Predictive model for retaining middle medical personnel in the region based on data from a medico-sociological survey and behavioral monitoring.

<jats:p>The relevance of developing a predictive model for retaining middle medical personnel is due to the need to stabilize the workforce of regional budgetary healthcare organizations. Objective: development of a scientifically grounded retention model for middle medical personnel in the region based on integrating data from a medico-sociological survey and organizational behavior monitoring methods to identify resignation risks at an early stage and reduce staff turnover in budgetary healthcare organizations. Materials and methods. A comprehensive study was conducted using quantitative sociological analysis methods and behavioral analytics. The sample consisted of 1331 questionnaires from employees belonging to middle medical personnel working in regional-level healthcare organizations. The reliability of the instrument was confirmed by Cronbach’s alpha coefficient equal to 0.66. Additionally, a method for monitoring organizational behavior with keeping individual accounting documents was used. Data processing was carried out using methods of descriptive statistics, factor, correlation and regression analysis. Results. Analysis identified three risk clusters for resignation with different motivation profiles – clusters «Segment of Emotional Burnout», «Segment of Work Overload», «Segment of Systemic Dissatisfaction». A regression model predicting the probability of resignation was built with an overall classification accuracy of 89.0%. Implementation of a behavioral monitoring system in the risk group allowed reducing staff turnover by 25% over one month. Conclusions. The transition from analyzing actual turnover to assessing intent to resign allows identifying risks at an early stage and taking preventive measures. The effectiveness of behavioral monitoring was confirmed by reduced staff turnover in pilot units when working with causes, not consequences, of changes in employee mood. Results are applicable for integrating surveys into workforce resource planning and developing personnel development programs.</jats:p>