Interrelationships between education, occupational social class, and income as determinants of disability retirement

Scand J Public Health. 2012 Mar;40(2):157-66. doi: 10.1177/1403494811435492. Epub 2012 Feb 6.


Aims: The association between a low socioeconomic position and disability retirement is established in the literature, but the interrelationships between various subdomains of socioeconomic position are poorly understood. We examined the independent and interdependent effects of education, social class, and income on disability retirement.

Methods: Using nationally representative register data we followed up over 260,000 Finns aged 30-63 at the end of 1995 for disability retirement from 1996 to 2004. Cox regression analysis was used to calculate hazard ratios (HR) and relative indices of inequality (RII).

Results: Each socioeconomic indicator had a linear negative association with disability retirement. The socioeconomic gradients were stronger in the younger age groups. The effect of education was largely mediated through succeeding social class. Social class was largely explained by preceding education, but was only moderately mediated through income. Income was largely explained by education, and even further by social class. The independent effects of education, social class, and income on disability retirement as measured by the RII were 1.74 (95% CI 1.60-1.90), 1.95 (1.78-2.15), and 1.35 (1.25-1.47) for men and 1.76 (1.61-1.92), 2.14 (1.95-2.34), and 1.14 (1.05-1.24) for women.

Conclusions: The effects of socioeconomic position on disability retirement may not be fully captured if the pathways between the various subdomains are disregarded. Our results suggest that efforts to delay and prevent disability retirement should focus on lifestyle and cognitive factors associated with education, as well as on factors associated with social class such as working conditions and power resources.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Age Distribution
  • Disability Evaluation
  • Disabled Persons / statistics & numerical data*
  • Educational Status*
  • Female
  • Finland
  • Humans
  • Income / statistics & numerical data*
  • Life Style
  • Male
  • Middle Aged
  • Occupations
  • Pensions / statistics & numerical data*
  • Proportional Hazards Models
  • Registries / statistics & numerical data
  • Risk Factors
  • Social Class*
  • Socioeconomic Factors