Dynamic forecasting model as a tool for projecting physician workforce capacity at the regional level.

<jats:p>Background. Persistent physician shortages in several Russian regions remain unresolved despite systematic organisational measures, suggesting structural causes beyond the reach of existing analytical tools. Aim: to develop a methodology for forecasting regional physician workforce availability based on a multilevel analysis of determinants (demographic trends, natural attrition, labor migration, and targeted specialty training), and to evaluate its analytical capabilities and practical applicability. Methods. Retrospective analysis of federal statistical monitoring data (Form No. 30) for 2018–2023; comparative analysis of three methodological approaches to physician workforce forecasting; testing of the PrognosMed-Kadr model on regional data (base year 2023, forecast horizon 2035). Results. The physician-to-population ratio in Vladimir Oblast has remained at 26–28 per 10,000 population for over 10 years – 27% below the national average. The current methodology (Order No. 974, 29.11.2019) is static: it relies on planned volumes of the territorial state health guarantees programme and does not forecast workforce supply. Under inertial conditions, the model projects a decrease from 27.9 to 19.9 per 10,000 by 2035, with a loss of 1,202 physicians (32.8%). Conclusions. The normative calculation reproduces the staffing deficit by omitting factors driving its long-term dynamics. Regions with persistent deficits require a dynamic forecasting model accounting for age-related attrition, labour migration, and the delayed effects of training programmes.</jats:p>