Measles infection, measles vaccination and the effect of birth order in the aetiology of hay fever.

It has recently been suggested that measles infection may reduce the risk of atopy. To study the independent effect of measles infection and measles vaccination on the occurrence of hay fever in a British national birth cohort. In over 6000 children born in 1970, details of immunizations and childhood diseases were collected by parental interviews at ages 5, 10 and 16 years, and hay fever within the past year at age 16 years. In univariate analysis, hay fever was less common in those contracting measles infection than in those not infected (OR 0.86, 95% CI 0.76-0.96), and more common in those given measles vaccination than in those not vaccinated (OR 1.16, 95% Cl 1.03-1.31). However, these effects were strongly confounded by birth order, which was closely associated with the likelihood of receiving measles vaccination and with the risk of hay fever. A strong interaction between the effects of measles vaccination and infection, and birth order was found, such that in those with many older sibling contacts, hay fever was significantly and independently reduced in relation to both measles infection and measles vaccination relative to those who were neither infected nor vaccinated. Both measles infection and measles vaccination in childhood appear to reduce the risk of hay fever in children with multiple older sibling contacts. Differential exposure or response to the measles virus may explain the effect of birth order on the occurrence of allergic disease.