This longitudinal prospective study started as a part of the International Study of Wheezing in Infants (EISL) study performed in Cartagena, Spain. Children have been followed up afterwards.10 Briefly, all primary care health centres monitoring children for nutrition, growth and development, and/or for vaccine administration from the health program in Cartagena were included as recruiting centres; and when the child attended to receive vaccination at 15 or 18 months of age (“survey 1.5”), parents or guardians were asked to complete the questionnaire, emphasising on nutritional aspects of the mother during pregnancy and respiratory/allergy symptoms in the offspring which occurred during the first 12 months. At age four years (“survey 4”) the participant families were contacted again to answer a similar questionnaire, emphasising nutrition and respiratory/allergy symptoms occurred in the offspring during the previous 12 months.

At survey 1.5, a standardised and validated questionnaire,10 including questions on epidemiological risk/protective factors, was completed. The questions were: age; gender; race; type of delivery; number of siblings; birth weight and height (by parental report); low birth weight (<2000g or <2500g); exclusive breastfeeding for six months (without any formula feeding or infant food); air pollution (living near to factories or roads with heavy traffic); mould stains on the household walls; dogs and cats at home, during pregnancy and at present; maternal age and educational level; oral contraceptive used before pregnancy; paracetamol used during pregnancy; number of colds during the first year of life; maternal smoking during pregnancy; paternal current smoking; parental asthma, rhinitis and eczema.

At survey 4, the data collected included: kindergarten attendance; parental current tobacco smoking; dogs and cats at home; mould stains on the household walls; type of fuel used in heating and cooking systems; physical exercises (hours/week); TV-video play watching (hours/day); and height and weight (by parental report).

Current wheezing was defined as a positive answer to the question: “Has your child had wheezing or whistling in the chest during the first 12 months of life?”. Current rhinitis was defined as a positive answer to the following question “Has your child had a problem with sneezing or a runny or blocked nose when he/she did not have a cold or the flu accompanied by itchy, watery eyes during the last 12 months?”. Current eczema was defined as a positive answer to the following question: “During the first 12 months of his/her life, has your child had an itchy rash which was coming and going in any part of his/her body except around the mouth and nose, and on the nappy area?”.

Questions were asked regarding the consumption of foods (never or occasionally, 1 or 2 times per week, and 3 or more times per week) by the mother during pregnancy at survey 1.5, and by the child at survey 4. The MedDiet score employed in the present study was previously developed by our group7,8 and is based on the score by Psaltopoulou et al.11: fruit, fish, vegetables, legumes, cereals, pasta, rice and potatoes are considered “pro- Mediterranean” foods and rated according to the frequency of their intake (0 points=never or occasionally, 1 point=1 or 2 times/week, or 2 points=≥3 times/week). Meat, milk and fast foods are considered “anti-Mediterranean” foods and are rated inversely (Table 1). We used MedDiet as quartiles of score and also as raw score. As in our three previous reports,4,7,8 olive oil was not included as part of the MedDiet score and was recorded apart. Therefore, a question about the type of oil used for cooking and dressing salads and vegetables was included; and a dichotomous variable: olive oil vs. others [margarine, butter or other oils] was built. Additionally, the frequency of industrial infant foods (e.g. yogurt, puddings, petit-suisse, commercial chips, jelly, chocolate, soft drinks and bottled/packed juices) consumed by the infants during their first year of life was recorded (as never, once per month, once per week and every day).

The Ethics Committee of the University of Murcia approved the study, and full informed and signed consent was obtained from parents before completing the questionnaire.

Three different outcomes were considered: current wheezing, current rhinitis and current dermatitis at survey 4. The bivariate analysis was performed by means of the Fisher's exact test for categorical variables; the chi square test for trend for ordinal variables; and the Student t-test for independent samples for continuous variables. Odds ratios (OR) and 95% confidence intervals (95% CI) were also calculated. Multivariate logistic regression analysis models were built for each outcome. Adjusted OR (aOR) and 95% CI were calculated from the logistic regression models. Statistical analyses were performed using statistical software (SPSS® v.17, IBM, Armonk, NY, US).

In terms of demographic and anthropometric characteristics the statistically significant differences between children with and without current wheezing were: length at birth; birth weight; number of colds during the first year of life; maternal tobacco smoking at survey 1.5; paternal rhinitis; mould stains on the household at survey 4; and age of wheezing onset (Table 2).

Among diet, MedDiet score adherence and olive oil consumption by the mother and by the child at survey 4 were not significantly different between children with and without current wheezing (Table 2). Also, when comparing separately each group of foods consumed by the mother and by the child, no significant differences were found. The only exception was the association between a higher prevalence of wheezing among those who never or occasionally took fruits and vegetables, and among those eating meat and fast food ≥3 times/week (Table 3a).

In the multiple logistic regression analysis parental rhinitis; birth weight below 2kg; maternal tobacco during pregnancy; mould stains in household walls at survey 4; having fruit never or occasionally; and consuming ≥3 times/week fast food by the child remained as independent risk factors associated to current wheezing. While consumption by the mother during pregnancy of meat 1 or 2 times/week, and never or occasionally pasta, remained as independent protective factors for current wheezing in the offspring, (Table 4).

There were significant associations with current rhinitis with the following factors in the univariate analyses (Table 2): male gender; maternal asthma; paternal rhinitis; maternal rhinitis; paternal dermatitis; dog/cat ownership at birth; dog/cat ownership at survey 1.5; mould stains on the household at survey 1.5 and at survey 4. Furthermore, as compared to those without current rhinitis, children with current rhinitis had a lower proportion of mothers with university degrees; a higher number of colds during the first year of life; and lower physical activity (Table 2).

MedDiet score adherence by the mother and olive oil consumption by the mother and by the child were not significantly different between children with and with current rhinitis (Table 2). However, children without current rhinitis had significantly lower MedDiet adherence at survey 4 (using quartiles of MedDiet) (Table 2). Separately considering each group of foods consumed by the mother and by the child at survey 4, when comparing children with and without rhinitis, those with current rhinitis had a significantly higher proportion of high consumption (≥3 times/week) of fruits, fish, vegetables, pasta, potatoes and meat; a higher proportion of low consumption (never or occasionally) of milk by the child; and a lower proportion of fast food consumption by the child (Table 3b).

In the multiple logistic regression analysis, pets at birth; maternal rhinitis; higher maternal education level; increased weight; and mould stains at survey 4 remained independent risk factors associated to current rhinitis; as did low consumption (never or occasionally) of fruits by the child. Female gender and consuming meat<than 3 times/week by the child were protective factors (Table 4).

There was a higher proportion of maternal asthma, maternal rhinitis, mould stains on the household at survey 4 among children with current dermatitis than without it (Table 2). Moreover, mothers of children with current dermatitis had a significantly higher frequency (>once per week) of paracetamol use during pregnancy; and children had more colds during the first year of life, and exercised less than those without dermatitis (Table 2).

MedDiet score adherence consumption by the mother during pregnancy; and olive oil consumption by the mother and by the child at survey 4 were not significantly different between children with current dermatitis than those without (Table 2). However, children with current dermatitis had higher MedDiet consumption (in quartiles and score) by child (Table 2). Comparing separately each group of foods consumed by the mother and by the child at survey 4, children with current dermatitis had a higher proportion of consumption of ≥3 times/week of fruits, fish, vegetables, potatoes and meat than those without dermatitis. Additionally, those without dermatitis had a lower proportion of never or occasional consumption of milk.

In the multiple logistic regression analysis, maternal rhinitis; mould stains at survey 4; ≥3 times/week fast food consumption by the mother; never or occasionally and ≥3 times/week fruits consumption by the child; and ≥3 times/week potatoes consumption by the child remained associated to higher risk of current dermatitis. Curiously, 1 or 2 times/week or ≥3 times/week fast food consumption by the child were a protective factor for dermatitis.

The authors declare that they have followed the protocols of their work centre on the publication of patient data and that all the patients included in the study have received sufficient information and have given their informed consent in writing to participate in that study.

The authors have obtained the informed consent of the patients and/or subjects mentioned in the article. The author for correspondence is in possession of this document.

The authors declare that no experiments were performed on humans or animals for this investigation.