**1. Introduction**

As indicated by Reference [1], workplace health management is crucial for improvements in psychosocial working conditions and health. Promoting healthy work environments is a matter of ethics as well as business interest, since the most competitive companies are those with mentally and physically healthy workers due to policies supporting and protecting their health [2]. There is no doubt that work is part of the social dimension of health. The National Institute for Occupational Safety and Health (NIOSH) recognizes the importance of the well-being of workers, their families and communities through a series of factors linked to the employment relationship, such as wages, hours of work, workload and stress levels, interactions with co-workers and supervisors, access to paid leave, and health-promoting workplaces [3]. Therefore, Total Worker Health promotes an integral intervention in health where measures aimed at the protection of the health of employees are combined with others supportive of well-being. From this perspective, Total Worker Health integrates the individual dimension of health with the organizational dimension and the environment [4]. The Job Demand-Control (JDC) model [5] and the Job Demand-Control-Support (JDCS) model [6,7] constitute the most widely used theoretical approaches to understanding and interpreting the relationships

between the characteristics of health, work and well-being [8]. In fact, according to Reference [9], the occupational stress literature is dominated by these models.

Undoubtedly, every effort to improve the integral health of workers translates into greater working efficiency, as shown by the meta-analysis by Reference [10]. The JDCS model warns that the greatest risks to physical and mental health are manifested among workers who experience a high isolation-strain (iso-strain) job—that is, those that are subject to high job demands in a context of low control or decision latitude and low social support (iso-strain hypothesis). However, empirical evidence of this three-way interaction effect is still limited, primarily by the variety that exists in terms of the characteristics and conditions of labor between different jobs and occupations. Many authors, such as Reference [11], indicate that the samples used to test the JDCS model should be as homogeneous as possible, although they are heterogeneous with respect to the level of exposure of workers to the labor environment variables. This circumstance suggests the need to carry out research in occupations with similar characteristics. In addition, not all professions are subject to the same degree of strain, but some occupations tend to combine certain conditions that make workers in those jobs more vulnerable in terms of their physical and mental health [12].

Accountants play an important role in the financial market because they provide accurate information for decision making. Expert judgement and mental equilibrium are required for sound decisions. Despite this, an important gap of research exists in relation to pressures on accountants and its effects. Recently, a professional publication has shown how leading firms of accountants have taken steps to improve their health at work: Training the team leaders on issues of occupational health, creating spaces for healthier work practices, knowledge of such risks, sharing experiences among team members, etc. [13].

A profession as important as accounting is subject to heavy job demands that may affect one's perceived occupational health (POH). Occupational health is the result of the confluence of a number of stress factors and mechanism moderators in organizations. The effect of stressors on accountants has been studied in terms of different effects: For example, dysfunctional behaviour [14–16], personal well-being [17,18], labor satisfaction [17], performance [17] or turnover intention [17,19–22]. This research used theoretical frameworks compatible with those that explain the effect of the job demands in occupational health deterioration and, in particular, one of its major manifestations: Burnout [17,19–26]. Thus, the literature on the accountancy profession has identified several main categories of stressors, which highlight role overload, role conflict and role ambiguity [23]. The Job Demand-Job Control-Social Support (JDCS) model provides a holistic framework to investigate the direct effects and moderators of these stressors on occupational health. To our knowledge, this model has not has been tested on a group of accountants, so this paper covers an important gap in the literature.

This paper tries to cover gaps in the research on this topic in the European context. Thus, the objective of this study is to investigate the effects of job demands on the occupational health of the accountants, as well as the role of the moderator that internal resources (locus of control) and external resources (social support) have in occupational health. To meet the objectives of this research, Section 2 presents the theoretical framework and hypotheses that are derived from the framework. Section 3 describes the empirical study design and methodology used, which are based mainly on the analysis of neural networks. Section 4 displays the main results. The article ends with a discussion of the results (Section 5) and the main limitations of the study and future lines of research (Section 6).
