Gastroesophageal reflux disease (GERD) affects nearly 30% of the US population.1 Clinicians recommend dietary and lifestyle modifications to prevent GERD symptoms, but no prospective data are available to inform these recommendations. We evaluated the joint association of dietary and lifestyle factors with the risk of GERD symptoms.
The Nurses’ Health Study II is an ongoing nationwide prospective cohort study established in 1989 with 116 671 female participants returning biennial health questionnaires, including information on smoking, body mass index (calculated as weight in kilograms divided by square of height in meters), physical activity, medication use, and history of diabetes and a validated, semiquantitative food frequency questionnaire every 4 years. Follow-up exceeds 90%.2 We queried about acid reflux or heartburn in 2005, 2009, 2013, and 2017.3 This study was approved by the Institutional Review Board at the Brigham and Women’s Hospital and the Harvard T. H. Chan School of Public Health. Return of the questionnaire was considered to imply consent. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
For this analysis, we excluded women at baseline if they reported GERD symptoms weekly or more, had cancer, regularly used proton-pump inhibitors (PPI) and/or histamine receptor antagonists (H2RA), had missing dietary data, or were lost to follow-up. We calculated an antireflux lifestyle score (range, 0-5) consisting of 5 dichotomized variables: normal weight (body mass index ≥18.5 and <25.0); never smoking; moderate-to-vigorous physical activity for at least 30 minutes daily; no more than 2 cups of coffee, tea, or soda daily; and a prudent diet4 (top 40% of dietary pattern score). Women were considered to have GERD symptoms if they reported acid reflux or heartburn at least weekly, as in previous studies.3
Person-years of follow-up accrued from the date of return of the 2007 questionnaire (starting June 1, 2007) to the first report of GERD, death, or end of follow-up (June 1, 2017), whichever came first. We computed multivariable hazard ratios (HRs) and 95% CIs using Cox proportional hazards models stratified by age and time and adjusted with time-varying covariates for total caloric intake; alcohol intake; use of menopausal hormones, PPIs, or H2RA; use of medications that may decrease lower esophageal sphincter pressure (ie, calcium channel blockers, benzodiazepines, and antidepressants); and history of diabetes.
We calculated the population-attributable risk for GERD symptoms attributable to all 5 antireflux lifestyle factors by estimating the relative risk from multivariable logistic regression models while controlling for other covariates. We used a significance threshold of P < .05 using 2-sided tests. SAS, version 9.4 (SAS Institute, Inc) was used for statistical analyses. Data were analyzed from September 1, 2019, to June 22, 2020.
Our cohort included 42 955 women aged 42 to 62 years (mean [SD] age, 52.0 [4.7] years). During 392 215 person-years of follow-up, we identified 9291 incident cases of GERD symptoms.
Compared with women without adherence to antireflux lifestyle factors, the multivariable HR for GERD symptoms was 0.50 (95% CI, 0.42-0.59) for those with 5 antireflux lifestyle factors (Figure). The proportion of cases of GERD symptoms that may be prevented by all 5 factors included in the antireflux lifestyle score was 37% (95% CI, 28%-46%).
Each of the 5 lifestyle factors were independently associated with GERD symptoms. The individual mutually adjusted multivariable HRs for nonadherence to each factor ranged from 0.94 (95% CI, 0.90-0.99; population-attributable risk, 3%) for smoking to 0.69 (95% CI, 0.66-0.72; population-attributable risk, 19%) for body mass index (Table).
We considered the possibility that initiation of PPI and/or H2RA treatment during follow-up may have influenced our results. We conducted analyses in which we considered use of these agents to indicate the presence of GERD symptoms and observed consistent results. Those with 5 factors had a multivariable-adjusted HR of 0.47 (95% CI, 0.41-0.54) for GERD symptoms or initiation of PPI and/or H2RA treatment compared with those with no antireflux lifestyle factors. In an analysis of 3625 women who reported regular use of PPIs and/or H2RAs and were free of GERD symptoms at baseline, those with 5 factors had a multivariable-adjusted HR of 0.32 (95% CI, 0.18-0.57) for GERD symptoms compared with PPI and/or H2RA users with no antireflux lifestyle factors.
Adherence to an antireflux lifestyle, even among regular users of PPIs and/or H2RAs, was associated with a decreased risk of GERD symptoms and may prevent nearly 40% of GERD symptoms that occur at least weekly. Possible explanations include decreases in lower esophageal sphincter tone, increases in gastroesophageal pressure gradients, and mechanical factors, including hiatal hernia.5
Study limitations include self-reported GERD symptoms. However, in clinical practice, GERD is primarily self-reported without confirmatory testing. Second, the population-attributable risk estimate is population specific and assumes a causal relationship. Third, our study was limited primarily to White women. However, GERD is more common in White women aged 30 to 60 years.6 These data support the importance of lifestyle modification in management of GERD symptoms.
Accepted for Publication: October 12, 2020.
Published Online: January 4, 2021. doi:10.1001/jamainternmed.2020.7238
Corresponding Author: Andrew T. Chan, MD, MPH, Clinical and Translational Epidemiology Unit, Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114 (achan@mgh.harvard.edu).
Author Contributions: Drs Mehta and Chan had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Mehta, Staller, Song, Chan.
Acquisition, analysis, or interpretation of data: Mehta, Nguyen, Ma, Staller, Chan.
Drafting of the manuscript: Mehta, Chan.
Critical revision of the manuscript for important intellectual content: Nguyen, Ma, Staller, Song, Chan.
Statistical analysis: Mehta, Ma, Song.
Obtained funding: Chan.
Administrative, technical, or material support: Staller, Chan.
Supervision: Staller, Chan.
Conflict of Interest Disclosures: Dr Staller reported receiving research support from AstraZeneca, Gelesis, and Takeda Pharmaceutical Company Limited; serving as a speaker for Shire Plc; and consulting for Arena Pharmaceuticals, Inc, Boston Pharmaceuticals, and Shire Plc. Dr Chan reported receiving grants and personal fees from Bayer Pharma AG and personal fees from Pfizer, Inc, and Boehringer Ingelheim outside the submitted work. No other disclosures were reported.
Funding/Support: This study was supported by grants U01 CA176726 (Dr Chan), K23 DK120945 (Dr Staller), and K23 DK125838 (Dr Nguyen) from the National Institutes of Health and by a Stuart and Suzanne Steele Massachusetts General Hospital Research Scholar Award (Dr Chan).
Role of the Funder/Sponsor: The funders had no role in the design of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Additional Contributions: We would like to thank the participants and staff of the Nurses’ Health Study II for their valuable contributions. They received no compensation for their contributions.