
This paper explores the architectural novelty of artificial intelligence (AI) and its implications for occupational health as both a field of practice and a domain of organisational design. Unlike previous general-purpose technologies such as databases, AI operates through stochastic modelling of normative patterns, generating predictions rather than retrieving stored information. This fundamental difference makes AI simultaneously powerful and limited: it can codify norms but struggles with specificity, nuance, and contextual variation central to occupational health decision-making. Drawing on management science, systems theory, and social theory, the paper contrasts two key organisational questions: how to use AI within existing structures, and how to reorganise occupational health given that AI exists. It argues that the latter question, aligned with Stafford Beer’s systems thinking, is essential for responsible and effective adoption. AI’s greatest potential lies not in automation, but in enhancing organisational perception—of human relations, meaning in research, social and spatial dynamics—functioning as a new kind of scientific instrument that amplifies reflexivity and end-to-end “flow.” The paper concludes that sustainable integration of AI in occupational health demands socio-technical design, responsible innovation frameworks, and a focus on organisational learning to maintain healthy, adaptive workplaces in an era of technological transformation.
Artificial intelligence is fundamentally reshaping occupational medicine through applications in automation, smart monitoring, extended reality training, algorithmic management, and digital work arrangements. The EASOM Summer School 2025 examined how AI technologies enhance occupational health systems by enabling real-time risk identification, predictive surveillance, and targeted interventions to prevent work-related accidents and diseases. Practical applications include smart personal protective equipment, wearable sensors, augmented reality training systems, and predictive diagnostics exemplified by noise-related hearing loss risk assessment. However, significant challenges accompany these opportunities: data quality concerns, new technological hazards, psychosocial risks including surveillance-related stress and work intensification, algorithmic bias, and accountability gaps. Implementation requires comprehensive governance frameworks emphasizing early worker involvement, privacy-by-design approaches, phased pilot programs, and human-centered practices that complement rather than replace traditional occupational safety measures. Professional responsibility and education demand new competencies in digital literacy, algorithmic critical thinking, and applied ethics while maintaining essential clinical judgment and empathy. AI offers unprecedented potential for protecting workers' health and safety when responsibly implemented within transparent frameworks, but poor management risks deepening inequalities and eroding established protections. Successful integration depends on multidisciplinary research, international cooperation, stakeholder engagement, and a commitment to human-centered occupational medicine that prioritizes worker dignity and wellbeing.
Noise in commercial aviation is a significant stressor in the work environment and can adversely affect health, safety, and the performance of flight and cabin crew duties. The aim of this paper was to compare noise levels in the passenger cabins of older and newer aircraft generations and to analyse the subjective perception of noise exposure and related health complaints among crew members. The experimental part of this article involved measuring A-weighted sound levels in the passenger cabins of Dash 8–Q400 and Airbus A220–300 aircraft during five flight phases. Measurements were conducted as short-term phase assessments, with results presented as minimum and maximum noise level values. A survey was also conducted among cabin crew (n = 55) and flight crew (n = 24) to assess noise exposure, subjective noise level assessment, frequency of health complaints, and use of hearing protection measures. The measurement results indicate higher noise levels and more pronounced peak acoustic loads during key flight phases in the older-generation turboprop aircraft (Dash 8–Q400), especially during take-off and landing, while the newer-generation turbofan aircraft (Airbus A220–300) exhibited lower and more uniform noise levels in the passenger cabin. Survey results confirm greater exposure and more frequent reported disturbances among cabin crew compared to flight crew, as well as the perception that newer-generation aircraft are quieter. In addition, structural–material differences between the examined aircraft types, particularly regarding fuselage configuration and material selection, may represent an additional factor influencing the observed cabin noise characteristics. The results highlight the need for continued monitoring of noise exposure in aviation and emphasise the importance of combining technical, organizational, and personal protective measures to reduce the potential negative effects of noise on crew health.
Artificial Intelligence (AI) is reshaping occupational health and safety (OHS), shifting prevention models, risk assessment, and organisational processes. While traditional safety systems rely on reactive approaches and post‑event analyses, AI enables predictive, adaptive, and data‑driven prevention based on continuous monitoring and machine learning. The integration of sensor technologies, computer vision, and intelligent personal protective equipment supports early detection of biomechanical risks, fatigue, unsafe behaviours, and environmental hazards. Furthermore, AI enhances the structuring and linkage of occupational electronic health records, integrating biometric, behavioural, and environmental information while ensuring compliance with privacy‑by‑design principles through advanced Privacy Enhancing Technologies. A human‑centred digital ecosystem is proposed, founded on three pillars - technical‑predictive, organisational‑decision‑making, and ethical‑social - to ensure transparency, fairness, and sustainability in AI adoption. The role of occupational physicians is reassessed considering emerging digital competencies and ethical requirements, emphasising that AI should augment, rather than replace, professional expertise. A successful digital transition requires continuous dialogue among stakeholders, updated ethical frameworks, and robust interdisciplinary cooperation to place human health, dignity, and rights at the core of technological innovation.
Musculoskeletal disorders (MSDs) remain a leading cause of occupational morbidity in Europe, particularly among workers performing physically demanding tasks. Despite the existence of regulatory frameworks and periodic health assessments, the burden of MSDs remains high in Croatia. This study aimed to assess the health risk levels related to physical workload in Croatian workers diagnosed with MSDs, focusing on those engaged in load handling and repetitive movements. The study included 250 employed individuals (62.8% male, 37.2% female) undergoing routine occupational health evaluations, all deemed fit for work. Participants completed a detailed questionnaire on physical workload, which was used to assess risk levels using the Key Indicator Method (KIM). Diagnoses were classified per ICD-10 codes and grouped by anatomical and etiopathogenetic characteristics. The study was conducted between October 2021 and October 2022 in occupational and sports medicine clinics. Most participants were exposed to elevated or very high health risks due to physical workload. Very high risk was present in 42.5% of those involved in lifting and carrying, while 49.5% of those performing repetitive movements were at elevated risk. Gender differences were significant for repetitive movements (P = 0.030), with women experiencing greater exposure. Back disorders were the most common MSDs (80.0%), followed by overuse syndromes (12.8%) and other MSDs (7.2%). The majority (90.4%) of MSDs had been diagnosed more than a year prior, and 69.0% of these workers had taken sick leave due to symptom worsening. The study highlights significant health risks in physically demanding occupations despite mandatory medical surveillance. Gender disparities and high chronic MSD prevalence underline the need for a paradigm shift toward proactive, functionally informed occupational health strategies. These should include individualized assessments, targeted rehabilitation, and workplace interventions adapted to enterprise capacity and worker demographics.