SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Aliment Pharmacol Ther 2010; 32: 645–654

Summary

Background  While there is evidence of ethnic variation in the prevalence of gastro-oesophageal reflux disease (GERD) symptoms, few population-based studies examine GERD symptom prevalence amongst the growing Hispanic minority in the US as well as Asians in the West.

Aim  To examine the prevalence, awareness and care patterns for GERD across different ethnic groups.

Methods  A population-based, cross-sectional survey was fielded in English, Chinese and Spanish that assessed self-reported GERD prevalence, awareness and care patterns in four ethnic groups (Caucasian, African American, Asian, Hispanic).

Results  A total of 1172 subjects were included for analysis: 34.6% experienced GERD symptoms at least monthly, 26.2% at least weekly and 8.2% at least daily. Statistically significant differences in raw prevalence rates between racial groups were found: 50% of Hispanics experienced heartburn at least monthly, compared with 37% of Caucasians, 31% of African Americans and 20% of Asians (> 0.0001). Significant differences in knowledge and care-seeking patterns by ethnicity were also observed.

Conclusions  This study confirms the high prevalence of GERD symptoms in the US and introduces Hispanics as the ethnicity with the highest prevalence rate. Asians in the US have higher rates of symptoms than in the Far East. These data demonstrate a need for culturally appropriate education about GERD symptoms and treatment.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Gastro-oesophageal reflux disease (GERD) is a chronic disorder, most commonly characterized by heartburn and acid regurgitation.1 There is evidence of ethnic variation in the prevalence of GERD symptoms in many parts of the world. Studies show higher prevalence rates in Western nations (20–40%), particularly amongst Caucasians, when compared with Spain, Italy and Asian countries.3–5 Other factors including genetics, demographics, nutritional status, cultural and behavioural influences have contributed to challenges to diagnosis, resulting in the under-reporting and over-reporting of GERD.3, 6

Studies confirm that medical care utilization rates for GERD vary in different countries from 16% to 56% and between ethnic groups.7, 8 Medical care was sought when the frequency and severity of symptoms increased impacting quality of life, when self-treatment was ineffective, when concerns about the condition increased, which could be resolved with care and prescription medication.9, 10 In addition, there are strong correlations between psychosocial factors like depression and anxiety and seeking care, which may serve as a catalyst for care rather than an aetiology.7 Cultural and ethnic factors, e.g. issues of trust, language, literacy and comprehension barriers influence health-seeking behaviours and may impede accurate and timely diagnosis, particularly where medical terminology is presented in language unfamiliar or unknown to patients.6, 11, 12 Beliefs about self-treating, lack of control of health, modesty, embarrassment, fear of illness and an attitude of fatalism influence decisions to seek care.8, 11

In the US, studies have identified comparable prevalence of GERD among African Americans and Caucasians; lower rates of symptoms were reported in Hispanic, Asian and North American populations.13, 14 An epidemiological survey of GERD characteristics was conducted among 450 customers (50 surveys/9 stores in different Philadelphia locations) for data reflecting representative prevalence, knowledge, complications and care-seeking behaviours in this urban centre.14 The findings showed similar frequencies of self-reported heartburn among racial groups, but acknowledged that the differences are ‘unknown’.14 This study found that 90.3% of medically diagnosed participants were familiar with GERD as opposed to 49.8% undiagnosed participants in an urban population, although there was minimal familiarity with other known symptoms and complications.14 Medical information was sought more frequently from the media and ‘word of mouth’ than from healthcare providers.14 Few prospective population-based studies exist that examine GERD symptom prevalence amongst the growing Hispanics and Asians living in the US. To understand better the prevalence, knowledge and health-seeking behaviours among ethnic populations in US, we conducted a small population-based observational study with African Americans, Asians, Hispanics and Caucasians in Philadelphia, PA.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This observational, cross-sectional, population-based study examined the prevalence, awareness and care patterns for GERD and related symptoms in a large urban setting, specifically assessing disparities across four different ethnic groups. We hypothesized that some of the ethnic minority groups would have higher prevalence of GERD, but have less knowledge about GERD and related symptoms. The rationale for conducting the study in the community setting rather than in clinical settings was to assess participants’ knowledge, awareness and behavioural intent prior to seeking medical care.

Study population

The study design designated inclusion of 200 adult (≥18 years) participants from each of the four ethnic groups: African American, Asian, Hispanic and Caucasian attending local community centres or faith-based community activities or health fairs in specific ethnic sections of Philadelphia. We targeted a minimum of 200 in each ethnic group to ensure that there were enough representatives from each group for analytical purposes, although a power analysis was not conducted. In addition, the recruitment strategy also considered that in many of the recruiting venues, the investigators would not be able to limit recruitment to just one ethnicity and designate specific numbers of surveys to collect.

Survey methods

This study was conducted in 2007 in Philadelphia, PA. To develop our community-based sample, letters (and follow-up calls) were sent to ethnic community and faith-based organizations in ethnically concentrated neighbourhoods to identify health events for our data collection.

A questionnaire, adapted from Srinivasan’s descriptive survey14 and Locke’s validated instrument,15 was developed that assessed self-reported familiarity with the term GERD, experiencing of heartburn and prevalence, triggers of heartburn and prevalence, diagnosis of GERD, treatment of heartburn, social impact, attitudes about seeking care, knowledge of heartburn/GERD, sources of information symptoms, care-seeking behaviour, knowledge of complications and demographic and socioeconomic status.8 Our questionnaire also included ordinal, nominal and Likert questions and was translated into and back-translated from Chinese and Spanish, with translators available to assist participants. The 36-question eight-page survey took respondents approximately 15–20 min to complete and a $5.00 gift card was given to respondents upon completion.

The domains of the questionnaire included: (1) familiarity with term GERD; (2) prevalence of heartburn; (3) diagnosis of GERD; (4) medication use; (5) social impact; (6) attitudes about seeking care; (7) knowledge about heartburn or GERD; (8) sources of information; (9) beliefs about symptomatic relief for heartburn; (10) beliefs about symptoms associated with heartburn; (11) intentions for treating heartburn; and (12) demographics (including gender, marital status, education, employment status, income, race/ethnicity, weight and height). Prevalence was based on terminology included in Srinivasan14 and Locke;15 and was worded as a ‘burning feeling in my lower chest rising up toward my neck (heartburn)’. Participants were also asked to identify for how long and at what times of the day they had experienced this feeling. Awareness of GERD was assessed by a question that asked if they ever heard of Gastro-oesophageal Reflux Disease, also known as GERD with the choice of responses as ‘Yes’ or ‘No’. This question was intended to differentiate an individual’s experience of symptoms (i.e. ‘a burning feeling’) from their knowledge of the term ‘GERD’. Other knowledge questions utilized the more generic term ‘heartburn’ and assessed belief(s) about items that would help control heartburn, symptoms possibly related to heartburn and actions they would take if they had severe heartburn. A complete copy of the survey is included in the Appendix S1.

Statistical methods and analysis

Survey data were entered into an Access database and checked for completeness, then analysed using sas 9.1 (SAS Institute Inc., Cary, NC, USA). Univariate and multivariate analyses were performed. The prevalence for GERD was calculated for each of the four ethnic groups (Caucasian, African American, Asian, Hispanic) and then compared, controlling for age, gender and other demographic variables. Body mass index (BMI) was calculated using self-reported height and weight and values were checked for statistical outliers. Knowledge, attitudes and care-seeking patterns for GERD symptoms were assessed for those reporting GERD symptoms and compared in four ethnic populations. The questionnaire featured two separate questions to assess the prevalence of heartburn after the subject had a meal and after the subject went to sleep.

A final measure for prevalence was created that accounted for all subjects who indicated experiencing heartburn either in general, after a meal, or after going to sleep. This measure was created because it became apparent that some participants did not indicate experiencing heartburn for the primary assessment question, but did indicate experiencing heartburn after a meal or after going to sleep. Statistical controls were used to ensure that no participant would be counted twice.

We chose to analyse monthly prevalence rates in the logistic regression models based on consultation with our physicians who determined that it was an appropriate timeframe to consider. The logistic model considers the impact of racial, socio-demographic and BMI variables upon monthly GERD symptoms. We also considered monthly prevalence rates in aggregate and adjusted the raw rates using direct adjustment methods for both age and gender, using US Census data for Philadelphia County as the standard population. We grouped individuals by age in ten-year increments and considered age and gender for each of the four ethnicity categories.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Demographics

Data were collected at 17 community-based events and seven faith-based events (Table 1). As the objective of this study was to assess disparities between African, Asian, Hispanic and Caucasian populations, we removed from any further analysis those participants who indicated their racial group to be ‘American Indian or Native American’, ‘Opt Not to Answer’, or a racial combination that did not clearly indicate one of the four target racial groups (40 surveys were excluded, representing 3.3% of N).

Table 1.   Study population demographics*
 AllAsianAfrican AmericanHispanicCaucasian
  1. Values are given as percentages, unless specified.

  2. * Total N varies due to missing values.

Total N1172211398250312
Age group (N = 1170), years
 18–3432.8 33.526.340.834.3
 35–4418.113.920.324.812.8
 44–5421.214.427.818.019.9
 55+27.938.325.616.433.0
Education (N = 1164)
 Up to grade 97.018.30.812.53.6
 HS graduate39.518.845.241.944.2
 College or more53.463.054.045.652.3
Household income (N = 1144), K
 <$2550.451.245.957.450.2
 $25–$5030.621.239.028.727.5
 $50–$8010.39.910.58.212.1
 >$808.717.74.65.710.2
Gender (N = 170)
 Male32.738.128.428.837.5
 Female67.361.971.671.262.5
Body mass index (N = 1124)
 Underweight2.78.61.31.32.0
 Healthy weight32.757.923.323.235.6
 Overweight32.926.431.835.037.0
 Obese31.77.143.740.525.4

A total of 1212 subjects completed surveys onsite; 1172 surveys were included in the analysis: African Americans (N = 398; 34%), Caucasians (N = 312; 27%), Hispanics (N = 250, 21%) and Asians (N = 211; 18%). Women constituted 67% of the study population. In self-reported BMI, we found that 32.9% of our populations were overweight and 31.7% obese, with African Americans and Hispanics reporting the highest proportions of obesity (43.7% and 40.5% respectively).

Population prevalence

We observed differences in population prevalence based on a positive response to the question ‘I get a burning feeling in my lower chest rising up toward my neck (heartburn)’ (Table 2). Thirty-five per cent (34.6%) of the subjects experienced GERD symptoms at least monthly, 26.2% at least weekly and 8.2% at least daily. Statistically significant differences in raw prevalence rates between racial groups were found: 50.0% of Hispanics experienced heartburn at least monthly, compared with 37.0% of Caucasians, 30.8% of African Americans and 19.9% of Asians (> 0.0001). This disparity in prevalence across ethnicities was maintained in other GERD symptom frequencies, i.e. weekly and daily, > 0.0001 and P = 0.0017 respectively.

Table 2.   Heartburn by racial group and frequency
Heartburn frequencyAllAsianAfrican AmericanHispanicCaucasianP-value
  1. Values are given as percentages, unless specified.

  2. * Based on positive response to ‘I get a burning feeling in my lower chest rising up towards my neck (heartburn)’.

  3. † Adjusted by age and gender, direct method, using Philadelphia County 2000 Census.

  4. ‡ Conditional prevalence based on a yes/no response to questions ‘After a meal, I get a burning feeling in my lower chest rising up toward my neck (heartburn)’ and ‘After I go to sleep I get a burning in my lower chest rising up toward my neck’.

  5. § ≥Monthly general occurrence of heartburn based on an indication of monthly symptoms of heartburn in general OR after a meal or after going to sleep.

Total N1172211398250312 
Crude rates*
 Any Freq46.532.243.462.447.3<0.0001
 ≥Monthly34.619.930.850.037.0<0.0001
 ≥Weekly26.214.722.138.029.9<0.0001
 ≥Daily8.24.36.813.68.40.0017
≥Monthly adjusted rates† 20.426.647.936.1 
Conditional prevalence‡
 After meals38.627.535.450.440.7<0.0001
 After sleep27.419.523.438.129.6<0.0001
≥Monthly general occurrence of heartburn§34.220.429.850.436.3<0.0001

The total unadjusted prevalence rate for heartburn at any frequency across all four racial groups was 46.5%. Here, the crude rate for Hispanics was the highest with 62.4% reporting heartburn at least rarely. Asians had the lowest crude rate with 32.3%. Hispanics were shown to have statistically significantly higher prevalence rates for heartburn across all frequencies when compared with other groups: 50.0% monthly, 38.0% weekly and 13.6% daily. The racial group reporting the lowest prevalence for heartburn was the Asian group, 19.9% reporting monthly symptoms, 14.7% weekly and 4.3% daily. Previous population-based and clinically based studies have varied in how prevalence of GERD frequency was reported, e.g. occasionally, monthly, weekly or daily.16 Our study found that regardless of the frequency used to compare prevalence rates amongst the racial groups, the same relationship trends were seen.

Adjusted monthly prevalence

Age and gender differences in prevalence were found between the racial groups. Age- and gender-adjusted monthly prevalence rates for heartburn were 47.9% in Hispanics, 36.1% in Caucasians, 26.6% in African Americans and 20.4% in Asians. The ratios of the age- and gender-adjusted rates comparing the unadjusted rates with adjusted rates (unadjusted/adjusted) for Hispanic, Caucasian, African and Asian groups were 1.04, 1.02, 1.16 and 0.97 respectively.

Conditional prevalence after eating or sleeping

The questionnaire featured two separate questions to assess the prevalence of heartburn for all subjects after a meal and after going to sleep. Statistically significant differences were found between the racial groups for both heartburn after a meal and after sleep (P < 0.0001). The rates after a meal were 50.4%, 40.7%, 35.4% and 27.5% for Hispanics, Caucasians, African Americans and Asians respectively. These rates include all frequencies. The prevalence rate for heartburn after having gone to sleep was lower for all racial groups: 38.1%, 29.6%, 23.4% and 19.5% for Hispanics, Caucasians, African Americans and Asians respectively.

General occurrence of heartburn

A final measure for prevalence was created that accounted for all subjects who indicated experiencing heartburn either in general, after a meal, or after going to sleep. The differences between the racial groups using this combined measure were very consistent with the differences in monthly prevalence rates for heartburn in general: 50.4% for Hispanics, 36.3% for Caucasians, 29.8% for African Americans and 20.4% for Asians (P < 0.0001).

Multivariate analysis of monthly GERD symptoms

We explored relationships between ethnicity, age, gender, education income and self-reported BMI and monthly GERD symptoms with a logistic regression (Table 3). In bivariate analyses of age, we observed that younger adults had a lower prevalence rate for heartburn monthly in comparison with all older age groups and that women had a higher prevalence of heartburn at least monthly when compared with men. However, these differences were not present in multivariate analyses. After controlling for other variables, Asian and African American ethnicities were significantly less likely to have monthly symptoms compared with Caucasians [OR = 0.51 (0.32, 0.80) for Asians and 0.62 (0.44, 0.87) for African Americans]; whereas Hispanics had a significantly greater likelihood of monthly symptoms [OR = 1.57 (1.09, 2.27)]. Individuals with lower self-reported income were twice as likely to have monthly symptoms [OR =2.00 (1.13, 3.57)]. Although we observed that those with very low education and very high education are protective against GERD, nearly half of those with very low education were Asians who were likely to be recent immigrants. Finally, obese individuals were one and a half times more likely to have monthly symptoms compared with those of healthy weight [OR = 1.53 (1.08, 2.16)].

Table 3.   Logistic regression of monthly gastro-oesophageal reflux disease symptoms
VariableOdds ratio95% CI
Ethnicity (reference = Caucasian)
 Asian0.510.32, 0.80
 African American0.620.44, 0.87
 Hispanic1.571.09, 2.27
Age group (reference = 55+), years
 18–340.730.51, 1.04
 35–441.440.97, 2.13
 45–541.380.95, 1.99
Male gender0.760.58, 1.01
Education (reference = 9–12 grade)
 Up to ninth grade0.590.34, 1.04
 College or more0.710.53, 0.96
Income (reference = >$80), K
 <$252.001.13, 3.57
 $25–$501.420.80, 2.53
 $50–$801.850.97, 3.54
BMI (reference = healthy weight)
 Underweight0.670.26, 1.75
 Overweight0.970.69, 1.37
 Obese1.531.08, 2.16

Awareness and knowledge of GERD

We assessed awareness, knowledge and care-seeking patterns for those who had GERD symptoms (N = 544). Seventy-four per cent (74.4%) of those with GERD symptoms were familiar with the term ‘Gastro-oesophageal Reflux Disease or GERD’ (Table 4). Research team members noted that many participants mentioned they had heard of ‘Reflux’ or ‘Reflux Disease’ and not necessarily ‘GERD’. Of those with GERD symptoms, 61.0% of men reported familiarity with GERD, compared with 80.2% of women (P < 0.0001). Although Hispanics had the highest prevalence rate for GERD, their familiarity with GERD was lower (72.4%) than that of Caucasians (78.2%) and Asians (74.4%).

Table 4.   Awareness and knowledge of gastro-oesophageal reflux disease (GERD) by racial group
 AllAsianAfrican AmericanHispanicCaucasianP-value
  1. Values are given as percentages, unless specified.

Total N54468172156147 
Heard of GERD74.475.072.772.478.20.6301
Serious heartburn can lead to ulcers
 Strongly agree/agree76.379.477.180.170.10.0918
 Neutral18.119.117.712.224.5 
 Disagree/strongly disagree5.61.55.37.75.4 
Serious heartburn can lead to cancer
 Strongly agree/agree44.847.040.254.838.80.0722
 Neutral35.437.937.330.337.4 
 Disagree/strongly disagree19.715.122.514.823.8 
I believe the following will help control heartburn
 Sleeping with head of bed raised43.936.836.448.751.00.0200
 Wearing loose fitting clothing18.927.915.618.619.10.1827
 Having smaller more frequent meals52.472.146.245.557.80.0005
 Choosing a low-fat diet48.545.649.157.140.10.0301
 Taking an antacid62.742.758.466.772.80.0001
 Avoiding fried foods61.651.562.462.864.00.3287
 Lying down for 2–3 h after eating16.714.719.121.89.50.0263

Seventy-six per cent (76.3%) of those with GERD symptoms believed that serious heartburn could lead to ulcers, but 5.6% disagreed. Hispanics were most likely to disagree, whereas Asians were least likely. Forty-five per cent (44.9%) of those participants with GERD symptoms believed serious heartburn could lead to cancer whereas 19.7% disagreed and 35.4% were neutral. Participants across all racial groups were more confident that heartburn could lead to ulcers and less confident that heartburn could lead to cancer.

Nineteen per cent (19.1%) of African American and 21.8% Hispanic participants with GERD symptoms falsely believed that lying down for 2–3 h after eating would actually help control heartburn, a significantly higher rate than Asians (14.7%) and Caucasians (9.5%) (P = 0.0263). Less than half of African American and Hispanic participants with GERD symptoms, 46.2% and 45.5% respectively, believed that smaller, more frequent meals might help control heartburn, far less than Asians (72.1%) and Caucasians (57.8%) (P = 0.0005). Unlike the other racial groups studied, less than half of Asians believed that antacids could help control heartburn (P = 0.0001).

Care-seeking patterns

Thirty-four per cent (33.7%, N = 166) of those with GERD symptoms had been diagnosed with GERD by a doctor and this did not vary significantly by ethnicity (Table 5). There was no significant difference in diagnosis rates by a doctor between gender, education levels or household income. When asked if they would see a doctor if they had lesser known symptoms of heartburn, responses varied by ethnic groups, with Caucasian participants more likely to seek care if they had pain or difficulty swallowing (P = 0.428).

Table 5.   Care-seeking patterns for gastro-oesophageal reflux disease by racial group
 AllAsianAfrican AmericanHispanicCaucasianP-value
  1. Values are given as percentages, unless specified.

Total N54468172156147 
Diagnosed by doctor33.730.729.837.935.60.4471
If you had the following experiences, would you see a doctor?
 Pain or difficulty swallowing79.167.279.779.084.30.0428
 Frequent vomiting69.567.271.765.672.00.5533
 Feel full after eating40.034.338.844.139.70.5621
 Chronic hoarseness/breathing problems79.964.271.474.773.30.4384
If I had severe heartburn, I would
 Take OTC medications44.138.241.642.351.70.1697
 See my doctor81.876.583.885.378.20.2329
 Go to a health clinic21.920.624.328.912.20.0045
 Go to an alternative/natural/complimentary medicine provider9.611.87.516.04.10.0031
 Go to an emergency room27.019.137.628.916.30.0001
 Go to a medical specialist in heartburn27.623.528.335.320.40.0293
 Ask about surgical procedures for treatment7.12.95.813.54.10.0033
Total N54468172156147 
I have learned about heartburn from the following sources
 My doctor58.152.957.865.453.10.1280
 My pharmacist14.910.317.314.115.00.5661
 My family37.019.128.949.441.5<0.0001
 Friends/co-workers29.027.923.131.434.00.1598
 Advertisements45.827.943.451.351.00.0054
 The internet17.123.59.818.621.10.0159
 Books/magazine articles30.532.428.930.132.00.9235
 Medical literature29.422.133.523.734.00.0683
 Alternative health sources8.35.98.17.110.90.5423
Issues that prevent you from seeing a doctor if you had heartburn symptoms
 No health insurance24.441.829.735.735.40.3185
 Cannot take time from work29.835.421.837.528.80.0142
 Cannot take time from family19.019.713.027.816.40.0067
 Does not have transportation24.721.523.527.225.20.8067
 Anxiety about seeing doctor23.222.711.733.826.0<0.0001
 Personal beliefs on healthcare15.421.912.419.711.60.0706

Only 9.8% of African Americans with GERD symptoms indicated that they learned about heartburn on the Internet, a rate much lower than Asians (23.5%), Hispanics (18.6%) and Caucasians (21.1%) (P = 0.0159). Likewise, African Americans were also least likely amongst racial groups to use the Internet to get more information about heartburn. Asians were least likely to have learned about heartburn from their doctor (52.7%) as well as least likely to go to a doctor to get more information about heartburn (70.6%). Forty-one per cent (40.8%) of Caucasians learned about heartburn from television ads, compared with 25.0% of Hispanics, 37.0% of African Americans and 16.2% of Asians (P = 0.0003). However, African Americans were significantly more likely to depend on television ads for more information about heartburn compared with other racial groups (P = 0.0003).

In the case of experiencing severe heartburn, Hispanics with GERD symptoms were more likely to go to a health clinic (28.9%) compared with African Americans (24.3%) Asians (20.6%) and Caucasians (12.2%) (P = 0.0045). African Americans and Hispanics were also significantly more likely to go the emergency room for severe heartburn (37.6% and 28.9% respectively) compared with Asians (19.1%) and Caucasians (16.3%) (P = 0.0001). Furthermore, Hispanics were most likely to see specialist for heartburn (P = 0.0293) and consider surgical procedures for heartburn (P = 0.0033).

Forty-two per cent (41.8%) of Asians and 35.7% of Hispanics with GERD symptoms indicated that cost and the lack of health insurance would prevent them from seeing a doctor, higher rates than Caucasians and African Americans, although this was not significant. Hispanics were more likely than the other racial groups to avoid going to a doctor because of not being able to take time away from work (P = 0.0142) or family (P = 0.0067). One third (33.8%) of Hispanics with GERD symptoms stated that ‘fear or anxiety about going to the doctor’ would prevent them from seeing a doctor, a rate significantly higher than other groups (P < 0.0001).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Our population-based community study explored the prevalence, knowledge and health-seeking behaviours of African American, Hispanic, Asian and Caucasian populations in Philadelphia. Data were solicited at community and faith-based organizational centres and health fairs through self-reported surveys.

All four ethnic groups reported experiencing heartburn, at least rarely. The prevalence rates for Caucasians and African Americans were consistent with other published reports.2 However, Hispanics reported the highest prevalence rates across all frequencies which were significantly higher than the other ethnic populations and higher than reported rates in the literature.3 The higher rates in Caucasians and African Americans compared with Asians are consistent with the literature citing a higher prevalence in Western populations than in Asian countries.16 The prevalence rates for the Asian population were the lowest in this study, but higher than the reported rates in East Asian countries.17–19 There were significant differences between ethnicities in their familiarity with ‘GERD’. Although Hispanics had the highest prevalence of heartburn, their familiarity with the term GERD was lower compared with African Americans and Caucasians with Asians having the least familiarity with GERD.

Hispanics in this study had the highest rate of diagnosis and were more likely to seek care at a health clinic compared with the other ethnic groups. Among all of the identified treatment options, Hispanics identified ‘see my doctor’ more frequently than the other options including ‘take OTC’, ‘go to a health clinic’ and ‘go to an alternative medicine provider’. This finding is consistent with studies reporting Hispanics choosing traditional care when concerned about the seriousness of a health condition. Fear or anxiety about going to the doctor, inability to take time from work or family were more significant deterrents for Hispanics than other groups. Cost, lack of health insurance and transportation were barriers as well. Television advertisements were the source of information about heartburn for Hispanics. While many Hispanics correctly identified behaviours to help control heartburn, other behaviours known to exacerbate heartburn were incorrectly identified as well.

Based on the respective prevalence rates, the data reflect a need for education about GERD for all populations. The prevalence rates for Asians in the US, although consistent in having the lowest rate among ethnic groups, were higher than the generally reported rate for Asians in Far East countries. The physical, psychological, social and economic impacts of GERD are significant as are associated complications. Increased knowledge will not only alert patients to symptoms, which may be indicative of GERD but also inform necessary behavioural changes including diet, substance use, knowing how to avoid and reduce the severity of symptoms, taking appropriate medications and seeking professional medical care at an earlier time. The need is most pressing in the Hispanic population reflected by the associated prevalence rate. Even though Latinos reported GERD symptoms more frequently, they are less aware of GERD as a disease. This lack of knowledge regarding GERD may delay their seeking treatment until symptoms are severe.

Education about GERD must be provided at medical visits and in the community. Soliciting information about classic and associated symptoms will raise patient awareness about symptoms that they may attribute to indigestion or seemingly insignificant disorders. A thorough medication history of all prescribed, OTC and alternative/home remedies may also provide a window of opportunity to assess concurrent GERD-like symptoms and intervene. Utilization of a validated culturally sensitive questionnaire with words and graphics will facilitate communication about symptoms and treatment.20 Healthcare providers should also educate patients at routine visits about the symptoms and treatment of GERD. Education should include identification of risk factors and triggers, symptoms associated with GERD, dietary and life-style modifications for reducing the severity and frequency of symptoms, appropriate OTC medications and guidelines for seeking care.

Limitations

There are a number of limitations with our study specific to subject definition, selection and exploration of physiological, psychological and social reasons for differences among ethnicities. The study was designed to obtain unsolicited feedback at a community level from members of the African American, Hispanic, Asian and Caucasian communities in selected ethnic neighbourhoods in Philadelphia. To meet our targeted number of subjects per group, certain neighbourhoods were specifically selected, possibly resulting in selection bias. In addition, to achieve the targeted 200 subjects per group, the study was conducted at community and faith-based facilities and affiliated health fairs held in various locations, possibly resulting in selection bias with participants seeking health information. Age was the only ‘inclusion’ criterion. There was no randomization in selection of participants or control groups for this survey seeking as large a response rate as possible. We acknowledge that the results of this survey may be limited in their generalizability, although we have adjusted prevalence rates by age and gender within ethnicity using US Census data for Philadelphia County in 2000. This study anecdotally documents important differences in awareness and care-seeking behaviours between racial groups and lays the ground for future work that more definitively reports on trends for these groups.

The study instrument was an eight-page, 34 question survey adapted from a population-based observational epidemiological validated instrument with English, Spanish and Chinese versions and translators to assist with comprehension.14 Although all translators were healthcare providers who are cognizant of survey research, we cannot be certain that they did not lead respondents in responding to the survey. Some of the limitations of the design include the necessity for the participant to be literate or have a literate companion assist with comprehension and the limitation of Asian participants to be able to read English and/or Chinese, possibly restricting the participation of the Asian community.

In addition, the data provided were self-reported without the benefits of soliciting a health history or medical records. Ethnicity was also self-selected by participants. If subjects did not identify themselves as a member of the African American, Hispanic, Asian or Caucasian population, their data were excluded from the analysis. This would include people who consider themselves ‘biracial’ or multi-ethnic. In addition, the study design and questionnaire did not accommodate for considering physiological, psychological and social reasons for differences among ethnicities.

Summary

The results of our study show that all four ethnic groups experienced heartburn, at least rarely. GERD symptoms are more frequent in Hispanics and lower in Asians living in the US compared with other ethnic groups. Prevalence rates for African American and Caucasian populations in this study were similar to findings in other population-based studies in Western countries. The reasons for these ethnic disparities require further study. We have identified additional areas of future research that might explore factors, which contribute to GERD as well as effective health promotion strategies to reduce the prevalence of GERD in all populations. Reasons for the increase in prevalence of GERD in Asians in the US compared with population-based studies in the Far East require investigation. The data from our study confirm existing prevalence trends, highlight new findings, inform clinical practitioners about the attitudes and behaviours impacting health-seeking behaviours and identify opportunities for education in the clinical and community settings.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Declaration of personal interests: Elaine J. Yuen, PhD, MBA, is a healthcare consultant and adjunct associate professor at Thomas Jefferson University, School of Population Health. Martha C. Romney, MS, JD, MPH, is an assistant professor; Nicole M. Cobb, MA, is a project manager; and Neil I. Goldfarb is vice dean for research; all are employed by Thomas Jefferson University, School of Population Health. At the time of this research, Richard W. Toner was a project manager in the Jefferson School of Population Health. Maya Spodik, MD, is employed by the Department of Gastroenterology, Thomas Jefferson University. Philip O. Katz, MD, is clinical professor of medicine, Thomas Jefferson University; and chairman, Division of Gastroenterology, Albert Einstein Medical Center. Declaration of funding interests: This study was funded in full by AstraZeneca, Grant Number IRUSESOM0448. AstraZeneca did not provide any outside writing support for this project, which was funded under their Independent Investigator Study (ISS) mechanism. This funding provided research support for Dr. Yuen, Mr. Toner, Ms. Moore, Mr. Goldfarb and Dr. Katz. Ms. Romney and Dr. Spodik received no funding for this project, and have no financial interests to declare other than their employment. The data were collected and analysed by Thomas Jefferson University and the study was approved by the Jefferson Institutional Review Board. AstraZeneca did not have any editorial control over the content of the study. Dr. Yuen has also received funding from the State of Pennsylvania, the Robert Wood Johnson Foundation, and the Regione Emilia Romagna, Italy, to support her work studying population-based patterns and costs of care, and care disparities for underserved populations. She serves on the board of SEAMAAC, a community-based organization serving SE Asians and other ethnic minorities in Philadelphia. Dr. Katz has served as a speaker for Takeda and a consultant for Eisai, Novartis, Takeda and Xenoport. Dr. Yuen, Dr. Goldfarb and Ms. Moore also serve as advisory board members for the Greater Philadelphia Schweitzer Fellows Program.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Appendix S1. Survey Instrument: Attitudes and knowledge about GERD.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
APT_4396_sm_AppendixS1.doc163KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.