Publication
Article
AJPB® Translating Evidence-Based Research Into Value-Based Decisions®
Author(s):
The perception and utilization of OTC medications among primary care providers for gastroesophageal reflux disease and chronic constipation shows a need for improved patient education on these medications.
ABSTRACT
Objectives: Primary care providers (PCPs) treat the vast majority of gastroesophageal reflux (GERD) and chronic constipation (CC) patients in the United States. The aim of this study was to evaluate how PCPs perceive and utilize store brand and brand name OTC products for GERD and CC.
Methods: A 33-question survey was electronically sent to 622 eligible PCPs from the Internal Medicine Insider Research panel. The survey asked respondents about their usage of store brand and brand name OTCs for patients with GERD and CC, and requested their opinions about the quality, efficacy, safety, and price of these medications. Chi-square and students' t tests were utilized for bivariate analysis.
Results: A total of 337 PCPs (54% response rate) completed the survey. For the typical GERD and CC patient, 64% and 98% of PCPs utilized OTC medications for first-line treatment, respectively. Proton pump inhibitors (PPIs) were the most common OTC utilized for GERD (59%), while OTC fiber supplements (45%) were utilized most commonly for CC. The vast majority of PCPs felt that OTC store brand and brand name PPIs (79%) and polyethylene glycol (PEG) 3350 (87%) were equally effective. Despite this, only 24.7% of physicians have “some or extensive” discussion with patients about the use of store brand products in the treatment of their gastrointestinal disease. PCPs believed that store brand products were less expensive than brand name products, but more than 60% underestimated the cost savings (>20%) for both PPIs and PEG 3350. Few physicians had knowledge regarding the regulation of store brands.
Conclusions: For PCPs, OTC drugs are the cornerstone of treating GERD and CC patients. Though the majority of PCPs feel that store brand and brand name treatments offer similar effectiveness and reduced cost for GERD and CC patients, few discuss these points with their patients. Systematic efforts are needed to ensure that patients are better educated about low-cost treatment options for GERD and CC.
Am J Pharm Benefits. 2017;9(4):116-121
Gastroesophageal reflux (GERD) and chronic constipation (CC) are common gastrointestinal complaints seen by primary care providers (PCPs). GERD affects at least 20% of the US population and is the fourth most common chronic condition seen in primary care.1,2 Constipation affects 16% of all adults and 33% of adults 60 years or later.3 Both disorders have high direct medical costs, with GERD being the leader of all digestive diseases at $12.1 billion yearly.4 For both disorders, medications contribute significantly to these costs.4
For both GERD and CC, the medication landscape has changed significantly in the last 10 years. This is primarily due to the availability of OTC formulations— either generic or store brand (ie, Target-brand omeprazole or Walmart-brand PEG 3350) or brand name (ie, Prilosec or Miralax)—of drugs previously available only by prescription. Brand name and store brand OTC drugs each contain the same active ingredient, but differ in price. Third-party payers are increasingly refusing to initially cover the cost of high-tier prescription medications for “lifestyle conditions.” This forces patients with conditions like GERD and CC to purchase OTC options out of pocket, placing a premium on economic value in addition to efficacy.
As PCPs see the most patients with GERD 5,6 and CC, 7 it is critical to understand their prescribing practices regarding prescription and OTC products in this era of cost consciousness. Therefore, the aim of this study was to assess how US PCPs perceive and utilize prescription as well as brand name and store brand OTC medications for GERD and CC, and to assess PCPs’ knowledge of the federal regulation of OTC medications. We hypothesized that US PCPs would be very familiar with and commonly recommend OTC medications for GERD and CC, but that they would have little knowledge of manufacturing, efficacy, and price differences between store brand OTC gastrointestinal medications and brand name alternatives.
MATERIALS AND METHODS
Sample Population
After attaining exemption through the University of Michigan Medical School institutional review board, a nationally representative sample of US internists was surveyed in 2015. For our sampling frame, we utilized the American College of Physicians (ACP) Research Center’s Internal Medicine Insider Research panel, which is a representative group of ACP members who have voluntarily agreed to participate in periodic physician surveys.8 The eligible population consisted of 622 internists of the 1286 ACP members who are representative of the United States ACP membership across multiple demographic characteristics. Panelists who completed surveys were awarded $10 electronic gift cards. Information was not collected on nonresponders. To attain a 95% confidence level with a 5% margin of error for our results, we required a response from at least 244 of the 662 members.
Survey Methods
The survey was developed by the authors after careful literature review and consulting with physicians regarding their prescribing practices for GERD and CC. All of the authors reviewed, tested, and edited the questions for readability. This survey was formulated without any influence from our pharmaceutical sponsor. Participants answered 33 questions assessing sociodemographics, including age, race/ethnicity, practice type, years of practice, and weekly patient volume; their usage of store brand and brand name OTCs for GERD and CC; and their knowledge about these products’ source, quality, efficacy, safety, and price, especially in comparison to one another (eAppendix, available at ajpb.com). Respondents were also assessed on their knowledge of the OTC regulatory process.
Statistical Analysis
Survey data were analyzed using descriptive statistics. Questions were examined to assess differences in respondent’s gender, practice type, practice region, years in practice, and patient volume. Practice years were categorized into 3 categories: 10 years or less, 11 to 20 years, and greater than 20 years. Primary outcome measures were the reported first-line medications used for routine patients with GERD or CC. We then performed bivariate analysis for demographic predictors including age, race/ethnicity, practice type, years in practice, and weekly patient volume for each of the survey’s questions using c2 and students' t tests. Two-sided P values >.05 were considered to be statistically significant. Statistical analyses were performed using SPSS version 23 (IBM Corp; Armonk, New York).
RESULTS
Of the 662 potential respondents, 337 internists (54% response rate) completed and returned the survey. Characteristics of the study population are described in Tables 1 and 2. Our study population was comparable with the current composition of US ACP members. The majority of participants were male (58.5%) and Caucasian (63%) with 47% in practice for more than 20 years. Respondents represented all 4 US geographic regions, with the largest response coming from the Western region (31%).
GERD
For the typical acid reflux patient, 64.5% of participants recommended an OTC medication as a first-line treatment (Table 3). Proton pump inhibitors (PPIs), both OTC and prescription, were the most commonly prescribed (59.4%) in the typical reflux patient. Almost two-thirds (63.5%) of the respondents who recommended a PPI suggested an OTC product. Thirty-one percent of practitioners started treatment for GERD with either an OTC (17.3%) or prescription (13.7%) H2 receptor antagonist (H2RA). Providers in the Midwest were least likely to utilize H2RAs (7%) for first-line treatment compared with other parts of the country (28.1% Northeast, 20.3% South, 29.7% West [P = .01]). The remaining practitioners (9.3%) recommended OTC antacids. There were no other predictors for the physicians who practiced “step-up” therapy, ie, used an antacid or H2RA as first-line medication.
More than three-fourths of respondents (76%) felt that store brand PPIs had an equal bioequivalence to brand name PPIs, and 82% of respondents felt that store brand PPIs were equally or more clinically effective than brand name PPIs. However, despite this belief, only 25.1% of respondents specifically recommended a store brand PPI “always” or “most of the time.” There were no defining provider or practice characteristics with this recommendation. Conversely, 19.1% of respondents specifically directed their patients to purchase a brand name PPI “always” or “most of the time.”
Respondents underestimated the cost savings of store brand products compared with brand name products for GERD. Overall, 61% of respondents did not realize that cost savings could exceed 20% by purchasing a store brand OTC instead of a brand name OTC for GERD treatment. However, 46.6% of respondents understood that store brand products offered some cost savings (1%-20%) in comparison with brand name products.
Chronic Constipation
For typical patients with CC, more than 98% of respondents utilized an OTC for first-line therapy, including fiber supplements, stimulants, osmotics, and stool softeners (Figure). Fiber supplements were the primary first-line therapy utilized for CC, with 45.7% of respondents recommending this therapy. Less than 2% of respondents recommended a prescription medication such as linaclotide as a first-line therapy. No respondents recommended the prescription medication lubiprostone as a first-line therapy. For patients who failed to improve after first-line therapy, nearly 88% of respondents continued to recommend OTCs as a second-line therapy, with osmotics as the first choice (29.8%) and OTC stimulants a close second (27%) (Figure). Only 12.5% of respondents recommended prescription medications for second-line therapies. For those who prescribed OTC fiber as first-line treatment, osmotics (OTC 44.2% and Rx 5.4%) were the most commonly prescribed second-line treatment. OTC stool softeners and stimulants were second-line choices of 32% and 16.3% of providers who initially prescribed fiber, respectively. Stool softeners were utilized by 23.3% and 19.3% of providers for first-line and second-line treatment, respectively. Providers aged 45 years or less were the most likely to use stool softeners (31%) for first-line treatment, compared with physicians aged 46-60 years (18.1%) and older than 60 years (14.3%) (P = .01). No other predictors for CC prescribing practices were seen.
Respondents were asked to assess the effectiveness of store brand polyethylene glycol (PEG) 3350 to its brand name counterpart PEG 3350 (Miralax). Eighty-seven percent of respondents believed that store brand PEG 3350 and Miralax were equally effective, and 2% believed that the store brand was slightly or significantly more effective than the brand name product. Similar to GERD drugs, respondents underestimated the cost savings associated with store brand PEG 3350 versus Miralax. About 30% answered that they thought the store brand product was more than 20% less expensive than the brand name product. A total of 53.4% of participants thought the store brand was a little less expensive (1%-20%) than the brand name.
OTC Regulatory Process
Internists were asked to evaluate their knowledge of the drug approval process for store brand OTC products. More than 70% of respondents reported they had “little” or “no” knowledge of the approval process for store brand products. Furthermore, more than 75% of respondents had “little” or “no” discussion regarding store brand OTC medications as a treatment option for GERD or CC with their patients. We then assessed the internists’ sense of bioequivalence of store brand medications from different retailers. Only 25.5% of internists “strongly” agreed that store brand medications from different retailers (Walgreens, Target, CVS, etc) were bioequivalent. Additionally, only 22.6% of respondents “strongly” agreed that production standards and FDA inspections were similar for facilities producing store brand OTCs and brand name products. Forty-three percent of participants correctly indicated that 1 to 2 manufacturers produce the majority of store brand OTC GI medications.
DISCUSSION
Our survey of US PCPs demonstrated some fascinating results on the reported medical treatment of GERD and CC. The majority of PCPs utilize OTC medications as the cornerstone of treatment for both GERD and CC. Although the majority of PCPs feel that store brand and brand name treatments offer similar effectiveness for both complaints, few directly discussed this opinion with their patients. Additionally, more than half the respondents recognized some cost savings associated with store brand versus brand name treatments. However, providers did not know the full extent of the savings.
For treatment of GERD, the most recent guidelines recommend empiric medical therapy with a PPI.9 Six of 10 internists chose a PPI, whether prescription or OTC, for initial use. Still, a significant portion of practitioners (40%) would utilize step-up therapy by starting with either an antacid or H2RA. This practice was common in the early 2000s with close to two-thirds of practitioners utilizing this strategy.10 Shortly thereafter, however, Howden and colleagues performed a clinical trial within a primary care setting, demonstrating that step-up therapy was inferior for control of GERD symptoms compared with initial PPI and continued dosing.11 A national survey of PCPs published in 2005 seemed to indicate this practice was largely abandoned, with only a minority of providers using this method at that time. However, our survey results suggest that a significant minority of PCPs still utilizes a “step-up” strategy for the treatment of GERD. This may be secondary to the emerging reports suggesting adverse effects associated with PPIs, such as the potential risk of fractures, clostridium difficile infections, hypomagnesemia, and pneumonia.12-19 Because this survey was performed in August 2015, and the associations of PPIs with chronic kidney disease and dementia were published and publicized nationally in 2016,20,21 we may continue to see increased use of step-up therapy in 2017 and beyond.
Internists utilized a variety of OTCs as their first- and second-line treatments for constipation. Even though OTC fiber supplements were the most commonly recommended first-line treatment, they seem to be underutilized by participating providers. National guidelines support the use of fiber supplements as first-line therapy for constipation.3 The use of OTC osmotics, specifically PEG 3350, is also a reasonable first-line choice for CC.3,22 Interestingly, OTC stool softeners were almost as commonly utilized for first-line treatment as OTC osmotics; these were used more than twice as commonly by physicians younger than 45 years compared with older physicians. This is troubling, as there is a lack of clinical evidence to support the efficacy of docusate for the management of constipation. The available data demonstrate that docusate is no more effective than placebo.23-25 For second-line treatment for constipation, OTC osmotics were the most popular choice, although only 30% of practitioners chose them. Additionally, there was a marked increase in the use of OTC stimulants from first- to second-line therapy (4.7%-27%), and the persistent use of OTC stool softeners for 1 in 5 providers continued in the second line. OTC stimulants work by propagating colonic contractions with effects on the balance between secretive and absorption.26,27 There were early concerns about the use of stimulants and potential damage to the enteric nerve system; however, this has not been seen in real-world practice.28,29 Stimulants are typically recommended as a third-line medication, as either a rescue therapy or as an adjunct to an osmotic regimen.
This survey provided evidence of some other interesting thought trends. A majority of the participating internists agreed that store brand and brand name PPIs and PEG 3350 were both bioequivalent and clinically effective. Additionally, the majority underestimated the extent of cost savings associated with store brand OTC PPIs and PEG 3350 compared with brand name medications. Importantly, only a minority had conversations with patients about the clinical effectiveness and cost savings associated with store brand products. Thus, significant cost savings continue to go unrealized by patients. These findings were similar to those that emerged from our survey of practicing gastroenterologists.30 It is worth noting that the cost savings for a particular store brand PPI is directly correlated to the length of time that PPI has been OTC. In addition to practitioners having knowledge deficits about this topic, it is also possible that conversations don’t happen because practitioners are under time constraints during patient visits, with too many other things to discuss. Lastly, most participants admitted to minimal knowledge about the OTC regulatory process, and just about half of internists understood the bioequivalence among store brands of different retailers and that the production standards and FDA inspections are the same for both store brand and brand name products.
This presents an opportunity for further education. In order for an OTC to be approved, the brand name pharmaceutical company has to demonstrate the following: 1) the drug is safe and effective, 2) label comprehension studies have been performed to ensure the product label is understandable, and 3) actual use studies have been performed to ensure that the general population can use the product correctly according to the label, without a prescriber. If the brand name drug is FDA-approved for OTC use, the brand name pharmaceutical company has 3 years of additional exclusivity after coming off patent. The approval process for the generic market takes approximately 5 to 8 years. Generic pharmaceutical companies must apply for an amended new drug application (ANDA), despite patents on brand name drugs. Fundamentally, brand name and store brand medications contain the same active ingredient. Although phase 3 efficacy trials are not required, bioequivalence studies are needed for FDA approval of store brand products.18 Those companies that file an ANDA may be sued by the brand name pharmaceutical company for patent infringement. If there is litigation and the generic company wins, they can then launch their market formulation. These OTC generic medicines are then sold to retailers and are labeled with a store name, ie, Walmart-brand omeprazole.
LIMITATIONS
Our study has some potential limitations. Our survey sampled physicians from the ACP’s membership list. It is possible that the responders may not represent the larger group of practicing PCPs, ie, an internist’s practice may not reflect a family-medicine physician’s practice. The generalizability of these findings to other PCPs is unknown. Additionally, our survey results relied on self-reported data, and so our findings may be subject to a number of unforeseen biases.
CONCLUSIONS
For internists, OTC drugs are the cornerstone of treating GERD and CC patients. Even though the majority of PCPs feel that store brand and brand name treatments offer similar effectiveness and reduced cost for GERD and CC patients, few discuss these points with their patients. Systematic efforts are needed to ensure that patients are better educated about cost-effective treatment options for GERD and CC.
Author Affiliations: Division of Gastroenterology, University of Michigan Health System (SWC, WDC, SDS, SBM), Ann Arbor, MI; Division of Gastroenterology, Ann Arbor Veterans’ Administration Hospital (SDS, SBM), Ann Arbor, MI; American College of Physicians (AW, LH), Philadelphia, PA; VA Center for Clinical Management Research (SDS), Ann Arbor, MI.
Source of Funding: This study was supported by an unrestricted grant from Perrigo. The sponsor played no role in design of the survey or analysis of the results.
Author Disclosures: Dr W.D. Chey has worked as a consultant for or received grants from Astra-Zeneca, Actavis, Ardelyx, IM Health, Ironwood, Nestle, Prometheus, OOL Medical, Salix, SK, Sucampo, Takeda, and Vibrant. Dr Saini was a paid consultant for FMS, Inc, on a modeling study examining the effects of PPI in aspirin users. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (WDC, SDS, SBM); acquisition of data (SDS, AW, LH); analysis and interpretation of data (SWC, WDC, SDS, AW, LH, SBM); drafting of the manuscript (SWC, SDS, SBM); critical revision of the manuscript for important intellectual content (SWC, WDC, SDS, SBM); statistical analysis (AW, LH);; obtaining funding (WDC); administrative, technical, or logistic support (SWC, WDC, SDS, LH); and supervision (SDS, SBM).
Address correspondence to: Stacy B. Menees, MD, 3912 Taubman Center, SPC 5362, Ann Arbor, MI 48109-5362. E-mail: sbartnik@med.umich.edu.
REFERENCES
1. El-Serag HB, Sweet S, Winchester CC, Dent J. Update on the epidemiology of gastro-oesophageal reflux disease: a systematic review. Gut. 2014;63(6):871-880. doi: 10.1136/gutjnl-2012-304269.
2. Ornstein SM, Nietert PJ, Jenkins RG, Litvin CB. The prevalence of chronic diseases and multimorbidity in primary care practice: a PPRNet report. J Am Board Fam Med. 2013;26(5):518-524. doi: 10.3122/jabfm.2013.05.130012.
3. Bharucha AE, Pemberton JH, Locke GR 3rd. American Gastroenterological Association technical review on constipation. Gastroenterology. 2013;144(1):218-238. doi: 10.1053/j.gastro.2012.10.028. Review.
4. Everhart JE, e., The Burden of Digestive Diseases in the United States, U.S.D.o.H.a.H.S. National Institute of Diabetes and Digestive and Kidney Diseases, Editor. 2008, NIH Publication: Bethesda, MD.
5. Friedenberg FK, Hanlon A, Vanar V, et al. Trends in gastroesophageal reflux disease as measured by the National Ambulatory Medical Care Survey. Dig Dis Sci. 2010;55(7):1911-1917. doi: 10.1007/s10620-009-1004-0.
6. Halpern R, Kothari S, Fuldeore M, et al. GERD-related health care utilization, therapy, and reasons for transfer of GERD patients between primary care providers and gastroenterologists in a US managed care setting. Dig Dis Sci. 2010;55(2):328-337. doi: 10.1007/s10620-009-0927-9.
7. Shah ND, Chitkara DK, Locke GR, Meek PD, Talley NJ. Ambulatory care for constipation in the United States, 1993-2004. Am J Gastroenterol. 2008;103(7):1746-1753. doi: 10.1111/j.1572-0241.2008.01910.x.
8. Butkus R, Weissman A. Internists’ attitudes toward prevention of firearm injury. Ann Intern Med. 2014;160(12):821-827. doi: 10.7326/M13-1960.
9. Katz PO, Gerson LB, Vela MF. Guidelines for the diagnosis and management of gastroesophageal reflux disease [published correction appears in Am J Gastroenterol. 2013;108(10)1672]. Am J Gastroenterol. 2013;108(3):308-328; quiz 329. doi: 10.1038/ajg.2012.444.
10. Wilcox CM, Heudebert G, Klapow J, Shewchuk R, Casebeer L. Survey of primary care physicians’ approach to gastroesophageal reflux disease in elderly patients. J Gerontol A Biol Sci Med Sci. 2001;56(8):M514-M517.
11. Howden CW, Henning JM, Huang B, Lukasik N, Freston JW. Management of heartburn in a large, randomized, community-based study: comparison of four therapeutic strategies. Am J Gastroenterol. 2001;96(6):1704-1710.
12. Khalili H, Huang ES, Jacobson BC, Camargo CA Jr, Feskanich D, Chan AT. Use of proton pump inhibitors and risk of hip fracture in relation to dietary and lifestyle factors: a prospective cohort study. BMJ. 2012;344:e372. doi: 10.1136/bmj.e372.
13. Targownik LE, Leslie WD, Davison KS, et al; CaMos Research Group. The relationship between proton pump inhibitor use and longitudinal change in bone mineral density: a population-based study [corrected] from the Canadian Multicentre Osteoporosis Study (CaMos) [published correction appears in Am J Gastroenterol. 2013;108(1):157]. Am J Gastroenterol. 2012;107(9):1361-1369. doi: 10.1038/ajg.2012.200.
14. Dial S, Delaney JA, Barkun AN, Suissa S. Use of gastric acid-suppressive agents and the risk of community-acquired Clostridium difficile-associated disease. JAMA. 2005;294(23):2989-2995.
15. Linsky A, Gupta K, Lawler EV, Fonda JR, Hermos JA. Proton pump inhibitors and risk for recurrent Clostridium difficile infection. Arch Intern Med. 2010;170(9):772-778. doi: 10.1001/archinternmed.2010.73.
16. Hess MW, Hoenderop JG, Bindels RJ, Drenth JP. Systematic review: hypomagnesaemia induced by proton pump inhibition [published correction appears in Aliment Pharmacol Ther. 2012;36(11-12):1109]. Aliment Pharmacol Ther. 2012;36(5):405-413. doi: 10.1111/j.1365-2036.2012.05201.x.
17. Luk CP, et al. Proton pump inhibitor-associated hypomagnesemia: what do FDA data tell us? Ann Pharmacother. 2013;47(6):773-780. doi: 10.1345/aph.1R556.
18. Laheij RJ, Sturkenboom MC, Hassing RJ, Dieleman J, Stricker BH, Jansen JB. Risk of community-acquired pneumonia and use of gastric acid-suppressive drugs. JAMA. 2004;292(16):1955-1960.
19. Sarkar M, Hennessy S, Yang YX. Proton-pump inhibitor use and the risk for community-acquired pneumonia. Ann Intern Med. 2008;149(6):391-398.
20. Lazarus B, Chen Y, Wilson FP, et al. Proton pump inhibitor use and the risk of chronic kidney disease. JAMA Intern Med. 2016;176(2):238-246. doi: 10.1001/jamainternmed.2015.7193.
21. Gomm W, von Holt K, Thomé F, et al. Association of proton pump inhibitors with risk of dementia: a pharmacoepidemiological claims data analysis. JAMA Neurol. 2016;73(4):410-416. doi: 10.1001/jamaneurol.2015.4791.
22. Ford AC, Moayyedi P, Lacy BE, et al; Task Force on the Management of Functional Bowel Disorders. American College of Gastroenterology monograph on the management of irritable bowel syndrome and chronic idiopathic constipation. Am J Gastroenterol. 2014; |109| (suppl 1):S2-S26; quiz S27. doi: 10.1038/ajg.2014.187. Review.
23. Fosnes GS, Lydersen S, Farup PG. Effectiveness of laxatives in elderly-a cross sectional study in nursing homes. BMC Geriatr. 2011;11:76. doi: 10.1186/1471-2318-11-76.
24. Tarumi Y, Wilson MP, Szafran O, Spooner GR. Randomized, double-blind, placebo-controlled trial of oral docusate in the management of constipation in hospice patients. J Pain Symptom Manage. 2013;45(1):2-13. doi: 10.1016/j.jpainsymman.2012.02.008.
25. Hawley PH, Byeon JJ. A comparison of sennosides-based bowel protocols with and without docusate in hospitalized patients with cancer. J Palliat Med. 2008;11(4):575-581. doi: 10.1089/jpm.2007.0178.
26. Ewe K, Hölker B. [The effect of a diphenolic laxative (Bisacodyl) on water- and electrolyte transport in the human colon (author’s transl)]. Klin Wochenschr. 1974;52(17):827-833.
27. Manabe N, Cremonini F, Camilleri M, Sandborn WJ, Burton DD. Effects of bisacodyl on ascending colon emptying and overall colonic transit in healthy volunteers. Aliment Pharmacol Ther. 2009;30(9):930-936. doi: 10.1111/j.1365-2036.2009.04118.x.
28. Dufour P, Gendre P. Ultrastructure of mouse intestinal mucosa and changes observed after long term anthraquinone administration. Gut. 1984;25(12):1358-1363.
29. Kiernan JA, Heinicke EA. Sennosides do not kill myenteric neurons in the colon of the rat or mouse. Neuroscience. 1989;30(3):837-842.
30. Menees SB, Guentner A, Chey SW, Saad R, Chey WD. How do US gastroenterologists use over-the-counter and prescription medications in patients with gastroesophageal reflux and chronic constipation? Am J Gastroenterol. 2015;110(11):1516-1525. doi: 10.1038/ajg.2015.156.