The aim of the study was to determine the model for predicting the risk of dissatisfaction of general practitioners (GPs) with their work in primary health care centers based on the results of a questionnaire.
Materials and methods. A survey of 89 GPs was conducted using a questionnaire to determine job satisfaction. Each question of the questionnaire was evaluated by the respondents on the Likert total ratings on a 5-point scale. Job satisfaction was assessed at 4 and 5 points, dissatisfaction at 1-3 points.
The logistic regression analysis was used to build a model for risk prediction of GPs’ job dissatisfaction. To assess the degree of each factor influence in the logistic regression model, the odds ratio (OR) and its 95 % CI (confidence interval) were calculated. Receiver Operating Characteristic analysis was used to assess the quality of the logistic regression model. The analysis of the research results was carried out using the license package the statistical analysis package MedCalc v.19.4.1 (MedCalc Software Inc, Broekstraat, Belgium, 1993-2020).
The results show that 76.06 % of GPs are satisfied with their work. The vast majority are satisfied with the attitude of the management staff, fellow GPs, patients, opportunities of professional development and advanced training, work schedule, working conditions, and team spirit. At the same time, more than half of GPs (54.92 %) are dissatisfied with the comfort of business relations with other subspecialists, 36.61 % of doctors are dissatisfied with the salary, 40.84 % believe that the salary does not correspond to the efforts made, and 45.07 % think that additional financial compensation does not affect the quality of medical care provided. The method of stepwise inclusion/exclusion of variables was used to select a set of features for the multivariate logistic regression model. As a result of this method, 2 factors were selected: a sense of pride in work and working conditions. We established that the more a GP is proud of his chosen specialty and the better working conditions he has, the less dissatisfied he feels with his work. We determined the sensitivity of the model as 76.5 % (95 % CI: 50.1 % – 93.2 %), the specificity as 85.2 % (95 % CI: 72.9 % – 93.4 %), and the diagnostic accuracy as 83.1 % (CI: 72.3 % – 91 %).
Conclusions. In the personnel management system of health care institutions, it is necessary to consider many different factors of GPs’ satisfaction with their work, including salaries, relationships with colleagues, patients, the opportunities of professional development, the attitude of management staff, a sense of respect, and working conditions. Forecasting the risk of GPs’ dissatisfaction with their work must be carried out taking into account the presence of doctors’ subjective sense of pride in their work and working conditions.