Construction and validation of a presenteeism prediction model for ICU nurses in China.

BACKGROUND

Presenteeism, also known as impaired health productivity, refers to the condition of impaired productivity of an individual due to physiological or mental health problems. ICU, as a place of intensive care for patients with acute and critical illnesses, nurses have long faced the nature of work with high loads, high pressures, and high intensities, which makes them a high prevalence group of presenteeism. Presenteeism not only affects the physical and mental health and work wellbeing of nurses but also reduces the quality of nursing services and affects the life safety of patients, such as increasing the risk of falls during hospitalization, increasing the risk of medication errors, and prolonging the hospitalization time of patients. Therefore, early identification and targeted interventions are crucial to reduce presenteeism among ICU nurses.

OBJECTIVE

This study aimed to construct and validate a predictive model for presenteeism among ICU nurses.

DESIGN

A cross-sectional study.

METHODS

1,225 ICU nurses were convened from January to April 2023 from 25 tertiary and secondary hospitals in Sichuan Province, China. ICU nurses were randomly divided into a development set ( = 859) and a validation set ( = 366) according to a 7:3 ratio. Univariate and multifactorial logistic regression analyses were used to determine the influencing factors for presenteeism, and R software was used to construct a column-line graph prediction model. The differentiation and calibration of the predictive model were evaluated by the area under the curve of subjects' work characteristics (ROC) and the Hosmer-Leme-show test, and the clinical decision curve evaluated the clinical validity of the predictive model.

RESULTS

The presenteeism rate of ICU nurses in the development set was 76.8%. Multifactorial logistic regression analysis showed that independent factors affecting ICU nurses' presenteeism included income per month, physical health status, job satisfaction, perceived work stress, perceived social support, transformational leadership, and occupational coping self-efficacy. In the development set and validation set, the area under the ROC curve was 0.821 and 0.786, respectively; the sensitivity and specificity were 80.6, 69.8 and 80.9%, 65.1%, respectively; the Hosmer-Lemeshow goodness-of-fit was  = 8.076 ( = 0.426) and  = 5.134 ( = 0.743), respectively, and the model had relatively good discrimination and consistency. The clinical decision curve showed that the model had good clinical validity.

CONCLUSION

The predictive model of presenteeism risk for ICU nurses constructed in this study has good predictive ability. The model can effectively identify ICU nurses with high presenteeism and provide a reference basis for developing targeted interventions to reduce presenteeism among ICU nurses.