Turnover in the nursing profession can be categorized into two types: organizational turnover, in which nurses change employers, and professional turnover, in which nurses leave the profession entirely. The latter is commonly referred to as "leaving the profession" or "nursing attrition" (Halter et al., 2017). One of the global challenges in nursing is the high rate of academic dropout among nursing students during their studies, and professional attrition among nurses after graduation—both of which affect workforce stability, the financial sustainability of health systems, and the quality of care (Lantz and Fagefors, 2025). The global rate of (Ren et al., 2024) nursing attrition has been reported to range from 8 % to 36.6 %, with a rate of 19 % in Asia. In Iran, 49.6 % of nurses have expressed an intention to leave the profession (Maleki et al., 2023). Annually, approximately 15–20 % of nursing students worldwide withdraw from undergraduate nursing programs (Coakley, 1999). In a study conducted in the United Kingdom, among 534 nursing and midwifery students, 55 had left and returned to their studies, and 281 had considered leaving or discontinuing their education (Thompson et al., 2025). In a study from Saudi Arabia, more than half of the surveyed nursing students reported an intention to leave the profession in the future (Kandil et al., 2021). In a study conducted in nursing schools in Tehran, Iran, only 18 % of students had a positive view of the nursing profession, while 69 % expressed a desire to leave it, 63.6 % planned to change their major, and 51.64 % were inclined to drop out (Joolaee et al., 2006). This situation has led to a shortage of nurses and decreased professional motivation among remaining students (Coakley, 1999), while the growing population continues to increase the demand for nursing care. According to the WHO, a global shortage of at least 12.9 million nurses is projected by 2035, and the current shortage is estimated to be around 7.2 million (McCarthy et al., 2020).

Nursing attrition, both professional and academic, is influenced by a range of factors that impact healthcare quality and workforce stability. Professionally, issues such as staff shortages, heavy workloads, burnout, limited opportunities for growth, poor work environments, salary dissatisfaction, and lack of support contribute to nurse turnover—particularly among younger and female nurses (Lantz and Fagefors, 2025, Mohamed et al., 2024). Additional contributing factors include workplace stress, unfavorable employment conditions, and limited resilience, especially among novice nurses (Lyu et al., 2024). Academic dropout among nursing students is similarly complex, leading to economic losses and a reduced number of professional entrants. Key reasons include a mismatch with the profession, low confidence, unmet expectations, negative clinical experiences, inadequate support, and abusive supervision (Canzan et al., 2022, Hong et al., 2024). Despite numerous studies on nurses’ intentions to leave the profession, a significant gap remains in the literature: to date, no study has simultaneously examined the predictors of both nursing student dropout and professional attrition within a coherent theoretical framework (Lantz and Fagefors, 2025, Bolt et al., 2022). Most previous research has focused on either academic or professional attrition in isolation, lacking the theoretical integration needed for a comprehensive understanding of this multifaceted issue. However, a simultaneous analysis of the academic and professional trajectories of nurses is essential for developing effective strategies to retain the nursing workforce (Halter et al., 2017, Bolt et al., 2022).

In response to this gap, the present study adopts the IPOD theoretical framework (Input–Process–Output–Development), which provides a comprehensive structure for evaluating educational quality and professional development (Luo et al., 2018). This framework, with its emphasis on academic and career growth during and after formal education, is particularly well-suited for nursing students and graduates. By encompassing the sequential stages of training and employment, it enables a coherent and targeted analysis of academic and professional pathways and supports the identification of factors influencing academic success, job retention, and career advancement. The IPOD model, through its four key dimensions—Input (e.g., personal characteristics and academic background), Process (educational experiences), Output (academic outcomes), and Development (career progression)—offers a structured and multi-layered platform for analyzing nurses’ professional trajectories (Luo et al., 2018, Jiabin et al., 2024). Given the aim of this study—to identify predictors of nursing student dropout prior to graduation and professional attrition after clinical entry—this model provides a relevant and practical theoretical foundation for examining the multifaceted nature of attrition in nursing. Its added value lies in its capacity to assess both academic performance and long-term professional development, making it particularly appropriate for investigating the causes of attrition across various stages of the nursing career, especially within the context of Iran’s healthcare system. Utilizing such a framework can yield practical and policy-relevant implications at both the educational and managerial levels of the health system (Luo et al., 2018). On the other hand, given the large volume and complexity of data related to attrition, recent studies have employed machine learning algorithms for data analysis (Ganthi et al., 2022, Iparraguirre-Villanueva et al, Mozaffari et al., 2023, Vaarma and Li, 2024, Huo et al., 2023). These methods are particularly effective in uncovering hidden patterns within large and complex datasets that may not be detectable using traditional statistical approaches (Arian et al., 2025). By simultaneously analyzing individual, educational, and environmental factors, machine learning models can accurately predict risk factors associated with leaving the profession and support managers in making faster, more targeted decisions to improve staff retention. Moreover, these models are both scalable and adaptable, enabling the development of personalized interventions for at-risk personnel (Alqahtani et al., 2024, Raza et al., 2022).

This study aims to identify key predictors of academic dropout and professional attrition among undergraduate nursing students in Iran, using the IPOD model and machine learning algorithms. Within this framework, the academic and professional trajectories of nurses are analyzed, and the roles of demographic, educational, environmental, and organizational variables in attrition are examined.