Article Text

Download PDFPDF
Education and health and well-being: direct and indirect effects with multiple mediators and interactions with multiple imputed data in Stata
  1. Mashhood Ahmed Sheikh,
  2. Birgit Abelsen,
  3. Jan Abel Olsen
  1. Health Services Research Unit, Department of Community Medicine, University of Tromsø, Tromso, Norway
  1. Correspondence to Mr Mashhood Ahmed Sheikh, Department of Community Medicine, University of Tromsø, 9037 Tromsø, Norway; senor_massao{at}


Background Previous methods for assessing mediation assume no multiplicative interactions. The inverse odds weighting (IOW) approach has been presented as a method that can be used even when interactions exist. The substantive aim of this study was to assess the indirect effect of education on health and well-being via four indicators of adult socioeconomic status (SES): income, management position, occupational hierarchy position and subjective social status.

Methods 8516 men and women from the Tromsø Study (Norway) were followed for 17 years. Education was measured at age 25–74 years, while SES and health and well-being were measured at age 42–91 years. Natural direct and indirect effects (NIE) were estimated using weighted Poisson regression models with IOW. Stata code is provided that makes it easy to assess mediation in any multiple imputed dataset with multiple mediators and interactions.

Results Low education was associated with lower SES. Consequently, low SES was associated with being unhealthy and having a low level of well-being. The effect (NIE) of education on health and well-being is mediated by income, management position, occupational hierarchy position and subjective social status.

Conclusion This study contributes to the literature on mediation analysis, as well as the literature on the importance of education for health-related quality of life and subjective well-being. The influence of education on health and well-being had different pathways in this Norwegian sample.

  • education
  • subjective social status
  • income
  • occupational hierarchy position
  • social epidemiology
  • health-related quality of life
  • subjective well-being

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Handling editor James Dunn

  • Contributors MAS developed the theory and analytical strategy, performed statistical analysis and data interpretation and wrote the manuscript. BA and JAO conceived the substantive research questions and contributed in revising the manuscript.

  • Funding This research was funded by UiT The Arctic University of Norway.

  • Competing interests None declared.

  • Ethics approval The Tromsø Study has been approved by the Regional Committee for Medical and Health Research Ethics, the Data Inspectorate and the Norwegian Directorate of Health. Written informed consent was obtained from all individual participants included in the study.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The process for applying for data is described at the website of University of Tromsø. See: The Stata code for assessing mediation with inverse odds weighting is included in eAppendix.