Fast Food Consumption in Pregnancy and Subsequent Asthma Symptoms in Young Children

Abstract

Background:

Recent cross-sectional studies suggested children’s current fast food consumption to be related to frequency of asthma and allergies. Prenatal diet has been suspected to contribute to children’s asthma and atopic disease risks.

Objectives:

We hypothesized that maternal fast food intake during pregnancy increases offspring’s risk for asthmatic symptoms.

Methods:

We conducted a population based study of 1201 mother/child pairs in Los Angeles, California. Detailed information about prenatal fast food intake and other dietary, lifestyle/environmental factors, and pregnancy was collected shortly after birth; further data were retrieved from birth certificates. Using the International Study of Asthma and Allergies in Childhood core questions, asthma and rhinitis symptoms were assessed, and doctor’s diagnoses were recorded in offspring 3.5 years after birth. Poisson regression with robust error variance using a log link function was used to estimate relative risks (RR). Models were adjusted using covariates or propensity scores.

Results:

Maternal prenatal fast food consumption increased their children’s risks for severe and current asthma symptoms (wheeze last 12 months combined with doctor’s diagnosis) in a dose-dependent manner: ‘once a month’: RR: 0.99 (95%CI: 0.36, 2.75), ‘once a week’: 1.26 (0.47, 3.34); ‘3–4 days a week’: 2.17 (0.77, 6.12); ‘every day’ 4.46 (1.36 14.6) compared to ‘never’, adjusting for potential confounders (P for trend=0.0025). Risks for rhinitis symptoms were also increased albeit less than for asthma symptoms.

Conclusions:

These findings suggest that in utero exposure to frequent fast food through maternal diet may be a risk factor for the development of asthmatic symptoms in young children.

Introduction

Asthma and atopic diseases continue to be among the leading childhood health problems worldwide.¹ Asthma prevalence increased over the last decades in many European countries and the US to 16% in girls and 21% in boys among six to seven year olds.¹ While a plateau seems to be reached in “western” countries, prevalence rates are beginning to increase in less affluent countries.¹ Measures for primary prevention of childhood asthma are still limited, and risk factors remain insufficiently understood. In addition to genetics, proposed key factors in early life include dietary factors,² farming environment³ and microbial exposures,⁴ obesity,⁵ maternal smoking,⁶ and viral infections.⁷

The proposed association between a “westernized” diet and lifestyle and higher rates of asthma and atopic disease⁸ prompted a few cross-sectional studies of children’s fast food consumption. In New Zealand, children’s hamburger consumption was positively associated with asthma symptoms, and takeaway food slightly increased bronchial hyper-responsiveness.⁹ In Spain, fast food intake in children and current asthma symptoms was positively associated, while a more “Mediterranean type” diet was protective.¹⁰ Recently, in the large worldwide International Study of Asthma and Allergies in Childhood (ISAAC), frequent current fast food intake in children was related to higher rates of asthma symptoms.¹¹ Although components of maternal diet may modulate fetal immune system development and may have epigenetic effects possibly predisposing to asthma,¹² data on maternal consumption of fast food in relation to their children’s risk for asthmatic symptoms are lacking.

We conducted a population based study, collecting extensive interview information about pregnancy, diet, environment, and lifestyle factors postpartum, and assessed asthmatic, allergic and respiratory outcomes in children 3.5 years after birth in Los Angeles County, California.

Methods

Study design and population

The UCLA Environment and Pregnancy Outcomes Study was originally designed to assess effects of air pollution on birth outcomes oversampling for preterm and low birth weight births as described previously.¹³ Briefly, we selected all 66,795 records for children born in 2003 to mothers who resided in 111 Los Angeles County zip codes (41% of all LA County births). Births excluded were those with recorded defects, extreme/missing gestational age or weight, multiple gestations, not reported to the state or outside LA County resulting in a cohort of 58,316 births. From this cohort, we selected all cases of low weight (<2500 gram) or preterm birth (<37 weeks) and an equal number of randomly selected controls (full-term ≥2,500 gram), from 24 zip codes located in close proximity to air monitoring stations, and randomly selected 30% of cases and an equal number of controls from 87 zip codes located near major roadways. Cases and controls were thus matched on zip code set and birth month. Of 6,374 women originally selected from the cohort 2,543 could be located and were interviewed 3–6 months postpartum in 2003. Of the 2,438 women who agreed being re-contacted, 1,201 could be located and participated in the follow-up survey (response 49.3%) in 2006–2007; a major reason for the attrition was the difficulty locating women after 3 years in Los Angeles, an area with high mobility. The UCLA Office of the Human Research Protection Program and the California State Committee for the Protection of Human Subjects approved this research and informed consent was obtained from the women.

Pregnancy assessment

The first maternal questionnaire after birth assessed detailed pregnancy related information. Dietary factors during pregnancy included the frequency of consumption of fast food on average (never, once a month, once a week, 3–4/week, daily), fish (salmon, tuna, mackerel), and well done meat or fish, and use of vitamins. Detailed information was also ascertained on maternal stress, maternal occupation, marital status, prepregnancy weight, smoking, and alcohol during pregnancy. The follow-up survey conducted 3.5 years after birth by phone interview, assessed duration of breast feeding, pets, mold, pests (cockroaches), and smoking in the home. Information about birth weight, gestational age, insurance type, parity, and demographics were retrieved from birth certificates. We described in detail the air pollution exposure assessment elsewhere.¹³

Asthmatic and respiratory symptoms/outcomes

Asthma and hay fever symptoms were assessed using the corresponding ISAAC core questions.¹⁴ We also recorded doctor’s diagnosed asthma, pneumonia, and otitis media, and asked whether the child had bronchitis, and has taken asthma medication in the past 12 months (please refer to footnote Table 3 for questions). We grouped children according to the ISAAC standard definition into “current asthma” (wheeze past 12 months and reported doctor’s diagnosed asthma), and “severe asthma” (positive response to the ISAAC questions referring to the following: ≥ 4 attacks of wheeze in last 12 months and/or ≥ 1 night per week sleep disturbance and/or wheeze affecting speech in the last 12 months); additionally, reported doctor’s diagnosed asthma was combined with dry cough at night without a cold as assessed by the corresponding ISAAC question.¹⁴

Table 3.

Maternal consumption of fast food and relative risks for asthmatic, allergic, and respiratory symptoms and disorders in the Environment and Child Health Outcomes Study in Los Angeles County, California, 2006 (n=1,201).

Outcome	Case N	Once / month Relative Risk (95% CI)	Once / week Relative Risk (95% CI)	3–4 times / week Relative Risk (95% CI)	Every day Relative Risk (95% CI)	P value for trend&
Current asthma a
Unadjusted	73	1.17 (0.43, 3.21)	1.56 (0.62, 3.94)	2.84 (1.09, 7.41)	7.47 (2.72, 20.48)
Adjusted#	69	0.99 (0.36, 2.75)	1.26 (0.47, 3.34)	2.17 (0.77, 6.12)	4.46 (1.36, 14.60)	0.0025
Severe asthma b
Unadjusted	71	1.77 (0.60, 5.16)	2.05 (0.74, 5.69)	3.03 (1.03, 8.87)	7.06 (2.16,23.12)
Adjusted#	68	1.71 (0.58, 5.03)	1.77 (0.64, 4.93)	2.59 (0.87, 7.71)	4.34 (1.22, 15.43)	0.0236
Asthma c
Unadjusted	118	1.09 (0.52, 2.31)	1.33 (0.67, 2.65)	2.28 (1.12, 4.64)	4.55 (2.04, 10.13)
Adjusted#	113	0.93 (0.44, 1.98)	1.05 (0.51, 2.13)	1.66 (0.78, 3.51)	2.47 (0.97, 6.29)	0.0162
Asthmac and dry coughd
Unadjusted	56	1.17 (0.43, 3.21)	0.85 (0.32, 2.29)	2.78 (1.07, 7.23)	3.90 (1.13, 13.41)
Adjusted#	53	1.07 (0.40, 2.87)	0.73 (0.26, 2.03)	2.24 (0.82, 6.11)	3.01 (0.67, 13.53)	0.0634
Asthma medication e
Unadjusted	190	1.14 (0.64, 2.05)	1.68 (0.99, 2.86)	1.86 (1.05, 3.29)	2.92 (1.45, 5.88)
Adjusted#	184	1.19 (0.65, 2.17)	1.73 (0.99, 3.03)	1.83 (1.00, 3.37)	2.38 (1.10, 5.14)	0.0027
Sneeze or runny nose without flu f
Unadjusted	189	1.32 (0.75, 2.34)	1.60 (0.94, 2.72)	1.74 (0.98, 3.10)	2.36 (1.11, 5.00)
Adjusted#	182	1.32 (0.75, 2.32)	1.54 (0.91, 2.61)	1.60 (0.90, 2.85)	1.92 (0.88, 4.17)	0.0491
Bronchitis g
Unadjusted	78	2.92 (0.88, 9.67)	2.90 (0.90, 9.29)	3.84 (1.15, 12.83)	5.33 (1.25, 22.68)
Adjusted#	77	2.90 (0.91, 9.27)	2.59 (0.82, 8.16)	3.01 (0.92, 9.79)	3.52 (0.74, 16.71)	0.1871
Pneumonia ever h
Unadjusted	85	0.89 (0.41, 1.92)	1.06 (0.53, 2.15)	1.14 (0.51, 2.53)	2.48 (0.95, 6.52)
Adjusted#	84	0.97 (0.44, 2.15)	1.18 (0.56, 2.48)	1.32 (0.58, 3.01)	2.94 (1.00, 8.67)	0.0982
Otitis Media recurrenti (>3)
Unadjusted	227	0.76 (0.54, 1.08)	0.82 (0.60, 1.12)	0.89 (0.62, 1.28)	0.54 (0.22, 1.31)
Adjusted#	221	0.81 (0.57, 1.15)	0.88 (0.64, 1.21)	1.04 (0.71, 1.52)	0.71 (0.30, 1.72)	0.7903

^#adjusted for maternal age, education, race/ethnicity, place of birth (US vs. non-US) preterm birth, child sex, mother or father history of atopy (asthma, hay fever, eczema), child age at interview, prenatal care, mother working during pregnancy, pregnancy average ozone and PM2.5. Reference group: “no fast food consumption”.

^&P value for trend derived from adjusted regression model using “fast food” as continuous variable coded 1–5.

^acorresponds to the ISAAC question: “Has your child had wheezing or whistling in the chest in the past 12 months?” and diagnosis of asthma.

^bcorresponds to positive response to the ISAAC questions referring to the following: ≥ 4 attacks of wheeze in last 12 months and/or ≥ 1 night per week sleep disturbance and/or wheeze affecting speech in the last 12 months (“How many attacks of wheezing has your child had in the last 12 months?”, “In the last 12 months, how often on average, has your child’s sleep been disturbed due to wheezing?”, “In the past 12 months, has wheezing been severe enough to limit your child’s speech to only one or two words at a time between breaths?”)

^cdoctor’s diagnosis of asthma.

^dcorresponds to the ISAAC question: “ In the past 12 months, has your child had a dry cough at night, apart from a cough associated with a cold or chest infection?”

^ecorresponds to the question: “In the past 12 months, has your child used any medications, pills, puffers, or other medication for wheezing or asthma?”

^fcorresponds to the ISAAC question: “In the past 12 months, has your child had a problem with sneezing, or a runny, or blocked nose Yes when he/she DID NOT have a cold or the flu?”

^gcorresponds to the question: “In the past 12 months, has your child had bronchitis?”

^hcorresponds to the question: “Has your child ever been seen by doctor or other health care practitioner for pneumonia?”

ⁱcorresponds to the question: “How many times has your child had ear infections?”

Statistical analysis

We estimated relative risks for maternal fast food consumption using Poisson regression models with robust error variance and a log link function. Single and multiple variable models were fitted; potential confounders were selected based on prior knowledge. The final models were adjusted for maternal age, education, race/ethnicity, place of birth, preterm birth, child sex, maternal/paternal history of atopy, child age at interview, prenatal care, mother working during pregnancy (variable definitions as in Table 1), and pregnancy average ozone and PM_2.5 (continuous variables). These variables were included because alone or in combination, they changed the estimates of interest at least 5–10%. Following variables were not retained in the final models because they did not further change the estimates of interest >10%: mother’s year’s lived in the US, duration of breast feeding, birth weight, gestational age, maternal prepregnancy weight, pregnancy complications, parity, marital status, maternal stress, pets, cockroaches or mold in the house, day care attendance, maternal smoking or alcohol consumption, passive smoke (prenatal or current), insurance type paid for delivery, prenatal fish consumption, vitamins before pregnancy, or consumption of grilled/well done meat or fish during pregnancy. Most potential confounders had few missing data (<1–2%), except income (missing: 11.6%) and prepregnancy weight (2.5%). Multiple imputations using standard SAS procedures with 5 imputation data sets were performed to replace missing values of all covariates in models adjusting for income or prepregnancy weight, and for the final adjusted models. Estimates of interest did not change appreciably thus we report the findings of the complete case analyses. P values for linear trend were derived from adjusted regression models using “fast food” as a continuous variable coded with values 1–5. To explore the role of parental history of atopy and child sex, respectively, on risks related to fast food we conducted stratified analyses. As another sensitivity analysis and because of small numbers for subgroups in the stratified analyses, we adjusted models with propensity scores based on regression of the “fast food” exposure estimate on all covariates included in the final models. Further sensitivity analyses were done excluding preterm births or subjects who were not breast-fed. All analyses were conducted with SAS 9.3.

Table 1.

Demographic and pregnancy characteristics by child asthma status in the Environment and Child Health Outcomes Study in Los Angeles County, California, 2006 (n=1,196^a).

	Doctor’s Diagnosed Asthma
	Yes n = 118 n (%$)	No n =1,078 n (%$)	Odds Ratio (95% CI)
Preterm birth (<37 weeks)	54 (45.7)	394 (36.6)	1.47 (1.00, 2.15)
Boys	76 (64.4)	530 (49.2)	1.87 (1.26, 2.78)
Maternal age
<20	10 (8.5)	66 (6.1)	1.70 (0.74, 3.90)
20–24	17 (14.4)	191 (17.7)	Referent
25–29	30 (25.4)	268 (24.9)	1.26 (0.67, 2.35)
30–34	41 (34.8)	313 (29.0)	1.47 (0.81, 2.66)
>35	20 (17.0)	240 (22.3)	0.94 (0.48, 1.84)
Parity >1	64 (54.2)	643 (59.7)	0.80 (0.55, 1.18)
Mother Atopy &	44 (37.3)	225 (20.9)	2.25 (1.51, 3.37)
Father Atopy &	26 (22.0)	162 (15.0)	1.60 (1.00, 2.55)
Race/ethnicity
Hispanic white	67 (56.8)	647 (60.0)	1.24 (0.76, 2.03)
Non-Hispanic white	23 (19.5)	275 (25.5)	Referent
African-American/Black	15 (12.7)	55 (5.1)	3.26 (1.60, 6.65)
Asian	6 (5.1)	45 (4.2)	1.59 (0.62, 4.13)
Other#	7 (5.9)	47 (4.4)	1.78 (0.72, 4.38)
Missing	0 (0.0)	9 (0.8)	_
Place of birth
Non-US born	53 (44.9)	574 (53.3)	0.73 (0.50, 1.07)
US born	64 (54.2)	503 (46.7)	Referent
Missing	1 (0.9)	1 (0.1)	_
Smoking
Former smokers	34 (28.8)	313 (29.0)	1.06 (0.69, 1.63)
Pregnancy smokers	10 (8.5)	43 (4.0)	2.27 (1.10, 4.70)
Non-smokers	74 (62.7)	722 (67.0)	Referent
Living with smoker in pregnancy
Yes	21 (17.8)	164 (15.2)	1.24 (0.75, 2.05)
Missing	3 (2.5)	4 (0.4)	_
Maternal education
<8	13 (11.0)	120 (11.1)	1.13 (0.57, 2.22)
9–11 years	14 (11.9)	171 (15.9)	0.85 (0.44, 1.64)
12 years	23 (19.5)	252 (23.4)	0.95 (0.54, 1.66)
13–15 years	36 (30.5)	177 (16.4)	2.12 (1.27, 3.52)
>16	32 (27.1)	333 (30.9)	Referent
Missing	0 (0.00)	25 (2.3)	_
Mother worked during pregnancy	74 (62.7)	627 (58.2)	1.25 (0.84, 1.86)
Missing	2 (1.7)	6 (0.6)	_
Prenatal care
Began 1st trimester	100 (84.8)	1,011 (93.8)	Referent
Began 2nd or 3rd trimester or none	16 (13.6)	62 (5.8)	2.61 (1.45, 4.69)
Missing	2 (1.7)	5 (0.5)	_
Marital status
Married or living together	92 (78.0)	894 (82.9)	Referent
Single, separated, divorced	26 (22.0)	181 (16.8)	1.40 (0.88, 2.22)
Missing	0 (0.0)	3 (0.3)	_
Breast feeding
Did not breast feed	26 (22.0)	179 (16.6)	1.38 (0.85, 2.25)
Breast fed for < 6 months	32 (27.1)	319 (29.6)	0.95 (0.61, 1.50)
Breast fed for ≥ 6 months	60 (50.9)	570 (52.9)	Referent
Missing	0 (0.0)	10 (0.9)	_
Day care	73 (61.9)	573 (53.2)	1.42 (0.96, 2.09)
Missing	0 (0.0)	5 (0.5)	_
Fast Food consumption in pregnancy
Never	9 (7.6)	125 (11.6)	Referent
Once a month	22 (18.6)	278 (25.8)	1.10 (0.49, 2.46)
Once a week	46 (39.0)	469 (43.5)	1.36 (0.65, 2.86)
3–4 days a week	30 (25.4)	166 (15.4)	2.51 (1.15, 5.48)
Every day	11 (9.3)	25 (2.3)	6.11 (2.29, 16.29)
Missing	0 (0.0)	15 (1.4)	_
Fish consumption+ in pregnancy
Never	48 (40.7)	402 (37.3)	Referent
Once a month	36 (30.5)	316 (29.3)	0.95 (0.60, 1.51)
Once a week	25 (21.2)	302 (28.0)	0.69 (0.42, 1.15)
3–4 days a week or every day	8 (6.8)	50 (4.6)	1.34 (0.60, 3.00)
Missing	1 (0.9)	8 (0.7)	_

^aFive children missed information on asthma diagnosis.

^#Race/ethnicity ‘other’ includes: Native American/American Indian, Indian, Filipino, Hawaiian, Guamanian, Samoan, Eskimo, Aleut, Pacific Islander, other (specified).

^&Atopy: defined as history of asthma, hay fever, or eczema.

⁺consumption of tuna, mackerel, sardines or salmon.

^$Differences from 100% due to rounding.

^aexcludes n=5 missing for “Doctor’s diagnosed asthma”.

Results

Responders’ characteristics are displayed by asthma status in Table 1. Among children with a doctor’s diagnosis of asthma more were male, had a family history of atopy, and African American/black racial identification. Characteristics of responders, non-responders and the baseline population are shown in Supplemental Table 1; frequency of disorders and symptoms are displayed in Table 2.

Table 2.

Frequency of asthmatic, allergic and respiratory symptoms and disorders in the Environment and Child Health Outcomes Study in Los Angeles County, California, 2006 (n=1,201).

Disorder and Symptoms	Outcomes	Missing
	N (%)	N (%)
Asthma doctor’s diagnosed	118 (9.8)	5 (0.4)
Emergency room visit asthma1	31 (26.3)	0
Wheeze ever	304 (25.4)	8 (0.7)
Wheeze past 12 months2	167 (54.9)	3 (1.0)
Wheeze during exercise2	57 (18.8)	3 (1.0)
Current asthmaa	73 (6.1)	5 (1.6)
Severe asthma2b	71 (5.9)	0
Asthma doctor’s diagnosed and dry cough at night	56 (4.7)	10 (0.8)
Asthma medication use past 12 months	192 (16.0)	2 (0.2)
Sneeze or runny nose without flu last 12 months	190 (15.8)	6 (0.5)
Itchy eyes3	91 (47.9)	7 (3.7)
Dry cough at night past 12 months	243 (20.3)	6 (0.5)
Bronchitis past 12 months	79 (6.6)	9 (0.8)
Pneumonia doctor’s diagnosed	85 (7.1)	0
Otitis media doctor’s diagnosed ever	604 (50.4)	4 (0.3)
Frequent otitis media doctor’s diagnosed (>3)4	229 (37.9)	12 (2.0)

¹⁼asked only among those who responded “yes” to asthma (doctor’s diagnosed)

²⁼asked only among those who responded “yes” to wheeze ever

³⁼asked only among those who responded “yes” to sneeze or runny nose without flu (last 12 months)

⁴⁼asked only among those who responded “yes” to otitis media (doctor’s diagnosed)

^acorresponds to the ISAAC question: “Has your child had wheezing or whistling in the chest in the past 12 months?” and diagnosis of asthma.

^bcorresponds to positive response to the ISAAC questions referring to the following: ≥ 4 attacks of wheeze in last 12 months and/or ≥ 1 night per week sleep disturbance and/or wheeze affecting speech in the last 12 months (“How many attacks of wheezing has your child had in the last 12 months?”, “In the last 12 months, how often on average, has your child’s sleep been disturbed due to wheezing?”, “In the past 12 months, has wheezing been severe enough to limit your child’s speech to only one or two words at a time between breaths?”)

Adjusted relative risk estimates for fast food consumption increased in an exposure-response manner (p for trend = 0.0025) for ”current asthma” and “severe asthma” with an over 4fold increase in the risk estimate for daily consumption (Table 3). Risk estimates were also increased albeit not as strongly for reported doctor’s diagnosed asthma, use of asthma medication, and doctor’s diagnosed asthma combined with dry cough at night, and also increased with increasing prenatal consumption of fast food. Rhinitis risk estimates increased moderately with increasing fast food consumption but all 95% confidence intervals included the null value.

Risk estimates for bronchitis and pneumonia were also increased up to around 3fold related to fast food intake although most 95% confidence intervals included the null value and there was no clear exposure-response trend. For “frequent otitis media” no increased risks were found. The estimates for asthma related measures and bronchitis in high exposure categories reduced after adjustment with largest changes after adding SES related variables. However, adjustment using propensity scores resulted in similar estimates (Supplemental Table 2) compared to models adjusted using covariates.

Maternal fish consumption was not consistently associated with any of the outcomes (data not shown). Sensitivity analysis stratified by family history of atopy (yes/no) indicated increased risks in both groups, but confidence intervals were wide (Supplemental Table 3). Sex stratified models suggested similar risks for asthmatic outcomes related to fast food in girls and boys; risk estimates for bronchitis and pneumonia were elevated mainly in boys but 95% confidence intervals were wide (Supplemental Table 4). Excluding preterm births or those never breast-fed did not change the findings appreciably (data not shown).

Discussion

In our population based study, frequent maternal consumption of fast food during pregnancy increased children’s risk for subsequently assessed wheeze/asthmatic outcomes in a dose dependent manner. Risks for symptoms of rhinitis were less increased. Bronchitis and pneumonia risks were increased in boys but confidence intervals were wide. No association was seen for otitis media.

Data is lacking on maternal fast food intake and asthma/allergy in offspring, but a few cross-sectional studies found children’s current consumption of fast food related to a higher prevalence of asthma or allergy symptoms, although effect sizes were generally smaller compared to the ones we estimated for maternal prenatal intake.^9,11 Children’s current high burger consumption was associated with a 40% increase in the odds ratio for “life-time asthma” in ISAAC phase II studies.¹⁵ Recently, the worldwide ISAAC phase III assessment suggested moderate associations between children’s and adolescents’ current fast food consumption and current and severe asthma and allergy symptoms with increases in odds ratio estimates around 10–30%.¹¹ Thus, the large 2.5 to over 4fold increase in risk for “current” and “severe asthma” related to frequent maternal fast food intake suggests that fast food consumption during fetal development is more important than intake in children. High energy and fast food in current diet of children was also associated with increased risk estimates for asthma in Taiwanese children at a similar magnitude as we estimated for maternal intake but the confidence intervals in the Taiwanese study were much wider.¹⁶ The associations with hay fever symptoms were weaker and similar in our study and the studies examining current consumption.^11,15 While the prior studies were limited due to their cross-sectional nature with reverse causation being a possible explanation, the notion that current fast food consumption can be a risk factor for asthma symptoms in children supports our findings.

Fast food is calorically dense, usually served in large portions featuring processed meat, high in refined carbohydrates, sodium, sugar, and cholesterol, contains additives such as preservatives and colorants, and is characterized by high concentrations of saturated and trans-fatty acids. While we could not examine specific components contained in fast food in our study, frequent maternal intake of fast food can lead to prenatal exposure to several compounds that have been suspected to interfere with the developing immune system, and shift the T-helper (Th) cell type 1/Th cell type 2 immune response balance towards a Th2 profile that has been linked with asthma.^12,17–19 Some studies suggested prenatal dietary fatty acids play a role in the development of atopic disease in children. Fatty acids and their derivatives can influence the early immune system development and maturation by regulating numerous metabolic processes and cytokine gene expression.¹² High intake of fast food during pregnancy likely leads to a high n-6-PUFA/low n-3-PUFA ratio^20,21 which may relate to the increased risks we observed. Positive associations between current intake of food such as margarine with high levels of trans-fatty acids (also high in fast food) and asthma/allergic diseases in children have been reported.^17,22 A small risk increase for asthma in children after use of margarine during lactation was reported,²³ while prenatal intake was not considered. Thus, our findings may be explained at least in part by the fatty acid exposure pattern related to high intake of fast food during pregnancy.

Another explanation of our findings may be related to the consumption of processed meat (high in fast food) in pregnancy which increased risks for wheeze in the first year of life.²⁴ In Japan, a “western” maternal diet with high intake of processed meat increased infant wheeze and eczema.²⁵ High fast food intake conversely may be related to low consumption of fruits and vegetables which has been linked to an increased risk for asthma and allergies in current diet of children,¹¹ or it may be a combination of both, high fast food/low fruits and vegetables. “Mediteranean” diet, containing high fruits/vegetables, is hypothesized to be protective in the etiology of asthma. However, during pregnancy it was not found to be protective against wheeze in infancy²⁴ and current intake in children exerted small protective effects on current wheeze only in girls.¹⁰

Further components of fast food that may contribute to the increased risks in offspring after maternal consumption are food additives such as colorants and preservatives. In vitro studies suggested that common food additives such as tert-butylhydroquinone, propionic acid, sodium benzoate, sodium sulfite, and sorbic acid can suppress the Th1 immune activation cascade and shift immune response toward a Th2 response,²⁶ and relate to immunological processes associated with atopic diseases.¹⁹ Soda beverages particularly are high in colorants and preservatives and typically a component of fast food meals. Current intake of sugar-sweetened soda related to asthma in adolescents independent of weight.²⁷ In the Danish National Birth Cohort positive associations were seen between maternal consumption of artificially sweetened soft drinks and offspring’s asthma and wheeze at 18 months and allergic rhinitis at age 7 years but no associations were seen for sugar-sweetened beverages.²⁸ We cannot disentangle specific components of the fast food intake of mothers in our study, but it also may be that the combination of the different food components typically consumed together increase the risk for asthmatic/respiratory outcomes we observed.

Fast food consumption is associated with increased BMI and obesity,²⁹ and children’s high weight, sedentary behavior and television watching has been associated with asthma and atopic disorders.⁵ Maternal fast food-consumption may be linked to child’s diet and child obesity which we could not assess in our study. However, neither maternal prepregnancy weight nor birth weight influenced our effect estimates for fast food. Possibly, obesity and asthma in children share common risk factors and immunological mechanisms, as similar immunological features were found involved in the development of obesity and asthma.³⁰

Our study has several limitations. We did not conduct a more detailed assessment of maternal diet, but rather assessed several indicators of dietary patterns which were shown to be proxies of maternal diet in earlier studies of atopic disease.¹⁵ Moreover, dietary patterns related to frequent intake of fast food have been identified in cardiovascular research as risk factor for metabolic and cardiovascular diseases irrespective of other dietary components.³¹ Further, we had a relatively high rate of attrition between the first and second assessment; among responding women, more were highly educated, older, white non-Hispanic and US-born. If response was differential and frequent fast food consumers with asthmatic/allergic children responded more frequently this may have biased our findings. However maternal fast food consumption was similar in responders and non-responders thus we have no indication for response bias. Fast food intake was assessed 3–6 months postpartum; while recall bias cannot be ruled out, the assessment of respiratory/asthmatic outcomes was conducted 3.5 years after birth thus differential recall bias is very likely not an issue. While residual confounding of unmeasured variables is possible, confounding is unlikely to explain our findings. We evaluated many potential confounders including air pollution, breastfeeding, and smoking, for which we had detailed information, and we controlled our final models for important confounders. A limitation is however, the lack of data on children’s diet and nutritional status. Another limitation is that children were relatively young to manifest asthma, thus part of the reported wheeze/asthma symptoms may be more consistent with transient wheeze. Strengths include the prospective nature of the study, the use of ISAAC core questions and a detailed assessment of many potential confounders.

In conclusion, our findings suggest that maternal fast food consumption during pregnancy may be a preventable risk factor for wheeze and asthma symptoms in young children. This is particularly important in view of rising fast food consumption worldwide.³² Future studies should be designed to disentangle which components of fast food may be causally related, and examine the possible role of related co-factors such as obesity. Recommending avoidance of fast food during pregnancy should be scrutinized as a potential protective public health measure against asthma in children.

Abbreviations

CI

Confidence interval

ISAAC

International Study of Asthma and Allergies in Childhood

LA

Los Angeles

PUFA

Poly unsaturated fatty acid

RR

Relative Risk

Th

T-helper

UCLA

University of California Los Angeles