Prevention by occupational physicians: a study from policy to practice

Work-related health complaints cause significant personal suffering and impose high costs on employers and society. To protect workers, the revision of the Dutch Working Conditions Act (Arbowet) in 2017 has strengthened the preventive role of occupational physicians (OPs). Yet in practice, many OPs spend little time on prevention, focusing mainly on sickness absence management. This dissertation therefore pursued two objectives: (1) to understand why preventive tasks are implemented so little in practice, and (2) to design and evaluate an intervention to increase OPs' engagement in prevention. Part 1: Reasons for limited implementation of prevention Research into behavioural determinants showed that OPs generally have a positive attitude towards preventive tasks. Nevertheless, their self-efficacy expectations regarding the execution of preventive tasks are relatively low and the perceived support from the environment varies by stakeholder. Both support from employers and higher self-efficacy expectations are associated with spending more time on preventive activities, while no associations were found for attitude (Chapter 2). Chapter 3 focuses on how OPs’ time allocated to preventive tasks is associated with their work experience. The results showed that more time spent on preventive tasks, relative to other tasks, was associated with a greater sense of variety in the work and a higher feeling that the work sufficiently utilizes skills and competencies. No significant associations were found for work pace and workload, autonomy, or work enjoyment. Self-employed workers (SEW) in the Netherlands technically have access to occupational health care but must arrange and pay for it themselves. In the study described in Chapter 4, only about 16% of surveyed SEW had ever consulted an OP preventively. The main reason for not doing so was a perceived lack of necessity. Lower perceived risk of work-related complaints and disability was associated with lower use of preventive care. A systematic review, described in Chapter 5, was conducted on the effectiveness of preventive interventions, carried out by occupational health professionals, on work-related stress complaints. Nine studies met the inclusion criteria, although the types of interventions, target groups, occupational health professionals involved and effects varied greatly. Preventive interventions appear to be especially effective when they are offered more intensively or frequently, although the evidence is limited and the outcomes are mixed. Part 2: Development and evaluation of an intervention To support OPs in their preventive role, we designed an intervention for OPs based on information from literature and input from stakeholders (Chapter 6). The intervention developed consisted of three peer group meetings, in which OPs created a personal action plan to address barriers in practice and increase their engagement in preventive tasks. The intervention for OPs was evaluated in the IM-PROmPt study (Implementation of Preventive Tasks by Occupational Physicians), a cluster-randomized controlled trial conducted between May 2023 and September 2024 with 41 peer groups. The process evaluation described in Chapter 7 was based on a questionnaire completed by 98 OPs and 17 individual interviews. The intervention was found to be well adopted by participating groups. The majority of OPs formulated personal goals related to prevention and started working on these in practice. OPs indicated that participation in the intervention led to good discussions with colleagues and greater awareness of (the importance of) preventive tasks. The effect evaluation, involving 227 OPs, found no impact on the overall percentage of working time spent on prevention: both groups increased from 14% at baseline to 17% at twelve months, possibly reflecting broader societal attention to prevention. However, the intervention did increase time spent on preventive advice during sickness absence management: OPs in the intervention group spent five more hours per month on this task than controls (Chapter 8).