Dietary and nutritional correlates of gastroesophageal reflux disease symptoms: A comprehensive island-wide study in Sri Lanka

Abstract

BACKGROUND

Diet is widely recognized as a significant contributor to the development and exacerbation of gastroesophageal reflux disease (GERD) symptoms. However, evidence on the impact of specific dietary factors remains inconsistent.

AIM

To identify the dietary and nutritional correlates of GERD symptoms in Sri Lankan adults.

METHODS

A cross-sectional analytic study was conducted on 1200 adults aged 18-70 years, recruited through stratified random cluster sampling across all 25 districts of Sri Lanka. Data was collected using interviewer-administered, country-validated tools: A GERD symptom screening tool and a food frequency questionnaire. Anthropometric measures (weight, height, abdominal and hip circumferences) were obtained using calibrated and sensitive instruments. Subjects having heartburn and/or regurgitation at least once per week were diagnosed to have GERD. They were compared with those without GERD (controls).

RESULTS

In our population (n = 1200, mean ± SD age: 42.7 ± 14.4 years, 46.1% males), significant dietary triggers of GERD symptoms included oily foods, bread, wheat, coffee, and sour or vinegar-based foods (P < 0.001, χ² test). Meal-related behaviors such as skipping breakfast, consuming late-night snacks (P < 0.001), and lying down within two hours after eating (P = 0.038) were also associated with GERD. No significant differences were observed between GERD and control groups in anthropometric measures, food portion sizes, or overall caloric and nutrient intake (P > 0.05). Total daily fat and fiber intake showed no correlation with GERD symptoms; however, bloating was associated with higher fat and fiber intake, greater fruit and vegetable consumption, and higher body mass index. These findings highlight dietary patterns and habits distinctive to the Sri Lankan population, where frequent consumption of oily and wheat-based foods may contribute to GERD symptom prevalence.

CONCLUSION

Some dietary items, such as oily foods, wheat, coffee, and sour/vinegar foods, and diet-related habits, such as fat intake, skipping breakfast, and lying down after meals, have a significant association with GERD symptoms as previously reported around the world. Bread is significantly associated with GERD symptoms in Sri Lankan adults, which has not been reported previously. Contrary to common belief, there was no significant association between GERD and nutrient intake or obesity. In addition, not only participants fulfilling criteria for GERD, but even controls were affected by food items triggering heartburn. Thus, lifestyle modifications based on diet and dietary habits are integral to managing GERD symptoms.

Key Words: Diet; Dietary habits; Gastroesophageal reflux disease; Nutrition; Heartburn

Core Tip: This nationwide study, the first of its kind in Sri Lanka, comprehensively evaluated dietary and nutritional factors associated with gastroesophageal reflux disease (GERD) symptoms among 1200 adults across all 25 districts. Oily foods, bread, wheat, coffee, and sour or vinegar-based foods, along with dietary habits such as skipping breakfast, consuming midnight snacks, and lying down soon after meals, showed significant associations with GERD symptoms. Interestingly, bread, identified as a novel trigger in this population, had not been reported previously in global literature. Contrary to popular belief, there was no association between GERD and overall caloric intake, nutrient composition, or obesity. These findings emphasize that effective GERD management should extend beyond pharmacological therapy to include culturally specific dietary and lifestyle modifications.

INTRODUCTION

Diet and nutrition play a significant role in the development and management of gastroesophageal reflux disease (GERD) and its symptoms[1]. However, studying the effects of diet on GERD is inherently complex due to multiple overlapping factors. This complexity can be broadly categorized into three main domains: Specific foods that trigger GERD symptoms, meal-related habits and behaviors, and the overall nutritional profile, including total caloric intake. Each of these factors interacts with individual's physiology, lifestyle, and cultural dietary practices, making the relationship between diet and GERD difficult to interpret.

Numerous dietary items have been identified as potential triggers of GERD symptoms, but the evidence remains inconsistent. Reported triggers vary widely, from spicy and oily foods to wheat products, citrus fruits, chocolate, and seafood[2]. This variability arises because food composition depends on multiple factors, including the ingredients used, their proportions, preparation and cooking methods, and regional or cultural dietary patterns. For example, spicy and oily foods vary widely across cultures, reflecting regional ingredients and cooking practices. In many Asian cuisines, spicy foods often include dishes prepared with chili peppers, black pepper, curry, or hot sauces - such as Sri Lankan curries, Indian masalas, Thai chili stir-fries, or Korean kimchi. In Western diets, examples include foods seasoned with hot sauces, jalapeños, or paprika-based spice blends. Oily foods typically refer to those high in fat or cooked using generous amounts of oil, butter, or ghee, such as deep-fried snacks, fast foods, and rich gravies. These foods may relax the lower esophageal sphincter (LES)[3] or delay gastric emptying[4], thereby aggravating gastroesophageal reflux symptoms. Even within the same cuisine, differences between households or personal taste preferences may alter the physiological effects of a meal. Furthermore, patients often modify their diet to avoid foods that provoke symptoms, introducing a recall bias and complicating the establishment of clear cause-and-effect relationships.

Dietary habits, such as meal timing, portion size, and eating speed, significantly influence GERD symptoms[2]. Large meals, late-night eating, or lying down soon after meals can increase intra-abdominal pressure and promote reflux[5]. However, quantifying such habits is challenging, as they vary between individuals and cultures and are influenced by social and occupational factors. This behavioral variability contributes to the complexity of dietary research, as it is difficult to isolate the effect of one behavior without accounting for others.

The overall diet composition - including total calorie intake and macronutrient balance - also impacts GERD. Diets high in fat may delay gastric emptying[4] and reduce LES tone[3]. Excessive caloric intake contributes to obesity, which is one of the strongest and most consistent risk factors for GERD[6,7]. Increased abdominal fat elevates intra-abdominal pressure and disrupts the gastroesophageal junction, facilitating reflux of gastric contents into the esophagus. Moreover, obesity is associated with anatomical changes such as hiatal hernia and functional impairments like transient LES relaxations and delayed gastric emptying, all of which predispose to reflux[8]. Epidemiological studies have shown a positive correlation between body mass index (BMI) and GERD prevalence, with central (visceral) obesity being a stronger predictor than overall body weight[7]. Importantly, weight reduction through caloric restriction or lifestyle modification has been shown to alleviate symptoms and reduce esophageal acid exposure[9]. Yet, these factors are often interrelated and distinguishing the independent role of nutritional composition from confounding variables, such as physical activity, metabolic rate, and comorbidities, remains methodologically challenging.

In summary, although diet and nutrition play a pivotal role in GERD, their precise effects are challenging to delineate due to the multifactorial nature of dietary patterns, behavioral influences, and overall nutritional balance. Understanding these complexities is essential for developing individualized dietary strategies in GERD management. This study aimed to conduct an in-depth analysis to identify the types of food and dietary habits that trigger GERD symptoms and, secondly, the effect of an overall nutrition profile on GERD.

MATERIALS AND METHODS

Study setting

We conducted a cross-sectional study on adult Sri Lankans aged 18 years to 70 years in all 25 districts of Sri Lanka. Any bedbound or wheelchair-bound patients or those with a history of major upper gastrointestinal surgery were excluded.

Sample size calculation

The sample size was calculated using accepted statistical methods. Here, the precision level was taken as 0.05, and the expected prevalence was taken as 50% (as the exact GERD prevalence is not known in Sri Lanka). With a 95% confidence interval, an allowance of 2 for the design effect of cluster sampling, and an anticipated non-response of 5%, a sample size of 810 was calculated. However, since it was intended to represent all 25 districts of the country, the number increased to 1200. Using census data from 2012 (the most recent) from the Sri Lanka Department of Census and Statistics, 40 clusters with a cluster size of 30 were distributed according to probability proportionate to the size of the population amongst the 25 districts. A cluster was defined as 30 adults living in a Grama Niladhari Division (the smallest administrative division in Sri Lanka). Simple random sampling of the Grama Niladhari Division lists was used to identify the clusters in each district.

Data collection method

The index house of a cluster was also identified randomly using maps, and every third house starting from it was approached. A member per household was selected (with informed consent) once again randomly, according to inclusion and exclusion criteria.

Instruments used in data collection

An interviewer-administered questionnaire was used to gather information. This was available in all three main languages of Sri Lanka, and the data collectors’ team was proficient in all three local languages (i.e., English, Sinhala, and Tamil). The questionnaire consisted of five parts. An interviewer-administered questionnaire used in this study included: A section for basic demographic details, a country-specific validated questionnaire for GERD[10], a section regarding dietary habits, a section related to food that triggers GERD, and a country-validated Food Frequency Questionnaire (FFQ) with food commonly used by Sri Lankans included in it[11].

Food items (e.g., spicy food, oily food) and dietary habits [quick eating (consuming a meal in 10 minutes), skipping breakfast (> 3 times per week), consuming snacks in the middle of the night (> 3 times per week)] included in our questionnaire were identified through a comprehensive literature review[1,12]. A previously country-validated GERD questionnaire was used to create a total score for seven common GERD symptoms (heartburn, regurgitation, chest pain, dysphagia, cough, burping, and bloating)[10]. Scores for individual symptoms were also calculated (frequency multiplied by severity).

The FFQ used in this study has been specifically developed to assess over 90 food items regularly consumed by different ethnic groups in Sri Lanka[11]. This questionnaire assesses both the frequency and quantity of different food items. For assessment of quantity, it contains color photographs of three different-sized portions of food, such as rice, vegetables, chicken, and dhal, together with open-ended response scales, and some other food items are assessed using cups, spoons, and item numbers. Frequency is calculated using the number of units taken at a time (‘per day’, ‘week’, ‘month’, or ‘year’) for each food item. The number of meals participants usually consume daily was also noted (e.g., 3 meals per day). The meals were added to the monthly total (e.g., 90 meals per month), with rice meals added to make up the deficit. Nutrient data for each food item have been obtained from the United States Department of Agriculture database[13], and the Indian Food Composition tables of the National Institute of Nutrition, India[14]. The volume of a meal was categorized as small volume (portion A of a rice plate on the FFQ) and large volume (portions B and C of a rice plate on the FFQ). Accepted protocols and worldwide standards were used to measure the participants’ weight, height, abdominal circumference, and hip circumference, and to calculate BMI[15].

Definitions used

GERD was defined as heartburn and/or reflux at least once a week, an internationally used criterion for screening GERD[1]. Participants were asked to consider the past 30-day period only when questioned. Controls are those who did not meet the above definition of GERD. GERD was identified using the above definition, designed to detect typical reflux symptoms, including heartburn and regurgitation. Although endoscopic or pH-metric confirmation was not performed, this symptom-based approach has been employed in multiple large-scale epidemiological studies to define GERD prevalence in community settings[1]. Hence, the term ‘GERD’ in this study refers to symptom-defined GERD.

Study duration

The research was conducted from May 2021 to July 2021.

Statistical analysis

Using Statistical Package for the Social Sciences (SPSS) 28, the results were analyzed according to standard statistical methods such as the Mann-Whitney test for ratio data, the χ² test for binary data, and the Spearman correlation coefficient. Multivariable analysis was performed using backward logistic regression to detect independent associations and the impact of confounders. Here, P < 0.05 was considered statistically significant.

RESULTS

A total of 1200 participants were recruited from all 25 districts of Sri Lanka (male:female, 1:1.16). All participants completed the questionnaires and were included in the analysis. The mean ± SD age was 42.7 ± 14.4 years. Heartburn and/or reflux at least once a week, an internationally used criterion for probable GERD, was used to calculate GERD prevalence. Participants who did not meet the above definition of GERD were considered controls.

Foods that trigger heartburn in the study population

Many food items are shown to trigger heartburn. Table 1 compares the self-reported prevalence of food triggers for heartburn and reflux in participants with GERD and controls. Table 2 compares the number of portions of different food items consumed per month between GERD and controls.

Table 1 Self-reported prevalence of food-related triggers for heartburn and reflux in subjects fulfilling criteria for gastroesophageal reflux disease and non-gastroesophageal reflux disease controls, n (%).

Food item	Total (n = 1200)	GERD (n = 304)	Controls (n = 896)	Adjusted OR	95%CI for OR	P-value1
Spicy foods	399 (33.3)	196 (64.5)	203 (22.7)	0.709	0.483-1.042	0.080
Oily foods	393 (32.8)	212 (69.7)	181 (20.2)	0.333	0.226-0.490	0.001
Bread	393 (32.8)	207 (68.1)	186 (20.8)	0.478	0.309-0.738	0.001
Wheat	305 (25.4)	173 (56.9)	132 (14.7)	0.508	0.330-0.784	0.002
Tea	146 (12.2)	85 (28)	61 (6.8)	1.190	0.738-1.919	0.475
Coffee	65 (5.4)	47 (15.5)	18 (2)	0.470	0.239-0.923	0.028
Chocolate	30 (2.5)	19 (6.3)	11 (1.2)	1.154	0.472-2.822	0.754
Fizzy drinks	150 (12.5)	88 (28.9)	62 (6.9)	0.647	0.417-1.003	0.052
Sour	263 (21.9)	154 (50.7)	109 (12.2)	0.510	0.346-0.752	0.001
Milk	32 (2.7)	19 (6.3)	13 (1.5)	0.758	0.329-1.749	0.516

¹Logistic regression analysis, gastroesophageal reflux disease vs controls.

GERD: Gastroesophageal reflux disease; OR: Odds ratio; CI: Confidence interval.

Table 2 Comparison of the number of portions of different food items consumed within a month between gastroesophageal reflux disease and controls, mean ± SD.

Food item	Portions consumed per month		P-value1
	GERD (n = 304)	Controls (n = 896)
Bread	19.8 ± 28.9	20.4 ± 27.2	0.608
Wheat	36.5 ± 35.4	38.6 ± 37.7	0.648
Tea	53.6 ± 45.1	58.9 ± 50.7	0.084
Coffee	5.4 ± 14	7.5 ± 22.5	0.255
Chocolate	3.7 ± 8.9	2.3 ± 5.7	0.392
Fizzy drinks	2.7 ± 7.5	2.8 ± 7.8	0.052
Sambol	18.2 ± 17.4	17.5 ± 16.6	0.624
Milk	32.9 ± 31.3	29.8 ± 29.5	0.176
Non-sweet biscuits	36.6 ± 45.3	36.9 ± 46.8	0.922
Milk rice (Kiribath)	4.2 ± 5.3	4.7 ± 5.9	0.260
Hoppers	5.8 ± 12	4.8 ± 9.4	0.183
Rice	65.8 ± 15.4	65.1 ± 16.4	0.733
Meat	21.4 ± 27.4	21.3 ± 25.6	0.558
Banana	29.4 ± 29	28.3 ± 30.7	0.335

¹Mann-Whitney U test, gastroesophageal reflux disease vs controls.

GERD: Gastroesophageal reflux disease.

Association between nutrition profile and GERD symptoms

Tables 3 and 4 compare the daily nutrient consumption and anthropometric parameters of the participants, such as calories, carbohydrates, lipids, protein, fiber, vegetable and fruit portions, BMI, and waist-hip ratio between GERD and controls. The volume of a meal (categorized as small volume or large volume) was not found to be significantly different between GERD and controls (χ² test, P = 0.620).

Table 3 Comparison of nutrient intake and anthropometric parameters between gastroesophageal reflux disease and controls, mean ± SD.

	Total (n = 1200)	GERD (n = 304)	Controls (n = 896)	P-value1
Total calorie intake per day (kcal)	2223 ± 532	2234 ± 524	2225 ± 535.5	0.792
Total carbohydrate intake per day (g)	392.5 ± 100	389.6 ± 99.9	393.4 ± 100.3	0.560
Total fat intake per day (g)	44.3 ± 17.4	45.1 ± 16.3	44.0 ± 17.8	0.356
Total protein intake per day (g)	60.1 ± 17.3	60.5 ± 16.5	60.0 ± 17.5	0.707
Total fiber intake per day (g)	21.5 ± 9	21.9 ± 9	21.4 ± 9	0.360
Fruits and vegetable portions per day (g)	4.07 ± 2.2	4.2 ± 2.2	4.0 ± 2.1	0.167
BMI (kg/m2)	24.6 ± 4.5	24.8 ± 4.3	24.5 ± 4.5	0.281
Waist-hip ratio	0.92 ± 0.1	0.93 ± 0.1	0.92 ± 0.1	0.348

¹Mann-Whitney U test, gastroesophageal reflux disease vs controls.

GERD: Gastroesophageal reflux disease; BMI: Body mass index.

Table 4 Correlation between gastroesophageal reflux disease screening tool score and nutrient intake and anthropometric parameters.

	Correlation coefficient	P-value1
Total calorie intake per day (kcal)	0.008	0.787
Total carbohydrate intake per day (g)	-0.022	0.445
Total fat intake per day (g)	0.074	0.010
Total protein intake per day (g)	0.031	0.277
Total fiber intake per day (g)	0.075	0.009
Fruits and vegetable portions per day (g)	0.040	0.163
BMI (kg/m2)	0.045	0.117
Waist-hip ratio	0.038	0.186

¹Spearman correlation coefficient.

BMI: Body mass index.

Association between nutritional profile and bloating

We also investigated the correlation between bloating score (severity multiplied by frequency in the GERD screening tool for bloating) and nutritional measurements (Table 5). The volume of a meal did not have a bearing on whether patients had bloating episodes during the month of questioning (χ² test, P = 0.288).

Table 5 Correlation between the bloating score and nutrient intake and anthropometric parameters.

	Correlation coefficient	P value1
Total calories intake per day (kcal)	0.046	0.114
Total carbohydrate intake per day (g)	0.019	0.503
Total fat intake per day (g)	0.076	0.009
Total protein intake per day (g)	0.057	0.049
Total fiber intake per day (g)	0.105	< 0.001
Fruits and vegetable portions per day	0.076	0.009
BMI (kg/m2)	0.087	0.002
Waist-hip ratio	0.040	0.162

¹Spearman correlation coefficient.

Bloating score: Severity multiplied by frequency of bloating reported in the gastroesophageal reflux disease screening tool. BMI: Body mass index.

Dietary habits associated with GERD symptoms

Table 6 compares undesirable dietary habits between GERD and controls. Skipping breakfast, lying down within 2 hours of having a meal, and consuming snacks in the middle of the night were significantly associated with GERD.

Table 6 Comparison of undesirable dietary habits between gastroesophageal reflux disease and controls, n (%).

	GERD (n = 304)	Controls (n = 896)	P value1
Quick eating (consuming a meal in less than 10 minutes)	166 (54.6)	492 (54.9)	0.926
Skipping breakfast	113 (37.2)	207 (23.1)	< 0.001
Lying down within 2 hours of having a meal	149 (49)	378 (42.2)	0.038
Consuming snacks in the middle of the night	36 (11.8)	51 (5.7)	< 0.001

¹Pearson χ².

GERD: Gastroesophageal reflux disease.

Association between wheat and bread, and GERD symptoms

Of the total population, 116 (9.67%), 28 (2.33%), and 277 (23.08%) had GERD symptoms for bread only, other wheat products only, or both bread and other wheat products, respectively. Of the 421 (35%) who had heartburn with either wheat or bread or both, 194 (46.1%) had bloating in the last month. Out of the 198 who have had symptoms for food items other than wheat or bread within their lifetime, only 71 (35.9%) suffered from bloating within the last month. There was a significant difference (P = 0.017) between those who have had symptoms of bread, wheat, or both in their lifetime vs those who have had bloating within the last month.

Association between alcohol and GERD symptoms

Only 257 participants of the 1200 population claimed to have used alcohol in the past month of questioning. Of them, 50 admitted to being frequent (daily or every other day) alcohol users. When considering the 257 participants, 19 (25%) of those with GERD had heartburn or regurgitation triggered by alcohol, while it was 12 (6.6%) amongst controls (P < 0.001, χ² test). There was no significant difference between the units of alcohol used per month between GERD and controls, with higher usage in GERD (mean ± SD: 80.9 ± 113) when comparing controls (mean ± SD: 69.2 ± 109) (P < 0.001, Mann-Whitney U test).

DISCUSSION

Identifying dietary factors associated with GERD is challenging due to regional, cultural, and disease-related variations in food habits, as well as the avoidance of triggers and the consumption of symptom-relieving foods. This nationwide study examined associations between GERD symptoms, diet, nutrient intake, and dietary behaviors in Sri Lankan adults.

Several foods triggered GERD symptoms in both GERD and non-GERD participants, though more frequently among those with GERD. The main triggers were oily foods, bread, wheat, coffee, and sour or vinegar-based items. Symptom occurrence was unrelated to portion size, and no significant differences were found in total caloric or nutrient intake. However, GERD symptoms correlated with total fat and fiber intake, while bloating correlated with fat, fiber, fruit, and vegetable intake, and BMI. Skipping breakfast, lying down within two hours after meals, and midnight snacking were also significantly associated with GERD. GERD is a common global disorder[16], including in Sri Lanka[17]. Dietary and related factors are recognized triggers, but their associations with GERD vary across regions, likely due to differences in dietary habits, food composition, meal timing, and genetic or microbiome influences on esophageal sensitivity.

Spicy food and bread were the most commonly reported triggers of heartburn and regurgitation. In our study, 33.3% of participants experienced heartburn after consuming spicy food, despite Sri Lankans’ high chili intake (2.1-2.3 kg per capita annually)[18]. Although more GERD participants reported symptoms related to spicy food, the difference was not statistically significant compared to controls. Chili, the main component of “spicy” food, is a well-known trigger in both Eastern and Western studies[19-22]. Capsaicin activates transient receptor potential vanilloid-1 receptors, causing a burning sensation[23], and may irritate the esophageal mucosa[24], delay gastric emptying[25], and increase acid secretion[26]. Individuals with functional gastrointestinal disorders may have increased transient receptor potential vanilloid-1 expression, heightening chili sensitivity[27]. Exact chili intake could not be assessed using the FFQ.

In our study, 32.8% of Sri Lankans were found to suffer from heartburn following the consumption of oily food, while there was a significant difference in the percentage of heartburn triggered among GERD and controls. Oily and fatty foods delay gastric emptying[4], increase transient LES relaxations[28], and reduce LES pressure[3]. While several international studies report a significant link between high-fat intake and GERD symptoms[29], findings remain variable across populations. Large population-based studies from East Asia have reported strong associations between high-fat diets and reflux symptoms; for example, a Chinese cohort study found an adjusted odds ratio of approximately eight for high-fat food consumption and GERD symptoms[30]. In contrast, a cross-sectional study from Iran found no significant association between total fat intake and GERD prevalence[31]. Western crossover trials have demonstrated that very high-fat test meals consistently increase symptom perception[32], although earlier United States epidemiological analyses suggested that obesity, rather than the proportion of dietary fat, was the main predictor of GERD-related hospitalization[33]. These discrepancies likely reflect differences in habitual diet (e.g., frying vs steaming), meal timing, BMI distribution, and the diagnostic methods used to define GERD, such as symptom-based questionnaires vs endoscopic or pH monitoring criteria. Like with spicy food, the FFQ used in this study was not sensitive to estimate the oil and fat consumption of participants.

In our study, 277 participants (23.1%) reported symptoms with wheat-based foods. Wheat can cause gastrointestinal symptoms that overlap with GERD via both immune and non-immune mechanisms[34]. However, evidence on wheat or gluten-related disorders in Sri Lanka is limited, and neither our findings nor those of previous studies have shown an association between wheat consumption and GERD. A novel finding in our study is that 9.67% of the study population experiences heartburn and/or regurgitation after consuming bread, but not other wheat products, a finding not previously reported. Bread is not considered a food that causes heartburn in the worldwide literature, except for a South Korean study that failed to show an association between GERD symptoms and bread[20] and an Iranian study that did show an association between GERD symptom prevalence and bread[35]. We hypothesized that yeast, added during bread making, is a possible contributor to the association between bread and GERD symptoms. This perception did not arise from a structured survey item but was spontaneously expressed by several participants during the interviews, who attributed their GERD symptoms to yeast in bread. These qualitative insights were recorded as part of the interviewers’ observations. Yeast, when used in higher quantities, can ferment starch, making the bread more acidic, which is suggested as a possible reason for inducing GERD symptoms. Several interviews with local Sri Lankan bakers also showed that bakeries use double or triple the recommended quantity of yeast during breadmaking to decrease production time. However, future research is needed to verify these claims. There was no significant difference in the number of bread portions consumed per month between GERD patients and controls. The association between bread GERD symptoms may be unique to Sri Lanka, since studies done in our neighboring countries, such as India, have not mentioned this association. There is no research comparing the bread consumption patterns between India and Sri Lanka. Bread is a commonly used food product in Sri Lanka due to its low cost and ease of purchase, largely facilitated by the culture of bread vans or tuk-tuks delivering bread daily to households.

Despite 87.2% of the population drinking tea as a habit, 12.2% of the Sri Lankan population experiences dyspeptic symptoms from it. Tea is also associated with GERD symptoms[36], which may be due to the action of theophylline, which is a major component of black tea[37], and has been shown to decrease LES pressure[36]. However, our results showed no statistically significant difference between GERD and controls in the percentage of individuals who experienced heartburn after consuming tea. In this study, those with GERD symptoms drank less tea than controls, which may be a result of avoiding trigger foods. Several studies from both the East and the West have reported that frequent tea consumption is positively associated with reflux symptoms[38,39]. However, the practice of tea drinking varies widely, depending on the type of tea (black, green, or milk tea), brewing strength, temperature of consumption, caffeine content, and cultural drinking habits. It may be postulated that consuming tea at very high temperatures may independently increase mucosal irritation and symptom perception, even in the absence of increased acid reflux. In this study, participants were asked to consider the consumption of any type of tea, e.g., black or green tea, with or without sugar or milk, and we did not differentiate between types of tea, added constituents or temperature. These need to be explored in future studies.

Coffee is believed to reduce LES pressure[40]. Only 5.4% of participants reported heartburn or regurgitation after coffee, which is fewer than after tea. Similar to tea, several Asian and Western studies have reported that frequent coffee intake is associated with increased reflux symptoms[38,39]. However, a meta-analysis has found no significant association between coffee intake and GERD[41]. Differences may be related to the type of coffee, caffeine content, temperature, and preparation methods.

Two-point seven percent of Sri Lankans reported experiencing heartburn or regurgitation triggered by milk consumption, although there was no significant difference in milk intake between individuals with GERD and controls. Several international studies have shown no association between milk consumption and GERD[39,42], while others have suggested a possible link between milk and GERD symptoms[43-45]. Lactose has been shown to increase transient LES relaxations[46], and experimental studies indicate that milk can stimulate gastric acid secretion[47]. The prevalence of lactase deficiency among Sri Lankan adults has been reported to range from 66.2% to 78.8%[48]. Although our study did not specifically inquire about diarrhea or bloating associated with milk, it is plausible that lactose intolerance may lead to bloating and thereby contribute to heartburn or regurgitation. As with many dietary factors, there remains considerable ambiguity regarding milk and dairy products - whether low-fat or high-fat, fermented, or mixed. Further research would be valuable to clarify the relationship between lactose intolerance and GERD symptoms.

Fizzy drinks, chocolate, and acidic foods are well-known triggers of GERD symptoms globally[49], and our study yielded similar results among Sri Lankans. Carbonated beverages and chocolate reduce LES pressure[50,51], while acidic foods lower esophageal pH, mimicking reflux[52]. Alcohol is a recognized risk factor for GERD[53], acting through mucosal injury, delayed gastric emptying, increased acid secretion, and LES dysfunction[54]. A meta-analysis showed drinkers had 1.5-fold higher odds of GERD than non-drinkers, with risk increasing with frequency[55]. Given cultural and religious beliefs, only 21.4% of our participants admitted to using alcohol, with hardly any females. Surprisingly, while alcohol triggered heartburn in GERD patients significantly, the consumption of alcohol units per month was also higher among GERD patients than among controls.

We anticipated differences in diet, nutrient intake, and food portions between GERD and controls, but none were statistically significant. GERD participants consumed less bread, wheat, tea, coffee, and fizzy drinks, while spicy dishes such as “sambol” were eaten more often, though this was not significant. Higher rice intake among GERD participants may reflect reduced wheat consumption. These findings may indicate symptom-driven dietary modification, warranting further investigation.

Obesity is considered one of the strongest predictors of GERD[7]. The postulated mechanisms include increased intra-abdominal pressure, vagal abnormalities, higher gastric acid secretion, weakening of LES by cytokines produced by visceral fat, and hiatal hernia[6,8]. However, in our study, we found no significant difference between GERD and controls for markers of obesity, including BMI and waist-to-hip ratio. There was also no evidence of increased consumption of calories or fat between GERD and controls. There was also no significant correlation between the GERD screening tool symptoms score and nutrient intakes, except regarding fat and fiber. Although obesity is widely recognized as a major risk factor for GERD in both Western and Asian populations, our study did not demonstrate a significant association between BMI and GERD symptoms in the Sri Lankan population. Interestingly, participants classified as having GERD reported lower caloric intake, although this difference was not statistically significant. As our study was cross-sectional in design, the presence of GERD symptoms may have influenced dietary intake, thereby obscuring a potential association between GERD and obesity.

Although nutrient intake and BMI were not associated with GERD symptoms, bloating symptom scores (calculated by multiplying bloating frequency and severity) correlated significantly with total fat, protein, fiber, fruit, and vegetable intake, as well as BMI. High fiber, fruit, and vegetable intake can cause bloating through bacterial fermentation[56], while protein- and fat-rich meals may delay gastric emptying, contributing to bloating[56-59]. Bloating is also common in functional gastrointestinal disorders and motility disturbances that can mimic GERD[60,61].

Our study examined both dietary factors and related habits, including consuming large meals, eating late or skipping meals, rapid eating, and midnight snacking. A shorter dinner-to-bed interval was significantly associated with GERD, consistent with previous findings[5,62], as the supine position promotes reflux[63]. Midnight snacking was more common among GERD participants, as reported globally[64], possibly linked to heartburn or GERD-related insomnia[65]. Skipping meals was also associated with GERD[64,66], likely because breakfast helps buffer early-morning gastric acidity[67]. Although large meals and rapid eating have been implicated in heartburn, thought due to problems related to gastric accommodation and delayed gastric emptying, respectively[20,68], our study found no significant association between meal size, eating speed, or bloating scores and GERD symptoms. A key strength of this study is its large, nationally representative sample of 1200 participants. It is the first in Sri Lanka to use a country-validated FFQ and standardized symptom-based criteria for GERD, allowing comparison with international data.

This study has several limitations. GERD was diagnosed solely on symptom criteria, which may overlap with other gastrointestinal disorders. In a previous unpublished Sri Lankan study, only 46.3% of patients with reflux symptoms had GERD confirmed by endoscopy or pH-impedance testing, while the majority had mimickers, such as reflux hypersensitivity, functional heartburn, or gastritis. Second, the FFQ used relied on participants’ one-month dietary recall, which may be affected by income, seasonal and religious variations, or social desirability bias - such as underreporting by obese or low-income individuals. In addition, the FFQ only identifies 90 items commonly used by Sri Lankans, and some items, like pizzas, local fruits, and other desserts, have not been included. Variations in cooking have not been accounted for, especially for vegetables, e.g., whether used as a salad or in curry. One factor we noted was that, during the FFQ, the total number of meals consumed might be underestimated. We attempted to rectify this by totaling the number of meals up to 90 per month after asking the participants whether they ate three meals daily. The number of meals unaccounted for was substituted with the rice meals that the participant regularly ate. Another limitation is the lack of data on the consumption of oil. Sri Lankan dishes often use varying amounts of oil, coconut milk, or grated coconut, which can lead to inaccuracies in estimating fat intake and overall caloric intake[69]. However, despite its limitations, the FFQ is the most comprehensive, objective, and informative tool available to assess diet and nutrient intake in Sri Lanka. Although data on comorbidities such as asthma and chronic obstructive pulmonary disease were collected, they were not analyzed in this study, which focused on dietary factors. Thus, their potential influence on GERD symptom expression cannot be excluded. Finally, GERD itself may prompt patients to modify their diets by adjusting food quantity or avoiding items like wheat, based on symptoms or advice. Such disease-related modifications can lead to false-negative associations.

CONCLUSION

This nationwide, community-based study, conducted across all 25 districts of Sri Lanka, highlights the dietary and lifestyle factors associated with GERD symptoms. Foods such as oily dishes, bread, wheat, coffee, and sour or vinegar-based items, as well as habits like high-fat intake, skipping breakfast, and lying down immediately after meals, were significantly associated with GERD symptoms. However, no association was found between GERD symptoms and nutrient intake or obesity, possibly because affected individuals made dietary adjustments. Notably, even controls reported food-related heartburn triggers. These findings emphasize that managing GERD requires lifestyle and dietary modification in addition to medication.