cleaning, object transfer, and lifting. For musculoskeletal rehabilitation patients, the C-REBA score could be defined as the benchmark for rehabilitation postures, with specific posture rating scores assigned to different rehabilitation activities. This facilitates autonomous and convenient execution of rehabilitation exercises by the patients. Our study offers a novel perspective in terms of seeking parameter adjustments to the REBA rules, which could serve as a reference for posture evaluations in different caregiving scenarios, potentially leading to the development of more targeted posture assessment methods.

Limitation

Due to the limitations of the experimental conditions, the sample size was small, and convenience sampling was used to recruit the participants; more sample data would be needed to determine the optimal parameter thresholds for scoring. Our sample size was comparable to that used in the studies by Law et al. [22] and Yuan et al. [49]. However, the fundamental considerations regarding the proposed influencing factors and the methodology for tuning the parameter thresholds could be extended to other types of caregiving movements. It would be worthwhile to recruit subjects of different heights and patients of different body types to improve the extensibility of these experimental results in further research. Also, although timely posture risk warning is necessary [50], exploring the long-term impact of this method on the musculoskeletal health of caregivers holds significant potential and value. Moreover, utilizing a binocular camera configuration can enhance the precision of measuring the body’s torsional motion [51]. It might also be more convenient to ascertain body torsion by detecting the vector displacement of OPENPOSE skeleton points such as the head and waist [52].

5. Conclusions

In this study, we explained a modified REBA method by adding new factors (COG trajectory, load-bearing time, and asymmetric load) as key parameters. Furthermore, how to determine the scores based on these factors was discussed in detail. As a result, compared with the original REBA method, the proposed C-REBA method addressed the risk overestimation of caregiving tasks and effectively differentiated the experienced and inexperienced caregivers. In the development of this method, its application could be extended to other scenarios by adjusting the parameters with the collected data. Furthermore, the developed analysis algorithm could be easily embedded into a camera IoT system for real-time risk assessment, and it could be also applied to the recorded video clips for providing visual guidance and warning of the risk of discomfort or injury. This kind of vision-based posture assessment and feedback system could provide ergonomic risk evaluations for caregiving postures without disrupting the caregivers’ normal work.

Author Contributions

Conceptualization, X.H., Z.J. and N.N.; methodology, X.H., Z.J. and N.N.; software, X.H.; validation, X.H., M.M. (Minoru Morita) and M.M. (Mao Mitsuda); formal analysis, X.H. and M.M. (Minoru Morita); investigation, X.H., M.M. (Minoru Morita) and M.M. (Mao Mitsuda); data curation, X.H.; writing—original draft preparation, X.H.; writing—review and editing, Z.J. and N.N.; visualization, X.H.; supervision, Z.J. and N.N. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

This study was approved by the ethics committee at the Center for Clinical Research of the co-authors’ hospital (H2019-182).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. The data are not publicly available because they are part of the ongoing doctoral research of the first author.

Acknowledgments

We would like to express our gratitude to the nurses and volunteers from Yamaguchi University who participated in our study. We also wish to thank Takeshi Nishimoto, Takashi Sakai, Kiminori Yukata and Satoshi Harada from the Department of Orthopedic Surgery, Yamaguchi University Hospital, for their assistance in the ergonomic assessment and validation.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Mitchell, S.E.; Laurens, V.; Weigel, G.M.; Hirschman, K.B.; Scott, A.M.; Nguyen, H.Q.; Howard, J.M.; Laird, L.; Levine, C.; Davis, T.C.; et al. Care Transitions from Patient and Caregiver Perspectives. Ann. Fam. Med. 2018, 16, 225–231. [Google Scholar] [CrossRef] [PubMed]
  2. Kokorelias, K.M.; Gignac, M.A.M.; Naglie, G.; Rittenberg, N.; Cameron, J.I. Caregivers’ decision-making for health service utilisation across the Alzheimer’s disease trajectory. Health Soc. Care Community 2021, 30, 1344–1352. [Google Scholar] [CrossRef] [PubMed]
  3. Rosqvist, K.; Schrag, A.; Odin, P. Caregiver Burden and Quality of Life in Late Stage Parkinson’s Disease. Brain Sci. 2022, 12, 111. [Google Scholar] [CrossRef] [PubMed]
  4. Secinti, E.; Wu, W.; Kent, E.E.; Demark-Wahnefried, W.; Lewson, A.B.; Mosher, C.E. Examining Health Behaviors of Chronic Disease Caregivers in the U.S. Am. J. Prev. Med. 2022, 62, 145–158. [Google Scholar] [CrossRef] [PubMed]
  5. Carneiro, P.; Martins, J.; Torres, M. Musculoskeletal disorder risk assessment in home care nurses. Work 2015, 51, 657–665. [Google Scholar] [CrossRef] [PubMed]
  6. Carneiro, P.; Braga, A.C.; Barroso, M. Work-related musculoskeletal disorders in home care nurses: Study of the main risk factors. Int. J. Ind. Ergon. 2017, 61, 22–28. [Google Scholar] [CrossRef]
  7. Faisting, A.L.R.F.; de Oliveira Sato, T. Effectiveness of ergonomic training to reduce physical demands and musculoskeletal symptoms-An overview of systematic reviews. Int. J. Ind. Ergon. 2019, 74, 102845. [Google Scholar] [CrossRef]
  8. Karhu, O.; Kansi, P.; Kuorinka, I. Correcting working postures in industry: A practical method for analysis. Appl. Ergon. 1977, 8, 199–201. [Google Scholar] [CrossRef]
  9. McAtamney, L.; Corlett, E.N. RULA: A survey method for the investigation of work-related upper limb disorders. Appl. Ergon. 1993, 24, 91–99. [Google Scholar] [CrossRef]
  10. Hignett, S.; McAtamney, L. Rapid Entire Body Assessment (REBA). Appl. Ergon. 2000, 31, 201–205. [Google Scholar] [CrossRef]
  11. Enez, K.; Nalbantoğlu, S.S. Comparison of ergonomic risk assessment outputs from OWAS and REBA in forestry timber harvesting. Int. J. Ind. Ergon. 2019, 70, 51–57. [Google Scholar] [CrossRef]
  12. Gallo, R.; Mazzetto, F. Ergonomic analysis for the assessment of the risk of work-related musculoskeletal disorder in forestry operations. J. Agric. Eng. 2013, 44, s2. [Google Scholar] [CrossRef]
  13. Norhidayah, M.S.; Mohamed, N.M.Z.N.; Mansor, M.A.; Ismail, A.R. A study of postural loading in Malaysian mining industry using rapid entire body assessment. MATEC Web Conf. 2016, 74, 00014. [Google Scholar] [CrossRef]
  14. Fouladi-Dehaghi, B.; Tajik, R.; Ibrahimi-Ghavamabadi, L.; Sajedifar, J.; Teimori-Boghsani, G.; Attar, M. Physical risks of work-related musculoskeletal complaints among quarry workers in East of Iran. Int. J. Ind. Ergon. 2021, 82, 103107. [Google Scholar] [CrossRef]
  15. Iridiastadi, H.; Vani, T.; Yamin, P.A.R. Biomechanical Evaluation of a Patient-Handling Technology Prototype. Int. J. Technol. 2020, 11, 180–189. [Google Scholar] [CrossRef]
  16. Hita-Gutiérrez, M.; Gómez-Galán, M.; Díaz-Pérez, M.; Callejón-Ferre, Á.J. An overview of REBA method applications in the world. Int. J. Environ. Res. Public Health 2020, 17, 2635. [Google Scholar] [CrossRef]
  17. Abdollahi, T.; Pedram Razi, S.; Pahlevan, D.; Yekaninejad, M.S.; Amaniyan, S.; Leibold Sieloff, C.; Vaismoradi, M. Effect of an Ergonomics Educational Program on Musculo-skeletal Disorders in Nursing Staff Working in the Operating Room: A Quasi-Randomized Controlled Clinical Trial. Int. J. Environ. Res. Public Health 2020, 17, 7333. [Google Scholar] [CrossRef] [PubMed]
  18. Abdollahzade, F.; Mohammadi, F.; Dianat, I.; Asghari, E.; Asghari-Jafarabadi, M.; Sokhanvar, Z. Working posture and its predictors in hospital operating room nurses. Health Promot. Perspect. 2016, 6, 17. [Google Scholar] [CrossRef]
  19. Davison, C.; Cotrim, T.P.; Gonçalves, S. Ergonomic assessment of musculoskeletal risk among a sample of Portuguese emergency medical technicians. Int. J. Ind. Ergon. 2021, 82, 103077. [Google Scholar] [CrossRef]
  20. Jahanimoghadam, F.; Horri, A.; Hasheminejad, N.; Nejad, N.H.; Baneshi, M.R. Ergonomic evaluation of dental professionals as determined by rapid entire body assessment method in 2014. J. Dent. 2018, 19, 155. [Google Scholar]
  21. Keskin, M.; Karadede, M.I.; Kaya, D.O. Spinal pain, curvature, and mobility comparisons according to spine region in dentists working in risky postures. Int. J. Ind. Ergon. 2023, 98, 103518. [Google Scholar] [CrossRef]
  22. Law, M.J.J.; Ridzwan, M.I.Z.; Mohd Ripin, Z.; Abd Hamid, I.J.; Law, K.S.; Karunagaran, J.; Cajee, Y. REBA assessment of patient transfer work using sliding board and Motorized Patient Transfer Device. Int. J. Ind. Ergon. 2022, 90, 103322. [Google Scholar] [CrossRef]
  23. Schwartz, A.H.; Albin, T.J.; Gerberich, S.G. Intra-rater and inter-rater reliability of the rapid entire body assessment (REBA) tool. Int. J. Ind. Ergon. 2019, 71, 111–116. [Google Scholar] [CrossRef]
  24. Yazdanirad, S.; Pourtaghi, G.; Raei, M.; Ghasemi, M. Developing and validating the personal risk assessment of musculoskeletal disorders (PRAMUD) tool among workers of a steel foundry. Int. J. Ind. Ergon. 2022, 88, 103276. [Google Scholar] [CrossRef]
  25. Tucker, C.A.; Ramirez, J.; Krebs, D.E.; Riley, P.O. Center of gravity dynamic stability in normal and vestibulopathic gait. Gait Posture 1998, 8, 117–123. [Google Scholar] [CrossRef] [PubMed]
  26. Chou, L.-S.; Kaufman, K.R.; Hahn, M.E.; Brey, R.H. Medio-lateral motion of the center of mass during obstacle crossing distinguishes elderly individuals with imbalance. Gait Posture 2003, 18, 125–133. [Google Scholar] [CrossRef] [PubMed]
  27. Gassett, R.S.; Hearne, B.; Keelan, B. Ergonomics and body mechanics in the work place. Orthop. Clin. N. Am. 1996, 27, 861–879. [Google Scholar] [CrossRef]
  28. Rosker, J.; Markovic, G.; Sarabon, N. Effects of vertical center of mass redistribution on body sway parameters during quiet standing. Gait Posture 2011, 33, 452–456. [Google Scholar] [CrossRef]
  29. Wu, G.; MacLeod, M. The control of body orientation and center of mass location under asymmetrical loading. Gait Posture 2001, 13, 95–101. [Google Scholar] [CrossRef]
  30. Ferland, P.M.; Comtois, A.S. Classic powerlifting performance: A systematic review. J. Strength Cond. Res. 2019, 33, 194–201. [Google Scholar] [CrossRef]
  31. Guo, L.; Xiong, S. Effects of working posture, lifting load, and standing surface on postural instability during simulated lifting tasks in construction. Ergonomics 2020, 63, 1571–1583. [Google Scholar] [CrossRef] [PubMed]
  32. Hwang, J.; Kuppam, V.A.; Chodraju, S.S.R.; Chen, J.; Kim, J.H. Commercially Available Friction-Reducing Patient-Transfer Devices Reduce Biomechanical Stresses on Caregivers’ Upper Extremities and Low Back. Hum. Factors 2019, 61, 1125–1140. [Google Scholar] [CrossRef] [PubMed]
  33. Kim, W.; Sung, J.; Saakes, D.; Huang, C.; Xiong, S. Ergonomic postural assessment using a new open-source human pose estimation technology (OpenPose). Int. J. Ind. Ergon. 2021, 84, 103164. [Google Scholar] [CrossRef]
  34. González, A.; Hayashibe, M.; Fraisse, P. Estimation of the center of mass with Kinect and Wii balance board. In Proceedings of the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura-Algarve, Portugal, 7–12 October 2012; pp. 1023–1028. [Google Scholar]
  35. Cotton, S.; Vanoncini, M.; Fraisse, P.; Ramdani, N.; Demircan, E.; Murray, A.P.; Keller, T. Estimation of the centre of mass from motion capture and force plate recordings: A study on the elderly. Appl. Bionics Biomech. 2011, 8, 67–84. [Google Scholar] [CrossRef]
  36. Clark, R.A.; Bryant, A.L.; Pua, Y.; McCrory, P.; Bennell, K.; Hunt, M. Validity and reliability of the Nintendo Wii Balance Board for assessment of standing balance. Gait Posture 2010, 31, 307–310. [Google Scholar] [CrossRef] [PubMed]
  37. Raman, V.; Ramlogan, S.; Sweet, J.; Sweet, D. Application of the Rapid Entire Body Assessment (REBA) in assessing chairside ergonomic risk of dental students. Br. Dent. J. 2020, 1–6. [Google Scholar] [CrossRef] [PubMed]
  38. Hecker, K.A.; Carlson, L.A.; Lawrence, M.A. Effects of the safety squat bar on trunk and lower-body mechanics during a back squat. J. Strength Cond. Res. 2019, 33, 45–51. [Google Scholar] [CrossRef]
  39. Falch, H.N.; Kristiansen, E.; van den Tillaar, R. A Biomechanical Comparison between Squatbar® and Olympic Barbell. Biomechanics 2023, 3, 258–266. [Google Scholar] [CrossRef]
  40. Hayes, J.; Chapman, P.; Young, L.J.; Rittman, M. The prevalence of injury for stroke caregivers and associated risk factors. Top. Stroke Rehabil. 2009, 16, 300–307. [Google Scholar] [CrossRef]
  41. Darragh, A.R.; Campo, M.; King, P. Work-related activities associated with injury in occupational and physical therapists in five practice areas. Work 2012, 42, 373–384. [Google Scholar] [CrossRef]
  42. Hartke, R.J.; King, R.B.; Heinemann, A.W.; Semik, P. Accidents in older caregivers of persons surviving stroke and their relation to caregiver stress. Rehabil. Psychol. 2006, 51, 150. [Google Scholar] [CrossRef]
  43. Darragh, A.R.; Sommerich, C.M.; Lavender, S.A.; Tanner, K.J.; Vogel, K.; Campo, M. Musculoskeletal discomfort, physical demand, and caregiving activities in informal caregivers. J. Appl. Gerontol. 2015, 34, 734–760. [Google Scholar] [CrossRef] [PubMed]
  44. Matsangidou, M.; Solomou, T.; Høegh Langvad, C.; Xynari, K.; Papayianni, E.; Pattichis, C.S. Virtual Reality Health Education to Prevent Musculoskeletal Disorders and Chronic Low Back Pain in Formal and Informal Caregivers. In Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization, New York, NY, USA, 26–29 June 2023; pp. 343–351. [Google Scholar]
  45. Chanchai, W.; Songkham, W.; Ketsomporn, P.; Sappakitchanchai, P.; Siriwong, W.; Robson, M. The Impact of an Ergonomics Intervention on Psychosocial Factors and Musculoskeletal Symptoms among Thai Hospital Orderlies. Int. J. Environ. Res. Public Health 2016, 13, 464. [Google Scholar] [CrossRef] [PubMed]
  46. Ziam, S.; Lakhal, S.; Laroche, E.; Lane, J.; Alderson, M.; Gagné, C. Musculoskeletal disorder (MSD) prevention practices by nurses working in health care settings: Facilitators and barriers to implementation. Appl. Ergon. 2023, 106, 103895. [Google Scholar] [CrossRef]
  47. Hernández, C.O.; Li, S.; Rivera, M.D.M.; Rodríguez, I.M. Does Postural Feedback Reduce Musculoskeletal Risk?: A Randomized Controlled Trial. Sustainability 2022, 14, 583. [Google Scholar] [CrossRef]
  48. Owlia, M.; Kamachi, M.; Dutta, T. Reducing lumbar spine flexion using real-time biofeedback during patient handling tasks. Work 2020, 66, 41–51. [Google Scholar] [CrossRef] [PubMed]
  49. Yuan, H.; Zhou, Y. Ergonomic assessment based on monocular RGB camera in elderly care by a new multi-person 3D pose estimation technique (ROMP). Int. J. Ind. Ergon. 2023, 95, 103440. [Google Scholar] [CrossRef]
  50. Hoe, V.C.; Urquhart, D.M.; Kelsall, H.L.; Zamri, E.N.; Sim, M.R. Ergonomic interventions for preventing work-related musculoskeletal disorders of the upper limb and neck among office workers. Cochrane Database Syst. Rev. 2018, 10. [Google Scholar] [CrossRef]
  51. Cai, S.; Shao, M.; Du, M.; Bao, G.; Fan, B. A Binocular-Camera-Assisted Sensor-to-Segment Alignment Method for Inertial Sensor-Based Human Gait Analysis. IEEE Sens. J. 2022, 23, 2663–2671. [Google Scholar] [CrossRef]
  52. Liu, W.; Bao, Q.; Sun, Y.; Mei, T. Recent advances of monocular 2d and 3d human pose estimation: A deep learning perspective. ACM Comput. Surv. 2022, 55, 1–41. [Google Scholar] [CrossRef]
Applsci 13 12699 g001
Figure 1. Representative pictures of the 5 stages of the caregiving task. These are representative pictures of the whole caregiving task completed by the one experienced nurse and the standard patient under the experimental conditions. This is the whole caregiving task, which is divided into 5 stages, and the above picture is a representative picture of each stage captured by the camera.
Figure 1. Representative pictures of the 5 stages of the caregiving task. These are representative pictures of the whole caregiving task completed by the one experienced nurse and the standard patient under the experimental conditions. This is the whole caregiving task, which is divided into 5 stages, and the above picture is a representative picture of each stage captured by the camera.
Applsci 13 12699 g001
Applsci 13 12699 g002
Figure 2. REBA scoring rules: (a) The details of the REBA rule. The upper left panel illustrates the scoring rule of Score A; the upper right panel illustrates the scoring rule of Score B; the lower left panel illustrates the generation of Score C by adding A and B. The lower right panel illustrates the final REBA score generation. (b) The pipeline of REBA scoring.
Figure 2. REBA scoring rules: (a) The details of the REBA rule. The upper left panel illustrates the scoring rule of Score A; the upper right panel illustrates the scoring rule of Score B; the lower left panel illustrates the generation of Score C by adding A and B. The lower right panel illustrates the final REBA score generation. (b) The pipeline of REBA scoring.
Applsci 13 12699 g002
Applsci 13 12699 g003
Figure 3. Evaluation results of caregiving posture with the REBA method; n = 8 in the Exp group, n = 10 in the Inexp group. NS = not significant. The red line represents a REBA score of 8 points.
Figure 3. Evaluation results of caregiving posture with the REBA method; n = 8 in the Exp group, n = 10 in the Inexp group. NS = not significant. The red line represents a REBA score of 8 points.
Applsci 13 12699 g003
Applsci 13 12699 g004
Figure 4. The difference in the trajectory of the pixel COG between the Exp and Inexp groups: (a) The representative COG trajectory map during different stages of caregiving. The pixel COG of each frame consists of pixel coordinates (x, z), where x represents the image width and z represents the image height. The COG coordinates of many frames were fused to obtain the COG trajectory. (b) Two representative trajectories of the pixel COG of the Exp group. (c) Two representative trajectories of the pixel COG of the Inexp group.
Figure 4. The difference in the trajectory of the pixel COG between the Exp and Inexp groups: (a) The representative COG trajectory map during different stages of caregiving. The pixel COG of each frame consists of pixel coordinates (x, z), where x represents the image width and z represents the image height. The COG coordinates of many frames were fused to obtain the COG trajectory. (b) Two representative trajectories of the pixel COG of the Exp group. (c) Two representative trajectories of the pixel COG of the Inexp group.
Applsci 13 12699 g004
Applsci 13 12699 g005
Figure 5. Difference analysis of the height change of the COG: (a) The height of the COG of the participants at the pixel level in different stages. N1−N8: nurses 1−8 (Exp group); V1−V10: volunteers 1−10 (Inexp group). The x-axis represents time, and the y-axis represents the normalized value of the COG height and the participant’s height. (b) The height difference between the COG of the previous stage and the next stage. Stage 1 represents the height difference between the COG of the participant standing upright and in Stage 1; n = 8 in the Exp group, n = 10 in the Inexp group. NS = not significant, ** p < 0.01, *** p < 0.001.
Figure 5. Difference analysis of the height change of the COG: (a) The height of the COG of the participants at the pixel level in different stages. N1−N8: nurses 1−8 (Exp group); V1−V10: volunteers 1−10 (Inexp group). The x-axis represents time, and the y-axis represents the normalized value of the COG height and the participant’s height. (b) The height difference between the COG of the previous stage and the next stage. Stage 1 represents the height difference between the COG of the participant standing upright and in Stage 1; n = 8 in the Exp group, n = 10 in the Inexp group. NS = not significant, ** p < 0.01, *** p < 0.001.
Applsci 13 12699 g005
Applsci 13 12699 g006
Figure 6. Left and right foot load difference range and the C-Extra A score: (a) The difference in left and right foot loads between the Exp group and the Inexp group, with the quartiles marked on the right; n = 8 in the Exp group, n = 10 in the Inexp group, NS = not significant. (b) C-Extra A scoring rule of C-REBA.
Figure 6. Left and right foot load difference range and the C-Extra A score: (a) The difference in left and right foot loads between the Exp group and the Inexp group, with the quartiles marked on the right; n = 8 in the Exp group, n = 10 in the Inexp group, NS = not significant. (b) C-Extra A scoring rule of C-REBA.
Applsci 13 12699 g006
Applsci 13 12699 g007
Figure 7. The load difference range between the left and right feet, matched to the distance range: (a) Schematic diagram of generating the fused COP. COP1 represents the data of the left foot measured by Wii Balance Board 1. COP2 represents the data of the right foot measured by Wii Balance Board 2. The ground reaction forces are F1 and F2, respectively. Based on COP1 and COP2, the fused COP could be calculated according to the center-of-pressure principle, where the fused COP ground reaction force is Fc, and the distance between the feet is L. (b) Representative trajectory of the fused COP and the distance range corresponding to the load difference range. The load difference data of each fused COP point are marked with corresponding colors (<9.99 kg, green; 9.99−24.54 kg, blue; >24.54 kg, red). The load difference score range (Figure 6b) could be converted into the distance score range of the fused COP trajectory on the x-axis (<0.99 kg: C−D; 9.99−24.54 kg: B−C and D−E; >24.54 kg: A−B and E−F). (c) Schematic diagram of generating the image distance scoring range. The distance between the feet was assumed as L at the visualized image level, while the position of the COG map on L was approximated as the position of the COP map on A−F. Then, the L percentile value for each individual within different score ranges could be calculated by Equation (3). Subsequently, the average of these L percentile values could be obtained using statistical methods to determine the score ranges (A−B−C−D−E−F) corresponding to the L-percentile-based values (0−0.19L−0.38L−0.75L−0.86L−L).
Figure 7. The load difference range between the left and right feet, matched to the distance range: (a) Schematic diagram of generating the fused COP. COP1 represents the data of the left foot measured by Wii Balance Board 1. COP2 represents the data of the right foot measured by Wii Balance Board 2. The ground reaction forces are F1 and F2, respectively. Based on COP1 and COP2, the fused COP could be calculated according to the center-of-pressure principle, where the fused COP ground reaction force is Fc, and the distance between the feet is L. (b) Representative trajectory of the fused COP and the distance range corresponding to the load difference range. The load difference data of each fused COP point are marked with corresponding colors (<9.99 kg, green; 9.99−24.54 kg, blue; >24.54 kg, red). The load difference score range (Figure 6b) could be converted into the distance score range of the fused COP trajectory on the x-axis (<0.99 kg: C−D; 9.99−24.54 kg: B−C and D−E; >24.54 kg: A−B and E−F). (c) Schematic diagram of generating the image distance scoring range. The distance between the feet was assumed as L at the visualized image level, while the position of the COG map on L was approximated as the position of the COP map on A−F. Then, the L percentile value for each individual within different score ranges could be calculated by Equation (3). Subsequently, the average of these L percentile values could be obtained using statistical methods to determine the score ranges (A−B−C−D−E−F) corresponding to the L-percentile-based values (0−0.19L−0.38L−0.75L−0.86L−L).
Applsci 13 12699 g007
Applsci 13 12699 g008
Figure 8. The load-bearing time and COG height difference: (a) The difference in load-bearing time between the Exp and Inexp groups; the quartiles are marked on the right; n = 8 in the Exp group, n = 10 in the Inexp group. (b) The difference in the height of the COG between Stage 1 and the highest point of lifting patients (Stage 3 or Stage 4); the quartiles are marked on the right; n = 8 in the Exp group, n = 10 in the Inexp group; *** p < 0.001.
Figure 8. The load-bearing time and COG height difference: (a) The difference in load-bearing time between the Exp and Inexp groups; the quartiles are marked on the right; n = 8 in the Exp group, n = 10 in the Inexp group. (b) The difference in the height of the COG between Stage 1 and the highest point of lifting patients (Stage 3 or Stage 4); the quartiles are marked on the right; n = 8 in the Exp group, n = 10 in the Inexp group; *** p < 0.001.
Applsci 13 12699 g008
Applsci 13 12699 g009
Figure 9. The differences between the REBA and C-REBA scoring results; n = 8 in the Exp group, n = 10 in the Inexp group. NS = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001. The red line represents a REBA score of 8 points.
Figure 9. The differences between the REBA and C-REBA scoring results; n = 8 in the Exp group, n = 10 in the Inexp group. NS = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001. The red line represents a REBA score of 8 points.
Applsci 13 12699 g009
Table 1. Demographics of the participants.
Table 1. Demographics of the participants.
DemographicExp Group (n = 8)Inexp Group (n = 10)p-Value
Age29.125 ± 3.4826.9 ± 6.110.348
Weight (kg)61.06 ± 9.562.97 ± 7.730.644
Experience (years)6.25 ± 3.150<0.001
Height (cm)167.5 ± 4.84169.89 ± 5.070.328
Table 2. The C-Extra C scoring rule of C-REBA.
Table 2. The C-Extra C scoring rule of C-REBA.
ScoreRule
+1Large body tipping due to COG instability
+1Load-bearing time over 3.87 s
+1The COG height difference between Stage 1 and the highest point exceeds 7.54 cm
Table 3. The influence of C-Extra A and C-Extra C on the REBA results.
Table 3. The influence of C-Extra A and C-Extra C on the REBA results.
StageREBAREBA with C-Extra AC-REBAp1p2p3
Exp Group
(n = 8)		Stage18.88 ± 0.356.38 ± 0.926.38 ± 0.92p < 0.001p < 0.0011.0
Stage29.00 ± 07.25 ± 0.897.25 ± 0.89p < 0.001p < 0.0011.0
Stage38.38 ± 0.926.13 ± 0.835.88 ± 0.99p < 0.001p < 0.0010.592
Stage48.00 ± 0.766.63 ± 1.066.38 ± 1.190.010.0050.663
Stage57.88 ± 0.836.63 ± 1.306.63 ± 1.300.0370.0371.0
Inexp Group
(n = 10)		Stage19.10 ± 0.577.40 ± 0.527.70 ± 0.67p < 0.001p < 0.0010.277
Stage29.20 ± 0.638.00 ± 0.8210.00 ± 1.150.0020.096p < 0.001
Stage38.90 ± 0.578.20 ± 0.429.60 ± 0.840.0060.061p < 0.001
Stage48.70 ± 0.678.10 ± 0.578.80 ± 0.790.0440.7790.035
Stage59.30 ± 0.677.30 ± 0.958.00 ± 1.33p < 0.0010.0220.192
Note: p1 represents the p-value of REBA and REBA with C-Extra A, p2 represents the p-value of REBA and C-REBA, and p3 represents the p-value of REBA with C-Extra A and C-REBA.
Table 4. The proportion of high-risk frames in caregiving work.
Table 4. The proportion of high-risk frames in caregiving work.
StageREBAC-REBAP
Exp Group
(n = 8)		Stage192% ± 8%20% ± 7%p < 0.001
Stage291% ± 6%18% ± 4%p < 0.001
Stage391% ± 5%6% ± 4%p < 0.001
Stage493% ± 7%6% ± 3%p < 0.001
Stage592% ± 5%13% ± 4%p < 0.001
Inexp Group
(n = 10)		Stage193% ± 8%50% ± 18%p < 0.001
Stage294% ± 6%82% ± 6%0.001
Stage394% ± 6%81% ± 13%0.019
Stage497% ± 2%74% ± 13%p < 0.001
Stage592% ± 8%46% ± 14%p < 0.001
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Han, X.; Nishida, N.; Morita, M.; Mitsuda, M.; Jiang, Z. Visualization of Caregiving Posture and Risk Evaluation of Discomfort and Injury. Appl. Sci. 2023, 13, 12699. https://doi.org/10.3390/app132312699

AMA Style

Han X, Nishida N, Morita M, Mitsuda M, Jiang Z. Visualization of Caregiving Posture and Risk Evaluation of Discomfort and Injury. Applied Sciences. 2023; 13(23):12699. https://doi.org/10.3390/app132312699

Chicago/Turabian Style

Han, Xin, Norihiro Nishida, Minoru Morita, Mao Mitsuda, and Zhongwei Jiang. 2023. "Visualization of Caregiving Posture and Risk Evaluation of Discomfort and Injury" Applied Sciences 13, no. 23: 12699. https://doi.org/10.3390/app132312699

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop