Format

Send to

Choose Destination
J Cancer Res Clin Oncol. 2010 Jan;136(1):71-7. doi: 10.1007/s00432-009-0638-9.

Level of education and the risk of lymphoma in the European prospective investigation into cancer and nutrition.

Author information

1
Division of Cancer Epidemiology (C020), German Cancer Research Center, Heidelberg, Germany.

Abstract

INTRODUCTION:

Lymphomas belong to the few cancer sites with increasing incidence over past decades, and only a few risk factors have been established. We explored the association between education and the incidence of lymphoma in the prospective EPIC study.

MATERIALS AND METHODS:

Within 3,567,410 person-years of follow-up, 1,319 lymphoma cases [1,253 non-Hodgkin lymphomas (NHL) and 66 Hodgkin lymphomas (HL)] were identified. Cox proportional hazard regression was used to examine the association between highest educational level (primary school or less, technical/professional school, secondary school, university) and lymphoma risk.

RESULTS:

Overall, no consistent associations between educational level and lymphoma risk were observed; however, associations were found for sub-groups of the cohort. We observed a higher risk of B-NHL (HR = 1.31, 95% CI = 1.02–1.68; n = 583) in women with the highest education level (university) but not in men. Concerning sub-classes of B-NHL, a positive association between education and risk of B cell chronic lymphatic leukaemia (BCLL) was observed only in women. In both genders, the risk of diffuse large B cell lymphoma (DLBCL) was significantly lower for subjects with university degree (HR = 0.46, 95% CI = 0.27–0.79) versus lowest educational level. No association was found for HL.

CONCLUSION:

We could not confirm an overall consistent association of education and risk of HL or NHL in this large prospective study; although, education was positively related to the incidence of BCLL and B-NHL (in women) but inversely to incidence of DLBCL. Due to limited number of cases in sub-classes and the large number of comparisons, the possibility of chance findings can not be excluded.

PMID:
19582474
DOI:
10.1007/s00432-009-0638-9
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center