Infections have been hypothesized to be a risk factor for Hodgkin disease and non-Hodgkin lymphoma. Because infections usually occur seasonally, the authors looked at the seasonal pattern of occurrence of these diseases, since seasonal variation in occurrence would lend support to the hypothesis of an etiologic link between infection and lymphoma.
The authors identified 575 patients with Hodgkin disease and 498 with non-Hodgkin lymphoma under the age of 20 years in the Danish Cancer Registry from 1943 to 1994. A periodic regression model was used to evaluate the seasonal distribution of month of diagnosis and month of birth.
For the month of diagnosis of Hodgkin disease, the peak month of occurrence was March. The peak-to-trough ratio (the estimated incidence in the month of the peak divided by the estimated incidence in the month of the trough) was 1.57. For birth date, the ratio of case births at the peak month to case births at the trough month was 1.21, with a peak in July. There was no meaningful seasonal variation in the diagnosis or month of birth of children with non-Hodgkin lymphoma.
These data corroborate a previously reported March peak in the diagnosis of Hodgkin disease.
From the Department of Epidemiology and Social Medicine, University of Aarhus, Aaarhus, Denmark (V.L.); Department of Clinical Epidemiology, Aarhus University Hospital & Aalborg Hospital, Aarhus, Denmark (B.N., L.P., H.T.S.); Danish Cancer Society, Institute of Cancer Epidemiology, Copenhagen, Denmark (L.M.); and School of Public Health, Boston University, Boston, Massachusetts, U.S.A. (K.J.R.).
Submitted for publication October 3, 2002; accepted March 7, 2003.
Funded by research grant 9610030 from the Danish Cancer Society.
Address correspondence and reprint requests to Vivian Langagergaard, MD, Department of Epidemiology and Social Medicine, Vennelyst Boulevard 6, DK-8000 Aarhus C, Denmark. E-mail: firstname.lastname@example.org.