Stress and sickness absence: Prediction and causal mechanisms of mental sickness absence

Maria Frederika Agnes van Hoffen

    Research output: PhD ThesisPhD-Thesis - Research and graduation internal

    49 Downloads (Pure)

    Abstract

    Summary   Chapter 1 introduces the context and importance of our topic. Mental disorders are the leading cause of sickness absence and disability pensions in European countries. We need more knowledge of predictors of mental sickness absence to identify those workers at high risk of mental long-term sickness absence (LTSA) before they report sick. In this thesis, prediction models for mental LTSA will be developed and validated using variables that are commonly addressed in occupational health surveys. This thesis also investigates causal mechanisms of mental long-term sickness absence. Chapter 2 describes the predictive performance of the Maslach Burnout Inventory (MBI-GS) and Utrecht Work Engagement Scale (UWES) for identifying workers at increased risk of mental LTSA. It was concluded that the MBI—GS and UWES predicted future mental LTSA in non-sicklisted employees, but discrimination was not practically useful. Chapter 3 investigates the ability of mental health symptoms to identify workers at risk of mental LTSA. Mental health symptoms were measured at baseline with the 4-DSQ (distress and depressed mood) and MBI–GS (fatigue). The symptom scores were analyzed against incident mental LTSA. Distress fairly discriminated between workers with and without mental LTSA, whereas the discriminative ability of both depressed mood and fatigue was poor. It was concluded that the 4-DSQ distress scale may be a promising tool to screen working populations for mental LTSA. Chapter 4 compares the discrimination by the 16-item 4-DSQ distress scale with discrimination by a distress screener with items on worrying, listlessness, and feeling tense, derived from the full 16-item distress scale. Discrimination between non-sicklisted workers with and without mental LTSA was found to be similar for the 16-item distress scale and the three-item screener. Thus, it is more convenient for healthcare providers to use the three key questions of the 16-item 4-DSQ distress scale to identify non-sicklisted employees at risk of future mental SA. Chapter 5 investigates psychosocial job demands and job resources for their predictions of mental LTSA among Norwegian nurses. Job demands and job resources were measured at baseline and linked to mental LTSA during 2-year follow-up. Harassment and social support were associated with mental LTSA, but the Cox regression model did not discriminate between nurses with and without mental LTSA. Chapter 6 describes predictions of mental LTSA by psychosocial job demands and job resources. Only performance feedback was associated with mental LTSA. A prediction model including psychosocial work characteristics poorly discriminated between workers with and without mental LTSA. Chapter 7 investigates the mediational effect of distress, burnout, work satisfaction, engagement, and work ability for the relation between psychosocial working conditions and mental health-related LTSA. Role clarity, cognitive demands, emotional demands, work variety, learning opportunities, social support from colleagues, and social support from family and friends were related to mental LTSA. Emotional demands had the strongest direct effect on mental LTSA. The relation between psychosocial working conditions and mental LTSA was mediated by distress, work satisfaction, and work ability. Distress was the most important mediator between psychosocial working conditions and mental LTSA. Chapter 8 presents the development of a multivariable prediction model for mental LTSA using distress together with other occupational health survey variables. An 11-predictor logistic regression model discriminated between workers with and without mental LTSA during 1-year follow-up. A 3-node decision tree equally well discriminated between participants with and without mental LTSA at follow-up. Chapter 9 describes the external validation of the logistic regression and decision tree prediction models. Both models fairly discriminated between participants with and without mental LTSA during follow-up. We recommend to use the decision tree based on distress, gender and work satisfaction in preventive consultations following occupational health surveys.
    Original languageEnglish
    QualificationDr.
    Awarding Institution
    • Vrije Universiteit Amsterdam
    Supervisors/Advisors
    • Twisk, J.W.R., Supervisor
    • Roelen, C.A.M., Co-supervisor
    • Norder, Giny, Co-supervisor, External person
    Award date26 Nov 2021
    Place of Publications.l.
    Publisher
    Print ISBNs9789083189338
    Publication statusPublished - 26 Nov 2021

    Keywords

    • Stress
    • Sickness absence
    • Prediction model
    • Psychosocial working conditions
    • Occupational Health
    • Sick leave
    • Prevention
    • Mental health

    Fingerprint

    Dive into the research topics of 'Stress and sickness absence: Prediction and causal mechanisms of mental sickness absence'. Together they form a unique fingerprint.

    Cite this