Use of machine learning to predict creativity among nurses: a multidisciplinary approach

Area of Science:

• Nursing Science
• Artificial Intelligence in Healthcare
• Organizational Psychology

Background:

• Nurses' creativity is crucial for patient care but understudied.
• Factors influencing nurse creativity require further investigation.
• Developing predictive models for nurse creativity is essential.

Purpose of the Study:

• To explore factors influencing nurses' creativity.
• To develop a machine learning-based decision support system (DSS) for predicting nurse creativity.
• To enhance understanding of creativity in the nursing profession.

Main Methods:

• A mixed-methods approach combining machine learning with a descriptive, cross-sectional, correlational design.
• Survey data collected from 191 registered nurses across eight hospitals in Jordan.
• Naïve Bayes and multiple linear regression models were utilized for analysis.

Main Results:

• Staff nurses reported high creativity levels (M=44.95).
• Machine learning models demonstrated high prediction accuracy for creativity predictors (e.g., psychological safety 99%, gender 98%).
• Humble leadership, psychological safety, experience, quality initiatives, and education level were key predictors of creativity.

Conclusions:

• Nurse managers should foster humble leadership and psychological safety to boost creativity.
• The developed DSS can assist in evaluating nurse creativity for better staff management.
• Machine learning offers a precise tool for predicting and understanding nurse creativity.