Background
The assessment and allocation of nursing manpower, as well as performance evaluation, heavily rely on nursing workload, which is also strongly associated with patient safety outcomes. Nevertheless, the majority of previous studies have utilized cross-sectional data collection methods, thereby impeding the precision of workload prediction. Static workload models fail to incorporate longitudinal changes in influential factors, potentially resulting in delayed or erroneous nursing management decisions and ultimately causing imbalances in nurses' workload.
Aim
To employ machine learning algorithms in order to facilitate the dynamic prediction of nursing workload based on patient characteristics.
Methods
This study was conducted as a prospective cohort quantitative study between March 2019 and August 2021 in two general hospitals located in China. Data pertaining to the characteristics of 133 patients over the course of 1339 hospital days, as well as nursing hours, were collected. A longitudinal investigation into nursing workload was carried out, employing multiple linear regression to identify measurable factors that significantly impact nursing workload. Additionally, machine learning methods were employed to dynamically predict the nursing time required for patients.
Results
Mean direct nursing workload varied greatly across hospitalization. Number of complications during hospitalization, age, income, SCS score, and ADL score were all significant factors contributing to increased care needs. Improving predictive performance through machine learning, with random forests performing the best, RMSE (989.67), R2 (0.76), and MSE (979451.24).
Conclusions
The variation in nursing workload during hospitalization is primarily influenced by patient self-care capacity, complications, and comorbidities. Random Forest, a machine learning algorithm, is capable of effectively handling a wide range of features such as patient characteristics, complications, comorbidities, and other factors. It has demonstrated exceptional performance in predicting workload.
Implications for Nursing Management:
This study introduces a quantitative model that evaluates nursing workload throughout the duration of hospitalization. The utilization of this model allows nursing managers to holistically consider multiple factors that impact workload, resulting in enhanced comprehension and interpretation of workload variations. By employing a random forest algorithm for workload prediction, nursing managers can anticipate and estimate workload in a proactive and precise manner, thereby facilitating more efficient planning of human resources.