Relationship Between Meal Frequency and Gastroesophageal Reflux Disease (GERD) in Iranian Adults

Abstract

Background:
The association between frequency of meals and snacks and gastroesophageal reflux disease (GERD) is less studied in Middle-Eastern countries.

Aim:
We aimed to determine the relationship between meal and snack frequency with GERD symptoms in a large sample of Iranian adults.

Methods:
In this cross-sectional study, 4669 individuals filled out a questionnaire about their number of meals and snacks. Frequency of total meals was defined by summing up the frequency of main meals and snacks, and participants were categorized into four categories: < 3, 3-5, 6-7 and ≥ 8 meals/day. GERD was defined as having heartburn sometimes or more during the last 3 months. The severity of disease was assessed.

Results:
The prevalence of GERD in the study population was 23.7%. There was no significant association between meal or snack frequency and GERD symptoms in the whole population. However, after adjustment of all potential confounders, we found that women who consumed 1-2 or 3-5 snacks per day, compared with those who never had snacks, had a 41% (OR 0.59; 95% CI 0.42-0.84) and 51% (OR 0.49; 95% CI 0.32-0.75) reduced risk of having GERD, respectively. Women who consumed 6-7 or ≥ 8 snacks and meals per day had a 38% (OR 0.62; 95% CI 0.41-0.96) and 43% (OR 0.57; 95% CI 0.34-0.95) risk reduction for GERD compared with those who ate < 3 snacks and meals per day.

Conclusion:
We found no significant association between meal frequency and GERD symptoms in the whole population. Gender-specific analysis revealed inverse associations between meal and snack frequency and GERD in Iranian women. Further prospective studies are required to confirm these associations.

Full text

Vol:.(1234567890)
Digestive Diseases and Sciences (2018) 63:2998–3008
https://doi.org/10.1007/s10620-018-5200-7
1 3
ORIGINAL ARTICLE
Relationship Between Meal Frequency andGastroesophageal Reux
Disease (GERD) inIranian Adults
MehrbodVakhshoori1· AmmarHassanzadehKeshteli1,2· ParvaneSaneei3,4· AhmadEsmaillzadeh5,6,7·
PeymanAdibi8
Received: 11 March 2018 / Accepted: 5 July 2018 / Published online: 16 July 2018
© Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract
Background The association between frequency of meals and snacks and gastroesophageal reflux disease (GERD) is less
studied in Middle-Eastern countries.
Aim We aimed to determine the relationship between meal and snack frequency with GERD symptoms in a large sample
of Iranian adults.
Methods In this cross-sectional study, 4669 individuals filled out a questionnaire about their number of meals and snacks.
Frequency of total meals was defined by summing up the frequency of main meals and snacks, and participants were cat-
egorized into four categories: < 3, 3–5, 6–7 and ≥ 8 meals/day. GERD was defined as having heartburn sometimes or more
during the last 3months. The severity of disease was assessed.
Results The prevalence of GERD in the study population was 23.7%. There was no significant association between meal
or snack frequency and GERD symptoms in the whole population. However, after adjustment of all potential confounders,
we found that women who consumed 1–2 or 3–5 snacks per day, compared with those who never had snacks, had a 41%
(OR 0.59; 95% CI 0.42–0.84) and 51% (OR 0.49; 95% CI 0.32–0.75) reduced risk of having GERD, respectively. Women
who consumed 6–7 or ≥ 8 snacks and meals per day had a 38% (OR 0.62; 95% CI 0.41–0.96) and 43% (OR 0.57; 95% CI
0.34–0.95) risk reduction for GERD compared with those who ate < 3 snacks and meals per day.
Conclusion We found no significant association between meal frequency and GERD symptoms in the whole population.
Gender-specific analysis revealed inverse associations between meal and snack frequency and GERD in Iranian women.
Further prospective studies are required to confirm these associations.
Keywords Gastroesophageal reflux· Meal frequency· Snack frequency· Feeding behavior
* Parvane Saneei
saneeip@yahoo.com
Mehrbod Vakhshoori
mehrbod10@yahoo.com
Ammar Hassanzadeh Keshteli
ahassanz@ualberta.ca
Ahmad Esmaillzadeh
a.esmaillzadeh@gmail.com
Peyman Adibi
payman.adibi@gmail.com
1 Integrative Functional Gastroenterology Research Center,
Students’ Research Committee, Isfahan University
ofMedical Sciences, Isfahan, Iran
2 Department ofMedicine, University ofAlberta, Edmonton,
AB, Canada
3 Department ofCommunity Nutrition, School ofNutrition
andFood Science, Isfahan University ofMedical Sciences,
Isfahan, Iran
4 Food Security Research Center, Isfahan University
ofMedical Sciences, Isfahan, Iran
5 Obesity andEating Habits Research Center, Endocrinology
andMetabolism Molecular–Cellular Sciences Institute,
Tehran University ofMedical Sciences, Tehran, Iran
6 Endocrinology andMetabolism Research Center,
Endocrinology andMetabolism Clinical Sciences Institute,
Tehran University ofMedical Sciences, Tehran, Iran
7 Department ofCommunity Nutrition, School ofNutritional
Sciences andDietetics, Tehran University ofMedical
Sciences, Tehran, Iran
8 Integrative Functional Gastroenterology Research Center,
Isfahan University ofMedical Sciences, Isfahan, Iran

2999Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
Abbreviations
GERD Gastroesophageal reflux disease,
GI Gastrointestinal
SEPAHAN Study on the Epidemiology of Psychologi-
cal, Alimentary Health and Nutrition
FGID Functional gastrointestinal disorder
LES Lower esophageal sphincter
FFQ Food Frequency Questionnaire
GPPAQ General Practice Physical Activity
Questionnaire
OR Odds ratio
95% CI 95% Confidence interval
US United States
IUMS Isfahan University of Medical Sciences
ANOVA Analysis of variance
SPSS Statistical Package for the Social Sciences
Introduction
Gastroesophageal reflux disease (GERD) is a common gas-
trointestinal disorder [1] characterized by heartburn and/
or regurgitation of gastric content from the stomach to the
esophagus [2]. Based on different diagnostic criteria, the
prevalence of GERD varies from 10 to 48% in different stud-
ies [1, 3–5]. Some studies in Iran have indicated that 21.2%
of the adult population is affected [5]. This condition affects
individual’s quality of life and causes a great economic bur-
den. The total cost of GERD per person per year was esti-
mated to be $195 [6].
Several factors have been reported to be involved in the
pathophysiology of GERD. Socio-cultural factors, obesity
[2, 7–10], cigarette smoking [2, 7, 8, 10], physical activ-
ity [2, 7], alcohol consumption [7, 8, 10] and psychologic
agents [11, 12] have been associated with GERD. Among
environmental factors, diet-related practices have received
great attention. Large meal sizes [13], late evening meals,
eating speed [14] and regularity/irregularity of meal pat-
terns [14–16] have previously been related to GERD. To our
knowledge, few studies are available linking meal frequency
with GERD symptoms. For example, in a cross-sectional
study in Korea, no significant correlation was found between
these two variables [16]. Another study done by Olive-
ria etal. [17] in the USA showed that 402 of 2000 (20%)
patients claimed that the number of daily meals caused their
heartburn as one of the most frequent GERD symptoms, but
this was not statistically significant.
Previous studies have been limited in sample size, have
been mostly done in western countries and have not consid-
ered the potential effect of other diet-related practices that
might confound the associations. In addition, socio-cultural
factors in the Middle-Eastern population are totally different
from those in western populations. Given the inconsistencies
concerning the association between meal frequency and
GERD, we aimed to investigate whether meal frequency is
related to the frequency of GERD symptoms in a large group
of Iranian adults.
Materials andMethods
Participants
This cross-sectional study was conducted in the context
of the Study on the Epidemiology of Psychological, Ali-
mentary Health and Nutrition (SEPAHAN) [19]. The main
aim of SEPAHAN was to evaluate the relation of different
psychologic and nutritional factors with functional gastro-
intestinal disorders (FGIDs) in Isfahan, Iran. To increase
the participation rate and accuracy of collected data and to
decrease participants’ fatigue, this project was done in two
main phases among the general adult population of Isfahan
Province consisting of medical university nonacademic staff,
including crews, employees and managers. In the first phase,
self-administered questionnaires were distributed among
10,087 subjects to obtain information on demographic, life-
style and nutritional factors, and 8691 of them completed
the questionnaires. In the second phase of the project, which
was aimed to collect information on the gastrointestinal and
psychologic profile of participants, self-administered ques-
tionnaires were sent to the same subjects. The response rate
in the first and second phase of SEPAHAN was 86.16 and
64.6%, respectively. After data merging from two phases,
complete information of 4669 participants was available for
the current analysis. Data for other participants could not be
used in the merging process because: (1) some participants
had no information at one of the phases, (2) some did not
complete their identification code in phase 1 or 2, which
prohibited us merging their data; (3) some had missing data
on exposure, outcome or covariate variables. This study was
approved by the Regional Bioethics Committee, affiliated to
Isfahan University of Medical Sciences.
Assessment ofMeal Frequency
To assess meal frequency, subjects were asked to quantify
the average number of main meals they consumed per day
(one, two or three meals). They were also asked to provide
an estimation of the numbers of snacks they consumed
(none, one or two, three to five or more than five snacks
per day). Total meal frequency was assessed by adding both
main meals and snacks and was categorized into four groups
(less than three, three to five, six to seven and eight or more
meals).

3000 Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
Assessment ofGERD
From an Iranian version of the Rome Ш criteria [19], a vali-
dated self-administered questionnaire was used to assess the
presence of different gastrointestinal symptoms including
heartburn in the preceding 3months (never or rarely, some-
times, often or always). Participants were also asked about
the severity of heartburn using a four-item rating scale (mild,
moderate, severe and very severe). GERD was defined as
the presence of heartburn sometimes or more during the
3months prior to the initiation of the study.
Assessment ofOther Variables
Information on other variables such as age, sex, marital
status, education level, smoking and diabetes was obtained
through a self-administered questionnaire. Data on weight
(in kilograms) and height (in centimeters) were obtained
using a self-reported questionnaire. Body mass index (BMI)
was calculated as weight in kilograms divided by height in
square meters (kg/m2). The General Practice Physical Activ-
ity Questionnaire (GPPAQ) was used for determining adults’
physical activity [20]. Individuals were categorized to the
active group (≥ 1h physical activity per week) and inactive
(< 1h per week). Regularity of meals was assessed using
a four-item scale (never, occasionally, often and always).
Data on the efficacy of food chewing were collected with
a question about how thoroughly subjects chew their food;
the answers could be not very well, well or very well. The
speed of eating was measured with the following questions:
How much time do you spend eating lunch (never eat lunch:
< 10, 10–20, > 20min)? How much time do you spend eat-
ing dinner (never eat dinner, < 10, 10–20, > 20min)? Par-
ticipants were also questioned about the number of breakfast
consumptions with a two-item scale (< 5, ≥ 5 times/week).
Data about intra-meal fluid intake were obtained from the
question, how often do you drink beverages before, within or
after meals? The answers were never, sometimes, often and
always. Participants were asked about the amount of fried
food and spicy food consumed per week. Chocolate, coffee,
tea and soft drink consumption was determined through a
106-item Food Frequency Questionnaire (FFQ).
Statistical Analysis
Comparison of continuous variables across different catego-
ries of meal or snack frequency was assessed using one-way
analysis of variance. The chi-square test was used to exam-
ine the distribution of participants in terms of categorical
variables across different categories of meal or snack fre-
quency. The relationship between meal or snack frequency
and GERD was examined using logistic regression in differ-
ent models. First, the relationship was assessed in a crude
model. Then, age (continuous) and sex (male, female) were
adjusted in the first model. In the second model, further
adjustments for physical activity (≥ 1, < 1h/week), smok-
ing (current smokers, ex-smokers, non-smokers) and self-
reported diabetes (yes, no) were done. Additional adjust-
ments for meal regularity (non-regular, regular), eating rate
(non-quick, quick or < 10min), breakfast consumption (≥ 5,
< 5 times/week), intra-meal fluid intake (never or some-
times, often or always), spicy food intake (never, 1–3, 4–6,
≥ 7 times/week), fried food intake (ordinal), frequency of
fluid intake (ordinal), chewing efficiency (not well, well) and
consumption of chocolate, tea, soft drinks and coffee were
done in the third model. In the last model, BMI was also
adjusted. In all these analyses, participants in the first cat-
egory of meal or snack frequency were considered the refer-
ence category. Stratified analysis by gender (male vs. female)
and BMI status (BMI < 25 vs. BMI ≥ 25kg/m2) was done.
The Statistical Package for Social Sciences (SPSS Inc., ver-
sion 18.0, Chicago, IL, USA) was used for all analyses, and
P < 0.05 was considered statistically significant.
Results
The mean age and weight of 4669 adult participants were
36.53 (years) and 68.89 (kg), respectively. General char-
acteristics of the study population are provided in Table1.
Compared with participants who ate one main meal per day,
those who consumed three main meals had lower mean age,
weight and body mass index (BMI), and most were men.
Most of these subjects who ate three main meals were edu-
cated, and the prevalence of smoking among them was lower
than among individuals eating one meal per day. Participants
who consumed more than five snacks per day were mostly
women, had lower weight and BMI, and were younger than
persons who did not eat snacks at all. Age, weight, BMI, sex,
marital status and education level had different distributions
in participants with different frequencies of main meals and
snacks per day.
The distribution of diet-related behaviors of study par-
ticipants across categories of meal frequency is provided in
Table2. Those who ate three main meals, compared with
participants who consumed just one meal/day, ate more
slowly, chewed their meals well and had more regularity in
meals and breakfast consumption and less intake of spicy
food. All diet-related behaviors were significantly different
between participants who consumed more than five snacks
per day and those with no snacks. For total meal frequency,
meal regularity, chewing food well and breakfast consump-
tion occurred more and eating speed and spicy and fried
food intake were less in participants with eight meals (or
more) than in those with less than three.

3001Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
The prevalence of GERD in the study population was
23.7%. The prevalence of GERD in different categories of
main meals, snacks and total meals is shown in Fig.1. Individ-
uals with eight (or more) meals and snacks per day, compared
with those who consumed less than three meals and snacks per
day, had a lower prevalence of GERD (21.8 vs. 26.8%); how-
ever, the difference was not statistically significant (P = 0.28).
Multivariable adjusted odds ratios (ORs) for GERD in dif-
ferent categories of meal frequency are provided in Table3.
After adjustment of all potential confounders, participants
in the highest category of main meals had a non-significant
reduced risk of GERD (OR 0.91; 95% CI 0.58–1.41) compared
with those in the lowest category. Also, we found no signifi-
cant risk reduction for GERD in participants who consumed
more snacks and total meals and snacks compared with those
in the reference category in both the crude and multivariable
adjusted models.
Multivariate adjusted ORs for GERD across categories of
meal frequency stratified by gender are provided in Table4.
Although after adjustment of all confounding variables there
was no significant relationship between meal frequency and
GERD in men, consuming more main meals and snacks was
associated with reduced risk of GERD in women. Those
women who ate 1–2 and 3–5 snacks per day, compared with
women who did not eat any snacks, showed 29% (OR 0.71;
95% CI 0.54–0.93) and 37% (OR 0.63; 95% CI 0.45–0.88)
reduced odds of GERD in the crude model and 41% (OR 0.59;
95% CI 0.42–0.84) and 51% (OR 0.49; 95% CI 0.32–0.75) in
the adjusted model, respectively. For total meal and snack fre-
quency, after adjustment for all confounding variables, women
who consumed eight meals (or more) had a 43% risk reduction
of having GERD compared with those who had less than three
meals (OR 0.57; 95% CI 0.34–0.95). Also, female participants
with 6–7 meals and snacks/day had a 38% reduced odds of
GERD compared with those women who consumed less than
three meals/day (OR 0.62; 95% CI 0.41–0.96).
Stratified analysis based on BMI status was done, and
multivariate adjusted ORs for GERD across different cat-
egories of BMI are provided in Table5. After adjustment
of all confounding variables, normal-weight participants
(BMI < 25kg/m2) with 3–5 snacks/day had decreased odds
(OR 0.55; 95% CI 0.34–0.88) for GERD compared with
those who did not consume snacks. There was no other sig-
nificant relation in normal-weight individuals. After adjust-
ment of all confounders, we did not find a significant associ-
ation in overweight or obese participants (BMI ≥ 25kg/m2).
Discussion
We found no significant association between frequency of
meals, snacks and total meals and snacks with GERD symp-
toms among Iranian adults. However, stratified analysis by
Table 1 General characteristics of study participants across categories of meal frequencya (n = 4669)
a All values are mean ± SD unless indicated
b Physically active: ≥ 1h/week
Frequency of main meals (times/day) Frequency of snacks (times/day) Total frequency of meals (times/day)
123P0 1–2 3–5 > 5 P< 3 3–5 6–7 ≥ 8 P
Age (years) 38.9 ± 7.92 36.6 ± 7.94 36.2 ± 8.13 < 0.001 38.7 ± 8.44 36.4 ± 7.94 34.4 ± 7.80 35.6 ± 8.41 < 0.001 38.9 ± 8.08 37.3 ± 8.12 36.1 ± 7.96 34.1 ± 7.96 < 0.001
Weight (kg) 69.2 ± 12.95 68.4 ± 13.12 69.0 ± 13.62 0.44 71.4 ± 13.31 68.5 ± 13.65 67.5 ± 12.28 69.3 ± 12.89 < 0.001 69.4 ± 13.34 69.6 ± 13.02 68.6 ± 13.99 67.3 ± 11.76 0.01
BMI (kg/m2)25.4 ± 4.10 25.2 ± 4.80 24.9 ± 4.65 0.20 25.7 ± 5.51 24.9 ± 4.53 24.7 ± 4.18 25.1 ± 3.48 < 0.001 25.4 ± 4.75 25.3 ± 5.08 24.9 ± 4.56 24.5 ± 3.74 0.003
Female (%) 60.7 63.4 52.6 < 0.001 36.2 57.8 69.7 65.5 < 0.001 52.9 52 56.4 66.2 < 0.001
Married (%) 80.4 79.5 81.9 0.06 84.5 81.8 74.2 77.2 < 0.001 80.7 82.8 81.7 74.8 < 0.001
Education
(% ≥ diploma)
81.8 86.5 86.5 0.01 75.7 86.9 95.4 94.5 < 0.001 79 82.1 88 94.9 < 0.001
Current smoker
(%)
25.2 15.2 13.7 < 0.001 20.2 13.8 12.4 15.5 < 0.001 24.2 15.6 13 11.9 < 0.001
Self-reported
diabetes (%)
2.9 1.9 1.7 0.37 2.5 1.7 1.3 3.4 0.22 2.8 1.9 1.7 1.3 0.34
Physically
activeb (%)
29.2 35.7 34.9 0.18 36 34.1 36.6 38.2 0.52 32.7 35.3 34.6 36.2 0.73

3002 Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
Table 2 Distribution of participants in terms of diet-related behaviors across categories of meal frequencya (n = 4669)
a All values are percentages
b Individuals who reported regular meal consumption often or always
c Individuals who reported chewing foods moderately or very well
d Individuals who spent < 10min for meal consumption
e Individuals who reported ≥ 3 glasses of beverages with meals
Frequency of meals (times/
day)
Frequency of snacks (times/day) Total frequency of meals (times/
day)
123P0 1–2 3–5 > 5 P< 3 3–5 6–7 ≥ 8 P
Regular mealsb32.6 46.5 64.9 < 0.001 48.1 59.7 65.7 58.6 < 0.001 31.0 51.9 64.9 68.4 < 0.001
Well-chewing of foodsc82.6 84.2 86.2 < 0.001 83.0 86.4 84.3 79.0 0.001 79.0 86.6 86.1 84.7 < 0.001
Rapid lunch intaked19.8 15.4 14.3 < 0.001 19.0 13.4 15.7 27.6 < 0.001 20.1 15.3 13.4 16.1 < 0.001
Rapid dinner intaked21.9 23.2 22.3 < 0.001 19.9 22.5 24.9 34.5 < 0.001 19.6 27.2 23.0 23.7 < 0.001
Breakfast consumption (times/week) < 0.001 < 0.001 < 0.001
< 5 times/week 45.5 46.5 12.5 30.7 21.0 21.2 24.6 50.8 35.1 13.7 12.1
≥ 5 times/week 54.5 53.5 87.5 69.3 79.0 78.8 75.4 49.2 64.9 86.3 87.9
Fluid intake along with meale5.4 3.9 4.4 < 0.001 4.9 4.1 4.2 10.4 < 0.001 5.8 1.5 1.3 1.7 < 0.001
Fried food intake 0.12 0.001 < 0.001
< 4 times/week 85.4 84.6 82.3 84.5 83.8 77.8 75.9 87.7 83.5 83.3 76.5
≥ 4 times/week 14.6 15.4 17.7 15.5 16.2 22.2 24.1 12.3 16.5 16.7 23.5
Spicy food intake (times/week) 0.001 < 0.001 < 0.001
Never 7.9 6.6 6.0 11.2 5.4 4.7 1.7 9.5 7.9 5.1 4.4
1–3 42.1 40.6 38.1 44.8 39.4 30.7 25.9 43.2 43.4 37.8 29.4
4–6 33.1 33.8 31.2 28.8 31.9 36.6 32.8 31.2 31.6 31.8 34.8
≥ 7 16.9 19.0 24.6 15.3 23.3 28.1 39.7 16.2 17.2 25.3 31.4
28.1
25.2
23.7
21
22
23
24
25
26
27
28
29
123
GERD Prevalence (%)
Meal frequency (times/day)
P=0.06
23.1 24.2 21.7
29.3
0
5
10
15
20
25
30
35
0 1_2 3_5 >5
GERD Prevalence (%)
Snack frequency (times/day)
P=0.39
26.8 24.3 23.2 21.8
0
5
10
15
20
25
30
<3 3_5 6_7 ≥8
GERD Prevalence (%
)
Total meal and snacks (times/day)
P=0.28
Fig. 1 The prevalence of GERD in different categories of main meals, snacks, and total meals and snacks

3003Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
Table 3 Multivariable adjusted odds ratios for GERD across categories of meal frequencya (n = 4669)
a GERD was defined as having heartburn (sometimes, often or always) in the preceding 3months
b Model 1: adjusted for age and gender
c Model 2: age, gender, physical activity, smoking and self-reported diabetes
d Model 3: further adjusted for meal regularity (non-regular, regular), eating rate (non-quick, quick or < 10 min), breakfast consumption, intra-meal fluid intake (never or sometimes, often or
always), spicy food intake (never, 1–3, 4–6, ≥ 7 times/week), fried food intake (ordinal), frequency of fluid intake (ordinal), chewing efficiency (not well, well), consumption of chocolate, tea,
soft drinks and coffee
e Model 4: further adjusted for BMI
Frequency of main meals (times/day) Frequency of snacks (times/day) Total number of meals and snacks (times/day)
123Ptrend 0 1–2 3–5 > 5 Ptrend < 3 3–5 6–7 ≥ 8 Ptrend
Crude 1.00 0.86 (0.63–
1.17)
0.75 (0.56–
1.01)
0.02 1.00 1.06 (0.88–
1.27)
0.92 (0.71–
1.18)
1.38 (0.76–
2.49)
0.95 1.00 0.87 (0.68–
1.12)
0.82 (0.65–
1.04)
0.76 (0.56–
1.03)
0.06
Model 1b1.00 0.93 (0.66–
1.30)
0.83 (0.60–
1.14)
0.10 1.00 0.98 (0.80–
1.20)
0.86 (0.65–
1.13)
1.29 (0.68–
2.42)
0.56 1.00 0.88 (0.67–
1.16)
0.82 (0.63–
1.05)
0.75 (0.54–
1.05)
0.06
Model 2c1.00 0.97 (0.68–
1.39)
0.84 (0.60–
1.18)
0.08 1.00 0.95 (0.77–
1.17)
0.83 (0.62–
1.11)
1.22 (0.63–
2.35)
0.43 1.00 0.89 (0.67–
1.18)
0.80 (0.61–
1.04)
0.74 (0.52–
1.04)
0.03
Model 3d1.00 0.98 (0.64–
1.49)
0.92 (0.60–
1.41)
0.53 1.00 0.95 (0.75–
1.19)
0.79 (0.58–
1.08)
1.05 (0.53–
2.10)
0.25 1.00 0.93 (0.68–
1.29)
0.84 (0.61–
1.17)
0.76 (0.50–
1.13)
0.11
Model 4e1.00 0.95 (0.62–
1.46)
0.91 (0.58–
1.41)
0.60 1.00 0.92 (0.73–
1.17)
0.78 (0.57–
1.07)
1.08 (0.54–
2.15)
0.24 1.00 0.94 (0.67–
1.31)
0.85 (0.61–
1.18)
0.76 (0.51–
1.15)
0.12

3004 Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
Table 4 Multivariable adjusted odds ratios for GERD across categories of meal frequency separated by gendera (n = 4669)
a GERD was defined as having heartburn (sometimes, often or always) in the preceding 3months
b Model 1: adjusted for age
c Model 2: age, physical activity, smoking, self-reported diabetes
d Model 3: further adjusted for meal regularity (non-regular, regular), eating rate (non-quick, quick or < 10 min), breakfast consumption, intra-meal fluid intake (never or sometimes, often or
always), spicy food intake (never, 1–3, 4–6, ≥ 7 times/week), fried food intake (ordinal), frequency of fluid intake (ordinal), chewing efficiency (not well, well), consumption of chocolate, tea,
soft drinks and coffee
e Model 4: further adjusted for BMI
Frequency of main meals (time/day) Frequency of snacks (times/day) Total number of meals and snack (times/day)
123Ptrend 0 1–2 3–5 > 5 Ptrend < 3 3–5 6–7 ≥ 8 Ptrend
Males
Crude 1.00 1.18 (0.69–
2.02)
1.05 (0.63–
1.75)
0.64 1.00 1.38 (1.06–
1.78)
1.14 (0.75–
1.73)
1.46 (0.51–
4.12)
0.14 1.00 1.19 (0.80–
1.77)
1.29 (0.88–
1.88)
1.09 (0.65–
1.83)
0.40
Model 1b1.00 1.48 (0.77–
2.84)
1.33 (0.72–
2.48)
0.90 1.00 1.35 (1.07–
1.80)
1.15 (0.72–
1.83)
1.23 (0.39–
3.84)
0.22 1.00 1.30 (0.82–
2.08)
1.42 (0.91–
2.22)
1.22 (0.67–
2.22)
0.30
Model 2c1.00 1.43 (0.72–
2.83)
1.28 (0.66–
2.45)
0.97 1.00 1.28 (0.95–
1.73)
1.05 (0.64–
1.71)
1.21 (0.38–
3.45)
0.41 1.00 1.29 (0.79–
2.10)
1.35 (0.85–
2.17)
1.14 (0.61–
2.14)
0.50
Model 3d1.00 1.76 (0.72–
4.26)
1.40 (0.59–
3.33)
0.70 1.00 1.34 (0.96–
1.86)
1.14 (0.67–
1.94)
0.87 (0.22–
3.34)
0.42 1.00 1.49 (0.83–
2.68)
1.47 (0.82–
2.63)
1.23 (0.58–
2.59)
0.73
Model 4e1.00 1.57 (0.64–
3.83)
1.33 (0.55–
3.19)
0.83 1.00 1.30 (0.93–
1.82)
1.25 (0.73–
2.14)
0.87 (0.22–
3.34)
0.34 1.00 1.56 (0.86–
2.85)
1.53 (0.84–
2.78)
1.44 (0.67–
3.08)
0.53
Females
Crude 1.00 0.72 (0.49–
1.05)
0.63 (0.44–
0.91)
0.02 1.00 0.71 (0.54–
0.93)
0.63 (0.45–
0.88)
1.02 (0.49–
2.12)
0.06 1.00 0.70 (0.51–
0.97)
0.59 (0.43–
0.79)
0.57 (0.39–
0.83)
0.001
Model 1 1.00 0.77 (0.52–
1.15)
0.67 (0.46–
0.98)
0.03 1.00 0.68 (0.51–
0.90)
0.61 (0.43–
0.87)
1.10 (0.51–
2.40)
0.07 1.00 0.71 (0.51–
1.00)
0.59 (0.43–
0.81)
0.57 (0.38–
0.85)
0.001
Model 2 1.00 0.82 (0.54–
1.25)
0.69 (0.46–
1.03)
0.03 1.00 0.68 (0.50–
0.92)
0.62 (0.42–
0.89)
1.02 (0.45–
2.30)
0.08 1.00 0.73 (0.51–
1.03)
0.59 (0.42–
0.82)
0.57 (0.37–
0.86)
0.001
Model 3 1.00 0.82 (0.50–
1.36)
0.85 (0.50–
1.44)
0.81 1.00 0.62 (0.44–
0.87)
0.53 (0.35–
0.80)
1.00 (0.43–
2.33)
0.04 1.00 0.75 (0.50–
1.13)
0.64 (0.42–
0.97)
0.61 (0.37–
1.00)
0.04
Model 4 1.00 0.82 (0.49–
1.37)
0.84 (0.49–
1.43)
0.74 1.00 0.59 (0.42–
0.84)
0.49 (0.32–
0.75)
1.04 (0.44–
2.45)
0.03 1.00 0.75 (0.49–
1.13)
0.62 (0.41–
0.96)
0.57 (0.34–
0.95)
0.02

3005Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
Table 5 Multivariable adjusted odds ratios for GERD across categories of meal frequency separated by BMI statusa (n = 4669)
a GERD was defined as having heartburn (sometimes, often or always) in the preceding 3months
b Model 1: adjusted for age and gender
c Model 2: age, gender, physical activity, smoking, self-reported diabetes
d Model 3: further adjusted for meal regularity (non-regular, regular), eating rate (non-quick, quick or < 10 min), breakfast consumption, intra-meal fluid intake (never or sometimes, often or
always), spicy food intake (never, 1–3, 4–6, ≥ 7 times/week), fried food intake (ordinal), frequency of fluid intake (ordinal), chewing efficiency (not well, well),and consumption of chocolate,
tea, soft drinks and coffee
Frequency of main meals (times/day) Frequency of snacks (times/day) Total number of meals and snack (times/day)
123Ptrend 0 1–2 3–5 > 5 Ptrend < 3 3–5 6–7 ≥ 8 Ptrend
BMI < 25 (kg/mb)
Crude 1.00 0.75 (0.47–
1.18)
0.65 (0.42–
1.01)
0.04 1.00 1.06 (0.80–
1.40)
0.90 (0.62–
1.31)
1.80 (0.78–
4.15)
0.95 1.00 1.04 (0.71–
1.53)
0.83 (0.58–
1.20)
0.88 (0.56–
1.37)
0.14
Model 1b1.00 0.81 (0.49–
1.34)
0.71 (0.44–
1.14)
0.10 1.00 0.91 (0.67–
1.25)
0.74 (0.49–
1.12)
1.60 (0.65–
3.93)
0.42 1.00 1.02 (0.67–
1.56)
0.79 (0.53–
1.18)
0.78 (0.48–
1.28)
0.05
Model 2c1.00 0.75 (0.44–
1.26)
0.63 (0.38–
1.04)
0.04 1.00 0.85 (0.61–
1.18)
0.68 (0.44–
1.05)
1.57 (0.63–
3.93)
0.29 1.00 0.95 (0.61–
1.47)
0.71 (0.47–
1.07)
0.70 (0.42–
1.17)
0.02
Model 3d1.00 0.74 (0.40–
1.39)
0.67 (0.35–
1.27)
0.24 1.00 0.76 (0.53–
1.09)
0.55 (0.34–
0.88)
1.43 (0.55–
3.72)
0.09 1.00 0.92 (0.56–
1.52)
0.67 (0.40–
1.12)
0.63 (0.34–
1.15)
0.03
BMI ≥ 25 (kg/mb)
Crude 1.00 0.98 (0.63–
1.52)
0.90 (0.60–
1.37)
0.44 1.00 1.04 (0.80–
1.34)
0.98 (0.68–
1.40)
1.07 (0.46–
2.49)
0.98 1.00 0.77 (0.55–
1.09)
0.86 (0.62–
1.20)
0.72 (0.46–
1.12)
0.48
Model 1 1.00 0.99 (0.62–
1.58)
0.95 (0.61–
1.47)
0.69 1.00 1.00 (0.76–
1.32)
0.97 (0.66–
1.44)
1.00 (0.40–
2.48)
0.94 1.00 0.79 (0.55–
1.15)
0.88 (0.62–
1.25)
0.76 (0.48–
1.22)
0.60
Model 2 1.00 1.15 (0.70–
1.90)
1.07 (0.66–
1.72)
0.84 1.00 1.00 (0.75–
1.34)
0.98 (0.66–
1.47)
0.88 (0.33–
2.33)
0.88 1.00 0.87 (0.59–
1.28)
0.94 (0.65–
1.36)
0.78 (0.48–
1.28)
0.66
Model 3 1.00 1.13 (0.62–
2.07)
1.22 (0.66–
2.24)
0.48 1.00 1.04 (0.76–
1.44)
1.02 (0.66–
1.59)
0.73 (0.25–
2.10)
0.88 1.00 0.91 (0.58–
1.43)
1.05 (0.66–
1.65)
0.87 (0.49–
1.55)
0.92

3006 Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
gender revealed that women with more meal frequency had a
lower risk of having GERD. Since this finding was obtained
in a post hoc secondary analysis, it should be interpreted
with caution. To the best of our knowledge, this was the first
study in a Middle-Eastern population to try to determine
the relation between meal frequency and GERD symptoms.
GERD is a globally increasing condition associated with
a poor quality of life, large economic burden [6] and morbid-
ity among individuals suffering from it. Improving dietary
habits might help prevent this condition. Our findings sug-
gest that increasing meal frequency might have a reverse
association with GERD symptoms, especially in women.
Also, increasing the frequency of meals and snacks might
be helpful in preventing the disease.
The current study showed that females consuming at least
six meals and snacks had fewer GERD symptoms. Although
having more than eight snacks and meals per day might be
unusual in many countries, such eating habits are common
in Iran. One important factor responsible for this great fre-
quency of meals and snacks among Iranian adults could be
the frequent tea consumption. Drinking tea, specifically
black tea, has a long history in the Iranian culture and has
become one of the most common dietary behaviors among
Iranian individuals [21]. Rezaee etal. [22] reported that
mean tea consumption among Iranian individuals was ten
times more than the global average. Although in the present
study we made adjustment for tea consumption in the analy-
sis, having such a high frequency of meals and snacks in the
Iranian population should be kept in mind when generalizing
the present findings to populations in other geographic areas.
Our study suggested that increasing the frequency of
main meals and snacks would decrease the likelihood of
having GERD symptoms in women, but not in men. This
result might be explained by different dietary habits exist-
ing between the genders. Several studies demonstrated that
women are more concerned about their diet and lifestyle.
For instance, in terms of choosing food for a healthy diet,
females pay more attention to selecting specific kinds of
nutrients that are suitable for their health status [23, 24]. In
contrast, despite awareness of healthy eating guidelines, men
tend to show more interest in other aspects such as food taste
[25]. Dietary and nutritional knowledge is another factor that
has been confirmed to be higher among women. Due to the
traditional role of females engaging in purchasing and cook-
ing foods, they tend to be more informed about all aspects of
nutritional field [23, 26]. Another possible factor influencing
dietary habits could be weight management. Weight control
and body perception are growing concerns these days, espe-
cially among women. To reach their desired body weight,
women will usually change their dietary patterns, while men
will usually increase their physical activities [23, 27]. Sev-
eral previous studies have suggested the effect of meal and
snack frequency as contributing to controlling body weight.
For instance, Azadbakht etal. [28] in a study on 265 Iranian
female adolescents showed that individuals consuming at
least four snacks per day had a significantly lower BMI and
waist circumference. Similarly, another study showed that
urban and rural Iranian females who did not eat daily snacks
had higher BMI levels and waist-to-hip ratios [29].
We found no significant association between meal or
snack frequency and GERD symptoms in the whole study
population. All relevant studies in the literature did not find
a significant relation between meal frequency and GERD
symptoms. Jarosz etal. [30] suggested that GERD patients
ate fewer meals than normal individuals; however, in mul-
tivariable logistic regression analysis, they did not find a
significant relationship. GERD patients might refuse to
consume multiple meals because of probable aggravation
of the symptoms. Increasing the meal volume might explain
their findings. Unlike our investigation, Jarosz etal. [30] did
not consider other dietary habits of patients such as eating
snacks, meal regularity and eating rate. In a case-control
study by Song etal. in Korea, 81 recently defined GERD
patients and 81 controls completed a questionnaire about
their symptoms and dietary habits. They found no statisti-
cally significant relationship between snack consumption
and GERD symptoms [16]. The sample in this study con-
tained individuals referred to a health care institution for
either reflux symptoms or a routine checkup. Furthermore,
the applied definition of GERD and small sample size might
be involved in their results. A cross-sectional survey of 2000
individuals with heartburn suggested that women might be
more aware of their daily number of meals as an effective
factor contributing to heartburn compared with men, but this
relation was not statistically significant [17].
The mechanism explaining the relationship between
increased meal frequency and reducing GERD symptoms
is yet unknown. One possible mechanism could be related
to meal volume. In individuals who eat fewer meals, each
meal will probably be large, which can lead to stomach wall
stretching and lower esophageal sphincter dysfunction [11],
which can result in GERD. By increasing meal frequency,
the volume of each meal will become smaller; therefore,
symptoms will be improved. Further prospective studies are
needed to find the exact mechanism.
This study was one of the first in the literature investigat-
ing the association of meal and snack frequency and GERD
symptoms. The large sample size was another strength of
the study. However, several limitations should be consid-
ered when interpreting our findings. The most important was
the cross-sectional design of the study, which did not allow
inferring the causality. GERD was defined with one question
about having heartburn in the 3months prior to the study.
Recall bias must be kept in mind in studies where remem-
bering events is necessary. The study population was chosen
from different job categories of university employers. We

3007Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
excluded some university teaching hospitals and research
centers to reduce the conflict of interest in the research.
Although the socioeconomic status of the study popula-
tion was representative of the general Iranian population,
extrapolating the findings to other populations must be done
cautiously. Furthermore, we tried to consider all confound-
ing variables; however, some of them such as psychologic
status might be able to affect the results. These initial find-
ings need to be confirmed with prospective assessments to
clarify the true causal association between meal and snack
frequency and GERD symptoms.
In conclusion, we found no significant association
between meal frequency and GERD symptoms in the Ira-
nian population. Gender-specific analysis revealed inverse
associations between meal and snack frequency and GERD
in Iranian women. Further prospective studies are required
to confirm these associations.
Acknowledgments This study was extracted from an MD disserta-
tion that was approved by Isfahan University of Medical Sciences (no.
292017). The financial support for this study comes from the Integra-
tive Functional Gastroenterology Research Center, Isfahan University
of Medical Sciences, Isfahan, Iran. The Integrative Functional Gas-
troenterology Research Center had no role in the conception, design,
data analysis and manuscript drafting. We thank all the staff of Isfahan
University of Medical Sciences who kindly participated in our study
and the staff of the Public Relations Unit and other authorities of IUMS
for their excellent cooperation.
Compliance with ethical standards
Conflict of interest None of the authors had any personal or financial
conflicts of interest.
References
1. Fazel M, Keshteli AH, Jahangiri P, Daneshpajouhnejad P, Adibi
P. Gastroesophageal reflux disease in Iran: SEPAHAN system-
atic review No. 2. Int J Prev Med. 2012;3:S10–S17.
2. Festi D, Scaioli E, Baldi F, etal. Body weight, lifestyle, dietary
habits and gastroesophageal reflux disease. World J Gastroen-
terol. 2009;15:1690–1701.
3. Badillo R, Francis D. Diagnosis and treatment of gastroesoph-
ageal reflux disease. World J Gastrointest Pharmacol Ther.
2014;5:105–112.
4. Matsuki N, Fujita T, Watanabe N, etal. Lifestyle factors associ-
ated with gastroesophageal reflux disease in the Japanese popu-
lation. J Gastroenterol. 2013;48:340–349.
5. Delavari A, Moradi G, Birjandi F, Elahi E, Saberifiroozi M.
The prevalence of gastroesophageal reflux disease (GERD) in
the Islamic Republic of Iran: a systematic review. Middle East
J Dig Dis. 2012;4:5–15.
6. Rezailashkajani M, Roshandel D, Shafaee S, Zali MR. A cost
analysis of gastro-oesophageal reflux disease and dyspepsia in
Iran. Dig Liver Dis. 2008;40:412–417.
7. Meining A, Classen M. The role of diet and lifestyle measures
in the pathogenesis and treatment of gastroesophageal reflux
disease. Am J Gastroenterol. 2000;95:2692–2697.
8. Locke GR III, Talley NJ, Fett SL, Zinsmeister AR, Melton LJ
III. Risk factors associated with symptoms of gastroesophageal
reflux. Am J Med. 1999;106:642–649.
9. Lagergren J. Influence of obesity on the risk of esophageal dis-
orders. Nat Rev Gastroenterol Hepatol. 2011;8:340–347.
10. Kaltenbach T, Crockett S, Gerson LB. Are lifestyle measures
effective in patients with gastroesophageal reflux disease? An
evidence-based approach. Arch Intern Med. 2006;166:965–971.
11. Dibley LB, Norton C, Jones R. Don’t eat tomatoes: patient’s
self-reported experiences of causes of symptoms in gastro-
oesophageal reflux disease. Fam Pract. 2010;27:410–417.
12. Bradley LA, Richter JE, Pulliam TJ, etal. The relationship
between stress and symptoms of gastroesophageal reflux:
the influence of psychological factors. Am J Gastroenterol.
1993;88:11–19.
13. Wu KL, Rayner CK, Chuah SK, etal. Effect of liquid meals
with different volumes on gastroesophageal reflux disease. J
Gastroenterol Hepatol. 2014;29:469–473.
14. Esmaillzadeh A, Keshteli A, Feizi A, Zaribaf F, Feinle-Bisset
C, Adibi P. Patterns of diet-related practices and prevalence
of gastro-esophageal reflux disease. Neurogastroenterol Motil.
2013;25:831-e638.
15. Yamamichi N, Mochizuki S, Asada-Hirayama I, etal. Lifestyle
factors affecting gastroesophageal reflux disease symptoms:
a cross-sectional study of healthy 19864 adults using FSSG
scores. BMC Med. 2012;10:45.
16. Song JH, Chung SJ, Lee JH, etal. Relationship between gas-
troesophageal reflux symptoms and dietary factors in Korea. J
Neurogastroenterol Motil. 2011;17:54–60.
17. Oliveria SA, Christos PJ, Talley NJ, Dannenberg AJ. Heartburn
risk factors, knowledge, and prevention strategies: a population-
based survey of individuals with heartburn. Arch Intern Med.
1999;159:1592–1598.
18. Adibi P, Keshteli AH, Esmaillzadeh A, Afshar H, Roohafza H,
Bagherian-Sararoudi R, etal. The Study on the Epidemiology of
Psychological, Alimentary Health and Nutrition (SEPAHAN):
overview of methodology. J Res Med Sci. 2012;17:S291–S297.
19. Sorouri M, Pourhoseingholi MA, Vahedi M, etal. Functional
bowel disorders in Iranian population using Rome III criteria.
Saudi Journal of Gastroenterology. 2010;16:154.
20. National Institute for Health and Clinical Excellence. Promoting
and creating built or natural environments that encourage and
support physical activity: scope. NICE: London; 2006.
21. Fazeltabar Malekshah A, KayediMajd S, Pourshams A, etal.
Assessment of food consumption pattern in high risk area of
esophageal cancer in northern Iran. Govaresh. 2008;13:13–17.
22. Rezaee E, Mirlohi M, Hassanzadeh A, Fallah A. Factors affect-
ing tea consumption pattern in an urban society in Isfahan, Iran.
J Educ Health Promot. 2016;5:13.
23. Arganini C, Saba A, Comitato R, Virgili F, Turrini A. Gender
differences in food choice and dietary intake in modern western
societies. InTech: Public Health Soc Behav Health; 2012.
24. Verbeke W, Vackier I. Profile and effects of consumer involve-
ment in fresh meat. Meat Sci. 2004;67:159–168.
25. Sobal J. Men, meat, and marriage: models of masculinity. Food
Foodways. 2005;13:135–158.
26. Parmenter K, Waller J, Wardle J. Demographic varia-
tion in nutrition knowledge in England. Health Educ Res.
2000;15:163–174.
27. Johnson F, Wardle J. Dietary restraint, body dissatisfaction, and
psychological distress: a prospective analysis. J Abnorm Psy-
chol. 2005;114:119.
28. Azadbakht L, Hajishafiee M, Golshahi J, Esmaillzadeh A.
Snacking behavior and obesity among female adolescents in
Isfahan, Iran. J Am Coll Nutr. 2016;35:405–412.

3008 Digestive Diseases and Sciences (2018) 63:2998–3008
1 3
29. Sotoudeh G, Khosravi S, Khajehnasiri F, Khalkhali HR. High
prevalence of overweight and obesity in women of Islamshahr,
Iran. Asia Pac J Clin Nutr. 2005;14:169.
30. Jarosz M, Taraszewska A. Risk factors for gastroesophageal reflux
disease: the role of diet. Prz Gastroenterol. 2014;9:297–301.