Non-standard employment and health : exploring the pathways to health inequality

In today's increasingly fragmented labour market, there is a growing reliance on non-standard forms of employment (NSE), often characterized by instability, low income, and limited access to rights and protections. While the term NSE captures a contrast from the Standard Employment Relationship (typically associated with stable, full-time, permanent contracts), the concept of precarious employment (PE) more accurately reflects disadvantageous conditions experienced by workers. The expansion of PE has raised concerns about the implications for workers' health and well-being. While previous research has established links between PE and adverse health outcomes, less is known about the mechanisms and the contextual factors that shape and modify the effects. This thesis addresses these gaps through a mixed-methods approach, drawing on four studies that examine how PE impacts health and well-being across different settings and study populations. Given the overlapping but distinct use of terminology across the studies, a brief clarification is warranted. This thesis focuses on the health implications of disadvantageous employment conditions. While the term NSE is used when referring to specific study populations in studies Il and III-reflecting the terminology applied in those original publications-the thesis adopts PE as the primary analytical concept, as it more directly captures the conditions of insecurity and vulnerability at the core of the investigation. Study I used cross-sectional survey data to investigate the association between PE and psychosocial work environment hazards (experiences of violence, sexual harassment, bullying, discrimination, high demands, and low control). Respondents (n=401) were recruited through web-based respondent driven sampling in Stockholm County (2016-2017). A precarity score was calculated using the Swedish version of the Employment Precariousness Scale (EPRES-Se) and participants were categorized according to level of PE (low/high). Prevalence ratios were calculated to analyze the relationship with psychosocial hazards, adjusting for gender, age, education, and country of birth. Results showed that high PE (compared to low PE) was associated with bullying (PR 1.07, 95% CI: 1.01- 1.13), discrimination (PR 1.52, 95% CI: 1.00-2.32), low control (PR 1.59, 95% CI: 1.30- 1.96), and passive work (PR 1.60, 95% CI: 1.23-2.08). Psychosocial hazards were more prevalent among women. Study Il is a qualitative multi-case study exploring workers' experiences of NSE and implications for health and well-being across six countries with different welfare state regimes: Belgium, Canada, Chile, Spain, Sweden, and the United States. The study draws on 250 in-depth interviews (Jan-Sept 2021) from a purposive sample of NSE workers. Interviews were analyzed thematically in two phases, first within countries and then across. Findings revealed that workers across all contexts faced multiple insecurities (employment, income, schedule) and imbalanced power relations that negatively impacted health, particularly for those with intersecting social disadvantages. Welfare state support helped buffer the most severe insecurities, but similar patterns appeared in all study countries. Study III draws on the same interview material and explores how NSE workers experience labour regulations and social protection policies, their coping strategies, and ideas for supportive reforms. Despite differences in formal protections, participants across countries described similar challenges related to policy gaps and access barriers. Many relied on personal savings or family support and used risk-averse coping strategies (e.g., work intensification, avoiding conflict), which often reinforced vulnerability. Participants struggled to imagine improvements and expressed low expectations for change. Responses reflected a desire for basic security (e.g., stable income, housing), highlighting the limitations of employment-based protections and the need for more comprehensive and inclusive social protections that extend beyond standard employment models. Study IV is a longitudinal register-based cohort study investigating if the association between PE and mental health is moderated by household disposable income and family type. The study population (n=2,509,229) included all individuals between 27-65 years who were employed in 2016. PE was assessed based on the Swedish Register-based Operationalization of Precarious Employment (SWE-ROPE). Data on household variables (2016), and mental health (2017-2019) were extracted from national registers. The effect of PE on mental health was estimated using Cox regression models, and by adding two- way interaction terms to the main effects model. All analyses were stratified on sex. Findings showed that workers in PE (compared to SE) had a higher risk of mental ill-health (HR 1.21 CI95% 1.18-1.23), consistently across household income levels and family types. High household income was protective in general, but less so for the PE group due to synergistic interaction effects for both men (HR 1.22 CI95% 1.04-1.43) and women (HR 1.25 CI95% 1.13-1.38). Compared with the reference category (couple without children), all other family types (couple with children, single, single parent) amplified the negative effects of PE on mental health among women in PE, especially single mothers (HR 1.27 CI95% 1.14-1.42). Findings highlight that higher household income does not fully buffer the risk of mental ill-health for workers in PE, and the negative impact of PE appears stronger for women, especially single mothers. Together, the studies shed light on how PE impacts health and well-being through intersecting material, psychosocial, and contextual mechanisms. Findings show how PE generates economic and social vulnerability, which in turn triggers coping strategies that may further deplete resources and reinforce disadvantage. Across diverse national contexts and populations, the findings reveal a consistent pattern of adverse health-related experiences, disproportionately affecting women and workers with limited individual resources or family support. While policy context made a difference in terms of severity, the convergence of experiences across countries suggests that broader labour market transformations-characterized by a shift of risk from employers to individuals-have created shared structural conditions that transcend welfare regime type. The thesis contributes new knowledge by providing insight into the pathways through which PE affects health, and by highlighting the role of contextual factors and social inequalities in shaping these outcomes. By deepening our understanding of how labour market fragmentation contributes to unequal health risks, the findings support future research and policy efforts aimed at reducing health inequalities in a rapidly evolving world of work. List of scientific papers I. Kvart, S., Jonsson, J., Bodin, T., Håkansta, C., Kreshpaj, B., Orellana, C., et al. (2021). Precarious employment and psychosocial hazards: A cross-sectional study in Stockholm County. International Journal of Environmental Research and Public Health. 18, 11218. https://doi.org/10.3390/ijerph182111218 II. Bosmans, K., Vignola, E.F., Álvarez-López, V., Julià, M., Ahonen, E.Q., Bolíbar, M., et al. (2023). Experiences of insecurity among non-standard workers across different welfare states: A qualitative cross-country study. Social Science & Medicine. 327, 115970. https://doi.org/10.1016/j.socscimed.2023.115970 III. Kvart, S., Cuervo, I., Gunn, V., Lewchuk, W., Bosmans, K., Davis, L., et al. (2025). Labour and social protection gaps impacting the health and well-being of workers in non-standard employment: An international comparative study. PLOS ONE. 20, e0320248. https://doi.org/10.1371/journal.pone.0320248 IV. Kvart, S, Mangot-Sala, L, Aronsson, A, Badarin, K, Bosmans, K, Gunn, V, et al. Precarious Employment Increases Mental Health Risks Regardless of Family Type and Household Income: A Swedish Register Study. [Manuscript]

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.