Prevalence and risk factors for gastroesophageal reflux disease in the North-Eastern part of Bangladesh

Despite a common disorder population-based data on gastro-esophageal reflux disease (GERD) in Bangladesh is lacking. This epidemiological study was designed to determine the prevalence of GERD and its association with lifestyle factors. This population-based cross-sectional study was done by door to door interview of randomly selected persons in both urban and rural areas of North Eastern part of Bangladesh by using a validated questionnaire. A cutoff point 3 was chosen as a valid and reliable scale to confirm GERD. Statistical analysis was done by SPSS-12 version and the level of significance was set at P≤0.05. A total of 2000 persons with an age range of 15 to 85 years were interviewed; 1000 subjects from urban area and 1000 from rural area. Among the study subjects 1064 were male and 936 were female. A total of 110 persons (5.5%) were found to have GERD symptoms and among them 47 were men and 67 were women. The monthly, weekly and daily prevalence of heart-burn and or acid regurgitation was 5.5%, 5.25% and 2.5% respectively. Female sex, increased age and lower level of education were significantly associated with GERD symptoms. Prevalence was found more among city dwellers (approximately 6.0% versus 4.8%), married (6.23%, n=86), widowed/widowers (16.83%, n=17) and day labourer (8.78%). Level of education inversely influenced the prevalence. No significant association of GERD was found with body mass index (BMI) and smoking. Prevalence of GERD in North-Eastern part of Bangladesh was lower than that of western world. Prevalence was found higher in urban population, women, married, widowed/widowers and in poor and illiterate persons. BMI and smoking had no significant association with GERD.


Introduction
Gastro Oesophageal reflux disease (GERD) is a potentially serious condition with risks of complications like stricture, Barrett's oesophagus and malignancy 1,2 .This chronic, painful disease substantially interferes with physical activity, impairs social functioning, disturbs sleep and reduces productivity at work [2][3][4] .Abnormal reflux of gastric contents to oesophagus is responsible for symptoms of GERD 1 .Various environmental or lifestyle risk factors like obesity, fatty food, smoking, alcohol and NSAIDS are thought to be associated with GERD 5, 6 .
GERD has traditionally been considered less common in Asian countries in com-parison to western world 5-12 .There are few population-based data on GERD in Asia 10- 12 .Studies from some Asian countries reported a rising trend of prevalence of GERD 11- 13 .
In Bangladesh we have little data on the magnitude of the problem 14, 15 .Only population based survey reported a prevalence of 40.9% 14 .According to Masud's report the monthly, weekly and daily prevalence rates of heart-burn and or acid regurgitation were 24.6%, 17.2% and 12.5% respectively in a hospital outpatient population of Bangladesh 15 .
Patients with GERD can have symptoms without objective evidence of oesophagitis 16 .Considering the limitations of objective medical testing, GERD symptoms play a pivotal role in the diagnosis of the problem.Heart burn and or acid regurgitation are considered to be reasonably specific symptoms for the diagnosis 5 .Further diagnostic evaluation is necessary 1,5, 17 , if there is no response to therapy, when there are alarm symptoms suggestive of complications or for confirmation of diagnosis prior to anti-reflux surgery 17 .Studies also reported association of GERD with asthma, hoarseness of voice, dyspepsia, atypical chest pain and nonobstructive dysphagia Knowledge on prevalence estimates and associated risk factors of GERD in our population might play a part in defining local health needs as well as to improve the management of these patients.This cross sectional population-based survey was designed to find out the prevalence of GERD symptoms as well as to find out the association of GERD with various socio-demographic variables in both urban and rural population of eastern part of Bangladesh.

Materials and Method
Subjects & study design:

Discussion
Gastro-esophageal reflux disease (GERD) is one of the most frequent health problems in the western world 13 and as a chronic condition, GERD places a substantial burden on patients and the health care delivery system 4 .There is no gold standard diagnostic test for GERD.Most frequently done investigations like upper GI endoscopy and 24 hour oesophageal pH monitoring lack sensitivity and have several limitations 21-23 .Questionnaire-based diagnosis of GERD is reliable with high sensitivity and specificity 18 .
In the present series the prevalence of GERD based on questionnaire 18 (cut-off point 3) was 5.5% which is similar to population-based survey in Korea 12 , Singapore 20 , China 11 , Iran 10, 24 and India 25,26 but is lower than that in the western population 4,27-29 .These Asian studies showed that prevalence varied with age, sex, education, occupation, smoking, alcohol intake and psychosomatic ailments etc.The prevalence of GERD in this study was lower than that of only population-based study in Dhaka city (40.9%) 14 .Only published study 15 involved OPD population of a referral center of Dhaka city and reported a GERD prevalence of 22.8%.This significant difference may be due to selected population who came for consultation in Gastroenterology outpatient clinic.The present population-based study included both rural and urban population.
The monthly, weekly and daily prevalence of heartburn and or acid regurgitation in our study were 5.5%, 5.25% and 2.5% respectively.The corresponding prevalence's reported by Masud 15 were 24.6%, 17.2% & 12.5% respectively.Studies from USA showed that heart burn and or acid regurgitation occurred in approximately in 7% of adult population on a daily basis, 14-20% on a weekly basis and 44% on monthly basis 6, 9 .A Chinese population survey reported that the annual, monthly and weekly prevalence rates of GERD were 29.8%, 8.9% and 2.5% respectively in their population 11 .The weekly prevalence of GERD on the basis of heart-burn and or acid regurgitation in Korea (3.0%) 12 is comparable to our study.Geographic differences in GERD prevalence estimates are difficult to interpret, but may be related to differences in study definitions, questionnaire used, population selection and sociocultural factors.
In this series prevalence estimates of GERD was higher among women than men based on symptom score (6.73% versus 4.41%) and by weekly prevalence of heart-burn and or acid regurgitation (table-II).This prevalence is comparable to the finding in India 25 , but higher than that of Korea 12 and lower than that of China 11 and western countries like USA 6 , Belgium 27 and Sweden 28 .Differences in GERD prevalence estimates may be due to differences in dietary habit, environmental and socio-cultural factors and work pattern.GERD prevalence was equal in both men and women in USA 6 but in Belgium prevalence was more in women (31.1% and 25.6% respectively) 27 .High prevalence of GERD in housewives in the present study may be related to their sedentary life style.
GERD is variably associated with age.On multiple logistic regression analysis an increasing prevalence of GERD was found with increasing age in our study.The finding conforms to the finding of Masud et al 15  BMI was found as a risk factor for GERD in the studies from China 30 , Spain 29 and Dhaka 14 .But studies from korea 12 and Iran 24 found no association between GERD and BMI.In this series, no significant difference in prevalence of GERD was found with increase in BMI which is consistent with Masud 15 .This may be due to relatively small number of over-weight and obese persons (only 12.5%) in our study population.
Association between smoking and GERD is controversial.Several reports showed direct relationship of GERD with smoking 32,33 .However, studies from Spain 29 and India 26 did not find any relation between smoking and GERD.In our study association of GERD with smoking was unclear (5.95% VS 5.32%) which is consistent with the finding of Masud 15 .
By multiple logistic analyses GERD was inversely related to level of education .This increased prevalence of GERD among persons with low level of education in our study is consistent with the report from Spain 29 and Masud 15 .Lack of knowledge regarding healthy life style and less ability to modify life style factors precipitating GERD symptoms might play a role 29 .Present study showed that GERD was more prevalent among day labourer (8.78%), farmer (8.42%) and house wives (5.32%).These findings are not in agreement with the reports from Bangladesh 14,15 and Korea 12 .Level of education and health related awareness may play a role here.Higher prevalence of GERD among married, widows and widowers in our study differs from the finding of Masud et al 15 .Psycho-social factor might have an influence here.Symptom analysis showed that the 5 most common GERD related symptoms were heart-burn (100%), regurgitation (90.9%),Chest pain, chronic cough, and globus (table-III).Wong et al 11 reported acid regurgitation as the most common symptom in a Chinese population, but heart-burn was the most common symptom in our study population and regurgitation occupied the 2nd position.Heart-burn and or acid regurgitation on daily and weekly basis was more prevalent in women (table-II).Among the other symptoms hoarseness of voice and chronic cough were more prevalent in men (table-III).Cigarette smoking may play a role.No significant variation was found in symptom pattern in different age group except heart-burn and or acid regurgitation.These two symptoms found to increase with increasing age.
In conclusion, this questionnaire based community survey showed that GERD is not an uncommon condition in our community though the disease burden is less than that of western world.GERD is more common in women, older age group (>45 years), urban population, and in lower socioeconomic groups with lower level of education.BMI and smoking were not found to be significantly associated with GERD.More research are required involving different population in different parts of country and preferably with some investigations for exact estimation of the disease burden and planning appropriate health care.

Table 1 :
Socio-demographic characteristics and prevalence of GERD symptoms and past history of ischaemic cardiac pain.iv)globus-eeling of a lump in the throat when not swallowing; (v) hoarseness of voice-harshness or roughness of voice not related to respiratory tract infection; and (vi) chronic cough-cough occurring 4-times in a day on four or more days per week.No laboratory investigation or other procedure was done due to lack of feasibility.Statistics: Statistical analysis was performed with SPSS-12 version.Distribution of the factors like age, sex, education, marital status, occupation, BMI, smoking & individual symptom were compared by Pearson's Chi-squared test.Multiple logistic regression analysis with age, sex marital status & occupation then performed to measure the association of these factors with GERD.Odds ratios with 95% confidence intervals (CIs) were computed.All the statistical comparisons were twosided using the 0.05 significance level.7.220) based on questionnaire (table-I) and as well as by weekly prevalence of heart-burn and or acid regurgitation (table-II).hoarseness of voice, chronic cough and asthma) of GERD (table-3), chronic cough (91.5 % VS 71.4%, P 0.008) and hoarseness of voice (66.0%VS 46.0%, p 0.029) were significantly more prevalent in men (tableIII).No significant age group difference was found in the prevalence of these symptoms (table-IV) except heart-burn.

Table II :
Prevalence of heart-burn and or acid regurgitation frequency by gender and age.

Table III :
GERD symptom pattern among male & female

Table IV :
Symptom pattern of GERD according to age group 60t Shahed found higher prevalence of GERD among younger age group (25-34years, OR 1.71)14.In our survey highest prevalence was found in the 55-65 years group, but in China30highest prevalence was found in 30-39 years age group and in Japan in 40-49 years age group was more vulnerable.On the other hand studies from USA6, Korea 12 Singapore 20 , Belgium 27 and Spain 29 failed to show any association of GERD with age.