Household Bird Ownership is Associated with Respiratory Illness among Young Children in Urban Bangladesh (CHoBI7 Program)

ABSTRACT.

There is limited evidence on the association between animal ownership and respiratory illness among young children in low- and middle-income countries. In this study, we examined the association between animal ownership and respiratory illness among children younger than 5 years of age enrolled in a prospective cohort study in urban Bangladesh. This prospective cohort study enrolled 884 participants younger than 5 years of age in Dhaka, Bangladesh. At baseline, trained research assistants administered caregivers of children younger than 5 years of age a questionnaire on household animal ownership. Animal ownership was defined as owning chickens, birds other than chickens, cats, and dogs. Respiratory surveillance was conducted monthly for children based on caregiver-reported coughing, rapid breathing, and difficulty breathing in the past 2 weeks during the 12-month study period. At baseline, 48% of children (424 of 884) had reports of coughing, 5% (40 of 884) had difficulty breathing, 3% (25 of 884) had rapid breathing, and 49% (431 of 884) had reports of any of these three respiratory symptoms. Seventeen percent of children (151 of 884) resided in a household that owned an animal. Children residing in households reporting bird ownership had a significantly greater odds of coughing (odds ratio, 1.14; 95% CI, 1.02–1.28) and any of the three respiratory symptoms in the past 2 weeks (odds ratio, 1.14; 95% CI, 1.02–1.28). Household bird ownership was associated with respiratory illness in young children. These findings suggest that interventions aiming to reduce young children’s exposure to domestic animals should extend to include birds other than chickens.

INTRODUCTION

Respiratory infections are the leading cause of death for children younger than 5 years of age globally.¹ Streptococcus, Haemophilus influenzae type B, and respiratory syncytial virus are the most common pathogens responsible for respiratory infections in young children.² Previous studies have shown that exposure to domestic animals, including pets and livestock, can cause respiratory illness and allergic airway disease resulting in coughing, difficulty breathing, and nasal discharge.^3–7 Both respiratory infections and allergic airway diseases have significant negative health consequences for children; however, overlapping symptoms can make distinguishing between the two challenging without a clinical diagnosis by a physician.^1,8 Respiratory infections and allergic airway disease are closely linked, with studies suggesting both are associated with the risk of asthma development.^9,10 The vast majority of studies investigating the association between animal presence in the home and respiratory illness have focused on pet animals such as cats and dogs, and were conducted in high-income countries. These studies often focus on the impact of animal ownership on allergic airway disease. Studies in Turkey and the United Arab Emirates found a significant association between respiratory symptoms and allergic airway disease among children living in homes with animals.^11,12 A recent study from Finland found that those owning domestic animals in urban environments were at greater risk of allergic respiratory symptoms than those in rural environments.¹³ However, a pooled analysis of data from European birth cohorts found no association between pet ownership in early life and risk of asthma or allergic rhinitis in children 6 to 10 years of age.¹⁴ An analysis of the International Study of Asthma and Allergies in Children data found that children 6 to 7 years of age in low- and middle-income countries (LMICs) who had been exposed to farm animals early in life had increased respiratory symptoms, but children in high-income countries did not.¹⁵

Previous studies have also found an association between exposure to animals and respiratory infections.^16–19 In the United States and Italy, chickens have been found to transmit respiratory influenza viruses to human hosts.^16–18 A study in Indonesia reported an association between household ownership of chickens and rabbits and increased risk for respiratory syncytial virus associated lower respiratory tract infections.¹⁹ This finding suggests that the risk of respiratory symptoms from exposure to domestic animals may vary across settings, and highlights the need for further research exploring this association in LMICs.

There is limited evidence, however, on the association between pet animal ownership and respiratory infections and symptoms among young children in LMICs. Most studies in LMIC settings have focused on the contribution of domestic animals residing in the household on childhood diarrheal illness or on the relationship between bird markets and avian influenza virus.^20–25 In slum areas of urban Dhaka, Bangladesh, as in many slum areas in LMICs globally, household members often reside in a single room.²⁶ Our previous prospective cohort study in this setting found that 10% of households reported keeping domestic animals in their child’s sleeping space.²⁷ This practice keeps children in close proximity to domestic animals and their feces, fur, feathers, and other allergens in indoor living spaces of the household. In our previous cohort study, we found that children younger than 5 years residing in the same sleeping space as domestic animals were at greater risk of linear growth faltering,²⁷ which is associated with a greater risk of mortality and adverse cognitive developmental outcomes later in life.^28–31 Our findings are consistent with the growing body of literature demonstrating the adverse health consequences of keeping domestic animals in the indoor living spaces of young children in LMIC settings.^{15,20,21,32–34}

To build evidence on the association between domestic animal ownership and respiratory illness among young children in an urban LMIC setting, the objective of this prospective cohort study was to investigate the association between animal ownership and respiratory symptoms among children younger than 5 years of age residing in slum areas of Dhaka, Bangladesh.

METHODS

Ethical approval.

Informed written consent was obtained from a parent or guardian for all study participants. The study procedures were approved by the Research and Ethical Review Committee of the International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) (PR-15133) and the Institutional Review Board of the Johns Hopkins Bloomberg School of Public Health (6785).

Study design.

This prospective cohort study was conducted from December 2016 to April 2019 in urban Dhaka, Bangladesh, as part of the Cholera Hospital-Based Intervention for 7 Days (CHoBI7) mobile health (mHealth) program randomized controlled trial. Patients with diarrhea presenting with three or more loose stools over a 24-hour period were recruited from two tertiary hospitals: the icddr,b, Dhaka Hospital (a private hospital) and Mugda General Hospital (a government hospital). The eligibility criteria for patients with diarrhea were the following: patients had 1) to have three or more loose stools during the past 24 hours, 2) to plan to reside in Dhaka for the next 12 months, 3) to not have a basin for running water in their home (mostly those residing in slum areas of Dhaka), 4) to have a child younger than 5 years of age in their household (including themselves), and 5) to have a working mobile phone in the household. After the recruitment of patients with diarrhea, their corresponding household members were also enrolled. All household members were enrolled, regardless of age or diarrhea patient status. The complete methodology of the CHoBI7 mHealth program and trial is published elsewhere.^35,36

To investigate the association between animal ownership and respiratory illness in young children, we compared household-level characteristics among 884 children younger than 5 years of age (both patients with and without diarrhea) enrolled in the CHoBI7 mHealth program trial. At baseline, trained research assistants administered caregivers of children younger than 5 years of age a questionnaire on sociodemographic factors, including household literacy; number of household members; type of latrine; refrigerator ownership; type of floor, roof, and walls; and animal ownership (not the location of animals).

Respiratory surveillance was conducted monthly for children younger than 5 years of age during the 12-month study period, including caregiver-reported coughing, rapid breathing, and difficulty breathing in the past 2 weeks assessed through the use of a structured questionnaire. The 2-week recall period was selected based on Bangladesh Demographic and Health Survey methods.³⁷ We did not collect data on clinical signs of severity of these respiratory symptoms. Prior to data collection, research assistants were trained on questionnaire administration for clinical surveillance by a study physician.

Statistical analysis.

Our primary objective was to determine whether animal ownership was associated with respiratory illness in young children. We investigated the association between animal ownership and caregiver-reported coughing, rapid breathing, and difficulty breathing within the past 2 weeks individually and together as “childhood respiratory illness.” We also investigated the association between animal ownership and pneumonia. Pneumonia was defined as caregiver-reported coughing and rapid breathing or rapid breathing and difficulty breathing using the integrated management of childhood illness guidelines.³⁸ Animal ownership was defined as the household reporting owning any animal. We separated chicken ownership and ownership of other birds such as pigeons, parrots, doves, quails, and mynah birds, given the literature demonstrating the association between chickens (with other birds excluded) and adverse child health outcomes.

To investigate the association between prevalence of respiratory symptoms and animal ownership, Poisson regression models were performed for the number of surveillance visits with caregiver-reported respiratory symptoms data available. Generalized estimating equations were used for clustering children residing in the same household. Household animal ownership was the primary predictor of interest. All Poisson regression models were adjusted for study arm, baseline age, number of sleeping rooms, roof type, wall type, refrigerator ownership, and whether the child was a patient with diarrhea at baseline. Stata 13.0 software (StataCorp LLC, College Station, TX) was used for the analysis.

RESULTS

In our analysis, 884 children younger than 5 years of age from 767 households were included. At the 12-month time point, 25% of households (n = 189) were lost to follow-up. The median baseline age for children was 11 ± 15 months (median ± SD) (range, 1–59 months) (Table 1). Forty-three percent of children (379 of 884) were female. Eighty percent of children (703 of 884) had a caregiver that reported being able to read and write, and 86% of children (762 of 884) resided in households with at least one household member who could read and write. The average number of individuals living in a household was 5 ± 1 (range, 2–12 individuals). Seventy-seven percent of children (677 of 884) resided in households with one room for sleeping. Thirty-one percent of children (270 of 884) resided in a household with a concrete roof, 72% (635 of 884) had concrete walls, and 96% (848 of 884) had a concrete floor. Forty-five percent of children (395 of 884) resided in households that owned a refrigerator. Eight percent of children (68 of 876) resided in households that used an unimproved latrine (flush or pour flush latrine to other than a sewer system or septic tank, a pit latrine without a slab or with a broken slab, or a hanging latrine). Seventeen percent of children (151 of 884) resided in a household that owned an animal: 9% (79 of 884) resided in households owning chickens; 9% (83 of 884) birds other than chickens; 2% (18 of 884) cats; < 1% (2 of 884) dogs; < 1% (5 of 884) goats; and < 1% (2 of 884) cows.

Table 1.

Population characteristics at baseline (N = 884)

Characteristic	n	%
Female gender	379	43
Literacy of household members*	762	86
Caregiver literacy	703	80
Unimproved latrine†	68	8
Household refrigerator ownership	395	45
Household animal ownership during the study period	151	17
Dog ownership	2	< 1
Cat ownership	18	2
Chicken ownership	79	9
Bird ownership (excluding chickens)	83	9
Goats	5	< 1
Cows	2	< 1
No. of sleeping rooms in household
1 room	677	77
> 1 room	207	23
Household with concrete roof	270	31
Household with concrete floor	848	96
Household with concrete wall	635	72
Respiratory symptoms
Coughing	424	48
Rapid breathing	25	3
Difficulty breathing	40	5
Childhood respiratory illness‡	431	49
Pneumonia§	23	3

^*At least one household member who could read and write.

^†An unimproved latrine is a flush or pour flush latrine to other than a sewer system or septic tank, a pit latrine without a slab or with a broken slab, or a hanging latrine.

^‡Childhood respiratory illness was defined as caregiver-reported coughing, rapid breathing, and difficulty breathing within the past 2 weeks.

^§Pneumonia was defined as coughing and rapid breathing or rapid breathing and difficulty breathing.

At baseline, 48% of children (424 of 884) had reports of coughing, 5% (40 of 884) had difficulty breathing, 3% (25 of 884) had rapid breathing, and 49% (431 of 884) had reports of any of these three respiratory symptoms (childhood respiratory illness). Three percent of children (23 of 884) had pneumonia. At the 12-month follow up, 38% of children (236 of 627) had reported childhood respiratory illness (cough, rapid breathing, or difficulty breathing), including 37% (234 of 627) with reported coughing, 2% (14 of 627) with difficulty breathing, and 1% (9 of 627) with rapid breathing. At the 12-month follow-up, 1% of children (9 of 627) had pneumonia.

Children living in households reporting bird ownership (excluding chickens) had significantly greater odds of having a cough (odds ratio, 1.14; 95% CI, 1.02–1.28) (Table 2) and childhood respiratory illness (odds ratio, 1.14; 95% CI, 1.02–1.28) in the past 2 weeks. There were no other significant associations found between animal ownership and child respiratory illness.

Table 2.

Associations between animal ownership and prevalence of respiratory illness for children younger than 5 years during 12 months of surveillance in Urban Dhaka, Bangladesh

Risk factor	Coughing, OR (95% CI)	P value	Rapid breathing, OR (95% CI)	P value	Difficulty breathing, OR (95% CI)	P value	Childhood respiratory illness,* OR (95% CI)	P value	Pneumonia,† OR (95% CI)	P value
Household animal ownership	1.07 (0.98–1.18)	0.15	0.80 (0.51–1.25)	0.33	1.08 (0.78–1.48)	0.65	1.07 (0.98– 1.17)	0.14	0.86 (0.55–1.35)	0.51
Household bird ownership (includes pigeon, parrot, dove, quail, mynah)	1.14 (1.02–1.28)	0.02	0.87 (0.50–1.51)	0.62	1.09 (0.74–1.62)	0.66	1.14 (1.02–1.28)	0.02	0.93 (0.53–1.62)	0.79
Household dog ownership	1.10 (0.61–2.00)	0.75	1.97 (0.27–14.42)	0.50	2.36 (0.58–9.72)	0.23	1.09 (0.60–1.97)	0.79	2.10 (0.29–15.41)	0.46
Household cat ownership	1.02 (0.78–1.33)	0.89	0.66 (0.16–2.76)	0.56	0.91 (0.36– 2.31)	0.85	1.02 (0.78–1.32)	0.91	0.69 (0.71–2.80)	0.60
Household chicken ownership	1.06 (0.94–1.19)	0.34	0.75 (0.41–1.36)	0.34	0.94 (0.61–1.44)	0.77	1.06 (0.94–1.19)	0.33	0.80 (0.44–1.46)	0.47

OR = odds ratio. CIs were estimated with generalized estimating equations accounting for clustering within households. Significant findings are in bold type. All associations were adjusted for study arm, refrigerator ownership, sleeping room, roof type, wall type, baseline age, and whether the index diarrhea patient was a child younger than 5 years old.

^*Childhood respiratory illness was defined as caregiver-reported coughing, rapid breathing, and difficulty breathing within the past 2 weeks.

^†Pneumonia was defined as coughing and rapid breathing or rapid breathing and difficulty breathing.

DISCUSSION

Few studies have examined the association between animal ownership and respiratory illness in urban LMIC settings. In this prospective cohort study, we assessed risk factors for respiratory illness among children younger than 5 years of age in urban slum areas of Dhaka, Bangladesh. To our knowledge, no previous studies in urban Bangladesh have explored this association. At baseline, nearly half of children had reports of respiratory illness, with reported cough being the highest at 48%. Bird ownership at baseline was found to be associated with childhood respiratory illness during the 12-month surveillance period for children younger than 5 years of age. In our study setting, in urban slums of Bangladesh, most households (> 70%) typically reside in a single room. Birds, such as pigeons, parrots, quails, and mynah birds, are usually kept in cages on the ground indoors where children play. This practice puts children in close proximity to bird dander (droppings, feathers, and other allergens). Most current intervention studies focused on childhood exposure to domestic animals focus on chickens. These findings suggest that interventions aiming to reduce young children’s exposure to domestic animals should extend to include birds other than chickens.

Most previous studies in Bangladesh exploring human health risks from exposure to birds have focused on the risk of pathogenic avian influenza (HPAI H5N1).^{22,23,39–41} Nasreen et al.⁴² discovered that poultry workers who fed, cleaned, and handled sick poultry without washing their hands afterward were at an increased risk of H5N1 infection. In addition, although raising backyard poultry provides important nutrition and income inputs for households in Bangladesh,⁴³ close contact with poultry is a risk factor for zoonotic transmission of H5N1.⁴⁴ A study in Dhaka, Bangladesh, found that young children with mild respiratory illness resulting from HPAI H5N1 had contact with poultry before onset of illness.²⁴ However, in our prospective cohort study, we found that owning birds other than chickens (e.g., pigeons, parrots, ducks, doves, quails, mynah birds) was associated with respiratory illness, whereas there was no association between chicken ownership and respiratory symptoms. This latter finding is similar to that reported by Ercumen et al.⁴⁵ in their study in Uganda, which also found no increased risk of respiratory illness from poultry ownership.⁴⁵ One potential explanation for this finding is that, in our study setting, birds other than chickens mostly stay in cages in the household living space throughout the day and night, whereas chickens roam outside the house and compound to forage for food during the day. Therefore, it is possible that children had more exposure to the bird dander from caged pet birds compared with chickens that spent less of their time in the home. Future studies should look at associations between ownership of specific bird types, and frequency of cleaning bird cages, and respiratory symptoms. Another potential explanation for the finding that chicken ownership was not associated with respiratory symptoms in children is the interplay between animal ownership and nutrition. In Bangladesh and other LMICs, animal ownership has been associated positively with reductions in childhood stunting.^46,47 Chickens are commonly used as a household animal protein source in our study setting, whereas other types of birds are less commonly used for consumption. It is therefore possible that differences in nutritional status with chicken ownership impacted the relationship between chicken ownership and respiratory illness in our study population.

There is a growing body of literature demonstrating an association between diarrheal diseases and respiratory infections in young children,^48,49 with the potential causal pathway being that children with chronic enteric infections may have a suppressed immune system that makes them more susceptible to respiratory infections. Studies from Ghana, Brazil, and Pakistan have all found diarrheal diseases and respiratory infections to be significantly associated.^48,49 This ties into the growing body of literature on the vicious cycle in which chronic enteric infections suppress immune function in young children, making them more susceptible to other infections, which worsens their nutritional status, resulting in growth faltering and poor cognitive outcomes.⁵⁰ We did not find a significant relationship between animals in the household and diarrhea in our cohort; therefore, we do not think this relationship between diarrhea and respiratory infections is driving our observed association between bird ownership and respiratory illness.⁵¹ However, it is important that studies investigating risk factors for respiratory illness in settings with high rates of diarrheal diseases consider the potential for diarrheal diseases to be driving their observed associations.

Most previous interventions around bird ownership have focused on chickens and not pigeons, parrots, doves, quails, or mynah birds.^52,53 In Bangladesh, previous public health interventions related to birds have focused mainly on reducing avian influenza transmission from poultry (chickens, ducks, and pigeons), including a national preparedness and response plan to prevent the spread of avian influenza.⁵⁴ A previous study in Peru found that corralling chickens was associated with greater Campylobacter-associated diarrhea in children than allowing the chickens to range freely.⁵³ Therefore, corralling chickens as an intervention should be approached with caution. In rural Democratic Republic of the Congo, indoor animal hutches for rabbits, guinea pigs, turkeys, and ducks were used by the The Reducing Enteropathy, Diarrhea, Undernutrition, and Contamination in the Environment (REDUCE) program to reduce child exposure to fecal pathogens in their indoor living spaces.⁵⁵ In our study setting, households own birds for several reasons, including having them as pets and as household food sources, and to sell at market. One potential intervention to reduce exposure to birds other than chickens is frequent cleaning and disinfection of bird cages.⁵ Bird cages could also be elevated away from where children play. If birds owned do not contribute positively to household income and food sources, recommendations might also be made not to keep pet birds in households with young children.⁵ Furthermore, the delivery of water, sanitation, and hygiene (WASH) WASH interventions to these households, such as promoting handwashing with soap, has the potential to reduce respiratory illness by interrupting pathogen transmission.^56–58

Our study has some limitations. First, we relied only on caregiver-reported respiratory symptoms over the past 2 weeks rather than clinical diagnosis. Second, we did not assess the frequency of child contact with animals. Third, coughing, difficulty breathing, and fast breathing are non-specific relative to respiratory illnesses and may be indicative of other conditions, such as allergic airway disease. Coughing, for example, has been associated with air pollution and allergic airway disease.⁵⁹

This study has several strengths. First, the prospective design of the study allowed us to investigate the association between animal ownership at baseline and subsequent respiratory symptom prevalence over the 12-month study period. Second, monthly surveillance visits during the 12-month study period allowed us to account for seasonal variability in risk factors. Third, we collected data on multiple respiratory symptoms, including coughing, difficulty breathing, and rapid breathing, and we also collected data on different types of animals owned, including poultry and other birds. The reported findings are novel and, to the best of our knowledge, this is the first study to report the association between animal ownership and respiratory illness among children younger than 5 years old living in slums in Dhaka, Bangladesh.

CONCLUSION

Ownership of birds other than chickens was a significant risk factor for developing child respiratory illness among young children living in slum areas of Dhaka, Bangladesh. This finding highlights the need for interventions to reduce young children’s exposure to domestic animals, and the need to consider birds in addition to chickens in these programs.