The association of education with long-term weight change in the EPIC-PANACEA cohort

S Rohrmann, A Steinbrecher, J Linseisen, S Hermann, A May, J Luan, U Ekelund, K Overvad, A Tjønneland, J Halkjær, G Fagherazzi, M-C Boutron-Ruault, F Clavel-Chapelon, C Agnoli, R Tumino, G Masala, A Mattiello, F Ricceri, N Travier, P AmianoE Ardanaz, M-D Chirlaque, M-J Sanchez, L Rodríguez, L M Nilsson, I Johansson, B Hedblad, M Rosvall, E Lund, T Braaten, A Naska, P Orfanos, A Trichopoulou, S van den Berg, H B Bueno-de-Mesquita, M M Bergmann, A Steffen, R Kaaks, B Teucher, N J Wareham, K-T Khaw, F L Crowe, A-K Illner, N Slimani, V Gallo, T Mouw, T Norat, P H M Peeters

Research output: Contribution to journalArticleAcademicpeer-review

13 Citations (Scopus)


BACKGROUND/OBJECTIVES: Cross-sectionally, educational attainment is strongly associated with the prevalence of obesity, but this association is less clear for weight change during adult life. The objective of this study is to examine the association between educational attainment and weight change during adult life in the European Prospective Investigation into Cancer and Nutrition (EPIC).

SUBJECTS/METHODS: EPIC is a cohort study with 361,467 participants and up to 10 years of follow-up. Educational attainment was categorized according to the highest obtained school level (primary school or less, vocational secondary training, other secondary education and university). Multivariate mixed-effects linear regression models were used to study education in relation to weight at age 20 years (self-reported), to annual change in weight between age 20 years and measured weight at recruitment, and to annual change in weight during follow-up time.

RESULTS: Higher educational attainment was associated with on average a lower body mass index (BMI) at age 20 years and a lower increase in weight up to recruitment (highest vs lowest educational attainment in men: -60 g per year (95% confidence interval (CI) -80; -40), women -110 g per year (95% CI -130; -80)). Although during follow-up after recruitment an increase in body weight was observed in all educational levels, gain was lowest in men and women with a university degree (high vs low education -120 g per year (95% CI -150; -90) and -70 g per year (95% CI -90; -60), respectively).

CONCLUSIONS: Existing differences in BMI between higher and lower educated individuals at early adulthood became more pronounced during lifetime, which possibly impacts on obesity-related chronic disease risk in persons with lower educational attainment.

Original languageEnglish
Pages (from-to)957-63
Number of pages7
JournalEuropean Journal of Clinical Nutrition
Issue number8
Publication statusPublished - Aug-2012
Externally publishedYes


  • Adult
  • Body Mass Index
  • Body Weight
  • Chronic Disease
  • Cohort Studies
  • Cross-Sectional Studies
  • Educational Status
  • Energy Intake
  • Europe/epidemiology
  • Female
  • Follow-Up Studies
  • Humans
  • Life Style
  • Linear Models
  • Male
  • Middle Aged
  • Nutritional Status
  • Obesity/epidemiology
  • Prevalence
  • Risk Factors
  • Surveys and Questionnaires

Cite this