“Autumn babies were three times as likely to have an allergy to milk and eggs as summer babies,” according to The Daily Telegraph. Researchers believe the variation is due to the foetus’s exposure to pollen at a critical time in pregnancy, it added.
The story comes from Finnish research that looked at whether the timing of birth or early development in the womb is associated with sensitivity to food allergens in childhood. It found that in children under four years old, positive food allergy tests were more common in babies born in October and November, and least common in those born in June and July. The association was particularly marked when looking at allergies to milk and eggs. The researchers suggest that the higher rate of allergies in autumn babies might be due to a high pollen count in spring coinciding with an important phase of foetal immune development.
While this large study might seem to suggest that food allergies in later life may be associated with foetal exposure to pollen in the early stages of pregnancy, the study does not prove that this is the case. In particular, it is possible that other factors, such as newborns’ susceptibility to winter infections, may have influenced the development of food allergies in children. More research is now needed to clarify the matter.
Where did the story come from?
The study was carried out by researchers from a number of Finnish research organisations: the University of Oulu, the University of Tampere, the Finnish Institute of Occupational Health, the Health Centre of Oulu and South Karelia District of Social and Health Services. It was funded by the Social Insurance Institution of Finland, the Academy of Finland and other organisations.
The study was published in the peer-reviewed Journal of Epidemiology and Community Health.
This research was reported widely and generally fairly by the media, although most reports tended to overstate the certainty of the study’s conclusions. Different newspapers focused on different angles. For example, the Daily Mail said that babies born in the autumn faced a higher risk of food allergies, while the Daily Express said that children conceived in the first few months of the year were at higher risk. Some papers, such as The Daily Telegraph, included the opinions of independent experts who recommended that couples should not time their efforts at pregnancy on the basis of this study.
What kind of research was this?
This was a cohort study looking at whether the timing of birth or foetal development were associated with later sensitisation to food allergens in children. Cohort studies are useful in following large numbers of people for several years to find out whether an event (in this case, season of conception or birth) is associated with health outcomes (in this case, sensitisation to food allergens). However, on their own, cohort studies cannot prove cause and effect.
The researchers say that children born in autumn or winter have a higher incidence of some allergies, and also a higher incidence of immunoglobulin E antibodies (which play an important role in the development of allergy), than children born in spring or summer. The reason for this is unknown, but they hypothesise that it could be related to maternal pollen exposure while the baby develops in the womb. They also point out that the foetus begins to produce immunoglobulin E antibodies in about the 11th gestational week of pregnancy.
What did the research involve?
The study involved 5,973 children born between April 2001 and March 2006 who were living in the province of Karelia in south east Finland in 2005 and 2006. During this period, researchers sent parents a questionnaire on duration of pregnancy, gender, relative ages of any siblings, history of hay fever or pollen allergy in the mother, and maternal smoking. Parents of 3,899 children (66%) returned the questionnaire.
The researchers also collected all available results from any food allergy tests that had been carried out at the province’s health clinics and hospitals between August 2001 and September 2006. The gathered data covered children from across the whole population, and included month of birth. The researchers also measured daily pollen counts in the area throughout the pollen season (from March to August) in 2002, and they measured mean monthly temperatures and monthly averages of sunshine hours.
The researchers looked at any positive results from all the different types of food allergy tests that had been conducted, including the presence of antibodies for specific food items, skin prick tests for food items, and open food challenges - a type of allergy test where people are physically exposed to items to check for a reaction. The researchers also conducted a separate analysis looking solely at allergic reaction to milk and eggs.
Using standard statistical methods, they analysed any association between a positive reaction to food allergy tests and the seasons of both birth and gestation. They took account of factors that might influence the results (called confounders), such as maternal pollen allergy, maternal smoking and the child's birth order.
What were the basic results?
Of the 5,920 children, 961 tested positive for food allergies. Over the four years:
- a positive food allergy test was recorded in 10% of children born in October and November, and 5% among those born in June and July.
- a positive food allergy test was recorded in 11% of children whose 11th gestational week was in April and May, and in 6% of those reaching that stage in December and January.
- Concentrations of alder and birch pollen in the area were highest in April and May.
- The seasonal effect was particularly pronounced in the incidence of allergic reaction to milk and eggs among children who had their 11th gestational week in April-May
How did the researchers interpret the results?
The researchers conclude that a positive food allergy test result was more likely in children whose 11th gestational week fell in April or May (and would therefore be likely to be born in autumn). One possible explanation for this would be their mothers’ increased exposure to pollen from leafy trees around this time, given the seasonal rise in pollen levels during springtime. They suggest that exposure to pollen allergens could “affect the immunological response by an unknown mechanism during the crucial first trimester of pregnancy”.
This was a large, well-designed study, which used carefully compiled data on laboratory tests for sensitivity to food allergies, linked with a questionnaire survey. However, it now shows on its own that food allergies are triggered by the timing of pregnancy and birth, or that they are related to seasonal pollen counts. The study had several limitations:
- There was some variation in the type of food allergy tests used, the ages of the children when tested, the laboratories used to process samples, and the testing procedures themselves. This may undermine the reliability of the findings, although the researchers argue that this diversity should not affect the overall results.
- Records on food allergy tests were identified for less than 20% of the children, and it is not certain that all data on relevant tests was available to the researchers.
- Pollen counts were sampled for only one year and at only one location. These might have varied throughout the region or in the other three years of the study when testing did not take place.
- Although the researchers tried to take account of confounders, it is possible that other unknown factors might have influenced the results. For example, babies born in winter are more likely to be exposed to viral infections, which could therefore affect their risk of having allergies. The researchers had no data on infections in this study.
Overall, the evidence from this study is not strong enough to provide any direction for couples concerned about reducing the risk of food allergies in their children.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Daily Telegraph, 20 October 2010
The Guardian, 20 October 2010
The Independent, 20 October 2010
Daily Mail, 20 October 2010
Daily Express, 20 October 2010
Links to the science
Journal of Epidemiology and Community Health, [Published Online First] October 19 2010