Risk Factors Associated with Bronchiolitis in Puerto Rican Children

INTRODUCTION

Bronchiolitis is the leading cause of respiratory compromise and hospitalization in infants younger than 2 years of age^1,2. Increasing evidence suggests that bronchiolitis early in life, with disregard to the etiologic agent, is a major risk factor for subsequent wheezing episodes and asthma^3,4. This may occur because viral infections in early childhood interact with dysfunctions in the immune system, increasing the susceptibility for subsequent development of allergies, atopy and asthma⁵; which may in turn inhibit antiviral responses and promote bronchial hyper-responsiveness, therefore triggering a more severe disease^6,7. Approximately 50% of children who experience severe respiratory syncytial virus (RSV) bronchiolitis have a subsequent asthma diagnosis^5,8. Whether viral bronchiolitis is causal or an early manifestation of future asthma remains uncertain^9,10.

Given that the incidence and risk for asthma is 1.5 times higher among children in Puerto Rico when compared to other regions of the United States¹¹; we hypothesize that they may experience specific risk factors early in life, making them more prone to bronchiolitis. Studies suggest that some risk factors associated with bronchiolitis overlap with atopy in its association with asthma^12–14; but little is known about atopy-related risk factors and its’ association with bronchiolitis frequency and severity. Therefore, the aims of this study are to: 1) determine the risk factor(s) associated with bronchiolitis in Puerto Rican children, analyze their distribution based on sex and age group, and correlate their relationship with previous episodes of bronchiolitis; 2) evaluate the association of atopic risk factors with bronchiolitis severity in Puerto Rican children; and 3) evaluate the association between atopic risk factors and environmental factors particular to Puerto Rican children.

MATERIALS AND METHODS

Study design and setting:

We performed a prospective cross-sectional study from June 2014 to May 2015. The study was performed in the Emergency Department (ED) of two community and two university-based hospitals of the Metropolitan area of Puerto Rico (PR). Each site research team consisted of two research coordinators. The research team was trained in recruitment of patients, inclusion and exclusion criteria, parental consent process, questionnaire administration, and use of the bronchiolitis severity assessment tool. The study was approved by the Institutional Review Board of each study site.

Identification of sample:

We included children equal or less than 24 months of age, who were receiving treatment at the ED due to an acute respiratory complaint, with a primary or secondary diagnosis of bronchiolitis, who were born and living in Puerto Rico at the time of recruitment, and whose parents/caretakers provided a signed informed consent. Bronchiolitis was defined as clinical evidence of lower respiratory tract involvement such as wheezing, rhonchi, crackles or chest wall retractions with or without upper respiratory tract involvement¹⁵. We excluded children older than 25 months of age, children known to have another reason for respiratory distress, such as chronic lung disease, bronchopulmonary dysplasia, bronchiectasis, patients with diagnosis of pneumonia by chest radiography, children born premature, with gastroenteritis, liver function impairment or congenital heart disease; and those who were not of Puerto Rican ethnicity as per study definition.

Data collection process

Demographic and clinical data:

Members of the research team administered the study questionnaires to the parents/caretakers of the participants to collect basic demographic information, medical history, family history, social history and household allergen exposure after the ED physician confirmed a diagnosis of bronchiolitis. The ED physician also assessed the bronchiolitis severity using a bronchiolitis severity score, which included respiratory rate, the presence of retractions, peripheral oxygen saturation, auscultatory findings, and total severity score. Detailed information about the bronchiolitis severity score and its implementation has been discussed elsewhere¹⁶.

Environmental data:

To evaluate environmental factors particular to Puerto Ricans, we used data collected by Dr. Benjamin Bolaños-Rosero from the Burkard Air Sampler stationed on the roof (30 meters above ground level) of the Medical Sciences Campus of the University of Puerto Rico, in San Juan. Detailed information about the air sampler, and the quantification of aeroallergens such as spores and pollen has been published elsewhere¹⁷. Daily weather conditions for precipitation (inches), dew point, and temperature (F˚) were provided by the National Oceanic and Atmospheric Administration from the primary weather station located at the San Juan International Airport¹⁸.

Study questionnaire and variable definitions:

The study questionnaire was modeled after the RSV Bronchiolitis in Early Life (RBEL) prospective cohort study^19,20, and consisted of 59 questions divided into 10 sections (plus demographics). The questionnaire assessed the following: a) country of nationality (7 questions), b) past medical history (PMH) per physician diagnosis (17 questions), c) prenatal history (3 questions), d) comorbidities (6 questions), e) family history (8 questions), f) social history (8 questions), g) household allergen exposure (7 questions), and h) socioeconomic information (3 questions). Demographic data included age, sex, municipality of origin, and ethnicity. Socioeconomic information included parental marital status, median household income, and type of insurance. We defined Puerto Rican ethnicity as a child born and living in Puerto Rico. The maternal, paternal and parental grandparents’ ethnicities were also recorded. We selected potential bronchiolitis risk factors based on the current literature^{7,12–13,15,18,21–34}. We defined risk factors as atopic when they were related to the development of atopy, atopic dermatitis/eczema, allergic rhinitis and/or asthma in the setting of bronchiolitis. These include: history of bronchiolitis^{12–14,19,21,23–25,34} (i.e. previous episode of bronchiolitis, prior hospitalization, intensive care admission); past medical history^{7,12–13,20,22,30,32–35} (i.e. asthma, atopic dermatitis, allergic rhinitis, and allergies); maternal smoking and asthma during pregnancy^19,29; family history^12,19,27 (i.e. asthma, atopic dermatitis, allergic rhinitis, and allergies); and household allergen exposure^{19–20,26,28,31} (i.e. cigarette or second hand smoke, cat, dog, bird, rodent, roach, and mold). We defined risk factors as non-atopic when they were related to bronchiolitis, without an association to atopy, atopic dermatitis/eczema, allergic rhinitis or asthma. The study questionnaire is available upon request.

Outcome measures:

The primary outcome measures were: 1) relationship between atopic risk factors and bronchiolitis frequency in Puerto Rican children; 2) association between atopic risk factors and bronchiolitis severity in Puerto Rican children. Other outcomes of interest included the association between atopic risk factors and environmental exposures.

Statistical Analysis:

A sample size of 673 patients was estimated to detect a correlation coefficient of 0.5 at an alpha error of 5% and beta error of 20%. For patients who contributed duplicate data as a result of multiple recruitments during the 12-month study period, only the last known ED visit was included in the study. A descriptive analysis was carried out to illustrate the characteristics of the population under study and the prevalence of atopic and non-atopic factors. An atopic factor score was calculated for this study, and measured as a conceptually continuous variable (1 point for each atopy-related factor that the parent/caretaker answered as positive, providing a score ranging from 0–16). The atopic factor score was not validated, and was used solely to quantify the cumulative effect of atopic risk factors. A normality test was used to assess quantitative variables (bronchiolitis severity score, episodes of bronchiolitis, atopic factor score and number of hospitalizations), and as this assumption was violated, non-parametric tests were chosen. Mann-Whitney U test was performed to compare the bronchiolitis severity scores with atopic and non-atopic risk factors. Spearman’s rank correlation was used to evaluate the strength of the association between episodes of bronchiolitis, total severity score, atopic factors score and number of hospitalizations. Chi-square test or Fisher’s exact tests were used to evaluate the relationship between age and sex with atopic and non-atopic factors. Logistic regression models were constructed to identify the predictors of patients having previous episodes of bronchiolitis. Univariate analyses were conducted first, with only the variables significantly associated with the outcome included in the final multivariate model. A critical p value of 0.05 was used for all tests.

A descriptive graphical visual analysis was performed to evaluate the association between atopic factors and environmental exposures. Means of aeroallergenic environmental exposures (i.e. temperature, precipitation, dew point, fungal spores and pollen) were calculated and plotted against the number of bronchiolitis cases recruited each month. A statistical test for this association was not conducted. All analyses were conducted in IBM SPSS Statistics 20 (IBM Corp., Armonk, NY, USA).

RESULTS

During the study period, we identified 673 ED visits with a diagnosis of bronchiolitis, of which 650 visits met inclusion criteria. From these visits, we identified 553 patients with one ED visit, 44 patients with 2 ED visits, and 3 patients with 3 ED visits, for a final sample of 600 patients. The patients’ demographic characteristics are detailed in Table 1. About 70% of bronchiolitis episodes occurred in infants younger than 1 year old, with more episodes occurring in the 4–12 month age group (48.2%). Most patients were male (60%), third generation Puerto Rican (60.4%), with married or cohabitating parents (67%).

Table 1:

Distribution of demographic characteristics and bronchiolitis risk factors of participating patients

Demographic Characteristics	Patients N=600 n (%)
Age distribution
0 to 3 months	130 (21.7)
4 to 12 months	289 (48.2)
> 12 months	181 (30.1)
Sex
Male	360 (60)
Female	240 (40)
Ethnicity (N=588) *
Third generation Puerto Rican	355 (60.4)
At least one generation Puerto Rican	233 (39.6)
Parental marital status
Single/ Separated	198 (33)
Married/ Cohabitating	402 (67)
Household income (N=524) ¥
≤ $10,000	246 (46.9)
$10,001-$30,000	190 (36.3)
> $30,000	88 (16.8)
Health insurance
Public	478 (79.7)
Private	114 (19)
None	8 (1.3)
Atopic Risk Factors
History of bronchiolitis
Previous episode	317 (52.8)
Hospitalization	225 (37.5)
PICU admission	41 (6.8)
Past medical history
Asthma	89 (14.8)
Atopic dermatitis (i.e. eczema)	102 (17)
Allergic rhinitis (i.e. seasonal allergies)	85 (14.2)
Allergies∫	56 (9.3)
Maternal history during pregnancy
Smoking	14 (2.3)
Asthma exacerbation	78 (13)
Family history $
Asthma	176 (29.3)
Atopic dermatitis	41 (6.8)
Allergic rhinitis	133 (22.2)
Allergies	91 (15.2)
Household allergen exposure
Cigarette or second hand smoke	146 (24.3)
Cat	53 (8.8)
Dog	228 (38)
Bird	61 (10.2)
Rodent	21 (3.5)
Roach	92 (15.3)
Mold	74 (12.3)
Non-Atopic Risk Factors
Past medical history
Hospitalization due to pneumonia (n=598)≠	43 (7.2)
No breastfeeding	126 (21)
Mode of delivery (n=599)
Vaginal delivery	300 (50.1)
Cesarean delivery	299 (49.9)
Social history
Overcrowding†	288 (48)
Daycare attendance	181 (30.2)
Spends time with 2 or more kids	370 (61.7)

PICU=Pediatric Intensive Care Unit.

^*Results reflect percentages of those who responded; unknown ethnicity for at least one family member was reported for 12 patients.

^¥Results reflect percentages of those who responded; multiple participants preferred not to answer (N=524).

^$Results reflect percentages based on having at least one family member (i.e. mother, father or sibling) with positive medical history for the condition.

^∫Refers to the presence of food or medicinal allergies.

^≠Results reflect percentages based on those who were previously hospitalized.

^†Results reflect percentages based on the number of household members, number of bedrooms and the parental marital status.

Risk factors associated with bronchiolitis in Puerto Rican children

The distribution of bronchiolitis atopic and non-atopic risk factors is detailed in Table 1. A previous episode of bronchiolitis (PEB) was reported in 11.5% of patients from the 0–3 month age group, in 56.7% of patients from the 4–12 month age group, and in 76.2% of patients in the >12 month age group. There was significant difference in the reported rate of PEB and prior hospitalization due to bronchiolitis in both males and females based on age group (p<0.001) (Table 2). However, differences in admission to the pediatric intensive care unit (PICU) due to bronchiolitis was only significant in females based on the 0–3, 4–12, and >12 month age group (0% vs 5.6% vs 10.7%, respectively; p=0.022). Among the atopic diseases reported in the PMH, atopic dermatitis had a higher prevalence at an earlier age; 11.7% in the 0–3 month age group, and 39.9% in the 4–12 month age group. PMH of asthma, atopic dermatitis (AD), and allergic rhinitis (AR) was significant across age groups for both males and females (p<0.05). Males had a higher rate of asthma and AD across all age groups when compared to females (p<0.05). Females, however, had a higher rate of allergic rhinitis than males in the 0–3 month age group (5.3% vs 4.1%; p<0.05). Maternal asthma during pregnancy was also statistically significant for females across age groups (p=0.015). Subgroup analyses showed no significant difference in the presence of atopic factors, non-atopic factors and household allergen exposure between males and females at 0–3 months of age. Males in the 4–12 month age group had higher rate of maternal asthma during pregnancy (16.6% vs 7.4%, P=0.048) and maternal smoking during pregnancy (3.9% vs 0%, P=0.048), than their female counterparts. Males in this age group were also found to have significantly higher rate of single/separated parents (42.5%; p=0.004). Lastly, males in the 4–12 month age group were significantly less likely to be exposed to cats when compared to females of the same age group (5.5% vs 15.7%; p=0.004). PMH of asthma was more prevalent in the >12 month age group (55%), compared to AR (49.5%) and AD (36.4%). The >12-month age group showed no significant difference in the presence of atopic factors or household allergen exposure between male and females. Males of this age group however, were more likely to have higher rate of asthma (28.3% vs 26.7%, P<0.001), allergic rhinitis (25.5% vs 24%, P<0.05), and public health insurance (ie, Medicare/Medicaid) (81.1% vs 66.7%; p=0.046), than females of the same age group.

Table 2.

Distribution of atopic factors in males and females per age group.

MALES
Atopic factors	Age group
	All patients (N=360) n (%)	0–3 months (N=73) n (%)	4–12 months (N=181) n (%)	>12 months (N=106) n (%)	P value*
Previous episode of bronchiolitis	207 (57.5)	11 (15.1)	110 (60.8)	86 (81.1)	<0.001
Hospitalization due to bronchiolitis	143 (39.7)	8 (11)	74 (40.9)	61 (57.5)	<0.001
Past medical history
Asthma	58 (16.1)	3 (4.1)	25 (13.8)	30 (28.3)	<0.001
Atopic dermatitis	71 (19.7)	6 (8.2)	42 (23.2)	23 (21.7)	0.021
Allergic rhinitis	52 (14.4)	3 (4.1)	22 (12.2)	27 (25.5)	<0.001
Household allergen exposure
Bird	38 (10.6)	7 (9.6)	13 (7.2)	18 (17)	0.032
FEMALES
Atopic factors	Age group
	All patients (N=240) n (%)	0–3 months (N=57) n (%)	4–12 months (N=108) n (%)	>12 months (N=75) n (%)	P value*
Previous episode of bronchiolitis	110 (45.8)	4 (7)	54 (50)	52 (69.3)	<0.001
Hospitalization due to bronchiolitis	82 (34.2)	3 (5.3)	39 (36.1)	40 (53.3)	<0.001
PICU admission due to bronchiolitis	14 (5.8)	0 (0)	6 (5.6)	8 (10.7)	0.022
Past medical history
Asthma	31 (12.9)	0 (0)	11 (10.2)	20 (26.7)	<0.001
Atopic dermatitis	31 (12.9)	2 (3.5)	18 (16.7)	11 (14.7)	0.049
Allergic rhinitis	33 (13.8)	3 (5.3)	12 (11.1)	18 (24)	0.005
Maternal asthma during pregnancy	27 (11.2)	4 (7)	8 (7.4)	15 (20)	0.015

PICU=Pediatric Intensive Care Unit.

^*P-values were computed using Chi-square or Fisher’s exact test, as appropriate.

^*All other atopic factors not shown were found to be statistically non-significant.

No difference was found between the mean atopic factor score for males (4.56; SD 2.51) and females (4.42; SD 2.58, respectively, p=0.474). There was no association between age and sex in relation to prenatal smoke exposure, family history of asthma, AD, AR, or allergies, and cigarette exposure. Biologic household allergens were not found to be significant in relation to bronchiolitis episodes. Analysis showed a positive association between atopy-related factors, maternal asthma during pregnancy, daycare attendance and second-hand smoke exposure with previous episodes of bronchiolitis. The strongest predictors of bronchiolitis frequency in multivariate analysis were older age, male sex, history of asthma, history of allergic rhinitis, and second-hand smoke exposure by a caretaker (Table 3).

Table 3.

Relationship between previous episode of bronchiolitis (PEB) and selected demographic characteristics, atopic and non-atopic factors^*.

	Univariate analysis		Multivariate analysis
Characteristic	OR	95% CI	OR	95% CI
Age
0–3 months	1		1
4–12 months	10.06	5.60–18.08	9.20	4.92–17.20
>12 months	24.61	13.00–46.56	18.38	9.24–36.58
Sex
Female	1		1
Male	1.60	1.15–2.22	1.67	1.12–2.49
PMH of asthma
No	1		1
Yes	16.38	7.02–38.18	8.96	3.65–22.0
PMH of atopic dermatitis
No	1		1
Yes	1.55	1.00–2.40	0.79	0.47–1.34
PMH of allergic rhinitis
No	1		1
Yes	5.55	3.05–10.09	3.67	1.81–7.43
Maternal asthma during pregnancy
No	1		1
Yes	2.55	1.51–4.29	1.67	0.89–3.14
Daycare attendance
No	1		1
Yes	1.59	1.11–2.26	1.10	0.72–1.68
Cigarette exposure
No	1		1
Yes	2.07	1.24–3.46	2.37	1.26–4.43

OR=Odds ratio; PMH=Past medical history.

^*All other variables excluded from the model were not significantly associated with PEB in the univariable analysis (p>0.05).

Association between atopic factors in Puerto Rican children and bronchiolitis severity

The correlation between atopic risk factors and components of the bronchiolitis severity score are detailed in Table 4. The clinical parameters that were most commonly associated with atopic risk factors were auscultatory findings, respiratory rate (RR), and total severity score. A PEB and PICU admission due to bronchiolitis were correlated with higher total score, RR, and auscultatory findings (p<0.05). PMH of asthma was associated with higher RR and total severity score (p<0.05), and PMH of AD was associated with auscultatory findings (p=0.048). Household allergen analysis showed that dog exposure was associated with higher RR, retractions, auscultation and total severity score (p<0.05), whereas roach exposure correlated only with auscultatory findings (p=0.030). Overcrowding was associated with higher RR (p=0.011); and a higher income was associated with higher auscultation score (p=0.046). Having at least 1 family member not born in PR, and married/cohabitating parents was associated with a higher auscultation and total score (p<0.05).

Table 4.

Correlation between atopic risk factor and components of the bronchiolitis severity score (N=600).

	Scores
Atopic risk factors	Respiratory rate		Retractions		Oxygen saturation		Auscultation		Total
	Median	Range	Median	Range	Median	Range	Median	Range	Median	Range
Previous episode of bronchiolitis	1	1–3	NS		NS		2	0–3	3	1–9
p-value	0.048						0.021		0.026
PICU admission	1	1–3	NS		NS		2	0–3	4	1–9
p-value	0.024						0.048		0.002
PMH asthma	1	1–3	NS		NS		NS		4	1–9
p-value	0.005								0.028
PMH atopic dermatitis	NS		NS		NS		2	0–3	NS
p-value							0.048
Dog exposure	1	1–3	0	0–3	NS		2	0–3	3	1–9
p-value	0.026		0.014				0.046		0.008
Roach exposure	NS		NS		NS		2	0–3	NS
p-value							0.030
Overcrowding	1	1–3	NS		NS		NS		NS
p-value	0.011
At least 1 family member not born in PR	NS		NS		NS		2	0–3	3	1–9
p-value							0.023		0.035
Married/cohabitating	NS		NS		NS		2	0–3	3	1–10
p-value							0.028		0.028

PICU=Pediatric Intensive Care Unit; PMH=Past medical history; PR=Puerto Rico.

^*P-values were computed using Mann-Whitney test. NS- Not significant at the 0.05 level.

A previous episode of bronchiolitis was correlated with a higher total severity score, atopic factor score and number of hospitalizations (Table 5). The atopic risk factor score was correlated with PEB and hospitalization due to bronchiolitis (p<0.001). There was no correlation between the atopic score and total severity score (p=0.230).

Table 5.

Correlation between having a previous episode of bronchiolitis (PEB) bronchiolitis severity, atopy and number of hospitalizations (N=600).

All patients		Episodes of bronchiolitis	Total severity score	Atopic factor score	Number of hospitalizations
Episodes of bronchiolitis	Correlation coefficient	1	0.084*	0.614*	0.680*
	p value	-	0.040	<0.001	<0.001
Total severity score	Correlation coefficient	0.084*	1	0.049	0.049
	p value	0.040	-	0.230	0.229
Atopic factor score	Correlation coefficient	0.614*	0.049	1	0.506*
	p value	<0.001	0.230	-	<0.001
Number of hospitalizations	Correlation coefficient	0.680*	0.049	0.506*	1
	p value	<0.001	0.229	<0.001	-

^*Spearman rank correlation coefficient is significant at the 0.05 level (2-tailed).

Association between atopic factors and environmental factors particular to Puerto Rican children

Temporal analysis between environmental factors and number of cases showed an inverse relationship between the highest peaks of mean temperature and number of bronchiolitis cases (Figure 1). It also showed a congruent distribution between increase and decrease of bronchiolitis cases in relation to precipitation and dew point, and a similar tendency with the aeroallergen distribution of fungal spores. Temporal analysis between bronchiolitis cases and pollen did not show any discernable distribution. There was no relationship between bronchiolitis severity, atopic score, and aeroallergens.

Figure 1.

Association between monthly cumulative frequency of bronchiolitis cases from participating patients and aerobiologic factors based on temporal distribution (N=600).

DISCUSSION

Our study shows that older age, male sex, history of atopy and allergies, maternal asthma during pregnancy, daycare attendance and second-hand smoke exposure are associated with previous episodes of bronchiolitis. These findings are consistent with other studies, especially those that have found a correlation between the number of bronchiolitis episodes and subsequent development of asthma^13,23,34. Other studies have evaluated the association between children with bronchiolitis and atopy and their subsequent development of asthma. A study by Waseem et al.¹² reported similar predictive factors of asthma, including atopy, allergies, and family history of asthma, in a cohort of 1991 children with bronchiolitis, of which 69% were Latino. The relationship between bronchiolitis and other atopic factors such as maternal smoking and asthma during pregnancy^6,24,29, and history of allergic rhinitis and atopic dermatitis^{6,13,24, 27}, have also been associated with asthma.

Our study showed a history of atopy, as early as 3 months compared to 6 months as previously reported³⁵, suggestive of an association with the atopic march, which establishes that AD is a prerequisite for the development of allergic rhinitis and a major risk factor for the development of asthma²². Asthma was the most predominant atopic disease in the >12 month age group. Older children (i.e >12 month age group) have had a longer lifespan in comparison to the other 2 age groups, and may have been more likely to have had previous episodes of bronchiolitis, which may explain a diagnosis of asthma at an early age; as well as longer period of exposure for the development of AD and AR. However, a study by Mikalsen et al.³⁶, where he evaluated the occurrence of asthma and atopy 11 years after hospitalization for bronchiolitis in the first year of life, argued that the development of asthma after bronchiolitis is not associated with atopy, but more so with gender and type of virus. Some studies have reported that at different ages, sex plays an important role in wheezing disorders³⁷, which may help explain why there is a variable distribution of atopic disorders among different age groups per sex. However, we did not find a statistical significance between sex and mean atopic score.

Over 50% of the recruited patients had a previous episode of bronchiolitis, for which 40% had been hospitalized. Our findings show that atopic factors were associated with higher frequency of bronchiolitis episodes and number of hospitalizations. Although there was no significant correlation between atopic factors and total severity score, there was an indirect correlation to severity through hospitalization as per Willson et al³⁸. These findings are supported by Murray et al.³⁹ who showed that although children in atopic groups do not suffer a higher incidence of viral infections; the allergen-sensitized group however, experienced a 47% increase in viral respiratory tract illness and severity, compared to nonsensitized groups. Although household allergens were not found to be significant in relation to previous episodes of bronchiolitis, dog and roach exposure were found to be associated with higher degree of bronchiolitis severity. This supports the theory that there is synergism between atopy and early-life respiratory viral infections, which is further exacerbated by the presence of allergy sensitization, atopy and exposure to aeroallergens, such as cockroach, dust mite, and cat allergens⁷. Our study shows that about 25% of patients were exposed to second-hand smoke. According to Martell et al.⁴⁰ Puerto Rico’s smoking prevalence is 29%; compared to CDC’s Morbidity and mortality report who published that in 2017, 9.9% of Hispanic adults in the US were smokers^41,42. This shows that smoking prevalence is higher in Puerto Rico and thus the children are experiencing higher rates of second-hand smoke exposure compared to Hispanic children living in other parts of the US. Lastly, our study showed that the presence of multiple primary caretakers in the household, and a higher household income may be associated with a higher tolerance for respiratory distress in the child with bronchiolitis, and a longer wait time until worsening respiratory status merits a visit to the ED. This suggests that lower socioeconomic status promotes health disparities in resource utilization⁴³.

Our findings suggest that preventive measures may be implemented to diminish the impact of modifiable atopic factors in the development of bronchiolitis and asthma later in life. Studies by Simpson et al.⁴⁴, and Lowe et al.⁴⁵ with 124 infants and 80 infants, respectively, showed that daily or twice daily emollient use reduced the incidence of atopic dermatitis in at-risk infants at six months. Maternal asthma during pregnancy is an easily identifiable risk factor, which can be ameliorated with family education, adequate follow up, compliance with maintenance medications, and reinforcement of avoidance measures associated with asthma exacerbation²⁹. Lastly, second-hand smoke exposure is a preventable factor that highlights the importance of efforts to prevent tobacco use and promote smoking cessation in population of all ages.

Environmental factors influence the efficacy of distribution of aerobiological factors, including the transmission of infectious diseases⁴⁶. In tropical and semitropical climates, transmission of respiratory viruses (i.e. RSV) occur throughout the year^37,47 with outbreaks associated with the rainy season^48–50. Our study shows that the number of bronchiolitis cases decreased with the highest peaks of mean temperature. As temperature rises, virus survival decreases⁴⁶. An increase in fungal spores can be observed after a 1-month lapse in a precipitation increase. Both increases are related to a surge in bronchiolitis cases. Due to this relationship, one may infer that bronchiolitis cases may be associated with aeroallergens, when in fact it may just be a coincidence based on meteorological factors that affect the dissemination of infectious particles. This information suggests a possible relationship, but due to the nature of our cross sectional study and variation of values in a month, we are unable to assess cause and effect. We did not evaluate the exact dates when patients were recruited (to see if they were preceded or not by a high or low value of X factor on that day or the previous ones), and considering all other factors that could be associated with bronchiolitis, we can not conclude that bronchiolitis cases occurred strictly because of the environmental “highs or lows”.

However, it is important to note that early life sensitization to aeroallergens has been associated with increased risk for development of asthma later in life⁶.

Limitations

The main limitation of our study was reporting and recall bias, which may have over- or underestimated the rate of previous episodes of bronchiolitis, since we did not verify the number of previous episodes of bronchiolitis with a medical chart review. The diagnosis of asthma, allergic rhinitis and atopic dermatitis was based on physician diagnosis as reported by the parent/caretaker, and not on evidence of allergic sensitization, which may have over- or underestimated the rate of atopic past medical history in our population. Ethnicity may have also been a confounder due to its definition based on nationality, for which we focused the population under study as third generation Puerto Rican. We limited our study to the main Metropolitan area, which has been shown to have higher number of hospitals, urbanization, and industrialization to better correlate for the presence of aeroallegens. We did not assess other regions of the island or variables that may have affected bronchiolitis severity, such as etiologic agent, laboratory values, and measurements of household allergens, among others. Because of nuances in the evaluation of environmental aeroallergens, we were unable to establish a correlation between bronchiolitis cases and environmental factors. Since our conclusions in this area are limited, we suggest that future studies develop a cohort design and study this relationship more thoroughly. This study however, benefited from being the first study that seeks to understand the prevalence and correlation of atopy in Puerto Rican children with bronchiolitis. We used a standardized instrument to assess bronchiolitis severity, and the study was performed throughout the year to assess epidemiologic behavior of bronchiolitis over a 12-month period. Therefore, we believe that the study is generalizable to Puerto Rican children with low sociodemographic status in urban areas.

Conclusion

We have identified that atopy-related risk factors are directly correlated with frequency and severity of bronchiolitis. Puerto Rican children present risk factors related to atopy earlier than reported in the literature, some of which may be modified to prevent the development of asthma later in life. This study may help identify Puerto Rican children who may present a severe clinical course of disease, without the traditional risk factors for severe bronchiolitis. It may also help improve prevention strategies, and treatment of bronchiolitis cases. Lastly, this study helps us better understand the epidemiology of bronchiolitis in Puerto Rico and its long-lasting repercussions in the healthcare system. Ultimately, the bronchiolitis-atopy-asthma triad requires further study, with special consideration of ethnic subgroups in their country of origin, which places them at higher risk of developing asthma.