Assessing Work Stressors in the Health Care Sector by Combining External Observation and Health Professionals’ Self-report in a Crosssectional Study Design

Objective: Health professionals are particularly affected by work stressors and various methods have already
been used to assess them. Linking health professionals’ self-report and external observations can provide a more
detailed assessment of stressors, since conclusions for interventions can be derived from their agreement. Since
there is a lack of studies in the health sector linking both data sources, the aim of this study is to identify the
convergence between health professionals’ self-reports and external observations.
Methods: Data were collected in general hospitals, nursing homes, psychiatric institutions and home-care
organizations in a cross-sectional study design. 110 health professionals were observed during one entire shift, by
one of eight trained external observers. Health professionals and observer separately filled out a questionnaire on
work stressors after the observation. For data analysis multiple regression models using bootstrap were calculated
considering possible observer effects.
Results: Convergent scores for 3 of 9 tested scales on ‘predictability’ of work, ‘social community’ and ‘social
relations’ (p>0.05) at work, were identified. However, health professionals rated their ‘quantitative’ (p=0.001),
‘sensorial’ (p=0.001) and ‘physical demands’ (p=0.001) significantly higher than the external observers did. On the
contrary, external observers perceived the ‘possibilities for development’ (p=0.007), ‘influence at work’ (p=0.032)
and ‘social support at work’ (p=0.002) as lower than did the health professionals. Results also indicate a significant
influence of different work settings (p<0.05) on the convergence of self-assessed and observed work stressors.
Conclusion: This study results reveal that results on work stressors can be influenced by the chosen method
for data collection, which should be considered when using one method only. Moreover, differences between the
settings indicate that results on work stressors from one health-care setting cannot be easily transferred to another.