PTH-131 Prevalence of dyspepsia in individuals with gastro-oesophageal reflux-type symptoms in the community: a meta-analysis

    loading  Checking for direct PDF access through Ovid



Dyspepsia and gastro-oesophageal reflux are highly prevalent in the general population, but the two conditions are felt to be separate entities. We conducted a systematic review and meta-analysis to estimate the prevalence of dyspepsia in individuals with gastro-oesophageal reflux symptoms (GORS), and to quantify the overlap between the two disorders


MEDLINE, EMBASE, and EMBASE Classic were searched (up until September 2016) to identify population-based studies reporting the prevalence of dyspepsia and GORS in adults (≥15 years), defined using specific symptom-based criteria or a questionnaire. The prevalence of dyspepsia and weekly GORS were extracted for all studies. Pooled prevalence, according to study location and criteria used to define weekly GORS or dyspepsia, as well as odds ratios (OR), with 95% confidence intervals (CIs) were calculated. The degree of overlap between the two was examined.


Of 14 132 papers evaluated, 79 reported prevalence of weekly GORS. Nineteen of these study populations, containing 1 11 459 participants, also reported the proportion of individuals with dyspepsia. The prevalence of dyspepsia in those with weekly GORS was 43.9% (95% CI, 35.1%–52.9%). The pooled OR for dyspepsia in individuals with weekly GORS, compared with those without, was 6.94 (95% CI 4.33 to 11.12). The OR for dyspepsia in weekly GORS remained significantly higher in all geographical regions studied, and across all diagnostic criteria used. The pooled degree of overlap between the two conditions was 25.9% (95% CI, 19.9%–32.4%), varying from 22% when the Bowel Disease Questionnaire was used to define weekly GORS, to 42.6% with the Mayo Reflux Questionnaire.


The OR of dyspepsia in individuals with weekly GORS was seven-fold that of individuals without GORS, and that there is overlap between the two conditions in up to one-quarter of individuals. Reasons for this remain speculative, but may include shared pathophysiological mechanisms or residual confounding.

Disclosure of Interest

None Declared

Related Topics

    loading  Loading Related Articles