Association of Work Schedules With Nurse Turnover: A Cross-Sectional National Study

Exploring Nurses' Behavioural Intention to Adopt AI Technology: The Perspectives of Social Influence, Perceived Job Stress and Human-Machine Trust

AIM

This study examines how social influence, human-machine trust and perceived job stress affect nurses' behavioural intentions towards AI-assisted care technology adoption from a new perspective and framework. It also explores the interrelationships between different types of social influence and job stress dimensions to fill gaps in academic literature.

DESIGN

A quantitative cross-sectional study.

METHODS

Five hospitals in Taiwan that had implemented AI solutions were selected using purposive sampling. The scales, adapted from relevant literature, were translated into Chinese and modified for context. Questionnaires were distributed to nurses via snowball sampling from May 15 to June 10, 2023. A total of 283 valid questionnaires were analysed using the partial least squares structural equation modelling method.

RESULTS

Conformity, obedience and human-machine trust were positively correlated with behavioural intention, while compliance was negatively correlated. Perceived job stress did not significantly affect behavioural intention. Compliance was positively associated with all three job stress dimensions: job uncertainty, technophobia and time pressure, while obedience was correlated with job uncertainty.

CONCLUSION

Social influence and human-machine trust are critical factors in nurses' intentions to adopt AI technology. The lack of significant effects from perceived stress suggests that nurses' personal resources mitigate potential stress associated with AI implementation. The study reveals the complex dynamics regarding different types of social influence, human-machine trust and job stress in the context of AI adoption in healthcare.

IMPACT

This research extends beyond conventional technology acceptance models by incorporating perspectives on organisational internal stressors and AI-related job stress. It offers insights into the coping mechanisms during the pre-adaption AI process in nursing, highlighting the need for nuanced management approaches. The findings emphasise the importance of considering technological and psychosocial factors in successful AI implementation in healthcare settings.

PATIENT OR PUBLIC CONTRIBUTION

No Patient or Public Contribution.

Intelligent Clinic Nurse Scheduling Considering Nurses Paired with Doctors and Preference of Nurses

The nurse scheduling problem (NSP) has been a crucial and challenging research issue for hospitals, especially considering the serious deterioration in nursing shortages in recent years owing to long working hours, considerable work pressure, and irregular lifestyle, which are important in the service industry. This study investigates the NSP that aims to maximize nurse satisfaction with the generated schedule subject to government laws, internal regulations of hospitals, doctor-nurse pairing rules, shift and day off preferences of nurses, etc. The computational experiment results show that our proposed hybrid metaheuristic outperforms other metaheuristics and manual scheduling in terms of both computation time and solution quality. The presented solution procedure is implemented in a real-world clinic, which is used as a case study. The developed scheduling technique reduced the time spent on scheduling by 93% and increased the satisfaction of the schedule by 21%, which further enhanced the operating efficiency and service quality.

Exploring nurses' experiences: Abandoning the profession and migrating for improved opportunities

AIM

This study explores nurses' experiences in migration for employment and professional abandonment in Barcelona (Spain).

METHODS

Employing a mixed-design approach comprising 1) a qualitative descriptive phenomenological study, followed by 2) a subsequent cross-sectional study, 20 and 225 nurses participated in each study, respectively. Qualitative data, gathered through 4 focus group discussions, underwent inductive thematic analysis, following the Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines, while quantitative data were descriptively analyzed.

FINDINGS

Three qualitative themes emerged: 1) Migration motives, such as improved job opportunities, permanent contracts, continuous training, and professional recognition; 2) Reasons for leaving or contemplating leaving the profession, including excessive workload, lack of recognition, limited development, and exhaustion; 3) Nurses' needs, encompassing more staffing, improved remuneration, permanent contracts, flexible schedules, greater autonomy, and career growth. The cross-sectional study revealed a 13.5 % professional abandonment rate at some point across all demographics and seniority levels. Migration trends varied by professional experience, with younger nurses seeking better conditions and opportunities elsewhere.

CONCLUSIONS

Multifactorial causes underlie job migration and professional abandonment, necessitating comprehensive interventions to improve nurses' working and professional conditions.

Uneven distribution of stressful working conditions among Japanese nurses: a secondary analysis of nurses with and without children

Supportive measures for employees raising children may have increased workloads on other nurses, causing psychological stress. This study aimed to clarify the differences in working conditions and psychological status among female Japanese nurses based on child-rearing attributes. We used data from 1,600 female nurses at 10 Japanese hospitals collected by the study of the Work Environment for Hospital Nurses in Japan conducted in 2016. The variables included work conditions (number of night shifts per month, daily overtime, number of paid holidays per year, and social support received), psychological status (sense of coherence, emotional exhaustion, and work engagement), and sociodemographic characteristics. An analysis of covariance was performed on the differences between the three groups (without children, with preschool-age children, and with children of other ages groups). The group without children had a relatively higher workload (p<0.01) and lower social support (p<0.01 and p<0.05). Additionally, they had higher emotional exhaustion and lower work engagement (p<0.01). This study confirmed the uneven distribution of work environment by work-life balance measures.

The Relationship Between Resilience and the Intention to Leave Among Staff Nurses at Governmental Hospitals in the Al-Baha Region of Saudi Arabia

Background Nurses with high intent to leave can cause substantial problems for healthcare organizations, such as staffing shortages and higher expenses due to hiring and onboarding new nurses. In light of the increasing demands placed on nurses in understaffed and overloaded healthcare systems, nurses frequently face various pressures and difficulties in their field of work, including high workloads, irregular hours, complicated patients, and infectious disease exposure; resilience is critical for handling stress and hardship at work. Nurses will thus retain their jobs for longer. This study aimed to determine the relationship between resilience and the intention to leave among staff nurses. Methods This study utilized a quantitative, cross-sectional correlation design. It comprised three Saudi Ministry of Health-affiliated facilities in the Al-Baha region (King Fahad Hospital, Prince Mashari Hospital, and Mikhwah General Hospital). The study sample comprised nurses employed in critical areas and inpatient and outpatient hospital departments using convenience sampling and inclusion and exclusion criteria. An online questionnaire involving three sections was given out. The first part collected sociodemographic data, the second part included the Connor-Davidson Resilience Scale 25 (CD-RISC-25), and the third included the Anticipated Turnover Scale (ATS). Results This study found a moderate degree of intention to leave and resilience. Most participants in the survey held a bachelor's degree (75.8%), and around 87.1% of the sample consisted of women. About half of the sample (57.2%) were married; 67.6% of the participants were not Saudi nationals; and regarding the number of children, the majority (53.8%) were childless. Overall, 318 nurses working in acute and outpatient departments and critical regions participated. According to the study, 73.3% of the participants reported a moderate intention to quit, whereas 50.9% had moderate resilience. Similarly, a significant negative relationship was found between nurses' intention to leave and resilience. Conclusions In the current study, resilience has a statistically significant negative relationship with the nurses' intention to leave. Hospital management must consider the amount of work and the excessive work schedule to reduce nurses' intentions to leave. One way to do this is by assigning tasks to employees, minimizing their workload through flexible work schedules and shorter duty hours, and fostering teamwork among coworkers by ensuring clear communication and cooperation. Interventions like orientation programs for new nurses, regular meetings, seminars, and training sessions can improve nurse resilience.

Predicting Nurse Turnover for Highly Imbalanced Data Using the Synthetic Minority Over-Sampling Technique and Machine Learning Algorithms

Predicting nurse turnover is a growing challenge within the healthcare sector, profoundly impacting healthcare quality and the nursing profession. This study employs the Synthetic Minority Over-sampling Technique (SMOTE) to address class imbalance issues in the 2018 National Sample Survey of Registered Nurses dataset and predict nurse turnover using machine learning algorithms. Four machine learning algorithms, namely logistic regression, random forests, decision tree, and extreme gradient boosting, were applied to the SMOTE-enhanced dataset. The data were split into 80% training and 20% validation sets. Eighteen carefully selected variables from the database served as predictive features, and the machine learning model identified age, working hours, electric health record/electronic medical record, individual income, and job type as important features concerning nurse turnover. The study includes a performance comparison based on accuracy, precision, recall (sensitivity), F1-score, and AUC. In summary, the results demonstrate that SMOTE-enhanced random forests exhibit the most robust predictive power in the classical approach (with all 18 predictive variables) and an optimized approach (utilizing eight key predictive variables). Extreme gradient boosting, decision tree, and logistic regression follow in performance. Notably, age emerges as the most influential factor in nurse turnover, with working hours, electric health record/electronic medical record usability, individual income, and region also playing significant roles. This research offers valuable insights for healthcare researchers and stakeholders, aiding in selecting suitable machine learning algorithms for nurse turnover prediction.