Burgers, Fast Foods, and Increased Associated Risk for Atopic Dermatitis: A Cross-Sectional Study of Dietary Habits among Young Chinese Adults in Singapore/Malaysia Open Access

As the second most common skin condition globally, atopic dermatitis (AD) is a multifactorial inflammatory disease with no cure [1, 2]. The skins of AD patients are hypersensitive towards common environmental aeroallergens such as house dust mites [3]. AD is marked with a persistent itch commonly distributed in the flexural regions with other clinical manifestations including eczematous lesions, inflamed, pruritus rashes with various degrees of remission [4, 5]. Although AD does not result in deaths and disabilities, AD causes sleeping difficulties, skin excoriations, increased healthcare costs, psychological stresses, and disrupted quality of life [6, 7]. AD typically occurs during early infancy and childhood from 6 months to 5 years old, affecting 5–20% of children worldwide [8]. However, AD can persist into adulthood as the incidence in the adult population is up to 2% [2, 8]. Moreover, AD may initiate other atopic comorbidities such as asthma and allergic rhinitis as part of the atopic march to complicate patients’ health further [9, 10].

The prevalence of AD in Singapore is up to 20%, but the actual reasons behind its high prevalence are not well described [11, 12]. Extensive literature has long suggested that the complex interaction between genetic and environmental components is complicit in AD pathogenesis [13]. These include the impairment of epidermal barriers, immunoregulatory abnormalities, and altered skin microbiome [14‒16]. Recently, dietary intake has been suggested to be a predictor of childhood immunity and allergic diseases [17]. The contemporary shift towards a westernized lifestyle and dietary habits in Asia has been associated with the increased AD prevalence regionally. In Singapore, the growing affluence shaped a diet structure that favours foods of higher sugar, fats, and sodium [18, 19].

To the best of our knowledge, very little documented research is available about the dietary habits and intake among adult AD patients specific to Malaysia and Singapore. A recent study among Malaysian AD adults associated a reduced intake of vitamin A, magnesium and energy, and increased saturated fat with severe AD, and mild AD, respectively [20]. Despite some research being done to investigate the role of diet and nutrients in AD, the current findings can vary and be controversial amongst different ethnicities and communities. Therefore, given the growing interest and importance of dietary intake/habits in the field of allergic diseases, we must gain a deeper understanding of the risks and benefits of potential dietary interventions to optimise AD treatments.

From our recent Singapore/Malaysia Cross-sectional Genetics Epidemiology Study (SMCGES), dietary habit is a pivotal risk factor associated with triggering AD presentation besides genetics [21]. Our diet consists of a combination of foods and beverages. Thus, an assessment of single foods or single nutrients will be inadequate to fully evaluate the impact of our diets on health outcomes. Even though there are many ways to measure dietary intake, the use of dietary indices allows for the measurement of an overall diet quality or variety while having a lower researcher and respondent burden [22]. This study assesses the relationship and pattern of the intake frequency of different food types and specifically glycaemic index (GI) with AD in our SMCGES population. We used a novel food score indicator termed Quality of Diet based on Glycaemic Index Score (QDGIS) to determine the association between dietary quality of high and low GI and AD. We also evaluated the association between QDGIS and 16 food types with AD chronicity and severity. This original study hopes to establish and promote research on the role of dietary quality and intake habits in influencing AD progression.

A volunteer sample of subjects was taken from our ongoing epidemiology collection in Singapore and Malaysia Universities. Missing or invalid data for age, gender, race, and dietary habits were excluded and while selecting only the Chinese subjects to form a final sample size of 11,494 for the epidemiological analysis. In our cross-sectional sequential study, the standardized and validated International Study of Asthma and Allergies in Childhood (ISAAC) core questions were used [23]. The questions in our investigator-administered questionnaires collected information on the AD clinical symptoms, sociodemographics, dietary habits, anthropometric measurements, and familial and personal medical history.

A skin prick test (SPT) was conducted by trained staff to assess the atopic status of our subjects. When a wheal diameter of ≥3 mm appeared with either Blomia tropicalis and/or Dermatophagoides pteronyssinus this is indicative of a positive SPT response [3]. We classified the subjects based on their SPT reactivity and responses to the questions that assess personal AD symptoms. We identified 1,550 AD case (SPT positive with AD symptoms), 2,978 AD non-atopic control (SPT negative without AD symptoms), and 6.386 AD atopic control (SPT positive but without AD symptoms). A smaller subset of our population with recurrent rashes in flexural areas was cross-validated by a certified doctor diagnosis with high sensitivity (90.6%) and specificity (85.5%) reported. Additionally, the AD cases were further classified into various phenotypes based on preceding questions found in the same section surveying AD symptoms. The distribution of AD phenotypes is described in Table 1 (refer to Lim et al. [21] for more details on SPT criteria and questions on AD symptoms).

In our questionnaire, a specific section modified from Ellwood et al. [24] asked the participants about their dietary habits on 16 different types of foods, on average, in the past 12 months. The 16 different food types are meat (e.g., beef, lamb, chicken, pork), seafood (including fish), margarine, butter, eggs, pasta, fruits, vegetables (green and roots), pulses (peas, beans, lentils), nuts, milk, Yakult^®/Vitagen^®/similar yoghurt drinks (collectively termed as probiotic drinks), burgers/fast food, cereals (including bread), rice, and potatoes. There are three possible responses: (1) never or only occasionally; (2) once or twice/week; or (3) most or all days. Any other responses were invalid and excluded from our analysis. The distribution of the subjects’ (AD case and non-atopic control) responses to each food type is described in online supplemental Table 1 (for all online suppl. material, see https://doi.org/10.1159/000533942).

Furthermore, we derived a novel dietary score (QDGIS) to study the association between GI intake and AD. QDGIS is based on the summation of the GI value and intake frequency of the low and high GI foods that were classified with reference to the international tables of GI and glycaemic load values by Atkinson et al. [25, 26]. A specific score was assigned for the frequency of GI foods consumption in terms of the number of times/week based on the rubrics by Manousos et al. [27]. Based on our preliminary distributional assessment of the population, QDGIS was categorized into three classes: poor (high GI intake with a score ≤0), moderate (moderate GI intake with a score of 1.00–8.00), and good (low GI intake with a score ≥9).

Data entry and analysis were performed with Microsoft Excel (http://office.microsoft.com/en-us/excel/) and R program version 2021.09.0.351 (RStudio Team, 2021). Binary logistic regression was used to model the association between disease outcomes and potential risk factors. Results were presented as odds ratios (ORs) with 95% confidence intervals (CI). OR with a p value <0.05 and 95% CI not including 1.000 is considered to be statistically significant.

The AD atopic control group (55.6%) formed the largest proportion in the combined database as compared to AD cases (13.5%) and AD non-atopic controls (25.9%). Across the three groups, no significant difference was observed between the mean age and the BMI index. Most subjects were young undergraduate students (mean age between 21.89 and 22.35 years old) and had normal BMI (mean BMI range between 20.64 and 21.04 kg/m²). Generally, there was a greater proportion of subjects from all three groups in the moderate and good QDGIS classes. Among the AD case and non-atopic control groups, most subjects had a good QDGIS class with 37.1% and 38.8%, respectively. This indicates that the majority of our cohort frequently consume low GI foods in their diets. Statistical analysis using χ² revealed that there was a significant difference between QDGIS classes, BMI, family history of AD, and total monthly household income per capita across all three groups. Thus, these potential confounders were controlled in a separate multivariable logistic regression subsequently (refer to online suppl. Table 2).

The multivariable comparison highlighted that having a moderate QDGIS class (adjusted OR (AOR): 0.839; AOR indicates the OR after age and gender adjustments) was sufficient to confer a protective association for AD and severe AD development (refer to Table 2 section [A]). In the same multivariable analysis, the significant association of a good QDGIS class with AD presentation and AD severity was captured in the combined moderate/good QDGIS class (AOR: 0.857). A separate comparison of AD cases with AD atopic controls revealed that QDGIS was not significantly associated with AD (refer to online suppl. Table 3 section [I]), highlighting the association between QDGIS and AD was dependent on subjects’ atopic statuses. These associations between QDGIS and all AD manifestations were also independent of total monthly household income per capita (refer to online suppl. Table 2). Finally, a good QDGIS class was significantly associated with decreased AD chronicity (AOR: 0.769). Our results altogether indicate that an improvement from a moderate intake frequency of GI foods to a lower intake frequency of GI foods is necessary to reduce the odds of AD development, and further accentuates the importance of controlling the GI intakes in diets.

Results from our QDGIS analysis suggested frequency of intake of high GI food types had a positive association with AD development, but burgers/fast food was the only high GI food type found to have a significant association. Among all the foods analysed, burgers/fast food intake for most or all days presented the highest odds for AD development in the multivariable analysis (AOR: 1.758) and its association was not confounded by age and gender (AOR: 1.758) (refer to Table 2 section [B]). Those who had an intake of most or all days for burgers/fast food were 2.431 times more likely to get chronic AD compared to the non-atopic controls. Frequent burgers/fast food intake also indicated that subjects were almost twice more likely to develop moderate-to-severe AD (AOR: 1.920) compared to those who did not. Overall, the evidence suggested that frequent intake of burgers/fast food was a strong risk factor associated with AD development.

Of the 6 low GI food types, only three offered protective associations on AD. The frequent intake of fruits (AOR: 0.706), vegetables (AOR: 0.597), and pulses (AOR: 0.789) for most or all days reduced the odds of AD significantly (refer to Table 2 section [C], [D], [E]), respectively). Remarkably, there was a greater dose-dependent decline in the AORs with an increased intake frequency of vegetables from once or twice per week to most or all days as compared to those in fruits and pulses. This suggested that the associations of vegetable intake as a protective factor were stronger than those of fruits and pulses. The protective associations of vegetables on AD presentation were not limited to non-atopic controls but also to atopic controls (refer to online suppl. Table 3 section II). Finally, the associations between frequent intake of vegetables (along with burgers/fast food and butter) and AD remained significant even controlled for other confounders (refer to online suppl. Table 2).

Despite extensive efforts in understanding AD pathogenesis and disease management, the knowledge gap about the detailed role and implication of diets in AD is not yet bridged. We investigated the dietary habits among young Singaporean and Malaysian Chinese AD adults and identified that frequent intake of GI foods and high energy-dense foods like burgers/fast food is associated with an increased odds of AD. Our findings were consistent with a Japanese adult cohort in highlighting higher intakes of carbohydrate in the diets of AD patients [28]. Additionally, nutritional studies have supported that the intake of carbohydrates from foods with a higher GI directly correlated to fatalities from inflammation in older patients and upregulation of inflammatory markers such as IL-6 [29, 30]. IL-6 is a pro-inflammatory cytokine capable of priming Th₁₇ differentiation in human skin. Interestingly, higher GI foods also increased plasma IL-6 concentration in participants experiencing puberty and IL-6 expression was shown to be dysregulated and persistent in the skin and blood of AD patients [31‒33]. Conversely, the disruption of IL-6 receptor signalling improved AD conditions [34]. More carefully planned research is needed to fully define the overall impact of GI on IL-6 induced inflammation and its association with AD.

Burgers/fast food is characterised to be a high-fat food in addition to its high GI content. There is increasing evidence of a dietary pattern for excessive dietary fat intake to enhance chronic low-grade inflammation and thus, promoting atopic diseases [35, 36]. Higher dietary intake of omega-6 PUFAs and TFAs are associated with atopic diseases in adults [37]. Several studies revealed that butter, margarine, along with fast foods are high in saturated fatty acids and linoleic acid content and have been associated with increased allergic diseases [24, 38, 39]. Thus, the association between dietary fat and AD should not be neglected and warrants our attention.

The consumption of a natural plant-based diet often contributed to a rich source of vital phytonutrients, vitamins, minerals, and dietary fibres. These beneficial bioactive micronutrients synergistically contribute to an overall antioxidative and immunomodulatory effect. Several studies demonstrated a reduced risk of allergic diseases and AD with high fruit and vegetable intake [40, 41]. In particular, a vegetarian diet improved the AD symptoms by moderating the pro-inflammatory prostaglandin E2 synthesis and eosinophil counts [42]. While a lower intake of fruits and vegetables and their associated antioxidants (vitamins A, C, E, and flavonoids) reduces protection from oxidative stresses and an increased free oxygen radicals level enhances the development of allergic diseases in susceptible individuals [43]. Severe AD conditions were dismissed in AD-prone mice models when challenged with flavonoids from persimmon leaves [44]. A German clinical trial described high intravenous doses of vitamin C as effective in improving allergy-related symptoms [45]. Furthermore, maternal diets containing rich natural source of vitamin C stimulated breastmilk with a higher vitamin C concentration that could protect high-risk infants from atopy [46]. Altogether these findings emphasized that the protective associations of these plant-based foods can be due to other essential micronutrients in them and not heavily dependent on their low GI statuses.

The main limitation in our cross-sectional study lies in its inability to assess causality but only restricted to correlation between dietary habits and AD. Although a prospective longitudinal study may allow us to examine AD disease progression with dietary habits, it is more costly and time consuming. To provide a stronger evidence between dietary habits and AD exacerbations, follow-up studies involving a randomized controlled feeding trial are necessary in the future. Nonetheless, this large-scale study established the groundwork for further investigation on various dietary habits and guide AD patients in pursuing prolonged dietary modifications.

Dietary quality and intake habits, particularly related to burgers/fast food, are associated with AD manifestation and progression.

We would like to sincerely thank all our participants for willing to participate in this study.

This study was conducted in accordance with the principles of the Declaration of Helsinki and Good Clinical Practices, and in compliance with local regulatory requirements. The cross-sectional studies in Singapore were conducted on the National University of Singapore (NUS) campus annually between 2005 and 2019 with the approval of the Institutional Review Board (NUS-IRB Reference Code: NUS-07-023, NUS-09-256, NUS-10-445, NUS-13-075, NUS-14-150, and NUS-18-036) and by the Helsinki declaration. The cross-sectional studies in Malaysia were held in the Universiti Tunku Abdul Rahman (UTAR), and Sunway University. Ethical approval was granted, respectively, from the Scientific and Ethical Review Committee (SERC) of UTAR (Ref. code: U/SERC/03/2016) and Sunway University Research Ethics Committee (Ref. code: SUREC 2019/029). Before the data collection, all participants involved signed an informed consent form.

F.T.C. reports grants from Singapore Ministry of Education Academic Research Fund, Singapore Immunology Network, National Medical Research Council (Singapore), Biomedical Research Council (Singapore), and the Agency for Science Technology and Research (Singapore), during the conduct of the study; and has received consultancy fees from Sime Darby Technology Centre, First Resources Ltd, Genting Plantation, and Olam International, outside the submitted work. The other authors declare no other competing interests.

F.T.C. received grants from the National University of Singapore (N-154-000-038-001), Singapore Ministry of Education Academic Research Fund (R-154-000-191-112; R-154-000-404-112; R-154-000-553-112; R-154-000-565-112; R-154-000-630-112; R-154-000-A08-592; R-154-000-A27-597; R-154-000-A91-592; R-154-000-A95-592; R154-000-B99-114), Biomedical Research Council (BMRC) (Singapore) (BMRC/01/1/21/18/077; BMRC/04/1/21/19/315; BMRC/APG2013/108), Singapore Immunology Network (SIgN-06-006; SIgN-08020), National Medical Research Council (NMRC) (Singapore) (NMRC/1150/2008; OFIRG20nov-0033), National Research Foundation (NRF) (Singapore) (NRF-MP-2020-0004); Singapore Food Agency (SFA) (SFS_RND_SUFP_001_04; W22W3D0006); and the Agency for Science Technology and Research (A*STAR) (Singapore) (H17/01/a0/008; and APG2013/108). The funding agencies had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

F.T.C. conceived and supervised the current research study. J.J.L. conducted the literature review, analysed and interpreted the data, and wrote the manuscript. M.H.L. assisted in the review and editing of the manuscript. J.J.L., Y.Y.E.L., J.Y.N., P.M., Y.T.N., W.Y.T., Q.Y.A.W., S.A.M., Y.Y.S., Y.R.W., K.F.T., S.M.R.S., K.R., and Y.H.S. assisted in the recruitment of participants, and collation of data. All authors read and approved the final manuscript.

Current Address: Kavita Reginald, Yee-How Say: Department of Biological Sciences, School of Medicine and Life Sciences, Sunway University, 47500 Petaling Jaya, Selangor, Malaysia.

Data are not publicly available due to ethical reasons. Further enquiries can be directed to the corresponding author (F.T.C.).