Results showed prevalence of asthma/hayfever (AS/HF) among the participants in this large cohort was 19-26% and prevalence of asthma (AS) was 10-14%, with higher prevalence in women, and these findings correspond closely with National Health Survey data from a similar time period
. The main finding of this cross-sectional study, however, was that dietary factors are associated with diagnosed AS and AS/HF in Australian adults of middle age or older. In looking at the analyses for both sexes, diets generally high in meat, particularly diets marked by greater consumption of poultry, seafood, and red and processed meats in females, and diets marked by greater amounts of red meat, processed meat, and cheese consumption in males, appear to be risk factors for AS and AS/HF diagnosis in this population. In comparisons of those in the highest category of these dietary factors to the lowest respective categories, most of the analyses showed increased odds in the 10-25% range. Additionally, the cheese/brown bread factor was shown to be a protective for AS diagnosis in women. Taken together, the overall pattern of these risk and protective factors suggests that adherence to a more meat-based diet may pose risk for AS and AS/HF in Australian adults.
Our findings are supported in previous literature on diet and asthma risk in both children and adults
[21, 25, 30, 35, 40], though adult studies on dietary patterns have been somewhat limited and equivocal to date
. In a large New Zealand children’s cohort study, eating hamburgers was associated with a higher lifetime prevalence of asthma and wheeze
. A high-meat diet, however, was not a risk factor for asthma or wheeze
. In a cross-sectional study of teenagers in Taiwan, consumption of liver, butcher’s meat, and fried foods were associated with increased risk of asthma, and liver also was shown to be a risk factor for allergic rhinitis
In the adult literature, several studies were in agreement with the primary findings of the current study
[35, 38, 40]. Although the current study did not examine Mediterranean, Cosmopolitan, or Prudent dietary patterns previously reported in adult literature on diet and asthma, our results support studies that suggest a contribution of a Western diet (characterized by high intakes of red meat, saturated fat, and refined grains, combined with low intakes of whole grains) to impaired lung function, increased asthma prevalence and worsened asthma control
Our findings failed to support, however, studies suggesting that a diet high in fruits and vegetables may have a protective association with allergic rhinitis
 and asthma in adulthood
[33, 39]. In contrast, we found weak and inconsistent associations for the fruits/vegetables factor in men, and significant positive relationships with AS/HF in women. The reasons underlying this surprising finding are unclear, but there are several possibilities. Our dietary measure was admittedly crude, and we had no way of differentiating frequently consumed high-fat versions of vegetables such as hot chips, other fried vegetables, or those mixed in casseroles and meat pies. Previous studies have shown deep-fried or high-fat foods may be a risk factor for asthma
[21, 23, 48, 49]. Also, total energy intake was not assessed, and this could contribute to potential confounding in analyses. Reverse causality is also a possibility
, such that those diagnosed with AS/HF might purposefully eat more fruits and vegetables in efforts to bolster health and control symptoms or attacks. Lastly, there are a number of fruits and vegetables that are high in histamine or other biogenic amines, and histamine intolerance may play a role in the observed relationship with fruits and vegetables. The common symptoms of histamine intolerance include sneezing, congestion, rhinorrhea, and asthma
. In some individuals, these reactions can occur even after small amounts of histamine are ingested
Our study findings are contrary to studies that have shown no relationship between Western dietary patterns or “vegetarian” dietary patterns and asthma
[30, 36], possibly due to differing populations and methods of dietary assessment. Lastly, our results did not agree with an observed protection associated with fish or seafood intake
. In contrast, our study found that fish or seafood was linked with poultry consumption, and together this factor posed a risk for AS and AS/HF in the majority of analyses, which supports earlier work from the Netherlands
Generally, Australians eat a calorically dense and nutrient-poor Western diet
[52, 53]. A typical Western diet is low in antioxidants, high in fat and saturated fat, and provides a surplus of calories
, any of which could contribute to increased asthma prevalence by modulating the innate immune response
. Among potential factors to consider in determining how a Western diet, or other meat-based diets, may pose a risk for asthma and/or hayfever are saturated fat consumption, exposures from processing or cooking, and fast food and takeaway consumption, marked by high intakes of sodium and fat.
The calorically dense, nutrient poor diet consumed by many Australians can result in post-prandial dysmetabolism and increased oxidative stress. The transient increase in free radicals may acutely trigger inflammation and endothelial dysfunction
. Consumption of meat, particularly red meat, has been associated with inflammation and oxidative stress, which may partly explain the association between meat consumption and chronic disease risk
[37, 56]. Also, fat content, including saturated fat in red meat, and dietary cholesterol may play an important role in the link to chronic disease
. A high fat intake and plasma triglyceride levels have been associated with airway hyper-responsiveness, asthma risk, and adult-onset wheeze
. High-fat meals have been shown to increase neutrophilic airway inflammation, and the effect is determined by the type of fat consumed
[48, 58]. High-fat meals also increase triglycerides and exhaled nitric oxide, a marker of airway inflammation, two-hours post-prandially, and suppress bronchodilator recovery in asthma
Processed meats contain mutagens, including N-nitroso compounds and heterocyclic amines (HCAs), and polycyclic aromatic hydrocarbons (PAHs) formed during high-temperature cooking and grilling
. Polycyclic aromatic hydrocarbons, in particular, have been linked to respiratory disorders and have been shown to trigger asthma attacks in asthmatics
. Eating meats cooked at higher temperatures, such as barbecues and grilling, also increases the formation of PAHs
. Previous studies have shown blood levels of PAHs are significantly correlated with grilled and smoked meat intake among asthmatic children
High-meat diets may be linked with fast food or takeaway consumption, typically high in sodium and saturated fat, and implicated in asthma
[53, 62, 63]. The issue of risk from saturated fat poses an explanatory problem with regard to cheese, however, as cheese loaded on a protective factor with brown and whole-meal bread for females, but on a risk factor with red and processed meats for males. It is possible that cheese is a marker of a more prudent diet for females, due to concerns regarding calcium intake and osteoporosis prevention. It is also possible that women consumed fat-free or low-fat cheeses that would not have been high in saturated fat. There is evidence, however, from a pediatric prospective cohort study that both brown bread and milk fat are protective against asthma symptoms
Strengths and limitations
Our findings should be viewed in light of a number of limitations, and foremost among these is the cross-sectional study design, which allows only the establishment of associations between variables, not temporal sequence or causal direction. In our study, we did not have access to information on relevant comorbidities, including anemia, cardiac dysfunction, systemic inflammation, gastro-esophageal reflux disease, sleep apnea, or chronic obstructive pulmonary disorder. Thus, we cannot rule out potential confounding associated with these comorbid conditions.
Selection bias is a potential concern for any study relying on voluntary participation from the population, and the 18% response rate obtained in The 45 and Up Study, combined with exclusion of participants without a full set of data in the current study warrants caution with regard to external validity. It is conceivable that the 156,035 participants in the current study could differ from non-participants in potentially meaningful ways. The 45 and Up Study, however, is the largest study of healthy aging to be carried out in the Southern Hemisphere, and is likely to be one of the more representative large-scale cohort studies conducted globally.
Self-report instruments are subject to a number of potential biases, and misclassifications of both exposure and outcome could result. In particular, our available measures of dietary habits were limited in scope, and generally did not provide information on portion size or other relevant attributes of foods consumed beyond the frequency of consumption. As such, we could not control for energy intake, although our covariates of body mass index and physical activity categories may have partially accounted for this potential influence.
Counterbalancing these limitations were several strengths. The large and diverse sample of the adult and older adult population of Australia’s most populous state allowed us to examine a group not often studied with regard to diet and asthma. Our study analyzed males and females separately, allowing us to establish dietary factors unique for each sex, and to assess the independent relationships between those factors and our outcomes of interest. Furthermore, we controlled for several potential key confounders: physical activity, smoking, age, education, and body mass index. Lastly, we were able to test and confirm our findings for relationships between diet and AS/HF in a large subset of participants with data on asthma, separated out from hayfever.
Future studies should include analysis of dietary patterns and asthma and hayfever over time, as longitudinal data become available from the 45 and Up study. It may be worthwhile to examine a subsample of this cohort using more rigorous measures of dietary intake, asthma, and hayfever. Additional epidemiological studies with more rigorous measures of exposures and outcomes should be undertaken in populations of differing ages, ethnicities, geographical locations, and in those with various co-morbidities. Lastly, dietary interactions with other environmental and genetic influences should be investigated with prospective designs.