Akademik Veri Yönetim Sistemi
BMC Health Services Research, cilt.26, sa.1, 2026 (SCI-Expanded, Scopus)
Background: Physician migration, as a salient dimension of brain drain, undermines service delivery capacity in source countries and threatens the sustainability of healthcare systems. Although Turkey has witnessed a marked increase in physician emigration in recent years, empirical research specifically examining physicians’ international migration remains limited. Addressing this gap, the present study integrates demographic characteristics and work-related conditions to provide a parsimonious account of physicians’ migration intentions in Turkey and to identify the predictors that most clearly distinguish physicians who intend to migrate from those who do not. Methods: Data were collected from 428 physicians working across different regions of Turkey (06–08/2025) via a nationwide online survey implemented using convenience and snowball sampling methods. Migration intention was defined as a binary outcome, and a Lasso logistic regression model was estimated using a broad set of demographic and occupational/work-life predictors. Results: More than half of the respondents reported an intention to migrate abroad. Male gender, foreign language proficiency, region of residence, and professional title were retained by the penalized model as informative predictors associated with migration intention. Among the work-related variables, perceived patient load, long working hours, workplace violence, concerns about the future of the profession, and dissatisfaction with the legal/regulatory framework were also retained and showed positive associations with migration intention. Conclusions: By jointly modeling demographic attributes and work-life problems within a unified analytical framework and applying a variable-selection approach, this study highlights the most informative predictors associated with migration intention among multiple correlated variables and contributes to the empirical literature on physician migration in Turkey. The findings should be interpreted as indicative patterns within the surveyed sample rather than nationally representative estimates.