Pulse Rate Variability in Emergency Physicians During Shifts: Pilot Cross-Sectional Study

Applying a Smartwatch to Predict Work-related Fatigue for Emergency Healthcare Professionals: Machine Learning Method

INTRODUCTION

Healthcare professionals frequently experience work-related fatigue, which may jeopardize their health and put patient safety at risk. In this study, we applied a machine learning (ML) approach based on data collected from a smartwatch to construct prediction models of work-related fatigue for emergency clinicians.

METHODS

We conducted this prospective study at the emergency department (ED) of a tertiary teaching hospital from March 10-June 20, 2021, where we recruited physicians, nurses, and nurse practitioners. All participants wore a commercially available smartwatch capable of measuring various physiological data during the experiment. Participants completed the Multidimensional Fatigue Inventory (MFI) web form before and after each of their work shifts. We calculated and labeled the before-and-after-shift score differences between each pair of scores. Using several tree-based algorithms, we constructed the prediction models based on features collected from the smartwatch. Records were split into training/validation and testing sets at a 70:30 ratio, and we evaluated the performances using the area under the curve (AUC) measure of receiver operating characteristic on the test set.

RESULTS

In total, 110 participants were included in this study, contributing to a set of 1,542 effective records. Of these records, 85 (5.5%) were labeled as having work-related fatigue when setting the MFI difference between two standard deviations as the threshold. The mean age of the participants was 29.6. Most of the records were collected from nurses (87.7%) and females (77.5%). We selected a union of 31 features to construct the models. For total participants, CatBoost classifier achieved the best performances of AUC (0.838, 95% confidence interval [CI] 0.742-0.918) to identify work-related fatigue. By focusing on a subgroup of nurses <35 years in age, XGBoost classifier obtained excellent performance of AUC (0.928, 95% CI 0.839-0.991) on the test set.

CONCLUSION

By using features derived from a smartwatch, we successfully built ML models capable of classifying the risk of work-related fatigue in the ED. By collecting more data to optimize the models, it should be possible to use smartwatch-based ML models in the future to predict work-related fatigue and adopt preventive measures for emergency clinicians.

Determination of Waste Management Workers' Physical and Psychological Load: A Cross-Sectional Study Using Biometric Data

Waste management workers experience high stress and physical strain in their work environment, but very little empirical evidence supports effective health management practices for waste management workers. Hence, this study investigated the effects of worker characteristics and biometric indices on workers' physical and psychological loads during waste-handling operations. A biometric measurement system was installed in an industrial waste management facility in Japan to understand the actual working conditions of 29 workers in the facility. It comprised sensing wear for data collection and biometric sensors to measure heart rate (HR) and physical activity (PA) based on electrocardiogram signals. Multiple regression analysis was performed to evaluate significant relationships between the parameters. Although stress level is indicated by the ratio of low frequency (LF) to high frequency (HF) or high LF power in HR, the results showed that compared with workers who did not handle waste, those who did had lower PA and body surface temperature, higher stress, and lower HR variability parameters associated with higher psychological load. There were no significant differences in HR, heart rate interval (RRI), and workload. The psychological load of workers dealing directly with waste was high, regardless of their PA, whereas others had a low psychological load even with high PA. These findings suggest the need to promote sustainable work relationships and a quantitative understanding of harsh working conditions to improve work quality and reduce health hazards.

Persuasive Design Techniques and App Design Recommendations to Improve Health Workforce Capability in Rural Health Professionals: What Do Users Want and How Does an App Help?

BACKGROUND

Health professionals' perceptions of persuasive design techniques for use in technological solutions to improve health workforce capability have not been previously explored.

OBJECTIVE

This study aims to explore rural health professionals' perceptions of health workforce capability and persuasive design techniques; and translate these into recommendations for designing a health workforce capability app to increase their impact and usefulness.

METHODS

Qualitative interviews with 13 rural health professionals were conducted. Subsequently, 32 persuasive techniques were used as a framework to deductively analyze the data. Persuasive design technique domains were Primary Task Support, Dialog Support, System Credibility Support, Social Support, and Cialdini's Principles of Persuasion.

RESULTS

Persuasive design techniques can be applied across the factors that influence health workforce capability including health and personal qualities; competencies and skills; values, attitudes, and motivation; and factors that operate outside of work and at the team, organizational, and labor market levels. The majority of the 32 persuasive design techniques were reflected in the data from the interviews and led to recommendations as to how these could be translated into practice, with the exception of scarcity. Many suggestions and persuasive design techniques link back to the need for tailored and localized solutions such as the need for country-specific-based evidence, the wish for localized communities of practice, learning from other rural health professionals, and referral pathways to other clinicians. Participants identified how persuasive design techniques can optimize the user experience to help meet rural health professionals needs for more efficient systems to improve patient access to care, quality care, and to enable working in interprofessional team-based care. Social inclusion plays a vital role for health professionals, indicating the importance of the Social Support domain of persuasive techniques. Overall, health professionals were open to self-monitoring of their work performance and some professionals used wearables to monitor their health.

CONCLUSIONS

Rural health professionals' perceptions of health workforce capability informed which persuasive design techniques can be used to optimize the user experience of an app. These were translated into recommendations for designing a health workforce capability app to increase likelihood of adoption. This study has also contributed to the further validation of the Persuasive Systems Design model through empirically aligning elements of the model to increase persuasive system content and functionality with real-world applied data, in this case the health workforce capability of rural health professionals. Our findings confirm that these techniques can be used to develop a future prototype of an app that may assist health professionals in improving or maintaining their health workforce capability which in turn may increase recruitment and retention in rural areas. Health professionals need to be central during the design phase. Interventions are needed to provide a supportive environment to rural and remote health professionals to increase their rural health workforce capability.

Psychological Effects of Heart Rate and Physical Vibration on the Operation of Construction Machines: Experimental Study

BACKGROUND

A construction method has emerged in which a camera is installed around a construction machine, and the operator remotely controls the machine while synchronizing the vibration of the machine with the images seen from the operator's seat using virtual reality (VR) technology. Indices related to changes in heart rate (HR) and physical vibration, such as heart rate variability (HRV) and multiscale entropy (MSE), can then be measured among the operators. As these indices are quantitative measures of autonomic regulation in the cardiovascular system, they can provide a useful means of assessing operational stress.

OBJECTIVE

In this study, we aimed to evaluate changes in HR and body vibration of machine operators and investigate appropriate methods of machine operation while considering the psychological load.

METHODS

We enrolled 9 remote operators (18-50 years old) in the experiment, which involved 42 measurements. A construction machine was driven on a test course simulating a construction site, and three patterns of operation-riding operation, remote operation using monitor images, and VR operation combining monitor images and machine vibration-were compared. The heartbeat, body vibration, and driving time of the participants were measured using sensing wear made of a woven film-like conductive material and a three-axis acceleration measurement device (WHS-2). We used HRV analysis in the time and frequency domains, MSE analysis as a measure of the complexity of heart rate changes, and the ISO (International Standards Organization) 2631 vibration index. Multiple regression analysis was conducted to model the relationship among the low frequency (LF)/high frequency (HF) HRV, MSE, vibration index, and driving time of construction equipment. Efficiency in driving time was investigated with a focus on stress reduction.

RESULTS

Multiple comparisons conducted via the Bonferroni test and Kruskal-Wallis test showed statistically significant differences (P=.05) in HRV-LF/HF, the vibration index, weighted acceleration, motion sickness dose value (MSDVz), and the driving time among the three operation patterns. The riding operation was found to reduce the driving time of the machine, but the operation stress was the highest in this case; operation based on the monitor image was found to have the lowest operation stress but the longest operation time. Multiple regression analysis showed that the explanatory variables (LH/HF), RR interval, and vibration index (MSDVz by vertical oscillation at 0.5-5 Hz) had a negative effect on the driving time (adjusted coefficient of determination R2=0.449).

CONCLUSIONS

A new method was developed to calculate the appropriate operating time by considering operational stress and suppressing the physical vibration within an acceptable range. By focusing on the relationship between psychological load and physical vibration, which has not been explored in previous studies, the relationship of these variables with the driving time of construction machines was clarified.