Predicting Quiet Quitting: Machine Learning Insights into Silent Resignations in Healthcare Industry

Abstract

Quite quitting, an increasingly widespread issue in healthcare, poses substantial problems to patient care and labor stability. This study presents a comprehensive review of quiet quitting in healthcare, addressing their ramifications and providi ng a machine learning model to predict and address this challenge. The technique adopted in this research involves decision trees, random forest, KNN, logistic regression, SVM (Support Vector Machine), Ensemble models, and neural networks. Based on organizations evolving in emerging markets, key results demonstrate a range of variables leading to quiet quitting, including, in this order of importance, fear of retribution, leadership styles, working conditions, meaningful jobs, level of bureaucracy, absence of career opportunities, and lack of trust. Surprisingly, salaries do not appear to be influential in quiet quitting, while paradoxically, years of experience are inversely correlated to silent quitting. Evaluating various machine learning models based on different metrics reveals notable performance differences. While Support Vector Machine (SVM) excels in precision and F1-score, Logistic Regression demonstrates high accuracy and performs well across multiple metrics, indicating its suitability for prediction.