Step-Down Management of Gastroesophageal Reflux Disease
JOHN M. INADOMI,*,‡ ROULA JAMAL,§,㛳 GLEN H. MURATA,§,㛳 RICHARD M. HOFFMAN,§,㛳LAURENCE A. LAVEZO,§,㛳 JUSTINA M. VIGIL,§,㛳 KATHLEEN M. SWANSON,§,㛳 and
AMNON SONNENBERG§,㛳
*Veterans Administration Center for Practice Management and Outcomes Research, Ann Arbor, Michigan;‡Division of Gastroenterology, Department of Medicine, University of Michigan Medical Center, Ann Arbor, Michigan;§VA Albuquerque Medical Center, Albuquerque, New Mexico; and㛳Department of Medicine, University of New Mexico Health Sciences Center, Albuquerque, New Mexico
Background & Aims: As the economic burden of
gastro-esophageal reflux disease (GERD) is largely weighted to maintenance as opposed to initial therapy, switching from more potent to less expensive medication once symptoms are alleviated (step-down therapy) may prove to be most cost-effective. This study aimed to prospec-tively evaluate the feasibility of step-down therapy in a cohort of patients with symptoms of uncomplicated GERD. Methods: Patients whose GERD symptoms were alleviated by proton pump inhibitors (PPIs) were re-cruited from outpatient general medicine clinics. After baseline demographic and quality of life information were obtained, PPIs were withdrawn from subjects in a stepwise fashion. Primary outcome was recurrence of symptoms during follow-up that required reinstitution of PPIs. Secondary outcomes included changes in quality of life and overall cost of management. Predictors of nonresponse to step-down were assessed. Results: Sev-enty-one of 73 enrolled subjects completed the study. Forty-one of 71 (58%) were asymptomatic off PPI ther-apy after 1 year of follow-up. Twenty-four of 71 (34%) required histamine 2-receptor antagonists, 5/71 (7%) prokinetic agents, 1/71 (1%) both, and 11/71 (15%) remained asymptomatic without medication. Quality of life did not significantly change, whereas management costs decreased by 37%. Multivariable analysis revealed younger age and a dominant symptom of heartburn to predict PPI requirement. Conclusions: Step-down ther-apy is successful in the majority of patients and can decrease costs without adversely affecting quality of life.
C
ontroversy exists concerning optimal management of patients with symptoms of gastroesophageal re-flux disease (GERD). Although proton pump inhibitors (PPIs) heal esophagitis more quickly and effectively than histamine 2-receptor antagonists (H2RAs),1–3as well asmaintain remission in a greater proportion of patients,4 – 6
they are considerably more expensive. Additionally, it is likely that the therapeutic success of alternative medica-tions depends on severity of disease.7 Thus, the key
management issue may not be determination of the most
efficacious therapy, but rather establishing which pa-tients may be effectively treated using less expensive medication.
Both initial and maintenance treatment of patients presenting with symptoms of GERD is debated. Advo-cates of “step-up” therapy argue that the most econom-ical strategy begins with less potent, less expensive med-ication before institution of more potent, more expensive medication if the initial therapy fails.8,9 Proponents of
“step-down” therapy, whereby less expensive medica-tions are instituted only after symptom relief has been achieved with PPIs, claim that this strategy serves the patient’s best interests.10,11 “Step-in” management,
whereby maintenance therapy is withheld unless relapses occur, has also been considered.12
Economic analyses have been performed to determine the most cost-effective strategy to treat GERD.10,12–15
These studies show that the economic burden of this disease is heavily weighted towards maintenance therapy, not initial management. Attempts to model this disease process have also revealed a deficit of information regard-ing the feasibility of step-down therapy in unselected populations. Studies evaluating relapse rates of GERD using maintenance PPIs or H2RAs have been limited to the subpopulation of GERD subjects with erosive esoph-agitis. It has been shown that in this subgroup, mainte-nance therapy with PPIs prevents relapse,16 and
alterna-tive agents such as H2RAs or prokinetic agents may not be effective.4,6,17
As the population with erosive esophagitis represents only a minority of all patients with GERD,18we wished
to evaluate a more inclusive population representative of patients managed in a primary care setting. The primary
Abbreviations used in this paper: GERD, gastroesophageal reflux disease; H2RA, histamine 2-receptor antagonist; PPI, proton pump inhibitor.
©2001 by the American Gastroenterological Association
0016-5085/01/$35.00 doi:10.1053/gast.2001.28649
goal of this study was to determine the feasibility of step-down therapy in patients with symptoms of GERD rendered asymptomatic with PPIs. We also aimed to determine whether there existed clinical factors that predicted the ability of patients to be maintained with less expensive medication.
Materials and Methods
The population studied was composed of outpatients in the general medicine clinics at the Veterans Administration (VA) Albuquerque Regional Medical Center. A list of patients in whom PPIs were prescribed was generated through the pharmacy module of VISTA, the computer database of the medical center. Because PPIs were also prescribed for non-GERD indications, such as ulcer disease or Helicobacter pylori eradication, only those prescriptions filled for greater than 8 weeks were considered. The local institutional review board approved the protocol for this project; informed consent was obtained for administration for quality of life questionnaires. However, because it was the policy of the medical center to initiate step-down therapy in patients with reflux symptoms rendered asymptomatic on PPIs, consent was not required for implementation of step-down therapy.
A clinic was formed in the outpatient general medicine area of the clinics staffed by clinical pharmacists where patients from the pharmacy list of long-term (⬎8 weeks) PPIs were evaluated for eligibility of intervention (the GERD clinic). During the initial visit, patients were interviewed and the chart reviewed to determine the indication for PPI initiation. Patients were considered eligible for step-down management if the indication for PPI therapy was heartburn or acid regurgi-tation. Dyspepsia defined by the Rome criteria19 could be
present in conjunction with heartburn or acid regurgitation and could in fact be the dominant symptom; however, patients in whom dyspepsia was the only complaint were excluded from this study. Finally, patients had to be free of symptoms of heartburn and/or acid regurgitation while on PPI therapy to be considered for entry into this study. Because this was a symp-tom-based protocol, neither an endoscopic diagnosis of erosive esophagitis nor abnormal ambulatory pH monitoring was required for entry into the study. Exclusion criteria con-sisted of the following: previous diagnosis of peptic stricture, extra-esophageal manifestations of GERD, anemia or occult gastrointestinal bleeding, intermittent PPI use, documented Barrett’s metaplasia, scleroderma, or previous gastric or esoph-ageal surgery.
Before intervention, eligible subjects were asked to com-plete 2 questionnaires, the SF-36 and a disease-specific assess-ment of disease severity. This latter instruassess-ment was chosen because of its previous validation in patients with GERD in a VA setting.20Baseline demographic information was also
ob-tained including age, gender, dominant or most bothersome symptom (heartburn, acid regurgitation, or dyspepsia), dura-tion of PPI therapy, previous GERD treatment, comorbid
medical conditions, concurrent medication use, height, and weight.
Step-down management was initiated in the following man-ner. The prescribed dose of PPI was halved; those already on the lowest dose available had PPIs discontinued. The subjects were seen again in the pharmacy GERD clinic 2 weeks after the initial step-down. If symptoms of heartburn or acid regur-gitation recurred in the interim, subjects were instructed to contact the GERD clinic before this visit. Symptoms were assessed and if heartburn or acid regurgitation had recurred, the patient was placed back on their original dose of PPI. If after 2 weeks, heartburn or acid regurgitation had not re-curred, PPIs were discontinued altogether if they had not already been stopped. Subjects were followed at 3-month intervals in the GERD clinic to assess for symptom recurrence. Additionally, subjects were instructed to contact the GERD clinic if heartburn or acid regurgitation recurred. For symptom recurrence, the following protocol was used: high-dose H2RAs were first prescribed (800 mg cimetidine twice daily or 300 mg ranitidine twice daily). Prokinetics (20 mg cisapride twice daily while available or 10 mg metoclopramide twice daily to 4 times daily) could be used in conjunction with H2RAs. If GERD symptoms persisted despite H2RAs and prokinetic therapy, the subjects were placed back on the dose of PPIs to which they had responded before onset of the study.
Six months after undergoing step-down, the SF-36 and disease severity instruments were readministered to subjects. Subjects continued to be followed for 1 year after initiation of step-down through the GERD clinic.
The primary outcome of the study was successful step-down defined as continued alleviation of GERD symptoms without the use of PPIs after 1 year of follow-up. Note that subjects in whom GERD symptoms had recurred but were eliminated by H2RA and/or prokinetic agents were considered to be success-fully stepped down. Secondary outcomes included quality of life and disease severity as measured by the SF-36 and disease severity instruments. Predictors of the inability to successfully step-down (defined as the requirement of PPIs to alleviate GERD symptoms) were also determined, as was the difference in costs required to manage patients between baseline and follow-up.
The primary outcome of successful step-down was measured by descriptive statistics. Changes in quality of life as measured by SF-36 and disease-specific instruments were analyzed by paired t tests. Univariate methods and multivariable logistic regression examined variables associated with unsuccessful step-down (the requirement of PPIs to alleviate symptoms). Potential predictors included in the logistic regression model were selected based on current understanding of GERD. They included age, gender, dominant symptom, alcohol intake, cigarette use, concurrent medication, and comorbid condi-tions. Of the potential predictor variables listed above, pre-liminary correlation analyses were performed to screen for multicolinearity. Stepwise procedure was used to arrive at the final model for the derivation set. Step selections were based on the maximum likelihood method with an alpha of 0.10 to
enter and 0.10 to remove. Interactions were examined for terms with significant main effects. The improvement at each step was tested by the likelihood ratio2analysis. The
good-ness-of-fit for the final method was tested by the Hosmer– Lemeshow method. A variable was considered predictive of nonresponse if it was included in the final model and the 95% confidence interval for its odds ratio did not include one.
Costs were limited in this analysis to the cost of medications assessed by actual expenditures at a VA Medical Center, as well as the direct cost of operating the GERD clinic, including salary and benefits of the clinical pharmacists staffing the clinic.
Results
Three hundred seventy-six patients were identi-fied as having been prescribed PPIs for 8 weeks or longer from our medical center. One hundred fifty-five consec-utive patients were evaluated in the GERD clinic. Of these, 82 were excluded (Table 1); the largest group of patients excluded were those still symptomatic despite PPI use. Subsequent evaluation of these excluded sub-jects revealed that 6 of 82 (7%) had GERD documented either by erosive esophagitis on esophagogastroduode-noscopy, 24-hour ambulatory pH monitoring DeMeester score ⬎14.72, or alleviation of symptoms on higher doses of PPI. Additional exclusions were due to the presence of peptic strictures, extra-esophageal manifesta-tions of GERD, or anemia. Also of note were 9 patients who had already decreased PPI use to an “as-needed” basis and 4 patients in whom a diagnosis of Barrett’s esophagus had been established.
The remaining 73 patients underwent step-down management. Baseline characteristics are shown in Table 2: 95% were men, mean age was 62.1 years (median, 64 years), 85% were on lansoprazole (the VA pharmacy contract PPI), and mean duration of PPI use was 21.3 months. Two thirds had been treated with an H2RA or prokinetic agent before prescription of a PPI. One hun-dred percent of subjects had complete follow-up to 6 months and 71 of 73 (97%) followed for 1 year. Incom-plete follow-up was because of death (1) and inability to contact (1).
After the 1-year follow-up, 41 of 71 (58%) of those in whom step-down was implemented were asymptomatic on either non-PPI therapy or no therapy to treat GERD (Figure 1). Of the subjects who remained off PPIs, 24 of 41 (59%) required H2RAs, 5 of 41 (12%) were on prokinetic agents, 1 of 41 (2%) were on both, and 11 of 41 (27%) were asymptomatic without the use of acid suppressive or prokinetic agents. Figure 2 illustrates the
Figure 1. Proportion of patients asymptomatic off PPIs.
Figure 2. Distribution of subjects on therapy. pro⫹ H2, prokinetic ⫹ histamine2-receptor antagonist; none, no medication to treat GERD. Table 1. Excluded Patients
Symptomatic despite PPI therapy
Heartburn or acid regurgitation 7
Dyspepsia 22
Peptic stricture 17
Extraesophageal GERD 11
Anemia or occult GI bleeding 10
PPI use intermittent 9
Barrett’s esophagus 4
Previous gastric surgery 2
GI, gastrointestinal.
Table 2. Baseline Demographics
Characteristic
Mean (range) or percentage
Age 62.1 (29–85)
Follow-up (days) 370 (167–426)
Duration of PPI use (mo) 21.3 (4–45)
Body mass index 27.8 (18–42.9)
% Treated with H2RA or prokinetic before PPI 69.9% % With weekly consumption of alcohol 32.8%
proportion of subjects on each therapy at the conclusion of the study.
Of those in whom step-down was unsuccessful (PPIs required to alleviate GERD symptoms), the median time to reinstitution of PPIs was 14 days (range, 2–210). Of note, only 1 subject developed symptoms requiring PPIs after 6 months of step-down initiation; the remaining subjects presented with recurrent symptoms requiring PPIs within 4 months of follow-up.
Quality of life and disease severity did not signifi-cantly change between baseline and 6-month follow-up intervals. SF-36 physical component summary scores were 37.9 (SD⫾ 7.9) at baseline and 37.1 (SD ⫾ 8.6) at follow-up (P⫽ 0.42), whereas mental component sum-mary scores were 45.0 (SD⫾ 6.2) and 43.9 (SD ⫾ 6.8), respectively (P ⫽ 0.25). Disease severity scores likewise did not significantly change, from 21.0 (SD ⫾ 1.2) at baseline to 21.8 (SD⫾ 1.7) at follow-up (note: score of 21 denotes absence of symptoms).
Univariate analysis of baseline demographic data and clinical information revealed that a dominant symptom of heartburn was significantly associated with PPI re-quirements (Table 3). Stepwise logistic regression ex-panded upon this finding: younger age and heartburn predicted unsuccessful PPI step-down management. The odds ratio for unsuccessful step-down (requirement of PPIs to alleviate symptoms) was 0.65 (95% confidence interval, 0.43– 0.98) per decade of life, whereas heart-burn as opposed to acid regurgitation or dyspepsia was associated with an odds ratio of 6.5 (95% confidence interval, 1.4 –29.6). Addition of cigarette smoking to the logistic model improved the goodness of fit (Hosmer– Lemeshow); however, the variable itself was not signifi-cantly associated with the success of step-down. Other factors, including duration of PPI administration, trial of H2RAs, or promotility agents before PPI institution, alcohol, comorbid medical illness, body mass index, or use of concurrent medication that could exacerbate
GERD symptoms, were not significantly associated with response to step-down.
Initial medication cost for this cohort was $43,994.36 per annum. Total medication cost at the end of study for this cohort was $2,724.77 per annum. Estimated direct costs for operating the GERD clinic were $25,200. Total cost saving for the institution per annum for this cohort was $16,069.59. Assuming that a similar distribution of eligibility and outcomes were realized for the entire population treated with PPIs at this center, expected cost savings would be $53,122.05 per annum.
Discussion
This study of patients with reflux symptoms treated with PPIs in a primary care setting revealed that more than half could be maintained in remission without PPI therapy. Twenty-seven percent required no medica-tion to treat their reflux symptoms, and the remaining 73% in whom step-down was successful required less expensive forms of anti-GERD therapy. Younger sub-jects required PPIs to alleviate recurrent symptoms more often than older subjects. A dominant symptom of heart-burn also predicted the need for PPIs and inability to successfully step-down. Neither quality of life nor disease severity was adversely impacted by this intervention as measured by validated instruments. Substantial eco-nomic benefit was realized in this cohort of patients.
Previous studies examining patients with endoscopi-cally documented erosive esophagitis have shown that the majority of patients require maintenance PPIs to prevent relapse.4,6Although erosive esophagitis is a firm
endpoint, it includes only a minority of the total popu-lation with gastroesophageal reflux disease.18 It was the
intent of this study to examine a cohort of patients with reflux symptoms managed in a general internal medicine outpatient setting. In this population, 24-hour pH mon-itoring and/or upper endoscopy are rarely used before empiric medical therapy. Our study was designed to base management on symptoms in a manner consistent with outpatient medical practice. Although every subject in the study experienced relief from symptoms of heartburn and/or acid regurgitation with PPI therapy before enroll-ment, the presence or absence of erosive esophagitis was not known. It is likely that our study population con-sisted of a heterogeneous population of subjects with GERD, and this may limit the application of the results of this trial to patients outside the scope of this study, such as those with known erosive disease. Additionally, it should be noted that this study was designed to examine the outcome of GERD patients undergoing step-down management. As such, no recommendations regarding
Table 3. Risk of Nonresponse to Step-down From PPIs: Univariate Analysis
Variable Odds ratio P value
Heartburn 2.7 0.025 Regurgitation 1.6 0.071 Dyspepsia 0.65 0.13 Alcohol use 1.2 0.46 Cigarette use 0.73 0.32 Concomitant medication 0.67 0.15 Comorbid disease 0.71 0.22
Age (per decade) 0.76 0.12
PPI duration 1.0 0.42
surveillance of symptomatic patients for Barrett’s esoph-agus or gastrointestinal malignancy can be made, and the decision to pursue upper endoscopy to evaluate GERD patients for the risk or presence of cancer must be made outside the context of this trial.
Age was a factor in predicting response to step-down therapy in the current study. We found advancing age to predict the success of step-down from PPIs. Fass et al.21
noted older patients’ perception of GERD symptoms to be less severe compared with younger patients. Their study noted that the frequency of heartburn and acid regurgitation were lower and less severe in subjects aged 60 years and older compared with those in younger subjects despite a higher frequency of erosive esophagitis in the older group. Additionally, time to symptom per-ception and sensory intensity scores were higher in the younger population. It is likely that we observed a similar effect in our study. Because the response to therapy was gauged by symptoms, older subjects less “sensitive” to acid exposure may have been tolerant of less potent acid suppression. The median age of subjects in our trial (64 years) may in part explain the success of our study. Due to the age-dependence of response, it may be found that attempts to implement step-down man-agement in cohorts consisting of younger patients may not yield as encouraging results.
Heartburn as opposed to acid regurgitation or dyspep-sia as a dominant symptom was also predictive, in this case, of nonresponse to step-down (requirement of PPIs to alleviate symptoms). The symptom of heartburn may signify more severe disease and thus the requirement for more potent acid suppression. Although resolution of GERD symptoms on PPI was an entry criterion for the study, it is also possible that in a subgroup of subjects, functional dyspepsia represented the underlying medical condition. Previous randomized trials have illustrated symptom resolution beyond placebo rates with PPIs in patients with documented functional dyspepsia.22–25It is
possible that subjects with a dominant symptom of dyspepsia may have less severe forms of GERD, or func-tional dyspepsia that may not require as potent acid suppression.
Concerning the use of promotility agents at the onset of this trial, cisapride was frequently used as a second-line drug for the treatment of GERD. However, after its removal from the market, prokinetics were not advocated because available replacements such as metoclopramide possessed side effect profiles that were substandard com-pared with alternative agents for the treatment of GERD. By study conclusion, only H2RAs and antacids were considered alternatives to PPI therapy for use in
GERD management; however, subjects in whom meto-clopramide was effective and devoid of side effects were maintained on this medication.
A considerable number of patients excluded from this study were symptomatic despite PPI use. It was found on further investigation that the majority of these patients were prescribed PPIs for indications unlikely to be asso-ciated with acid production. Specifically, these patients were not found to have evidence of either peptic ulcer or GERD after having undergone upper endoscopy, 24-hour ambulatory pH monitoring, and a therapeutic trial of high-dose PPI therapy.26 Although not included as
“successes” in our final analysis, it further illustrated the usefulness of a disease-oriented pharmacy clinic to opti-mally manage GERD patients. The use of nonphysicians to implement clinical practice guidelines was unique as it applied to this setting of reflux patients. In this case, in contrast to hypertension, anticoagulation, or diabetes clinics, in which laboratory or clinical parameters such as blood pressure, prothrombin time, or hemoglobin A1C
are followed, the pharmacists involved in this project were able to evaluate symptoms to manage a cohort of GERD patients.
Cost savings were realized in this cohort of veterans. As the VA contract price for the formulary PPI was $1.25 per standard dose at the time of the study, it is likely that substantially greater savings could be gener-ated in environments where PPIs are discounted less heavily. Generic PPIs will soon be available, thus de-creasing the cost to most consumers; however, it is unlikely that even generic drugs will be priced lower than the VA contract amount, and a significant differ-ence in costs between PPIs and H2RAs will persist. The cost analysis was limited to the incremental expenditures incurred by the medical center for implementation of the step-down protocol, including the proportion of salary required to operate the pharmacy GERD clinic and expenditures for dispensed medication. Neither indirect costs for items such as office space, ancillary staff, and administrative support, nor direct costs of additional health care visits, investigations related to the care of symptomatic subjects, or sick leave were considered, which may limit the generalizability of our results to other health care settings.
In conclusion, our study revealed that more than half of patients rendered asymptomatic on PPI therapy for reflux symptoms can be successfully stepped-down to less expensive medication. The need for PPIs is increased in the younger population, as well as in those who present with a dominant symptom of heartburn. In the course of monitoring step-down therapy of gastroesophageal reflux
disease, it is likely that the majority of patients who require PPIs will declare themselves within the first month of step-down, and there are few patients who recur with symptoms requiring PPIs beyond 6 months of initial step-down. Quality of life and disease severity do not seem to be adversely affected by step-down manage-ment, and costs clearly can be decreased.
References
1. Jansen JB, Van Oene JC. Standard-dose lansoprazole is more effective than high-dose ranitidine in achieving endoscopic heal-ing and symptom relief in patients with moderately severe reflux oesophagitis. The Dutch Lansoprazole Study Group. Aliment Phar-macol Ther 1999;13:1611–1620.
2. Robinson M, Campbell DR, Sontag S, Sabesin SM. Treatment of erosive reflux esophagitis resistant to H2-receptor antagonist therapy. Lansoprazole, a new proton pump inhibitor. Dig Dis Sci 1995;40:590 –597.
3. Feldman M, Harford WV, Fisher RS, Sampliner RE, Murray SB, Greski-Rose PA, Jennings DE. Treatment of reflux esophagitis resistant to H2-receptor antagonists with lansoprazole, a new H⫹/K(⫹)-ATPase inhibitor: a controlled, double-blind study. Lanso-prazole Study Group. Am J Gastroenterol 1993;88:1212–1217. 4. Hallerback B, Unge P, Carling L, Edwin B, Glise H, Havu N,
Lyrenas E, Lundberg K. Omeprazole or ranitidine in long-term treatment of reflux esophagitis. The Scandinavian Clinics for United Research Group. Gastroenterology 1994;107:1305– 1311.
5. Dent J, Yeomans ND, Mackinnon M, Reed W, Narielvala FM, Hetzel DJ, Solcia E, Shearman DJ. Omeprazole v ranitidine for prevention of relapse in reflux oesophagitis. A controlled double blind trial of their efficacy and safety. Gut 1994;35:590 –598. 6. Vigneri S, Termini R, Leandro G, Badalamenti S, Pantalena M,
Savarino V, Di Mario F, Battaglia G, Mela GS, Pilotto A, Plebani M, Davi G. A comparison of five maintenance therapies for reflux esophagitis. N Engl J Med 1995;333:1106 –1110.
7. Richter JE. Long-term management of gastroesophageal reflux disease and its complications. Am J Gastroenterol 1997; 92(suppl 4):30S–34S.
8. Eggleston A, Wigerinck A, Huijghebaert S, Dubois D, Haycox A. Cost-effectiveness of treatment for gastro-oesophageal reflux dis-ease in clinical practice: a clinical database analysis. Gut 1998; 42:13–16.
9. Hixson L J, Kelley CL, Jones WN, Tuohy CD. Current trends in the pharmacotherapy for gastroesophageal reflux disease. Arch In-tern Med 1992;152:717–723.
10. Sridhar S, Huang J, O’Brien BJ, Hunt RH. Clinical economic review: cost-effectiveness of treatment alternatives for gastro-oesophageal reflux disease. Aliment Pharmacol Ther 1996;10: 865– 873.
11. Zagari M, Villa KF, Freston JW. Proton pump inhibitors versus H2-receptor antagonists for the treatment of erosive gastro-esophageal reflux disease: a cost-comparative study. Am J Manag Care 1995;1:247–255.
12. Harris RA, Kupperman M, Richter JE. Prevention of recurrences of erosive reflux esophagitis: a cost-effectiveness analysis of main-tenance proton pump inhibition. Am J Med 1997;102:78 – 88. 13. Fennerty MB, Castell DO, Fendrick AM. The diagnosis and
treat-ment of gastroesophageal reflux disease in a managed care environment. Arch Intern Med 1996;156:477– 484.
14. Hillman AL, Bloom BS, Fendrick AM, Schwartz JS. Cost and quality effects of alternative treatments for persistent gastro-esophageal reflux disease. Arch Intern Med 1992;152:1467– 1472.
15. Sonnenberg A, Inadomi JM, Becker LA. Economic analysis of step-wise treatment of gastro-oesophageal reflux disease. Ali-ment Pharmacol Ther 1999;13:1003–1013.
16. Robinson M, Lanza F, Avner D, Haber M. Effective maintenance treatment of reflux esophagitis with low-dose lansoprazole. Ann Intern Med 1996;124:859 – 867.
17. Koelz H, Birchler R, Bretholz A, Bron B, Capitaine Y, Delmore G, Fehr HF, Fumagalli I, Gehrig J, Gonvers JJ, Halter F, Nammer B, Kayasseh L, Kobler E, Miller G, Munst G, Pelloni S, Realini S, Schmid P, Voirol M, Blum AL. Healing and relapse of reflux esophagitis during treatment with ranitidine. Gastroenterology 1986;91:1198 –1205.
18. Sontag S. Gastroesophageal reflux disease. Aliment Pharmacol Ther 1993;7:293–312.
19. Talley NJ, Stanghellini V, Heading RC, Koch KL, Malagelada JR, Tytgat GN. Functional gastroduodenal disorders. Gut 1999; 45(suppl 2):II37–II42.
20. Williford WO, Krol WF, Spechler SJ, and the VA Cooperative Study Group on Gastroesophageal Reflux Disease. Development for and results of the use of a gastroesophageal reflux disease activity index as an outcome variable in a clinical trial. Control Clin Trials 1994;15:335–348.
21. Fass R, Pulliam G, Johnson C, Garewal HS, Sampliner RE. Symp-tom severity and oesophageal chemosensitivity to acid in older and young patients with gastro-oesophageal reflux. Age Ageing 2000;29:125–130.
22. Blum AL, Arnold R, Stolte M, Fischer M, Koelz HR. Short course acid suppressive treatment for patients with functional dyspep-sia: results depend onHelicobacter pylori status. Gut 2000;47: 473– 480.
23. Veldhuyzen van Zanten SJ, Cleary C, Talley NJ, Peterson TC, Nyren O, Bradley LA, Verlinden M, Tytgat GN. Drug treatment of functional dyspepsia: a systematic analysis of trial methodology with recommendations for design of future trials. Am J Gastroen-terol 1996;91:660 – 673.
24. Talley NJ, Meineche-Schmidt V, Pare P, Duckworth M, Raisanen P, Pap A, Kordecki H, Schmid V. Efficacy of omeprazole in func-tional dyspepsia: double-blind, randomized, placebo-controlled trials (the Bond and Opera studies). Aliment Pharmacol Ther 1998;12:1055–1065.
25. Farup PG, Hovde O, Torp R, Wetterhus S. Patients with functional dyspepsia responding to omeprazole have a characteristic gas-tro-oesophageal reflux pattern. Scand J Gastroenterol 1999;34: 575–579.
26. Fass R, Ofman JJ, Gralnek IM, Johnson C, Camargo E, Sampliner RE, Fennerty MB. Clinical and economic assessment of the ome-prazole test in patients with symptoms suggestive of gastro-esophageal reflux disease. Arch Intern Med 1999;159:2161– 2168.
Received March 2, 2001. Accepted July 12, 2001.
Address requests for reprints to: John M. Inadomi, M.D., VA Ann Arbor Medical Center (111-D), 2215 Fuller Road, Ann Arbor, Michigan 48105. e-mail: jinadomi@umich.edu; fax: (734) 761-7549.
Supported by a Junior Faculty Development Award from the Amer-ican College of Gastroenterology (to J.M.I.) and by a grant from the Centers for Disease Control (to A.S.).
