OBJECTIVES: To investigate the effect of dog and cat contacts on the frequency of respiratory symptoms and infections during the first year of life.
METHODS: In this birth cohort study, 397 children were followed up from pregnancy onward, and the frequency of respiratory symptoms and infections together with information about dog and cat contacts during the first year of life were reported by using weekly diaries and a questionnaire at the age of 1 year. All the children were born in eastern or middle Finland between September 2002 and May 2005.
RESULTS: In multivariate analysis, children having dogs at home were healthier (ie, had fewer respiratory tract symptoms or infections) than children with no dog contacts (adjusted odds ratio, [aOR]: 1.31; 95% confidence interval [CI]: 1.13–1.52). Furthermore, children having dog contacts at home had less frequent otitis (aOR: 0.56; 95% CI: 0.38–0.81) and tended to need fewer courses of antibiotics (aOR: 0.71; 95% CI: 0.52–0.96) than children without such contacts. In univariate analysis, both the weekly amount of contact with dogs and cats and the average yearly amount of contact were associated with decreased respiratory infectious disease morbidity.
CONCLUSIONS: These results suggest that dog contacts may have a protective effect on respiratory tract infections during the first year of life. Our findings support the theory that during the first year of life, animal contacts are important, possibly leading to better resistance to infectious respiratory illnesses during childhood.
- aOR —
- adjusted odds ratio
- CI —
- 95% confidence interval
- GEE —
- generalized estimating equations
What's Known on This Subject:
Respiratory infectious symptoms are common during the first year of life. Day care attendance, older siblings, and lack of breastfeeding have been considered as possible factors influencing early respiratory tract infections.
What This Study Adds:
Children with early dog contacts seem to have fewer infectious respiratory symptoms and diseases, especially otitis, during the first year of life.
Respiratory infectious symptoms and diseases are frequent during the first of year life. They are usually caused by a virus, with human rhinovirus being the most common pathogen.1–3 In previous reports, the frequency of respiratory tract infections has been estimated to vary between 3 and 6 episodes within the first year of life.4–6
A variety of factors, such as day care attendance,7,8 older siblings7,9,10 and lack of breastfeeding,10–12 have been considered as possible risk factors for early respiratory tract infections. In addition, a parental history of asthma13,14 and smoking10,15–18 are thought to have a role in the child’s susceptibility to infections and to respiratory symptoms. Earlier reports on the role of animal contacts on the prevalence of respiratory tract infections are sparse. Some previous studies have shown that dog contacts,19,20 although not cat contacts,15 seem to decrease the number of common cold episodes during childhood. However, pet ownership has also been considered a possible risk factor for frequent childhood respiratory tract infections in some reports.21
In the last 10 years, a large body of literature has evaluated the role of early childhood animal exposure, including household pets, in the risk of asthma or allergy in children.22–24 We have previously reported a decreased tumor necrosis factor α–producing capacity at birth and in the first year of life in children exposed to indoor dogs in early life,25 suggesting that exposure to dogs during early life may reduce innate immune responses already at birth. At the same time, virological studies have shown large differences in immune responses among the children exposed to viruses that cause a common cold, such as human rhinovirus; it seems to cause common colds for most children during childhood, whereas atopic children with the higher risk of persistent asthma have had more frequently severe infections with wheezing and altered immune response.26 Thus, a better understanding of the interplay between pet-related exposures and the development of early respiratory tract infections may provide indirect insight regarding the factors affecting the maturation of immune responses and its disturbances, such as asthma. In addition, recognizing the risk factors for respiratory tract infections during childhood is important because there might be a connection between childhood respiratory tract infections and chronic airway diseases later in adulthood.21
The aim of this study was to describe the effect of domestic animal contacts on respiratory tract infection morbidity during the first year of life. The study was based on a prospective birth cohort study with diary-based data on respiratory tract infections and animal exposures during the first year of life.
The primary study population consisted of Finnish participants in a prospective birth cohort study, PASTURE (Protection Against Allergy in Rural Environments), which is an ongoing international study in 5 different European countries (Austria, Finland, France, Germany, and Switzerland). The Finnish study population consists of 208 children whose mothers were followed up from the third trimester of pregnancy. The mothers lived in rural areas, either on a farm or in a nonfarming environment. The cohort has been described in more detail elsewhere.27 Another 216 mothers and their children (the extended Finnish cohort) were followed up through the same extensive protocol as the primary study population. The mothers lived in either rural or suburban environments, and all gave birth at Kuopio University Hospital. All the study children were born between September 2002 and May 2005.
The children’s parents were given diary questionnaires consisting of questions related to infectious symptoms and health care attendance as well as on the children’s dog and cat contacts. The diary questionnaires were filled in weekly, beginning from the ninth postnatal week and continuing up to the 52nd week, providing a total of 44 weeks of diary entries. In the diary questionnaires, parents were asked if their children had been “hale and hearty” during the previous 7 days. If the child had not been completely healthy, the parents also filled in a weekly questionnaire concerning different infectious diseases and symptoms (whether the child had had cough, wheezing of breath, rhinitis, fever ≥38.5°C, middle ear infection, diarrhea, urinary tract infection, itchy rash, or some other illness during these last 7 days). In this analysis, we evaluated respiratory infectious symptoms (cough, wheezing of breath, rhinitis, and fever) and infections (middle ear infection). Any children for whom there were less than one-half the number of possible diary entries during the study period (<23 weeks) were excluded (n = 27). Thus, 397 children were included in the following analyses.
Families were asked in the weekly questionnaires whether they had a dog or a cat at home and how much time it had spent inside daily. For the following analysis, dog and cat contacts indoor at home were grouped into: (1) no contact at all; (2) low contact (pet inside at home up to a maximum of 6 hours daily); (3) medium contact (pet inside from 6–16 hours daily); and (4) high contact (pet inside >16 hours daily). Breastfeeding was reported in 3 categories weekly: the child had either been solely breastfed, partly breastfed, or not breastfed at all.
We further evaluated the estimated average amount of daily dog and cat contacts during the study period by using a self-administered retrospectively collected 1-year questionnaire, which was collected from the mothers when the children were 12 months old. The estimated average amount of daily dog and cat contacts was grouped as follows: (1) no dog/cat at all or dog/cat never inside at home; (2) dog/cat occasionally inside at home; (3) dog/cat often inside at home; or (4) dog/cat mostly inside at home. The analysis also included those children with both dog and cat contacts.
Additional data were collected from questionnaires during pregnancy and early childhood. Children were divided into 4 groups depending on the birth month: summer (June–August), autumn (September–November), winter (December–February), and spring (March–May). The birth weight was categorized into tertiles: <3480g, 3480 to <3810 g, and ≥3810 g. The number of older siblings was also categorized into 3 groups: no siblings, 1 sibling, or ≥2 siblings. Information concerning maternal smoking 2 months after delivery was requested in the 2- month questionnaire: yes versus no. Parental atopy was categorized as yes if the mother or father had ever been diagnosed with asthma, allergic eczema, or rhinitis, or no if neither of the parents had ever had any of those diagnoses. The choice of families of whether to keep pets was categorized yes versus no. Both maternal and paternal educational levels were categorized into 3 groups: (1) elementary or vocational school; (2) high school or college; or (3) university.
When data on animal contacts or information on the health, respiratory symptoms, and infections were missing from a diary, the week in question was excluded from the analysis. The same was done with other missing data if the information was not reliably derivable from other sources (eg, missing breastfeeding status from previous or subsequent weeks).
Data analysis was performed by using PASW statistics version 18.0 (SPSS Inc, Chicago, IL). The comparison of the frequencies of the healthy weeks and duration of the various respiratory diseases or symptoms with different baseline and risk factors was performed by using the χ2 test or the Kruskall-Wallis test. Generalized estimating equations (GEE) were used to investigate the relationships between the prevalence of respiratory disease or symptoms during the follow-up and various predictive factors. The working correlation matrix was AR(1). In the multivariate GEE models, potential confounding variables were selected a priori on the basis of biological plausibility. These variables included: gender (male versus female), living environment (farm, rural nonfarming versus suburban), number of siblings (0, 1, vs >1), maternal smoking (yes versus no), parental atopy (yes versus no), breastfeeding (solely, partly versus no), birth weight (<3480, 3480 to <3810 vs ≥3810 g), season of birth (winter, spring, summer, autumn), diary month, and cohort (Finnish PASTURE [Protection Against Allergy in Rural Environments] versus extended Finnish cohort). Contacts with dogs or cats at home were also determined according to the amount of the time that the animals had spent inside the house daily. The cutoff level for significance was set at 0.05.
The parents of all the children involved in the study gave their written informed consent. The research protocol was approved by the Research Ethics Committee of the Hospital District of Northern Savo (Kuopio, Finland).
In total, information was received for a total of 17 124 follow-up weeks recorded from the 397 children. More weekly questionnaires were completed at the beginning of the follow-up: the number varied from 396 forms in weeks 7, 11, 12, 17, and 18 to 347 forms in week 44. Of the 44 follow-up weeks, the mean number of completed weeks was 43.0 per child (range: 23–44 weeks), with 94.2% of families filling in at least 40 forms.
Table 1 shows the baseline characteristics and animal contacts of the 397 study children presented as diary weeks in relation to living environment. There were significant differences in all shown confounding variables between different living environments, and thus the living environment variable was further selected as a confounder in the multivariate analysis.
Infectious Respiratory Symptoms
Four children were reported as having been healthy during the whole study period, and 15.6% of the children (n = 62) were healthy for less than one-half of the follow-up weeks. In total, 285 (71.8%) children had had fever, 157 (39.5%) a middle ear infection, 384 (96.7%) rhinitis, 335 (84.4%) cough, and 128 (32.2%) wheezing during any time of the study period. In addition, nearly one-half of the children (n = 189 [47.6%]) needed systemic antibiotics in the course of the 44-week study period. The most frequently reported disease was rhinitis, which occurred in 17.0% of the follow-up weeks. Cough occurred in 10.4%, fever in 4.0%, wheezing in 2.0%, and middle ear infection in 2.5% of the follow-up weeks. Two children had pneumonia during the follow-up.
Cat and Dog Contacts
Of 397 children, 245 (61.7%) had had dog and 136 (34.3%) had had cat contact at home at some time during the study period (range: 1–44 weeks). However, dog contacts at home were not stable: during the study weeks 1, 22, and 44, the percentages of children living with no dog or no cat contact in the home varied from 66.1% to 69.3% and 76.1% to 77.4%, respectively. According to the 1-year retrospective questionnaire at the end of the study period, 65.2% of the children lived mainly in homes with no dog contact and 75.5% in homes with no cat contact. During the pregnancy, 22.7% of families reported pet animal avoidance due to allergic symptoms in the family.
If children had dog or cat contacts at home, they were significantly healthier during the study period in univariate tests (both P < .001) (Table 2, Fig 1) and had fewer weeks with cough, otitis, and rhinitis and also needed fewer courses of antibiotics than children with no cat or dog contacts at all. The number of study weeks with different respiratory tract symptoms and infections was also compared with 1-year questionnaire data on cat and dog contacts (Table 3), and the results were comparable to the results of the animal contact data from the diary (Table 2).
The multivariate analysis was conducted longitudinally (Tables 4 and 5) and cross-sectionally (Tables 6 and 7), by using GEE models. Even after adjusting for possible confounders, children having a dog at home were significantly healthier, had less frequent otitis, and tended to need fewer courses of antibiotics during the study period than children without dog contacts. Both the weekly amount of dog contacts (according to diary data) and the average amount of yearly contact (according to 1-year questionnaire data) with dogs was associated similarly with decreasing respiratory infectious disease morbidity. The highest protective association between dog ownership and healthiness, as well as lower risk for antibiotic use, otitis and rhinitis was detected among children who had a dog inside at home for <6 hours daily (according to diary data) or had a dog temporally or often inside (according to retrospective data) compared with those who did not have any dogs or dogs were not inside (Tables 4, 5, 6, and 7). In sensitivity analysis, the associations did not change after we removed from analysis those children whose families reported avoidance of pets due to some allergic causes (data not shown).
According to our results, dog and cat contacts during early infancy may be associated with less morbidity in general (indicated as more healthy weeks) and concomitantly may have a protective effect on respiratory tract symptoms and infections. In comparisons between cat and dog contacts, dog contacts showed a more significant protective role on respiratory infectious disease morbidity. To our knowledge, this is the first study that has evaluated the significance of pet contacts during childhood for the development of respiratory tract infections and, furthermore, has made analyses by using 2 different data collection methods for information on animal contact. This study also made possible evaluation of the effects of different lengths of animal contacts, by using short time variations, on the risk of respiratory tract illnesses.
The results support those of Hatakka et al,20 who reported that furred pets would decrease the risk of recurrent acute respiratory tract symptoms with children 1 to 6 years old, although some reports have found no association between respiratory tract infections and pet keeping.28 Our current results also support indirectly those of Grüber et al,19 who found that having a dog at home during the first 2 years of life decreased the number of common cold episodes among 0- to 2-year-olds. We showed that children who had dog contacts at home had less otitis and rhinitis and more healthy weeks than children without dog contacts at home, but having a dog at home during the first postnatal year had no significant role in the occurrence of wheezing and cough, which is in line with earlier results,29 even though the reason for these findings is obscure.
Cat ownership seemed to also have an overall protective effect, although weaker than dog ownership, on the infectious health of infants. Similarly, in some earlier studies, exposure to cats has been considered a protective factor toward wheezing in the first year of life,13 as well as for croup with older children.30 On the contrary, many studies have shown that cat contacts have no effect on infectious symptoms15,29 or wheeze among children.31 In line with this, the association between exposure to cats and respiratory tract symptoms and infections during childhood was decreased in our study after confounding in multivariate analysis. It is unsure why cat exposure is less significant compared with dog exposure. However, Heyworth et al32 have shown that cat and/or dog ownership is also associated with a reduced risk of gastroenteritis. This finding could indicate a real relationship exists between animal contacts and the frequency of infections among children.
We also showed that children living in houses in which dogs spend only part of the day inside (defined as <6 hours or temporally) had the lowest risk of infectious symptoms and respiratory tract infections. A possible explanation for this interesting finding might be that the amount of dirt brought inside the home by dogs could be higher in these families because they spent more time outdoors. In other words, less dirt is brought indoors by dogs who mainly live indoors. The living environment could also affect the amount of dirt and animal contacts. Hence, we did a subanalysis and evaluated the effect of animal contacts on overall healthiness separately for children living in rural and suburban environments. The directions of associations did not change, although some of the associations weakened slightly, as well as some got stronger. Furthermore, we included the area of living as a covariate in the multivariate analysis, and the associations between dog exposure and health variables remained. The amount of dirt is likely to correlate with bacterial diversity in the living environment, possibly affecting the maturation of the child’s immune system and further affecting the risk of respiratory tract infections. However, in this article, we could not objectively analyze the actual role of bacterial diversity, which will have to be the subject of further studies.
There are several possible reasons why the reported associations between pet ownership and respiratory tract infections during childhood are inconsistent. First, the different types of pet animals have not been evaluated separately in the analyses.5,16,20,21,28 Second, some studies enrolled older children. For example, Burr et al16 found that pet ownership increased the risk of wheezing, rhinitis, and the number of cold episodes per year for 12- to 14-year-olds. Third, studies have evaluated the associations between respiratory tract infections and animal contacts during childhood retrospectively.21 Our study was prospective in nature, and we also analyzed results by using data collected with 2 different methods, thus making our results more reliable.
Lastly, it is possible that the effect of animal contacts on the frequency of respiratory tract symptoms is mediated by a choice among atopic parents to not keep pets.20,33 We included parental atopy as a confounder in the multivariate analysis, and the association was not diminished. When we repeated the analyses without subjects with a reported family history of pet avoidance, associations were not changed. As a whole, predictions on pet-keeping data are not straightforward, and there are studies showing that atopic families do not necessarily avoid keeping pets.34 However, the influence of possible atopic predisposition of the study children on the results cannot be completely ruled out.
Evidence suggests that animal contacts, especially during early life, might be crucial in immunity developing along a nonallergic route24,35 and in ensuring effective responses to respiratory viral infections in early life.35,36 We speculate that animal contacts could help to mature the immunologic system, leading to more composed immunologic response and shorter duration of infections. We offer preliminary evidence that dog ownership may be protective against respiratory tract infections during the first year of life.
To our knowledge, this is the first prospective study in which the frequencies of respiratory infectious symptoms and diseases are compared with weekly amounts of pet contacts in a diary follow-up manner during early infancy. Our results suggest that dog contacts protect children from respiratory tract infections during the first year of life.
We thank the families for participating in the study, the field nurses for their dedicated work, and Pekka Tiittanen for data management and statistical assistance.
- Accepted April 5, 2012.
- Address correspondence to Eija Bergroth, MD, Department of Pediatrics, Kuopio University Hospital, Puijonlaaksontie 2, 70210 Kuopio, Finland. E-mail:
Dr Bergroth substantially contributed to analysis and interpretation of data, drafted the article, and approved the final version to be published; Dr Remes substantially contributed to design and interpretation of data, revised the article for important intellectual content, and approved the final version; Dr Pekkanen substantially contributed to conception, design and interpretation of data, revised the article for important intellectual content, and approved the final version; Mr Kauppila substantially contributed to design and acquisition of data, revised the article for important intellectual content, and approved the final version; Dr Büchele substantially contributed to conception, design and interpretation of data, revised the article for important intellectual content, and approved the final version; and Dr Keski-Nisula substantially contributed to analysis and interpretation of data, revised the article for important intellectual content, and approved the final version.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: This study was supported by Foundation for Pediatric Research, Kerttu and Kalle Viikki, Päivikki and Sakari Sohlberg, and Juho Vainio Foundations, EVO funding, Farmers’ Social Insurance Institution–Mela, Academy of Finland (grant 139021), and Kuopio University Hospital, Finland and European Union (QLK4-2001-00250 and FP7-211911).
- Koehoorn M,
- Karr CJ,
- Demers PA,
- Lencar C,
- Tamburic L,
- Brauer M
- Cushing AH,
- Samet JM,
- Lambert WE,
- et al
- Quigley MA,
- Kelly YJ,
- Sacker A
- Belanger K,
- Beckett W,
- Triche E,
- et al
- Burr ML,
- Anderson HR,
- Austin JB,
- et al
- Håberg SE,
- Stigum H,
- Nystad W,
- Nafstad P
- Karvonen AM,
- Hyvärinen A,
- Roponen M,
- et al
- Gold DR,
- Burge HA,
- Carey V,
- Milton DK,
- Platts-Mills T,
- Weiss ST
- Copyright © 2012 by the American Academy of Pediatrics