Abstract
Background
While sexual dimorphism in wheeze and asthma prevalence are well documented, sex-specific risk factors for wheeze and longitudinal wheeze phenotypes have not been well elucidated.
Objective
Using a large pre-birth cohort, this study aimed to identify sex-specific risk factors for wheeze from birth through mid-childhood, and identify distinct longitudinal wheeze phenotypes and the sex-specific risk factors associated with these phenotypes.
Methods
Mothers reported child wheeze symptoms over the past year, approximately yearly on 9 occasions starting at age 1 year. We identified sex-specific predictors of wheeze, wheeze phenotypes, and sex-specific predictors of these phenotypes using generalized estimating equations, latent class mixed models, and multinomial logistic analysis, respectively.
Results
A total of 1623 children had information on wheeze at 1 or more time points. Paternal asthma was a stronger predictor of ever wheeze in boys (OR=2.15, 95% CI 1.74, 2.66) than in girls (OR=1.53, 95% CI 1.19, 1.96, p for sex by paternal asthma interaction=0.03), while being Black or Hispanic, birthweight for gestational age z-score, and breastfeeding duration have stronger associations among girls. We identified 3 longitudinal wheeze phenotypes: never/infrequent wheeze (74.1%), early transient wheeze (12.7%), and persistent wheeze (13.1%). Compared to never/infrequent wheeze, maternal asthma, infant bronchiolitis, and atopic dermatitis were associated with persistent wheeze in both sexes, but paternal asthma was associated with persistent wheeze in boys only (OR=4.27, 95% CI 2.33, 7.83, p for sex by paternal asthma interaction=0.02) while being Black or Hispanic was a predictor for girls only.
Conclusion
We identified sex-specific predictors of wheeze and longitudinal wheeze patterns, which may have important prognostic value and may allow for a more personalized approach to wheeze and asthma treatment.
Keywords: asthma, sex differences, parental asthma, bronchiolitis
Background
Sexual dimorphism in wheeze and asthma prevalence and their associated morbidities is well-documented, with a higher male-to-female ratio in childhood and a reversal of this ratio during adolescence, which persists into adulthood(1-7). It has been suggested that dysynapsis and the relative collapsibility and size of airways may predispose male infants to wheeze compared to females(8). Other hypotheses include female and male hormonal influences (9-11) and differences in consultation practices(12).
While several large cross-sectional and longitudinal studies have identified risk factors for wheeze and asthma, few have looked specifically at sex-specific risk factors. Instead, most simply adjust for sex in their multivariate analytic models. The identification of sex-specific risk factors has important implications for asthma prediction and prevention, as well as in our understanding of the genetic determinants of asthma. Using a large birth cohort of children with a parental history of allergy or asthma and follow-up to age 14 years, we found that paternal asthma and infant bronchiolitis were risk factors for wheeze in boys only, while maternal asthma was a risk factor for both sexes(13). Interestingly, the magnitude of the associations of paternal asthma and infant bronchiolitis with wheeze in boys persisted throughout childhood, suggesting the long-term impact of these factors on the longitudinal trajectory of wheeze may also be sex-specific.
Recently, trajectories of wheeze have been identified using longitudinal latent class analyses in large birth cohorts(14). These trajectories are consistent with previously documented patterns(15) and a few studies have identified risk factors associated with these trajectories(16-18). For example, in models including boys and girls, parental asthma and aeroallergen sensitization were a stronger predictor of persistent wheeze while early respiratory infections, breastfeeding, and daycare attendance were predictors of early transient wheeze(17, 18). However, sex-specific risk factors, which are important from a prognostic point of view, have not been examined.
Using a large birth cohort of children not selected for atopy risk, this study examined sex-specific risk factors for wheeze from birth through mid-childhood. Furthermore, distinct longitudinal phenotypes of wheeze were identified and sex-specific risk factors for these trajectories were examined.
Methods
Subjects
Project Viva is a prospective pre-birth cohort study whose goal is to examine associations of prenatal and perinatal factors with maternal and child health(19). Mothers were recruited at their first prenatal appointment at Atrius Harvard Vanguard Medical Associates, a multispecialty group practice in Massachusetts. This cohort consisted of 2128 singleton liveborn infants. At in-person and mailed visits, mothers were invited to respond to a questionnaire about the child's health at birth, six months, one year and at approximately one-year intervals thereafter.
Ascertainment of wheeze and exposures
For this study, we used maternal reports of child wheeze for questionnaire year 1 through 9 inclusively, corresponding approximately to yearly assessments between ages 1 and 9 years. Wheeze was considered present if the parent answered “yes” to the question: “Since your child was [age at last questionnaire], had he/she ever had wheezing (or whistling in the chest)?” Children with at least 1 wheeze outcome available from questionnaire year 1 through 9 were included (n=1623). We ascertained most demographics data, parental medical history, and the child's characteristics through maternal reports at the enrollment into the study or at the child's birth and assessed bronchiolitis and atopic dermatitis before 1 year of age at the questionnaire year 1. Environmental smoke exposure before 1 year was considered present if the parents reported that the child was exposed to at least 1h of cigarette smoke per week on the 6 months or 1 year questionnaire.
Statistical analysis
We performed a descriptive analysis of the baseline characteristics of the entire cohort and by sex. We used generalized estimating equations (GEE) to identify predictors of wheeze longitudinally, stratified by sex, and incorporating all observed outcomes. An independent correlation structure was used and this model was compared to the unstructured and exchangeable structures using the Quasi-Akaike Information Criterion (QIC). We included the following exposures in the final model, based on our previous study and results from other studies (13, 20-23): maternal and paternal asthma, household income (>$70,000/year vs. ≤$70,000/year), child's age at the time of wheeze ascertainment, child's race/ethnicity, season of birth, gestational length, birthweight for gestational age z-score, breastfeeding duration, environmental smoke exposure before age 1 year, and bronchiolitis and atopic dermatitis before 1 year. Given the variation in ages at the time of yearly questionnaires, we used the exact age in years for each ascertainment time point for all analyses. Other covariates that we considered but did not include in the final model were: maternal smoking during pregnancy, maternal and paternal eczema and hay fever, and child care attendance at 6 months of age. These did not independently predict wheeze (parental eczema, maternal smoking during pregnancy) and/or were correlated with other variables in the model (maternal smoking during pregnancy with environmental smoke exposure, child care attendance with bronchiolitis, parental eczema and hay fever with parental asthma). Additionally, in a sub-analysis, we analyzed the age-specific effects of the covariates on wheeze by stratifying the cohort by age (<5 years and ≥5 years) and sex. If the results suggested an age-specific effect, we further examined the interaction between age and those risk factors on wheeze.
Additionally, we applied latent class mixed modeling to identify distinct phenotypes of wheeze. To increase the validity of the analysis, we included only children who had at least one wheeze measure between 1 year and 5 years and at least one wheeze measure between 6 years and 9 years. We initially performed latent class mixed modeling stratified by sex to compare wheeze phenotypes between boys and girls and assessed the goodness of fit for the number of latent classes using the Bayesian Information Criterion. Based on previous literature(14, 24), we specified 3, 4, and 5 latent classes and 3 latent classes provided the best fit. Because the similarity of the latent classes in both sexes, we subsequently performed this analysis in the entire cohort using the same parameters. We examined sex-specific risk factors for individual patterns through multinomial logistic regression stratified by sex. Covariates included in the multinomial logistic regression were the same as those for the GEE.
To optimize power, we imputed missing exposures in the final model. We used Proc MI in SAS version 9.3 (Cary, NC) and imputed 50 values for each missing observation to create 50 “completed” datasets. We combined multivariable modeling estimates using Proc MIANALYZE. Characteristics were similar in the unimputed and imputed datasets (see Table E1 and E2 in the Online Repository). Multiple imputation also did not substantially change the associations resulting from our analyses. P values are two-sided. We performed analyses using SAS version 9.3 (Cary, NC) and R, version 3.2.1 (www.r-project.org). We performed multiple imputation, GEE, and multinomial logistic regression using proc MI, proc GENMOD and MIANALYZE, and proc LOGISTIC in SAS, respectively. We performed longitudinal latent class analysis using the package “lcmm”(25) in R.
Results
A total of 1623 children in Project Viva had at least 1 wheeze outcome available for questionnaire year 1 through 9; 1574 (97.0%) had at least one wheeze measure between 1 and 5 years and 1344 (82.8%) has at least one wheeze measure between 6 and 9 years (see Table E3 in the Online Repository for detailed distribution of data availability by questionnaire year). A total of 1295 children with at least one wheeze outcome available ≤5 years and >5 years were included in the latent class analysis. Baseline characteristics of the children, their family history of atopy, and important environmental exposures for these 2 subgroups are presented in Table 1 and 2, respectively. These characteristics are similar to those of the entire Project Viva cohort (n=2128, Table E4 in the Online Repository). This is a predominantly white cohort (64.6%) with 13.5% and 11.8% of children having a history maternal and paternal asthma, respectively. The prevalence of bronchiolitis before age 1 year was statistically significantly higher in boys (12.3% in girls and 17.2% in boys). The prevalence of wheeze was higher in boys at each questionnaire year, though it had a decreasing trend with time for both sexes with possibly a slight increase in prevalence at the questionnaire year 9 (Figure 1).
Table 1.
Baseline characteristics of the Viva cohort (children with at least 1 wheeze outcome from questionnaire year 1 to 9) (n=1623).
| Total | Girls | Boys | |
|---|---|---|---|
| n=1623 | n=778 | n=845 | |
| N (%) | |||
| Maternal smoking status | |||
| Never | 1119 (69.0) | 545 (70.0) | 575 (68.0) |
| Former | 321 (19.8) | 156 (20.1) | 165 (19.5) |
| Smoked during pregnancy | 183 (11.3) | 77 (9.9) | 105 (12.5) |
| Maternal asthma | 219 (13.5) | 110 (14.2) | 109 (12.9) |
| Maternal hay fever | 487 (30.0) | 232 (29.8) | 255 (30.1) |
| Maternal eczema | 210 (12.9) | 96 (12.4) | 114 (13.4) |
| Paternal asthma | 191 (11.8) | 85 (10.9) | 106 (12.6) |
| Paternal hay fever | 430 (26.5) | 209 (26.8) | 221 (26.2) |
| Paternal eczema | 96 (5.9) | 42 (5.3) | 55 (6.5) |
| Household income >$70,000/year | 966 (59.5) | 456 (58.6) | 510 (60.3) |
| Child race/ethnicity | |||
| Black | 254 (15.7) | 114 (14.7) | 140 (16.6) |
| Hispanic | 86 (5.3) | 35 (4.5) | 51 (6.0) |
| White | 1049 (64.6) | 518 (66.5) | 532 (62.9) |
| Other | 233 (14.4) | 111 (14.3) | 122 (14.5) |
| Season of birth | |||
| Winter | 410 (25.3) | 199 (25.6) | 211 (25.0) |
| Spring | 430 (26.5) | 207 (26.6) | 223 (26.4) |
| Summer | 426 (26.2) | 201 (25.8) | 225 (26.6) |
| Fall | 357 (22.0) | 171 (22.0) | 186 (22.0) |
| Mean (SD) | |||
| Gestation length, weeks | 39.5 (1.9) | 39.5 (1.8) | 39.4 (1.9) |
| Birthweight for gestational age, z-score | 0.20 (0.96) | 0.18 (0.94) | 0.21 (0.98) |
| Duration of breastfeeding, months | 6.0 (4.6) | 6.1 (4.6) | 5.9 (4.6) |
| N (%) | |||
| Environmental smoke exposure before 1 year | 149 (9.2) | 71 (9.1) | 78 (9.2) |
| Bronchiolitis before age 1 year | 241 (14.9) | 96 (12.3) | 146 (17.2) |
| Atopic dermatitis before age 1 year | 369 (22.8) | 162 (20.8) | 207 (24.5) |
Table 2.
Baseline characteristics of children with at least one wheeze outcome <5 years and one wheeze outcome ≥5 years (n=1295)
| Total | Girls | Boys | |
|---|---|---|---|
| n=1295 | n=630 | n=665 | |
| N (%) | |||
| Maternal smoking status | |||
| Never | 910 (70.3) | 456 (72.4) | 454 (68.2) |
| Former | 262 (20.2) | 124 (19.6) | 138 (20.8) |
| Smoked during pregnancy | 123 (9.5) | 50 (7.9) | 73 (11.0) |
| Maternal asthma | 166 (12.8) | 85 (13.6) | 80 (12.1) |
| Maternal hay fever | 392 (30.2) | 191 (30.3) | 201 (30.2) |
| Maternal eczema | 167 (12.9) | 77 (12.3) | 89 (13.4) |
| Paternal asthma | 155 (12.0) | 76 (12.0) | 80 (12.0) |
| Paternal hay fever | 343 (26.5) | 171 (27.1) | 173 (26.0) |
| Paternal eczema | 79 (6.1) | 36 (5.7) | 43 (6.5) |
| Household income >$70,000/year | 804 (62.1) | 382 (60.7) | 422 (63.5) |
| Child race/ethnicity | |||
| Black | 172 (13.3) | 82 (13.1) | 90 (13.5) |
| Hispanic | 52 (4.0) | 21 (3.3) | 31 (4.7) |
| White | 881 (68.0) | 433 (68.7) | 448 (67.4) |
| Other | 190 (14.7) | 94 (14.9) | 96 (14.4) |
| Season of birth | |||
| Winter | 332 (25.6) | 159 (25.2) | 173 (26.0) |
| Spring | 328 (25.3) | 163 (25.9) | 165 (24.8) |
| Summer | 350 (27.0) | 168 (26.7) | 182 (27.4) |
| Fall | 285 (22.0) | 140 (22.2) | 145 (21.8) |
| Mean (SD) | |||
| Gestation length, weeks | 39.5 (1.8) | 39.5 (1.8) | 39.4 (1.8) |
| Birthweight for gestational age, z-score | 0.19 (0.97) | 0.19 (0.93) | 0.18 (0.99) |
| Duration of breastfeeding, months | 6.4 (4.6) | 6.4 (4.6) | 6.4 (4.6) |
| N (%) | |||
| Environmental smoke exposure before 1 year | 109 (8.4) | 55 (8.7) | 54 (8.1) |
| Bronchiolitis before age 1 year | 188 (14.5) | 74 (11.7) | 114 (17.1) |
| Atopic dermatitis before age 1 year | 297 (22.9) | 135 (21.4) | 162 (24.3) |
Figure 1.
Prevalence of parent-reported wheeze in the past 12 months by age stratified by sex. The curves are calculated using local regression (LOESS).
Sex- and age-specific predictors of wheeze
Risk factors for wheeze among girls included being black (OR=2.36, 95% CI 1.84, 3.04), being Hispanic (OR=3.24, 95% CI 2.13, 4.93), maternal asthma (OR=1.92, 95% CI 1.54, 2.41), paternal asthma (OR=1.53, 95% CI 1.19, 1.96), bronchiolitis before age 1 year (OR=3.51, 95% CI 2.76, 4.47) and atopic dermatitis before age 1 year (OR=1.72, 95% CI 1.39, 2.13) (Table 3). These were also predictors of wheeze in boys, but the effect of being Black (OR=1.54, 95% CI 1.23, 1.93) and Hispanic (OR=1.63, 95% CI 1.15, 2.29) were less important while the effect of paternal asthma (OR=2.15, 95% CI 1.74, 2.66) was more important. There was a significant statistical interaction between sex and being Black (p=0.002), sex and being Hispanic (p=0.01), and sex and paternal asthma (p=0.03).
Table 3.
Predictors of wheeze, stratified by sex*
| Girls | Boys | |
|---|---|---|
| n=778 | n=845 | |
| OR (95% CI) | OR (95% CI) | |
| Maternal asthma | 1.92 (1.54, 2.41) | 1.80 (1.46, 2.22) |
| Paternal asthma** | 1.53 (1.19, 1.96) | 2.15 (1.74, 2.66)† |
| Household income >$70,000/y | 0.84 (0.68, 1.02) | 0.77 (0.64, 0.92) |
| Child race/ethnicity, vs. white | ||
| Black** | 2.36 (1.84, 3.04) | 1.54 (1.23, 1.93) |
| Hispanic** | 3.24 (2.13, 4.93) | 1.63 (1.15, 2.29) |
| Other | 1.22 (0.96, 1.56) | 1.03 (0.83, 1.28) |
| Season of birth, vs. winter | ||
| Spring | 0.93 (0.74, 1.16) | 1.05 (0.86, 1.29) |
| Summer | 0.65 (0.51, 0.83) | 0.85 (0.69, 1.04) |
| Fall | 0.72 (0.56, 0.92) | 1.09 (0.88, 1.34) |
| Gestation length, weeks | 0.94 (0.90, 0.98) | 0.95 (0.91, 0.98) |
| Birthweight for gestational age, z-score** | 0.91 (0.82, 1.00) | 1.07 (0.98, 1.15) |
| Duration of breastfeeding, months** | 0.96 (0.94, 0.97) | 1.01 (0.99, 1.02) |
| Environmental smoke exposure before 1 year | 1.36 (0.97, 1.90) | 1.45 (1.09, 1.92) |
| Bronchiolitis before age 1 year | 3.51 (2.76, 4.47)† | 4.23 (3.40, 5.25)† |
| Atopic dermatitis before age 1 year | 1.72 (1.39, 2.13) | 1.54 (1.28, 1.85)† |
| Child age at outcome, years | 0.94 (0.91, 0.97) | 0.92 (0.90, 0.94) |
Multivariable model adjusted for all variables presented in the table.
Statistically significant interaction with sex on wheeze.
Statistically significant interaction with age on wheeze.
Older age at outcome time point and a longer gestational length were associated with lower risk of wheeze in both sexes. Environmental smoke exposure before 1 year was associated with higher risk wheeze in boys (OR=1.45, 95% CI 1.09, 1.92), although the effect size was similar in girls. Being born in the summer and fall was associated with lower risk of wheeze in girls only. While household income >$70,000 per year was significantly associated with lower risk of wheeze in boys (OR=0.77, 95% CI 0.64, 0.92), the effect sizes were similar though with confidence intervals that straddled the null in the opposite sex. Longer duration of breastfeeding associated with lower wheeze risk in girls (OR=0.96 per month, 95% CI 0.94, 0.97).
In addition to stratification by sex, we also stratified the cohort by age in order to examine the age-specific effects of risk factors in a sub-analysis (Table E5 in the Online Repository). The effect of infant bronchiolitis on wheeze was greater among subjects aged <5 years, compared to those ≥5 years for both sexes (age by infant bronchiolitis interaction p<0.0001 for both boys and girls) while paternal asthma and atopic dermatitis before age 1 year were associated with higher odds of wheeze in those ≥5 years in boys only (age by paternal asthma interaction p<0.0001, age by atopic dermatitis interaction=0.01).
The GEE analyses were performed using an independent correlation structure, which had a high goodness-of-fit (as measured by the QIC) that is comparable to the exchangeable structure and higher than the unstructured model. Of note, GEE provides robust unbiased estimates even in case of misspecification of the correlation structure(26).
Longitudinal wheeze phenotypes and sex-specific predictors of wheeze phenotypes
A total of 1295 subjects had at least one wheeze measure at ≤5 years and at least one wheeze measure >5 years and were included in the latent class analysis, with the majority(n=1135, 87.6%) having 5 or more observations. Three distinct wheeze phenotypes were identified and patterns were similar in girls and boys: never/infrequent wheeze (n=960, 74.1% overall; n=497, 78.9% in girls and n=665, 69.6% in boys), early transient wheeze (n=165, 12.7% overall; n=64, 10.2% in girls and n=101, 15.2% in boys), and persistent wheeze (n=170, 13.1% overall; n=69, 11.0% in girls and n=101, 15.2% in boys). Consistent with the prevalence of wheeze throughout the ages, there were more boys in the early transient wheeze and persistent wheeze groups. These phenotypes were labelled based on their temporal pattern (Figure 2) and based on patterns reported in the previous literature.
Figure 2.
Longitudinal wheeze phenotypes for the entire study cohort. Sex-specific wheeze trajectories were similar in boys and girls. The majority of subjects were in the never or infrequent wheeze group (n=960, 74.1%), with minorities in the early transient wheeze group (n=165, 12.7%) and persistent wheeze group (n=170, 13.1%).
In the multinomial analysis, infant bronchiolitis (OR=4.84, 95% CI 2.34, 9.98 in girls and OR=10.17, 95% CI 5.47, 18.93 in boys) was associated with higher odds of early transient wheeze in boys and girls, compared to never/infrequent wheeze (Table 4). Maternal asthma was also associated with early transient wheeze in boys (OR=3.59, 95% CI 1.90, 6.75) and showed a borderline significance in girls (OR=2.03, 95% CI 0.97, 4.25). Interestingly, the association of infant bronchiolitis with early transient wheeze appeared greater in boys, though the sex by infant bronchiolitis interaction was not statistically significant (p=0.15). The sex by maternal asthma and by atopic dermatitis interactions were also not significant (p=0.41 and 0.32, respectively). In girls only, duration of breastfeeding (OR=0.92, 95% CI 0.86, 0.98) and atopic dermatitis (OR=2.06, 95% CI 1.04, 4.09) were associated with a lower and higher odds of early transient wheeze, respectively.
Table 4.
Multinomial logistic regression for wheeze phenotypes stratified by sex (vs. never/infrequent wheeze)*
| Early transient wheeze | Persistent wheeze | |||
|---|---|---|---|---|
| Girls | Boys | Girls | Boys | |
| n=64 | n=101 | n=69 | n=101 | |
| Maternal asthma | 2.03 (0.97, 4.25) | 3.59 (1.90, 6.75) | 2.13 (1.04, 4.35) | 2.21 (1.11, 4.40) |
| Paternal asthma* | 1.29 (0.59, 2.83) | 0.85 (0.35, 2.09) | 1.46 (0.68, 3.15) | 4.27 (2.33, 7.83) |
| Household income >$70,000/year | 0.78 (0.42, 1.46) | 0.85 (0.48, 1.50) | 0.72 (0.38, 1.36) | 0.72 (0.41, 1.26) |
| Child race/ethnicity, vs. white | ||||
| Black* | 0.56 (0.21, 1.53) | 1.25 (0.57, 2.76) | 3.23 (1.55, 6.75) | 1.50 (0.73, 3.09) |
| Hispanic | 0.69 (0.13, 3.59) | 1.11 (0.35, 3.53) | 3.60 (1.06,12.17) | 1.38 (0.47, 4.08) |
| Other | 0.83 (0.38, 1.82) | 1.27 (0.64, 2.51) | 1.39 (0.63, 3.11) | 0.96 (0.46, 2.01) |
| Season of birth, vs. winter | ||||
| Spring | 0.93 (0.46, 1.88) | 1.74 (0.88, 3.43) | 0.98 (0.46, 2.09) | 0.87 (0.45, 1.69) |
| Summer | 0.54 (0.25, 1.19) | 1.36 (0.69, 2.70) | 0.85 (0.40, 1.84) | 0.82 (0.43, 1.57) |
| Fall | 0.52 (0.23, 1.19) | 1.30 (0.63, 2.70) | 0.67 (0.30, 1.52) | 0.86 (0.44, 1.70) |
| Gestation length, weeks | 0.93 (0.81, 1.07) | 1.05 (0.91, 1.21) | 0.97 (0.84, 1.12) | 0.92 (0.81, 1.04) |
| Birthweight for gestational age, z-score* | 0.76 (0.55, 1.03) | 1.13 (0.87, 1.47) | 0.83 (0.61, 1.13) | 1.11 (0.86, 1.42) |
| Duration of breastfeeding, months | 0.92 (0.86, 0.98) | 0.98 (0.92, 1.03) | 0.97 (0.91, 1.03) | 1.02 (0.97, 1.08) |
| Environmental smoke exposure before 1 year | 1.73 (0.67, 4.44) | 1.02 (0.39, 2.64) | 1.96 (0.81, 4.73) | 2.16 (0.96, 4.89) |
| Bronchiolitis before age 1 year | 4.84 (2.34, 9.98) | 10.17 (5.47,18.93) | 4.38 (2.03, 9.45) | 6.49 (3.29,12.79) |
| Atopic dermatitis before age 1 year | 2.06 (1.04, 4.09) | 1.24 (0.69, 2.23) | 2.88 (1.54, 5.38) | 2.01 (1.16, 3.49) |
Statistically significant interaction with sex on persistent wheeze.
While maternal asthma, infant bronchiolitis, and atopic dermatitis were associated with increased odds of persistent wheeze in both girls and boys (sex by maternal asthma interaction p=0.76, sex by infant bronchiolitis interaction p=0.25, sex by atopic dermatitis interaction p=0.33), paternal asthma was a predictor of persistent wheeze in boys only (OR=4.27, 95% CI 2.33, 7.83, p for sex by paternal asthma interaction = 0.02). Being black (OR=3.23, 95% 1.55, 6.75 in girls vs. OR=1.50, 95% CI 0.73, 3.09 in boys) or Hispanic (OR=3.60, 95% CI 1.06, 12.17 in girls vs. OR=1.38, 95% CI 0.47, 4.08 in boys) were risk factors for persistent wheeze in girls but not boys, with a statistically significant interaction between sex and being black on wheeze (p=0.02).
To further characterize the atopic phenotype within the longitudinal wheeze phenotypes, we documented that the presence of a positive specific IgE to one of 5 inhalant allergens (dust, cat, dog, cockroach, and mold) at age 3 years was more than twice as high for those with persistent wheeze (45%), compared to those with never/infrequent wheeze and early wheeze (20% and 16%, respectively, Table E6). Similarly, the prevalence of mid-childhood (year 7) inhalant sensitization prevalence was 62%, 33%, and 40% among children with persistent, early transient, and never/infrequent wheeze, respectively. For both of these periods, inhalant sensitization prevalence is up to 1.9 times in children with persistent wheeze, compared to the other wheeze phenotypes, while the prevalence of sensitization is more similar among those with early transient and never/infrequent wheeze. If food allergen sensitization and any atopy (total IgE ≥100 IU/mL) is included in mid-childhood, the prevalence of sensitization is similar between the wheeze phenotypes (Table E6). For those with wheezing and with current asthma at year 7, the prevalence of any sensitization was similar (approximately 65%).
Discussion
In this large pre-birth cohort study, we identified sex-specific predictors of wheeze over the first 9 years of life, described distinct longitudinal phenotypes of wheeze using latent class analysis, and explored the sex-specific predictors of these wheeze phenotypes. There were three key findings. First, while some risk factors were associated with longitudinal reporting of wheeze in both boys and girls in patterns consistent with previous studies, paternal asthma had a greater association with wheeze in boys while black or Hispanic race/ethnicity had a greater association in girls. Second, three longitudinal phenotypes of wheeze were identified using latent class analysis: never/infrequent wheeze, early transient wheeze, and persistent wheeze. Finally, as well as finding sexual dimorphism in longitudinal predictors of annual reports of wheeze, we identified the same sex-specific risk factors for the phenotype of persistent wheeze as for longitudinal reporting of any wheeze, namely paternal asthma for boys and being black or Hispanic for girls.
Sexual dimorphism in asthma prevalence and asthma-related morbidities is well-documented (1-7). However, few longitudinal cohorts have examined sex-specific risk factors for wheeze and asthma. Furthermore, sex-specific risk factors for longitudinal wheeze phenotypes have not been examined. In a smaller recent study (n=499) of a birth cohort of children with a parental history of atopy(13), we found that while maternal asthma was a risk factor for wheeze in girls and boys, paternal asthma and infant bronchiolitis were risk factors for boys only. In the current study, using a similar analysis but a cohort of children not selected for risk of atopy and using multiple imputation to account for missing covariates, paternal asthma, but not infant bronchiolitis, showed a statistically significant sex-specific effect on wheeze, though the effect sizes were higher in boys for bronchiolitis. The differences between the two studies may be partly due to analytical technique (no imputation vs. imputation of covariates), characteristics of the birth cohort (high risk for atopy vs. general population), and differences in sample sizes (larger cohort in Project Viva, the current study).
Paternal history of asthma as a predictor of child wheeze: sexual dimorphism
Interestingly, paternal asthma was not only a predictor of any wheeze in boys, it was also a predictor of persistent wheeze in boys, but not in girls. Consistent with the findings of our current and previous study(13), another birth cohort documented that paternal asthma was associated with asthma in boys(27). However, while both our studies found that maternal asthma was associated with wheeze in both sexes, they documented this association with asthma among girls only, though the statistical interaction between sex and maternal asthma was not significant. These differences may reflect differences in the cohort composition or methodological differences, including the use of parental reports of asthma as the outcome (vs. wheeze in our studies) and a longer follow-up to 18 years in their study (27). A recent study demonstrated that paternal imprinting of ADAM33, a gene that has been associated with asthma, was associated with the severity of airway hyperreactivity in children with asthma (28), suggesting a parent-of-origin-specific effect. However, this effect was not examined in the context of the child's sex. On the other hand, Naumova et al recently documented that sex-specific DNA methylation patterns in the promoter region of zona pellucida binding protein 2 (ZPBP2), a gene in the 17q12-21 region, were associated with childhood asthma development(29). Taken together, these studies suggest that the sex-specific effect of paternal asthma may result from complex interactions between genetic and epigenetic factors. Our age-specific analysis demonstrated that the association between paternal asthma and wheeze was greater for children ≥5 years, further reinforcing the notion of this risk factor for the persistent wheeze phenotype. The notion that paternal asthma has a longer lasting effect on wheeze has been reported recently, although sex-specific effects were not examined(30). Further studies are needed to elucidate the mechanisms underlying this observation.
Infant bronchiolitis as a predictor of child wheeze
Our findings indicated that infant bronchiolitis was associated with wheeze in both boys and girls and that this association was strongest in the early ages for both sexes. Although the interaction between sex and bronchiolitis was not statistically significant, the effect of infant bronchiolitis on wheeze in boys was consistently higher than for girls. This is true of infant bronchiolitis as a predictor of wheeze and of the wheeze phenotypes (specifically for early transient wheeze and persistent wheeze). This finding is consistent with our previous study, where we reported an association between infant bronchiolitis and wheeze in boys only. While studies document a higher incidence of bronchiolitis in male infants worldwide (31, 32) and a higher prevalence of wheeze among infants who had bronchiolitis(33, 34), few studies examined the sex-specific effects of infant bronchiolitis on future wheeze. Consistent with our findings, Valkonen et al. documented a higher risk of asthma development at 3-year follow-up in boys with a history of bronchiolitis compared to girls(35). Another study reported increased airway hyperresponsiveness in boys 11 years after infant bronchiolitis(36). Further studies are needed to elucidate the mechanisms underlying this association.
Black and Hispanic race/ethnicity as predictors of wheeze: sexual dimorphism
We found that that being black or Hispanic had a greater effect on wheeze among girls than boys. It was also a predictor of persistent wheeze in girls only, although further studies are needed to confirm this finding given the relatively small numbers of racial/ethnic minorities in each wheeze phenotype group after stratification by sex. The higher prevalence of wheeze and asthma is well documented among Black compared to White children(37-40). Among Hispanic groups, a wide range in asthma prevalence has been documented, with Puerto Ricans and Mexicans having higher and lower prevalence, respectively(39). Interestingly, among Puerto Rican children with asthma, a higher percentage of African ancestry has been associated with decreased lung function (41). Of note, Borrell et al showed that the effect of obesity on asthma control differed by race/ethnicity in girls, with obese African American girls having better controlled asthma compared with their normal-weight counterparts whereas Mexican American girls had worse asthma control(42). This interaction between obesity and race/ethnicity was not seen in boys. Another study in children with asthma showed larger lung function deficits in Hispanic girls compared to non-Hispanic white girls, while this was not statistically significant in boys(43). Taken together, these studies support our findings by suggesting that race/ethnicity may have a sex-specific effect on wheeze and asthma, likely interacting with other environmental and genetic factors.
Immune phenotypes play an important role in the overall asthma phenotype of the patient, with atopy being a well-documented risk factor of persistent wheeze. In this study, we used infant atopic dermatitis as a measure of the immune phenotype of the child. Atopic dermatitis was associated with any wheeze and persistent wheeze in both sexes. Interestingly, while atopic dermatitis was associated with wheeze across the ages in girls, it was more strongly associated with wheeze in later ages in boys, suggesting that other factors, such as bronchiolitis, may be more important in early life. Furthermore, while we did not have data on cord blood IgE, consistent with existing literature(44), we documented a doubled prevalence of inhalant sensitization in early and mid-childhood among children with persistent wheeze, compared to never/infrequent or early wheeze. Specifically, inhalant allergen sensitization may be more indicative of the wheeze phenotype as the prevalence of sensitization was similar between the wheeze phenotypes when food allergen sensitization was also considered.
Three wheeze phenotypes were identified in Project Viva. While this is consistent with other birth cohorts, some have identified intermediate-onset and late-onset wheeze as additional phenotypes, such as the ALSPAC and PIAMA cohorts (14). While the probability of wheeze in the persistent wheeze group remains relatively high (50-90%, Figure 2), we noted that this probability also increased with age. Thus, it is possible that we were unable to differentiate children with intermediate- and late-onset wheeze from those identified as having persistent wheeze in our cohort. This may be due to the relatively smaller sample size compared to the reported cohorts. Furthermore, differences in identification of subgroups may stem from the composition of the birth cohort and methodological differences. While latent class mixed models analysis allows the use of exact ages at the time of outcome assessment, other similar analyses, such as longitudinal latent class analysis, do not. Furthermore, since latent class analysis relies on the availability of longitudinal data, differences in the completeness of the data and the length of follow-up of the cohort may partly explain the discrepancies.
While we focused on the outcome of wheeze in this study, it would be interesting to examine sex-specific risk factors of physician-diagnosed asthma and wheeze without an asthma diagnosis. In the Project Viva cohort, at questionnaire year 7, the prevalence of wheeze in the past 12 months and current asthma were concordant (16.4% and 19.2%, respectively). That is to say, by age 7, very few children with wheeze did not have an asthma diagnosis. Thus, we expect that an analysis with physician-diagnosed asthma would yield similar results, but further studies are needed to confirm this hypothesis. Since the prevalence of potentially undiagnosed asthma is likely low overall, an analysis using this outcome may not be adequately powered in this cohort.
Our study has several strengths. First, we analyzed data from a large pre-birth cohort with regular assessment of respiratory symptoms. Compared to our previous study using the Home Allergens cohort(13), children in Project Viva are not selected for high risk of atopy, making the results of this study are more generalizable to a general population. The use of longitudinal data has several advantages, including regular data collection with short intervals minimizing recall bias and the possibility of following an individual's symptoms over time. In addition, while predictors of wheeze are of epidemiological importance, this study also identified distinct phenotypes of wheeze based on each individual's reports and examined the sex-specific predictors of these phenotypes. Being able to predict the trajectory of a child's symptoms using socio-demographic factors, family history, and early life events is of clinical and prognostic value. Finally, we used multiple imputation for predictors and covariates in this study to deal with missing data, which maximizes statistical power and improves generalizability of the results to the population sampled by this study. Additionally, the results of the study, including the effect sizes of the risk factors, were similar with and without imputation, further validating the appropriate use of the multiple imputation technique. We chose not to impute the outcome of wheeze because it is difficult to be certain that missing outcomes are not Missing Completely At Random in a longitudinal study, in which case multiple imputations would not be valid. To further validate the absence of obvious bias, we found no correlation between missing data at year 9 and previous reports of wheeze at years 1, 3, 5, and 7, and found that the baseline characteristics of the subjects with at least 1 wheeze outcome (n=1623) and those with ≥1 wheeze outcome ≤5 years and ≥1 wheeze outcome >5 years (n=1295) were similar to those of the Project Viva cohort (n=2128).
One limitation of this study is that the cohort was mostly white and relatively well off (almost 60% had a yearly household income >$70,000 at enrollment during pregnancy), which may limit the generalizability of the findings to other populations with a different composition. The assessment of the outcome was based on parental reports but not validated with clinical records. In this cohort, more boys wheezed than girls, with a total of 64 and 69 girls identified as having early transient wheeze and persistent wheeze, respectively. Thus, it is possible that there was limited power to examine predictors of distinct wheeze phenotypes and future larger studies are needed to confirm the current findings. Finally, given that infant bronchiolitis was associated with wheeze, having data on the viruses associated with the bronchiolitis may further provide insight on the mechanism leading to wheeze in children. In particular, respiratory syncytial virus and more recently rhinovirus C have been associated with recurrent wheezing illness(45, 46) and an increased prevalence of subsequent asthma (34, 47). However, we did not have data on specific viruses that caused the bronchiolitis in infancy.
In conclusion, using a large birth cohort of children followed through pre-adolescence, we identified sex-specific risk factors for wheeze and for distinct phenotypes of wheeze. These predictors may have clinical and prognostic significance. Future studies are needed to examine the mechanisms underlying these sex-specific effects.
Supplementary Material
Clinical implications.
This study found sex-specific predictors of wheeze and longitudinal wheeze patterns, including paternal asthma and race/ethnicity. These findings may have prognostic value in the care of children with asthma.
Capsule summary.
While risk factors for wheeze have been extensively studied, we document sex-specific effects of these predictors of longitudinal wheeze phenotypes. This may have important prognostic value in the practice of precision medicine.
Acknowledgments
This paper is subject to the NIH Public Access Policy (http://publicaccess.nih.gov). Project Viva is funded by R01 HL064925, R01 AI102960 and R37 HD 034568. Dr. Oken received funding from NIH (K24 HD069408 and P30 DK092924).
Abbreviations
- OR
Odds ratio
- GEE
Generalized estimating equations
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Venn A, Lewis S, Cooper M, Hill J, Britton J. Questionnaire study of effect of sex and age on the prevalence of wheeze and asthma in adolescence. Bmj. 1998;316(7149):1945–6. doi: 10.1136/bmj.316.7149.1945. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Skobeloff EM, Spivey WH, St Clair SS, Schoffstall JM. The influence of age and sex on asthma admissions. Jama. 1992;268(24):3437–40. [PubMed] [Google Scholar]
- 3.Fu L, Freishtat RJ, Gordish-Dressman H, Teach SJ, Resca L, Hoffman EP, et al. Natural progression of childhood asthma symptoms and strong influence of sex and puberty. Ann Am Thorac Soc. 2014;11(6):939–44. doi: 10.1513/AnnalsATS.201402-084OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schatz M, Camargo CA., Jr. The relationship of sex to asthma prevalence, health care utilization, and medications in a large managed care organization. Ann Allergy Asthma Immunol. 2003;91(6):553–8. doi: 10.1016/S1081-1206(10)61533-5. [DOI] [PubMed] [Google Scholar]
- 5.Almqvist C, Worm M, Leynaert B. Impact of gender on asthma in childhood and adolescence: a GA2LEN review. Allergy. 2008;63(1):47–57. doi: 10.1111/j.1398-9995.2007.01524.x. [DOI] [PubMed] [Google Scholar]
- 6.Debley JS, Redding GJ, Critchlow CW. Impact of adolescence and gender on asthma hospitalization: a population-based birth cohort study. Pediatr Pulmonol. 2004;38(6):443–50. doi: 10.1002/ppul.20108. [DOI] [PubMed] [Google Scholar]
- 7.Anderson HR, Pottier AC, Strachan DP. Asthma from birth to age 23: incidence and relation to prior and concurrent atopic disease. Thorax. 1992;47(7):537–42. doi: 10.1136/thx.47.7.537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hibbert M, Lannigan A, Raven J, Landau L, Phelan P. Gender differences in lung growth. Pediatr Pulmonol. 1995;19(2):129–34. doi: 10.1002/ppul.1950190208. [DOI] [PubMed] [Google Scholar]
- 9.Dratva J, Schindler C, Curjuric I, Stolz D, Macsali F, Gomez FR, et al. Perimenstrual increase in bronchial hyperreactivity in premenopausal women: results from the population-based SAPALDIA 2 cohort. J Allergy Clin Immunol. 2010;125(4):823–9. doi: 10.1016/j.jaci.2009.12.938. [DOI] [PubMed] [Google Scholar]
- 10.Kharitonov SA, Logan-Sinclair RB, Busset CM, Shinebourne EA. Peak expiratory nitric oxide differences in men and women: relation to the menstrual cycle. British heart journal. 1994;72(3):243–5. doi: 10.1136/hrt.72.3.243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yu CK, Liu YH, Chen CL. Dehydroepiandrosterone attenuates allergic airway inflammation in Dermatophagoides farinae-sensitized mice. Journal of microbiology, immunology, and infection = Wei mian yu gan ran za zhi. 2002;35(3):199–202. [PubMed] [Google Scholar]
- 12.Wright AL, Stern DA, Kauffmann F, Martinez FD. Factors influencing gender differences in the diagnosis and treatment of asthma in childhood: the Tucson Children's Respiratory Study. Pediatr Pulmonol. 2006;41(4):318–25. doi: 10.1002/ppul.20373. [DOI] [PubMed] [Google Scholar]
- 13.Tse SM, Coull BA, Sordillo JE, Datta S, Gold DR. Gender- and age-specific risk factors for wheeze from birth through adolescence. Pediatr Pulmonol. 2014 doi: 10.1002/ppul.23113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Savenije OE, Granell R, Caudri D, Koppelman GH, Smit HA, Wijga A, et al. Comparison of childhood wheezing phenotypes in 2 birth cohorts: ALSPAC and PIAMA. J Allergy Clin Immunol. 2011;127(6):1505–12. e14. doi: 10.1016/j.jaci.2011.02.002. [DOI] [PubMed] [Google Scholar]
- 15.Stein RT, Holberg CJ, Morgan WJ, Wright AL, Lombardi E, Taussig L, et al. Peak flow variability, methacholine responsiveness and atopy as markers for detecting different wheezing phenotypes in childhood. Thorax. 1997;52(11):946–52. doi: 10.1136/thx.52.11.946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Caudri D, Savenije OE, Smit HA, Postma DS, Koppelman GH, Wijga AH, et al. Perinatal risk factors for wheezing phenotypes in the first 8 years of life. Clin Exp Allergy. 2013;43(12):1395–405. doi: 10.1111/cea.12173. [DOI] [PubMed] [Google Scholar]
- 17.Lodge CJ, Zaloumis S, Lowe AJ, Gurrin LC, Matheson MC, Axelrad C, et al. Early-life risk factors for childhood wheeze phenotypes in a high-risk birth cohort. The Journal of pediatrics. 2014;164(2):289–94. e1–2. doi: 10.1016/j.jpeds.2013.09.056. [DOI] [PubMed] [Google Scholar]
- 18.Midodzi WK, Rowe BH, Majaesic CM, Saunders LD, Senthilselvan A. Predictors for wheezing phenotypes in the first decade of life. Respirology. 2008;13(4):537–45. doi: 10.1111/j.1440-1843.2008.01284.x. [DOI] [PubMed] [Google Scholar]
- 19.Oken E, Baccarelli AA, Gold DR, Kleinman KP, Litonjua AA, De Meo D, et al. Cohort profile: project viva. International journal of epidemiology. 2015;44(1):37–48. doi: 10.1093/ije/dyu008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chipps BE. Determinants of asthma and its clinical course. Ann Allergy Asthma Immunol. 2004;93(4):309–15. doi: 10.1016/S1081-1206(10)61388-9. quiz 15-8, 80. [DOI] [PubMed] [Google Scholar]
- 21.Burke H, Leonardi-Bee J, Hashim A, Pine-Abata H, Chen Y, Cook DG, et al. Prenatal and passive smoke exposure and incidence of asthma and wheeze: systematic review and meta-analysis. Pediatrics. 2012;129(4):735–44. doi: 10.1542/peds.2011-2196. [DOI] [PubMed] [Google Scholar]
- 22.Turner S. Perinatal programming of childhood asthma: early fetal size, growth trajectory during infancy, and childhood asthma outcomes. Clinical & developmental immunology. 2012;2012:962923. doi: 10.1155/2012/962923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sonnenschein-van der Voort AM, Jaddoe VW, van der Valk RJ, Willemsen SP, Hofman A, Moll HA, et al. Duration and exclusiveness of breastfeeding and childhood asthma-related symptoms. Eur Respir J. 2012;39(1):81–9. doi: 10.1183/09031936.00178110. [DOI] [PubMed] [Google Scholar]
- 24.Chen Q, Just AC, Miller RL, Perzanowski MS, Goldstein IF, Perera FP, et al. Using latent class growth analysis to identify childhood wheeze phenotypes in an urban birth cohort. Ann Allergy Asthma Immunol. 2012;108(5):311–5. e1. doi: 10.1016/j.anai.2012.02.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Proust-Lima C, Philipps V, Diakite A, Liquet B. lcmm: Estimation of latent class mixed models, joint latent class mixed models and mixed models for curvilinear outcomes. R package version 163. 2013 http://CRANR-projectorg/package=lcmm.
- 26.Pan W, Connett J. Selecting the working correlation structure in generalized estimating equations with application to the lung health study. Statistica Sinica. 2002;12:475–90. [Google Scholar]
- 27.Arshad SH, Karmaus W, Raza A, Kurukulaaratchy RJ, Matthews SM, Holloway JW, et al. The effect of parental allergy on childhood allergic diseases depends on the sex of the child. J Allergy Clin Immunol. 2012;130(2):427–34. e6. doi: 10.1016/j.jaci.2012.03.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kopriva F, Godava M, Markova M, Vodicka R, Dusek L, Muzik J, et al. Possible control of paternal imprinting of polymorphisms of the ADAM33 gene by epigenetic mechanisms and association with level of airway hyperresponsiveness in asthmatic children. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2013;157(4):367–73. doi: 10.5507/bp.2013.025. [DOI] [PubMed] [Google Scholar]
- 29.Naumova AK, Al Tuwaijri A, Morin A, Vaillancout VT, Madore AM, Berlivet S, et al. Sex- and age-dependent DNA methylation at the 17q12-q21 locus associated with childhood asthma. Human genetics. 2013;132(7):811–22. doi: 10.1007/s00439-013-1298-z. [DOI] [PubMed] [Google Scholar]
- 30.Paaso EM, Jaakkola MS, Rantala AK, Hugg TT, Jaakkola JJ. Allergic diseases and asthma in the family predict the persistence and onset-age of asthma: a prospective cohort study. Respiratory research. 2014;15:152. doi: 10.1186/s12931-014-0152-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Green CA, Yeates D, Goldacre A, Sande C, Parslow RC, McShane P, et al. Admission to hospital for bronchiolitis in England: trends over five decades, geographical variation and association with perinatal characteristics and subsequent asthma. Arch Dis Child. 2015 doi: 10.1136/archdischild-2015-308723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hasegawa K, Tsugawa Y, Brown DF, Mansbach JM, Camargo CA., Jr. Temporal trends in emergency department visits for bronchiolitis in the United States, 2006 to 2010. Pediatr Infect Dis J. 2014;33(1):11–8. doi: 10.1097/INF.0b013e3182a5f324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Midulla F, Pierangeli A, Cangiano G, Bonci E, Salvadei S, Scagnolari C, et al. Rhinovirus bronchiolitis and recurrent wheezing: 1-year follow-up. Eur Respir J. 2012;39(2):396–402. doi: 10.1183/09031936.00188210. [DOI] [PubMed] [Google Scholar]
- 34.Sigurs N, Aljassim F, Kjellman B, Robinson PD, Sigurbergsson F, Bjarnason R, et al. Asthma and allergy patterns over 18 years after severe RSV bronchiolitis in the first year of life. Thorax. 2010;65(12):1045–52. doi: 10.1136/thx.2009.121582. [DOI] [PubMed] [Google Scholar]
- 35.Valkonen H, Waris M, Ruohola A, Ruuskanen O, Heikkinen T. Recurrent wheezing after respiratory syncytial virus or non-respiratory syncytial virus bronchiolitis in infancy: a 3-year follow-up. Allergy. 2009;64(9):1359–65. doi: 10.1111/j.1398-9995.2009.02022.x. [DOI] [PubMed] [Google Scholar]
- 36.Mikalsen IB, Halvorsen T, Oymar K. The outcome after severe bronchiolitis is related to gender and virus. Pediatr Allergy Immunol. 2012;23(4):391–8. doi: 10.1111/j.1399-3038.2012.01283.x. [DOI] [PubMed] [Google Scholar]
- 37.Wegienka G, Havstad S, Joseph CL, Zoratti E, Ownby D, Woodcroft K, et al. Racial disparities in allergic outcomes in African Americans emerge as early as age 2 years. Clin Exp Allergy. 2012;42(6):909–17. doi: 10.1111/j.1365-2222.2011.03946.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Kumar R, Tsai HJ, Hong X, Gignoux C, Pearson C, Ortiz K, et al. African ancestry, early life exposures, and respiratory morbidity in early childhood. Clin Exp Allergy. 2012;42(2):265–74. doi: 10.1111/j.1365-2222.2011.03873.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Akinbami LJ, Moorman JE, Bailey C, Zahran HS, King M, Johnson CA, et al. Trends in asthma prevalence, health care use, and mortality in the United States, 2001-2010. NCHS Data Brief. 2012;(94):1–8. [PubMed] [Google Scholar]
- 40.Gold DR, Wypij D, Wang X, Speizer FE, Pugh M, Ware JH, et al. Gender- and race-specific effects of asthma and wheeze on level and growth of lung function in children in six U.S. cities. Am J Respir Crit Care Med. 1994;149(5):1198–208. doi: 10.1164/ajrccm.149.5.8173760. [DOI] [PubMed] [Google Scholar]
- 41.Brehm JM, Acosta-Perez E, Klei L, Roeder K, Barmada MM, Boutaoui N, et al. African ancestry and lung function in Puerto Rican children. J Allergy Clin Immunol. 2012;129(6):1484–90. e6. doi: 10.1016/j.jaci.2012.03.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Borrell LN, Nguyen EA, Roth LA, Oh SS, Tcheurekdjian H, Sen S, et al. Childhood obesity and asthma control in the GALA II and SAGE II studies. Am J Respir Crit Care Med. 2013;187(7):697–702. doi: 10.1164/rccm.201211-2116OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Zhang Y, McConnell R, Gilliland F, Berhane K. Ethnic differences in the effect of asthma on pulmonary function in children. Am J Respir Crit Care Med. 2011;183(5):596–603. doi: 10.1164/rccm.200912-1863OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, Morgan WJ. Asthma and wheezing in the first six years of life. The Group Health Medical Associates. N Engl J Med. 1995;332(3):133–8. doi: 10.1056/NEJM199501193320301. [DOI] [PubMed] [Google Scholar]
- 45.Cox DW, Bizzintino J, Ferrari G, Khoo SK, Zhang G, Whelan S, et al. Human rhinovirus species C infection in young children with acute wheeze is associated with increased acute respiratory hospital admissions. Am J Respir Crit Care Med. 2013;188(11):1358–64. doi: 10.1164/rccm.201303-0498OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Stein RT, Sherrill D, Morgan WJ, Holberg CJ, Halonen M, Taussig LM, et al. Respiratory syncytial virus in early life and risk of wheeze and allergy by age 13 years. Lancet. 1999;354(9178):541–5. doi: 10.1016/S0140-6736(98)10321-5. [DOI] [PubMed] [Google Scholar]
- 47.Sigurs N, Bjarnason R, Sigurbergsson F, Kjellman B. Respiratory syncytial virus bronchiolitis in infancy is an important risk factor for asthma and allergy at age 7. Am J Respir Crit Care Med. 2000;161(5):1501–7. doi: 10.1164/ajrccm.161.5.9906076. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.