- Top of page
- Material and methods
Background: It is still unclear how early-life exposure to pets is related to children's risk of developing atopy-related diseases. We estimated associations between early-life exposure to pets and atopy-related diseases at 0-4 years of life in a cohort of Norwegian children.
Methods: A population-based cohort of 2531 children born in Oslo, Norway, was followed from birth to the age of 4 years. Information on early-life exposure to pets, a number of possible confounders, and atopy-related diseases was mainly collected by questionnaire.
Results: In logistic regression analysis adjusting for potential confounders, the odds ratio for being exposed to pets in early life (reference category: not exposed) was, for bronchial obstruction at 0–2 years of life, 1.2 (95% confidence interval 0.9, 1.8); for asthma at the age of 4 years, 0.7 (0.5, 1.1); for allergic rhinitis at the age of 4 years, 0.6 (0.4, 1.0); and for atopic eczema at 0–6 months of life, 0.7 (0.5, 0.9).
Conclusions: The results indicate that early-life exposure to pets or lifestyle factors associated with exposure to pets reduce the risk of developing atopy-related diseases in early childhood. However, these findings might also be explained by selection for keeping pets.
Studies have shown that exposure to furred pets can increase the occurrence of atopy-related diseases and conditions (1–9). Whether early exposure to pets could increase the risk of developing atopic diseases is more uncertain (1, 10–11). Even so, it has been common to advise families disposed to allergic diseases to avoid keeping pets. Results from recent studies have challenged this practice, as they show negative associations between early-life exposure to domestic pets and the occurrence of atopy-related conditions (6, 12, 13). This is further supported by the findings of reduced risks of developing atopy among children raised on farms with livestock (14–16). It seems that early exposure to pet allergens could be favorable or that growing up in an environment with pets or other animals could change children's exposure to other factors in a protective direction. So far, most of these studies have collected information on exposure to pets retrospectively, and the time from exposure to the assessment of the health outcomes has been long. We used data from the Oslo Birth Cohort Study (17–19) to estimate associations between early-life exposure to pets and atopy-related diseases during the first 4 years of life.
- Top of page
- Material and methods
Pets were kept in 22.4% of the homes when the children were born (7.1% of homes had cats, and 9.1% dogs). The associations between early-life exposure to pets and atopy-related conditions at 4 years of age were mainly negative. This was also the case with atopic eczema at 0–6 months of life. The adjusted odds ratio for early-life atopic eczema if exposed to pets at home was 0.7 (95% CI 0.5, 0.9); if exposed to cats, it was 0.5 (95% CI 0.3, 0.9).
Unlike other researchers (6, 12–13), we were able to estimate associations between early-life exposure to pets and atopy-related diseases from prospectively collected information on pet keeping, as well as for a number of potential confounders. This approach reduced the chance of recall bias. Keeping pets could have been associated with conditions that could confound the association between exposure to pets and atopic disorders. Environmental tobacco smoke exposure might have been such a confounder, as it was associated with keeping pets and was likely to influence some of the health outcomes. However, we found no convincing evidence of confounding factors. Due to a large study population, the health outcomes were mainly assessed by questionnaires. This could have led to some misclassification. A cohort study reduces the chance of systematic misclassification. However, it could be that parents who keep pets underreport diseases in general. This did not seem to be the case, as associations between diseases and exposure to pets varied, and negative associations seemed to be specific for the diseases we considered to be the most atopic. Nonsystematic misclassification may have weakened the associations, but could not have changed their direction.
Information on pet allergens and keeping pets was not collected at exactly the same time. Even so, the agreement between the results of two different methods for measuring exposure to dog and cat allergens was quite good. Few families kept pets (1, 2, 5–7). This could indicate that exposure to pet allergens in general was low, resulting in a larger difference in exposure between homes with and without pets than in populations with a higher prevalence of keeping pets. However, we also found that cat and dog allergens were present in homes without pets, demonstrating that it is difficult to establish a truly unexposed reference group (23, 24). For practical and economic reasons it is seldom feasible to monitor allergen levels on a large scale in house dust over time. Therefore, keeping pets is the most feasible way to assess exposure over time in large cohort studies. Due to this and to low statistical power, we have not focused on the associations between concentrations of pet allergens and the studied health outcomes.
The results of this study are generally in agreement with earlier findings of negative associations between early-life exposure to pets and atopy-related conditions (6, 12, 13). The positive association between dog exposure and early-life bronchial obstruction could be explained by the fact that early-life bronchial obstruction is a more infectious and less atopic condition than asthma later in childhood, and that keeping dogs increases the risk of contracting infections. The negative associations could have several explanations. First, early exposure to pet allergens could have a general or specific protective effect, resulting in a reduced development of all or several types of allergic or atopy-related diseases/conditions later in life. Second, exposure to pets could indicate other exposures or lack of exposures of importance for the development of these diseases (25). Third, the associations could have been the result of selection for keeping pets both after and before childbirth, including previous generations. To single out the right explanation is difficult. Allergic reactions to pets in early life might lead to a reduction in children's exposure to pets that could reduce the occurrence of atopic disorders later in life. A lengthy period between the measurements of exposure and outcome would favor such an explanation. However, if the protective effect is of short duration, it could be overlooked. A follow-up of a birth cohort would not necessarily exclude these possibilities. The short periods between pet exposure and the occurrence of the outcomes in this study (from 0–6 months to 4 years) support a true negative association. Furthermore, the association was specific for early-life atopic eczema, and not for diaper eczema. However, it is still possible that the tendency to keep pets has been under selection for prebirth experiences. The selection mechanism may not be captured by a simple control by parental atopy. It is likely that the decision to keep pets is often based on other conditions than a family history of asthma or hay fever. To reduce the chance of selection bias, one could try to control for reports of refraining from keeping pets (12), but the reasons for not keeping pets may not always be obvious and easy to report.
The effect of early-life exposure to pets on the development of atopic disorders is difficult to study. The negative associations shown in this study could have several explanations. They could indicate a protective effect of early-life exposure to pets, or that keeping pets is associated with a lifestyle that protects against the development of atopic diseases. However, they might also be explained by prebirth selection with regard to keeping pets.