Advertisement

Cost-Utility of Laparoscopic Nissen Fundoplication versus Proton Pump Inhibitors for Chronic and Controlled Gastroesophageal Reflux Disease: A 3-Year Prospective Randomized Controlled Trial and Economic Evaluation

      Abstract

      Background

      Very few randomized controlled trials (RCTs) have compared laparoscopic Nissen fundoplication (LNF) to proton pump inhibitors (PPI) medical management for patients with chronic gastroesophageal reflux disease (GERD). Larger RCTs have been relatively short in duration, and have reported mixed results regarding symptom control and effect on quality of life (QOL). Economic evaluations have reported conflicting results.

      Objectives

      To determine the incremental cost-utility of LNF versus PPI for treating patients with chronic and controlled GERD over 3 years from the societal perspective.

      Methods

      Economic evaluation was conducted alongside a RCT that enrolled 104 patients from October 2000 to September 2004. Primary study outcome was GERD symptoms (secondary outcomes included QOL and cost-utility). Resource utilization and QOL data collected at regular follow-up intervals determined incremental cost/QALY gained. Stochastic uncertainty was assessed using bootstrapping and methodologic assumptions were assessed using sensitivity analysis.

      Results

      No statistically significant differences in GERD symptom scores, but LNF did result in fewer heartburn days and improved QOL. Costs were higher for LNF patients by $3205/patient over 3 years but QOL was also higher as measured by either QOL instrument. Based on total costs, incremental cost-utility of LNF was $29,404/QALY gained using the Health Utility Index 3. Cost-utility results were sensitive to the utility instrument used ($29,404/QALY for Health Utility Index 3, $31,117/QALY for the Short Form 6D, and $76,310/QALY for EuroQol 5D) and if current lower prices for PPIs were used in the analysis.

      Conclusions

      Results varied depending on resource use/costs included in the analysis, the QOL instrument used, and the cost of PPIs; however, LNF was generally found to be a cost-effective treatment for patients with symptomatic controlled GERD requiring long-term management.

      Keywords

      1. Introduction

      Gastroesophageal reflux disease (GERD) is a common disorder managed by both family physicians and specialists. Affecting up to 30% of the developed world [
      • Dent J.
      • El-Serag H.B.
      • Wallander M.A.
      • Johansson S.
      Epidemiology of gastro-oesophageal reflux disease: a systematic review.
      ,
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ], patients with GERD exhibit a full spectrum of symptoms varying from mild, which require little medical attention, to severe, which requires continuous medication therapy, diagnostic investigations, and perhaps corrective surgery. It is in the intermediate symptom group where GERD treatment is most prevalent and which imposes the largest impact on the health care system. In the United States in 2002, GERD was the most common reason for visits to physicians (more than 5 million visits) for gastrointestinal disorders [
      • Shaheen N.J.
      • Hansen R.A.
      • Morgan D.R.
      • et al.
      The burden of gastrointestinal and liver diseases, 2006.
      ], and as a result GERD was a costly disease costing more than $10 billion [
      • Sandler R.S.
      • Everhart J.E.
      • Donowitz M.
      • et al.
      The burden of selected digestive diseases in the United States.
      ]. In Canada, more than CDN$670 million is spent on proton pump inhibitor (PPI) and histamine-2 receptor antagonist (H2RA) therapy each year [
      • Armstrong D.
      • Marshall J.K.
      • Chiba N.
      • et al.
      Canadian Consensus Conference on the management of gastroesophageal reflux disease in adults - update 2004.
      ]. GERD can also have a large affect on health-related quality of life (QOL) by disrupting sleep, diet patterns, and workplace productivity. For example, studies have shown QOL is statistically significantly lower in GERD patients compared to non-GERD populations [
      • Eslick G.D.
      • Talley N.J.
      Gastroesophageal reflux disease (GERD): risk factors, and impact on quality of life-a population-based study.
      ,
      • Revicki D.A.
      • Wood M.
      • Maton P.N.
      • Sorensen S.
      The impact of gastroesophageal reflux disease on health-related quality of life.
      ].
      Goals of GERD treatment are to resolve symptoms, heal the esophagitis, prevent recurrences, and prevent complications such as stricture, Barrett's esophagus, and bleeding [
      • Tessier D.
      Medical, surgical, and endoscopic management of gastroesophageal reflux disease.
      ]. Patients with frequent and severe symptoms, particularly if they are associated with moderate or severe esophagitis, generally need long-term maintenance therapy. Treatment of patients with GERD includes medical and surgical therapies that have varying success rates. There are two general classes of medical therapy for acid suppression: H2RAs and PPIs. PPIs have been proven to be more effective in symptom resolution and healing of the esophagitis is quicker and more reliably achieved with PPIs across all grades of GERD severity [
      • Chiba N.
      • De Gara C.J.
      • Wilkinson J.M.
      • Hunt R.H.
      Speed of healing and symptom relief in grade II to IV gastroesophageal reflux disease: a meta-analysis.
      ]. In addition, almost all economic evaluations comparing H2RAs to PPIs for patients with symptomatic GERD have found that on-demand (i.e., taken as needed) PPIs are less costly and more effective than H2RAs for mild to severe GERD [
      • Ofman J.J.
      The economic and quality-of-life impact of symptomatic gastroesophageal reflux disease.
      ].
      Despite studies reporting varying effectiveness for PPIs (80–95%), it has also been demonstrated that up to 50% of patients continue to experience abnormal esophageal acid exposure on 24-hour pH testing [
      • Gerson L.B.
      • Boparai V.
      • Ullah N.
      • Triadafilopoulos G.
      Oesophageal and gastric pH profiles in patients with gastro-oesophageal reflux disease and Barrett's oesophagus treated with proton pump inhibitors.
      ] and up to 37% experience a relapse of symptoms within 5 years of treatment [
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • et al.
      Continued (5-year) follow-up of a randomized clinical study comparing antireflux surgery and omeprazole in gastroesophageal reflux disease.
      ]. An alternative to long-term medical therapy with PPIs is open or laparoscopic Nissen fundoplication (LNF) surgery. Although both surgical approaches are equally effective in controlling reflux, the open technique is associated with longer hospitalization and significantly higher rates of wound and respiratory complications [
      • Chrysos E.
      • Tsiaoussis J.
      • Athanasakis E.
      • et al.
      Laparoscopic vs open approach for Nissen fundoplication A comparative study.
      ]. In addition to shorter hospitalization, LNF can reduce hospital costs and can result in faster return to work and normal activities than open surgery [
      • Coster D.D.
      • Bower W.H.
      • Wilson V.T.
      • et al.
      Laparoscopic Nissen fundoplication--a curative, safe, and cost-effective procedure for complicated gastroesophageal reflux disease.
      ]. It has also been shown that LNF can result in significant improvements in QOL over open surgery in a number of domains [
      • Velanovich V.
      Comparison of symptomatic and quality of life outcomes of laparoscopic versus open antireflux surgery.
      ] and, therefore LNF has emerged as the gold standard for surgical treatment of GERD patients [
      • Gould J.C.
      • Melvin W.S.
      Computer-assisted robotic antireflux surgery.
      ,
      • Hartmann J.
      • Jacobi C.A.
      • Menenakos C.
      • et al.
      Surgical treatment of gastroesophageal reflux disease and upside-down stomach using the Da Vinci robotic system A prospective study.
      ].
      Whereas medical therapies provide symptom relief, the aim of surgery is to correct the underlying physiological defect [
      • Swanstrom L.L.
      Motion-laparoscopic Nissen fundoplication is more cost effective than oral PPI administration: arguments for the motion.
      ]. In a systematic review comparing medical therapy with LNF, the authors found that surgically treated patients experienced significantly less esophageal acid exposure; however, there were insufficient data to determine whether LNF was superior for maintaining healing of esophagitis or in reducing the incidence of Barrett's esophagus or esophageal adenocarcinoma [
      • Ip S.
      • Bonis P.
      • Tatsioni A.
      • et al.
      Comparative Effectiveness of Management Strategies for Gastroesophageal Reflux Disease Comparative Effectiveness Review No. 1.
      ]. Unfortunately, there have been few randomized controlled trials (RCTs) comparing LNF to PPIs for treating patients with chronic GERD. The UK REFLUX trial found that LNF resulted in improved reflux, QOL, EuroQol 5 Dimmensions (EQ5D), and well-being scores compared to PPIs at 1 year [
      • Grant A.M.
      • Wileman S.M.
      • Ramsay C.R.
      • et al.
      Minimal access surgery compared with medical management for chronic gastro-oesophageal reflux disease: UK collaborative randomised trial.
      ,
      • Mahon D.
      • Rhodes M.
      • Decadt B.
      • et al.
      Randomized clinical trial of laparoscopic Nissen fundoplication compared with proton-pump inhibitors for treatment of chronic gastro-oesophageal reflux.
      ]. The Nordic GORD Study Group found that LNF improved rates of symptomatic recurrence but did not find differences in QOL compared to PPIs at 3 [
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • et al.
      Long-term management of gastro-oesophageal reflux disease with omeprazole or open antireflux surgery: results of a prospective, randomized clinical trial The Nordic GORD Study Group.
      ], 5 [
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • et al.
      Continued (5-year) follow-up of a randomized clinical study comparing antireflux surgery and omeprazole in gastroesophageal reflux disease.
      ], or 12 years [
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • Hatlebakk J.G.
      • Wallin L.
      • Engstrom C.
      • et al.
      Comparison of outcomes twelve years after antireflux surgery or omeprazole maintenance therapy for reflux esophagitis.
      ] of patient follow-up. And finally, interim analysis of the LOTUS trial found improved QOL with LNF but no difference in recurrence rates compared to PPIs at 3 years of follow-up [
      • Lundell L.
      • Attwood S.
      • Ell C.
      • et al.
      Comparing laparoscopic antireflux surgery with esomeprazole in the management of patients with chronic gastro-oesophageal reflux disease: a 3-year interim analysis of the LOTUS trial.
      ]. Clearly the RCT evidence to date has reported mixed results with respect to symptom control, rate of recurrence, and affect on patient QOL.
      Similarly, economic evaluation studies have reported mixed results comparing the two types of therapy. There has been one trial-based economic evaluation based on the REFLUX trial, which reported LNF to be very cost-effective at 1 year [
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ]. Modeling study results have varied from PPIs dominating LNF [
      • Arguedas M.R.
      • Heudebert G.R.
      • Klapow J.C.
      • et al.
      Re-examination of the cost-effectiveness of surgical versus medical therapy in patients with gastroesophageal reflux disease: the value of long-term data collection.
      ], to PPIs being more effective but more costly than LNF [
      • Romagnuolo J.
      • Meier M.A.
      • Sadowski D.C.
      Medical or surgical therapy for erosive reflux esophagitis: cost-utility analysis using a Markov model.
      ], to LNF being more effective and more costly but not considered to be cost-effective [
      • Comay D.
      • Adam V.
      • da Silveira E.B.
      • et al.
      The Stretta procedure versus proton pump inhibitors and laparoscopic Nissen fundoplication in the management of gastroesophageal reflux disease: a cost-effectiveness analysis.
      ,
      • Heudebert G.R.
      • Marks R.
      • Wilcox C.M.
      • Centor R.M.
      Choice of long-term strategy for the management of patients with severe esophagitis: a cost-utility analysis.
      ], to LNF being more effective and more costly and considered to be cost-effective [
      • Bojke L.
      • Hornby E.
      • Sculpher M.
      A comparison of the cost-effectiveness of pharmacotherapy or surgery (laparoscopic fundoplication) in the treatment of GORD.
      ,
      • Epstein D.
      • Bojke L.
      • Sculpher M.J.
      Laparoscopic fundoplication compared with medical management for gastro-oesophageal reflux disease: cost-effectiveness study.
      ].
      Given the paucity of RCT data comparing LNF to PPIs and in light of the mixed clinical results from the existing trial data on remission rates and affect on QOL, additional comparative trial data is needed to help determine the clinical effectiveness and affect on patient QOL of LNF compared to PPI management. In addition, because most of the existing RCTs are relatively short in duration (i.e., 1 year), the longer-term effectiveness of LNF has not been adequately explored with existing trials. For example, one of the proposed benefits of LNF surgery is the reduction in future drug use, doctor visits, and investigative testing. However, 1 year of patient follow-up, like in the two UK RCTs [
      • Grant A.M.
      • Wileman S.M.
      • Ramsay C.R.
      • et al.
      Minimal access surgery compared with medical management for chronic gastro-oesophageal reflux disease: UK collaborative randomised trial.
      ,
      • Mahon D.
      • Rhodes M.
      • Decadt B.
      • et al.
      Randomized clinical trial of laparoscopic Nissen fundoplication compared with proton-pump inhibitors for treatment of chronic gastro-oesophageal reflux.
      ], may be too short to adequately capture these benefits or determine if the observed short-term benefits from LNF surgery are maintained over longer periods of time. Another concern with some of the existing trial data is that GERD patients are recruited from specialist centers and a number of these patients are referred to specialists because their GERD symptoms are uncontrolled with PPI treatment. A trial that focuses specifically on chronic and controlled GERD patients would address the effectiveness of LNF in patients who are managed primarily by a family physician. Finally, given the wide variation in reported economic evaluation results, the paucity of trial-based economic evaluations that are based on actual patient data as opposed to modeling assumptions, and the absence of any trial-based economic evaluations in Canada, the purpose of this study was to determine the incremental cost-effectiveness (specifically cost-utility) of LNF compared to PPI for the treatment of patients with chronic and controlled GERD over a 3-year time period from a societal perspective based on a RCT in an experienced Canadian center.

      2. Methods

      2.1 Overview of the trial

      A detailed description of the study design and 1-year clinical outcomes have been previously reported [
      • Anvari M.
      • Allen C.
      • Marshall J.
      • et al.
      A randomized controlled trial of laparoscopic Nissen fundoplication versus proton pump inhibitors for treatment of patients with chronic gastroesophageal reflux disease: one-year follow-up.
      ]; however, a brief overview is provided here. Between October 2000 and September 2004, patients with GERD who were stable and symptomatically controlled on long-term medical therapy were recruited into a two-arm, parallel, non-blinded prospective RCT. Patients were recruited from a GERD clinic run by one of the principal investigators (MA) in a tertiary level teaching hospital in Hamilton, ON, Canada. Patients were referred to the clinic for diagnosis and management by family physicians, gastroenterologists, general surgeons, pulmonologists, otolaryngologists, allergists, and internists. Patients recruited into the study had to be between age 18 and 70 years, were required to have had prior long-term (minimum 1 year) treatment with a PPI, were required to be controlled on medical therapy, and were required to have chronic symptoms of GERD necessitating long-term therapy anticipated to last at least another 2 years.

      2.2 Interventions

      Patients were randomized to receive either PPI therapy or LNF plus medication therapy as needed. Medically treated patients received the same PPI dose they received before the trial to control their symptoms. Patients' symptoms were assessed at baseline using a symptom visual analogue scale (VAS) and a disease-specific gastroesophageal reflux disease symptom scale (GERSS) [
      • Allen C.J.
      • Parameswaran K.
      • Belda J.
      • Anvari M.
      Reproducibility, validity, and responsiveness of a disease-specific symptom questionnaire for gastroesophageal reflux disease.
      ] and if a patient's symptoms were poorly controlled, medical therapy was adjusted until symptoms were controlled. For patients randomized to surgery, the LNF was performed by one of four laparoscopic surgeons, each with experience on more than 50 cases, and each using their own preferred method of wrap. Surgery patients also received medications as needed to control their GERD symptoms.

      2.3 Primary and secondary clinical outcomes

      The primary clinical outcomes were reflux symptoms measured using the GERSS [
      • Allen C.J.
      • Parameswaran K.
      • Belda J.
      • Anvari M.
      Reproducibility, validity, and responsiveness of a disease-specific symptom questionnaire for gastroesophageal reflux disease.
      ] and heartburn (HB)-free days based on the week before regular assessment visits. Secondary clinical outcome measures included esophageal function, gastroscopy, 24-hour pH, esophageal manometry, and patient symptom control using a VAS score.

      2.4 Economic evaluation—resource use and costs

      In addition to clinical outcomes, resource use and QOL data were collected on each patient every 3 months for 3 years. Costing was conducted from a societal perspective. Resource use collected included: emergency room visits, hospitalizations, family physician visits, specialist visits, clinic visits, other health care professional visits, tests, procedures, surgeries, and prescription and over-the-counter medications (by drug name and dosage as reported in the patient expense log). Productivity loss information was also collected. The base case analysis included all costs, but in a sensitivity analysis costs were restricted only to those that patients self-reported were attributable to GERD. For costs such as emergency room visits and hospitalizations, the attribution to GERD was determined by clinical opinion and review of medical records. Productivity losses were estimated separately as days lost from paid employment for the patient or caregiver, days lost from homemaking activities, and days lost from volunteer activities.
      Unit prices used in the analysis came from various sources. Professional fees were obtained from the Ontario Schedule of Benefits for physician services [
      Schedule of Benefits for Physicians Services under the Health Insurance Act.
      ]; other professional fees (e.g., acupuncturist, massage therapy) were obtained through a survey of local commercial providers. Clinic visit costs, laboratory and procedure costs, emergency room visits, and hospitalization costs were obtained from a hospital participating in the Ontario Case Costing Initiative [
      Ministry of Health and Long-Term Care
      Ontario Case Costing Initiative (OCCI).
      ] and the appropriate professional fee code came from the Ontario Schedule of Benefits [
      Schedule of Benefits for Physicians Services under the Health Insurance Act.
      ]. Over-the-counter medication costs were obtained from a survey of local retail pharmacies; prescription medication costs were obtained from the Ontario Drug Benefit Formulary Plan [
      Ministry of Health and Long-Term Care
      Ontario Drug Programs Formulary.
      ] along with a pharmacy mark-up allowance of 8% [
      Ministry of Health and Long-Term Care
      Public Information: Notice to Pharmacies regarding regulation changes made to the Ontario Drug Benefit Act (ODBA) and the Drug Interchange and Dispensing Fee Act (DIDFA) (Bill 102, Transparent Drug System for Patients Act.
      ] and average dispensing fees were obtained from a survey of retailers [
      Community Pharmacy in Canada Executive Summary: Trends and Insights 2007 Survey of Pharmacists.
      ]. Productivity losses were valued using average hourly wages from Statistics Canada [
      Statistics Canada average hourly wages of employees by selected characteristics and profession, unadjusted data, by province (monthly) (Ontario).
      ]. As there were a large number of drug/dosage combinations, tests and procedures, and hospitalizations reported in the patient case report forms, only selected unit costs used in the analysis are presented in Table 1. All costs are expressed in 2009 Canadian dollars and costs in years 2 and 3 were discounted at 5% in the base case analysis. As a result of recent changes in drug prices in Ontario in July 2010, an additional set of analyses was conducted using 2010 drug prices for the main analysis and for each analysis based on key methodologic assumptions.
      Table 1Selected unit costs by type of resource, 2009 Canadian dollars.
      ResourceUnit cost (2010 drug prices for sensitivity analysis)Source
      Specialist consult$132.50Ontario SoB
      Specialist follow-up visit$82.90Ontario SoB
      Family physician visit$56.10Ontario SoB
      Physiotherapy$24.40Ontario SoB
      Chiropractor visit$77.00Ontario Chiropractic Association
      Massage therapy$50.00Survey of local commercial rates
      Walk-in clinic visit$60.77OCCI
      Nutrition clinic$80.00Survey of local commercial rates (initial visit/assessment)
      Emergency room visit$246.00OCCI
      LNF surgery cost per day$2340.00OCCI
      LNF redo surgery cost per day$1279.00OCCI
      Bowel Obstruction cost per day$1489.00OCCI
      Chest pain care cost per day$995.00OCCI
      Chest Inflammation cost per day$843.00OCCI
      CBC$8.27OCCI
      Helicobacter pylori$15.51 + $21.70Ontario SoB, OCCI
      Upper GI series$79.96 + $98.53Ontario SoB, OCCI
      PPI – Omeprazole 20 mg$1.10 ($0.55)Ontario DBF
      PPI – Pantoprazole 40 mg$1.90 ($0.51)Ontario DBF
      PPI – Rabeprazole 20 mg$1.30 ($0.33)Ontario DBF
      PPI – Lansoprazole 15 mg$2.00 ($0.50)Ontario DBF
      Rantidine 150 mg$0.40 ($0.30)Ontario DBF
      Average hourly wage$21.87Statistics Canada
      SoB, schedule of benefits; OCCI, Ontario Case Costing Initiative; DBF, Drug Benefit Formulary; CBC, complete blood count; PPI, proton pump inhibitor; LNF, laparoscopic Nissen fundoplication.

      2.5 Economic evaluation—utility measurement and valuation

      Utility values were estimated based on QOL questionnaires administered at baseline and every 3 months. Although the Health Utility Index 3 (HUI3) was identified as the primary utility measure for the economic evaluation, for comparative purposes Short Form 6D (SF6D) and EQ5D scores were also collected on each patient over time for use in a sensitivity analysis. The HUI3 was used as the primary QOL instrument because of the popularity of the HUI3 in Canada and because this instrument is the most sensitive in terms of attributes and levels. The SF6D was administered every 3 months and the EQ5D annually. Average treatment group utility scores were estimated at each time point and treatment group utility profiles were estimated over time based on linear interpolation between assessment intervals. Three-year quality-adjusted life years (QALY) estimates were calculated as the area under the utility curve for each patient and treatment group. Despite randomization, average utility scores at baseline were not equivalent between treatment groups, and differences in QALYs across the 3-year period were therefore estimated as the difference in mean QALYs between the two treatment groups adjusted for baseline differences [
      • Manca A.
      • Hawkins N.
      • Sculpher M.J.
      Estimating mean QALYs in trial-based cost-effectiveness analysis: the importance of controlling baseline utility.
      ].
      Dominance (i.e., whether one treatment was superior on both costs and effects) was first assessed and in the case of a trade-off between higher costs and improved QOL with LNF, an incremental cost-utility ratio (ICUR) was calculated as the ratio of the incremental cost of LNF divided by incremental QALYs (i.e., incremental cost per QALY gained). In sensitivity analyses, the ICUR calculation was repeated using the SF6D and EQ5D utility scores. If dominance occurred, average costs and effectiveness were reported and an ICUR was not calculated. As with costs, effectiveness in years 2 and 3 were discounted at 5% per year in the base case analysis.

      2.6 Data analysis for missing and censored data

      For patients with incomplete costs or QOL questionnaires during the 3-year follow-up period (i.e., those who withdrew, [for PPI n = 3, for LNF n = 2], were lost to follow-up [for PPI n = 3, for LNF n = 0], or missed intermittent assessments), any missing information was estimated using multiple imputation. Hot-deck multivariate normal switching regression multiple imputation [
      • Royston P.
      Multiple imputation of missing values.
      ] was implemented using all available data from all patients as potential covariates. Covariates included age, sex, smoking status, presence of cardiac history, design block, as well as health status (individual HUI3, SF6D, and EQ5D overall scores) and costs (individual GERD and non-GERD costs) from all available past and subsequent visits. On each occasion, 10 imputations were obtained using STATA SE version 9.2 [
      StataCorp
      Stata Statistical Software: Release 9.
      ] and the imputations were averaged for estimating the missing information.

      2.7 Stochastic uncertainty and sensitivity analyses

      Following imputations for any missing information, the estimated cost and outcome data for each patient were used for assessing stochastic uncertainty by estimating 1000 bootstrapped samples (i.e., sampling with replacement). For a visual representation of this patient-level uncertainty, these results were plotted on the cost-effectiveness plane. Cost-effectiveness acceptability curves (CEACs) were calculated showing the probability of LNF being more cost-effective than PPI as a function of society's willingness-to-pay (WTP) for a QALY gained. For assessing methodological uncertainty, deterministic sensitivity analyses were conducted. Assessed were alternative discount rate assumptions (5% vs. 0% and 10%), different cost perspectives (society vs. third-party payer), inclusion of different resource use items (all costs vs. self-assessed GERD-related only), and different instruments for assessing utility (HUI3 vs. SF6D and EQ5D). Two sets of sensitivity analyses are provided, one with 2009 prices for drugs and an additional analysis with 2010 prices for drugs.

      2.8 Statistical analyses

      Resource use data are presented as counts and percentages, whereas continuous cost and utility variables are presented as means ± SD. Differences in counts for resource use were compared using chi-squared tests. Because costs and QOL estimates tend to be skewed, these variables were fitted using gamma distributions and arithmetic means were compared across groups. CEACs are typically interpreted as the probability (based on stochastic uncertainty) that one treatment has a higher probability of being cost-effective at a given WTP threshold. As CEACs are based on expectations, no statistical analyses are conducted on these results.

      3. Results

      Of the 1666 patients screened for the study, 180 were eligible for randomization and 104 agreed to participate in the study. Baseline demographics of the study participants have been reported elsewhere and were comparable between the two treatment groups with 53% being men and with a mean age of 42 years [
      • Anvari M.
      • Allen C.
      • Marshall J.
      • et al.
      A randomized controlled trial of laparoscopic Nissen fundoplication versus proton pump inhibitors for treatment of patients with chronic gastroesophageal reflux disease: one-year follow-up.
      ]. Of the 104 patients randomized (52 to each group), two PPI patients and one LNF patient withdrew from the study immediately and 93 patients were available for the full 3-year follow-up.

      3.1 Clinical outcomes

      In terms of the primary clinical outcome of GERSS, both PPI and LNF patients demonstrated improved symptoms over time, with no statistically significant difference between groups at 3 years (mean difference 2.66; 95% confidence interval −1.11–6.43; P = 0.1660). HB-free days and patient symptom control using a VAS score were significantly better at 3 years for LNF treated patients (HB days difference −1.35; P = 0.0077; and VAS score difference -10.16; P = 0.0093).

      3.2 Resource use and costs for study participants

      Selected resource use information collected on the 104 patients during the 3-year follow-up period is presented in Table 2. Not only are total resources presented, but also shown are the patient's attributions of whether they thought these were related to their GERD condition (i.e., for use in a sensitivity analysis). Other than hospitalizations for LNF, there were no differences across groups in terms of other hospitalizations or emergency room visits. As shown in Table 2, the one patient who withdrew from the study did not receive LNF, and as such the percentage of patients receiving LNF is 98%. LNF treated patients had more total visits to specialists, more clinic visits, more hours of lost productivity, but fewer family physician visits and hours lost from normal activity. In terms of GERD-related resource use, family physician visits (51 vs. 12) and the use of other health care professionals were lower for LNF-treated patients (41 vs. 0 visits).
      Table 2Selected resource use over 3 years of follow-up (percentage), by treatment group
      Does not include multiple imputations.
      .
      Resource useTotalGERD-related only
      PPI (n = 52)LNF (n = 52)PPPI (n = 52)LNF (n = 52)P
      LNF or repeat LNF11 (15)58 (98)<0.00111 (15)58 (98)<0.001
      Other hospitalizations15 (14)12 (15)0.481
      Emergency room visits60 (63)74 (62)0.22723 (25)29 (21)0.405
      Family physician visits567 (90)441 (94)<0.00151 (40)12 (15)<0.001
      Specialist visits239 (88)317 (96)0.00197 (69)88 (75)0.508
      Clinic visits32 (38)51 (44)0.0373 (6)4 (6)0.705
      Other HC professionals955 (65)934 (54)0.62941 (77)0 (0)<0.001
      Productivity hours lost20.4 (69)37.7 (79)0.02320.4 (69)37.7 (79)0.023
      Normal activity hours lost41.4 (69)29.0 (75)0.13941.4 (69)29.0 (75)0.139
      Total
      Note that totals may not equal sums due to rounding.
      61.7 (88)66.7 (92)0.65361.7 (88)66.7 (92)0.653
      PPI, proton pump inhibitor; LNF, laparoscopic Nissen fundoplication; HC, health care; GERD, gastroesophageal reflux disease.
      low asterisk Does not include multiple imputations.
      Note that totals may not equal sums due to rounding.
      Mean costs by type of expense, year of follow-up, and treatment group are presented in Table 3. Costs by expense category and the first set of total cost estimates do not include multiple imputations for missing or incomplete resource use and cost information. The total cost estimates at the bottom on Table 3 include estimates following multiple imputations. As can be seen from Table 3, the biggest difference in 3-year costs were related to hospitalizations (i.e., higher for LNF-treated patients) and medications (i.e., lower for LNF-treated patients) such that overall costs were $3205 higher for patients receiving LNF. Considering estimates of GERD-related costs only, this difference still remained higher by about $3100 for LNF-treated patients. As can also been seen from Table 3, there was a strong time pattern in costs with LNF costs higher in year 1 where the surgery occurred and then decreasing over time to be about $1700 lower in year 3 for total resource use and costs and $1200 lower for GERD-related costs.
      Table 3Mean (median) costs per patient, by type of expense, year of follow-up and treatment group
      Mean and median costs by expense category and totals across categories do not include multiple imputations for missing or incomplete resource se and cost information.
      .
      Expense categoryLNF (n=52)PPI (n=52)3 year difference, LNF–PPI ($)
      Year 1 ($)Year 2 ($)Year 3 ($)Year 1 ($)Year 2 ($)Year 3 ($)
      Hospitalizations7710 (4680)1274 (0)507 (0)719 (0)1618 (0)1264 (0)5889
      Emergency room visits164 (0)138 (0)100 (0)90 (0)122 (0)98 (0)91
      Family doctor visits142 (70)150 (112)134 (56)199 (148)225 (105)187 (105)−184
      Specialist visits253 (176)216 (116)123 (0)162 (116)179 (87)175 (0)77
      Clinic visits48 (0)43 (0)14 (0)17 (0)12 (0)14 (0)62
      Other health care professionals169 (0)198 (0)191 (0)412 (0)161 (0)272 (8)−287
      Lab tests236 (101)92 (8)179 (12)126 (21)135 (8)262 (8)−17
      Medications979 (564)849 (369)714 (96)1765 (1136)1839 (1175)1813 (1423)−2876
      Productivity losses464 (153)224 (5)185 (0)198 (15)170 (0)170 (0)334
      Usual activity losses335 (153)235 (0)128 (0)372 (29)216 (0)173 (0)−63
      Total costs
      Mean and median costs by expense category and totals across categories do not include multiple imputations for missing or incomplete resource se and cost information.
      10499 (8371)3418 (1083)2276 (1089)4060 (1962)4678 (2212)4429 (2331)3027
      GERD-related total costs
      Mean and median costs by expense category and totals across categories do not include multiple imputations for missing or incomplete resource se and cost information.
      8938 (6630)655 (165)549 (144)1972 (1175)2717 (1268)1983 (1086)3470
      Imputed total costs
      These total costs and GERD-related total costs include multiple imputations.
      10326 (8811)3540 (1908)2083 (1308)4237 (2175)4730 (2274)3776 (2326)3205
      Imputed GERD-related total costs
      These total costs and GERD-related total costs include multiple imputations.
      8601 (7353)765 (407)572 (365)1979 (922)3117 (1195)1770 (1066)3073
      LNF, laparoscopic Nissen fundoplication; PPI, proton pump inhibitor; GERD, Gastroesophageal reflux disease.
      low asterisk Mean and median costs by expense category and totals across categories do not include multiple imputations for missing or incomplete resource se and cost information.
      These total costs and GERD-related total costs include multiple imputations.

      3.3 QoL and utility

      Based on patient QOL assessments during the 3-year follow-up period, utility scores by treatment group are presented in Figure 1 for the HUI3, SF6D, and EQ5D instruments. Although patients in the study were randomized, there were differences in baseline utility scores, especially for the HUI3 and EQ5D. As can be seen from Figure 1, patients' QOL improved over time in both treatment groups and this was observed across all three utility instruments. The changes in QALYs for each treatment group during 3 years based on the HUI3, and after adjusting for baseline differences, are presented in Table 4. LNF patients had an area under the curve improvement of 0.276 compared to PPI patients at 0.167 (mean difference 0.109; P = 0.159). In other words, during the 3-year period LNF patients experienced a 0.109 gain in QALYs compared to PPI patients.
      Figure thumbnail gr1
      Fig. 1Quality of life scores over time using the Health Utilities Index version 3 (HUI3), Short Form 6 Dimensions (SF6D), and EuroQOL 5 Dimensions (EQ5D) by treatment group (unadjusted for baseline imbalances).
      Table 4Mean costs, quality-adjusted life years (QALYs) and cost-effectiveness of laparoscopic Nissen fundoplication (LNF) compared to proton pump inhibitor (PPI) treatment.
      LNF (n = 52)PPI (n = 52)LNF-PPI
      Total costs$15,948 (11,281)$12,743 (12,488)$3,205 (16,828)
      Change in QALYs from baseline (based on HUI3)0.276 (0.517)0.167 (0.589)0.109 (0.784)
      ICUR$29,404/QALY gained
      Values are mean (SD).
      ICUR, incremental cost-utility ratio; HUI3, Health Utilities Index version 3.

      3.4 Cost utility and stochastic uncertainty

      Based on an incremental cost of $3205 and gain in QALYs of 0.109 for LNF-treated patients, the base case ICUR for LNF compared to PPI is $29,404 per QALY gained (Table 4). The results from the 1000 bootstrap estimates are plotted on the cost-effectiveness plane in Figure 2, which shows a concentration of uncertainty in the northwest quadrant indicating a cost and utility trade-off but with a fair amount of this stochastic uncertainty crossing into all four quadrants of the cost-effectiveness plane. Through the use of the net-benefit framework, the CEAC for LNF was calculated and is presented in Figure 3 for the HUI3 utility scores (small dashed line). The CEAC shows that for societal WTP values above $30,000 per QALY gained, LNF has the highest probability of being the most cost-effective treatment alternative.
      Figure thumbnail gr2
      Fig. 2Cost-effectiveness stochastic uncertainty results based on 1,000 bootstrap samples. Proton pump inhibitor at origin and each dot represents mean incremental cost and quality-adjusted life year (QALY) for lapasoscopic Nissen fundoplication for each simulation.
      Figure thumbnail gr3
      Fig. 3Cost-effectiveness acceptability curves. HUI3, Health Utilities Index version 3. EQ5D, EuroQOL 5 dimensions. SF6D, Short Form 6 Dimensions. WTP, willingness-to-pay; QALY, quality-adjusted life year.

      3.5 Methodologic uncertainty

      Presented in the first panel in Table 5 are the expected costs, QALYs, and cost-utility results of alternative methodologic assumptions using 2009 drug prices (i.e., base case year of analysis). Using SF6D utility scores compared to HUI3 scores had a minimal influence on QALYs and cost-utility ($29,404 per QALY gained vs. $31,117 per QALY gained). However, the estimated QALYs gained using the EQ5D scores were less than half that estimated for the HUI3 or SF6D, and therefore had a much larger influence on the cost-utility results ($76,310 per QALY gained). For comparative purposes the CEACs for all three utility measures are presented in Figure 3. The sensitivity analysis results in the first panel of Table 5 also show that using GERD-related costs only, or adopting a third-party payer perspective and using alternative discount rate assumptions (0% and 10%) did not have a substantial influence on the cost-utility results. The second panel of Table 5 presents sensitivity analysis results that reflect recent drug price changes for 2010. As can be seen from Table 5, the cost-utility results increase for the base case analysis (i.e., $43,422 per QALY gained) and all other sensitivity analysis variants. When using 2010 drug prices, most of the cost-utility results increased approximately $13,000 to $15,000 per QALY gained. The exception is the analysis based on EQ5D utility weights, which increased the cost-utility results to $112,690 per QALY gained.
      Table 5One-way sensitivity analysis around key methodologic assumptions.
      Modeling assumptionLNFPPIICUR
      CostsQALYsCostsQALYs
      Analysis based on 2009 drug prices
       Base case
      Base case uses Health Utilities Index version 3 utility weights, uses total costs from a societal perspective, and discounts costs/effects at 5%.
      $15,9480.276$12,7430.167$29,404/QALY
       SF6D utility weights$15,9480.159$12,7430.057$31,117/QALY
       EQ5D utility weights$15,9480.216$12,7430.174$76,310/QALY
       GERD-related costs only$9,9390.276$6,8600.167$28,193/QALY
       3rd party payer perspective$14,0700.276$11,5360.167$23,248/QALY
       0% discounting$16,3380.290$13,3660.178$26,536/QALY
       10% discounting$15,6030.263$12,1920.157$32,179/QALY
      Analysis based on 2010 drug prices
       Base case
      Base case uses Health Utilities Index version 3 utility weights, uses total costs from a societal perspective, and discounts costs/effects at 5%.
      $16,0080.276$11,2750.167$43,422/QALY
       SF6D utility weights$16,0080.159$11,2750.057$46,402/QALY
       EQ5D utility weights$16,0080.216$11,2750.174$112,690/QALY
       GERD-related costs only$9,7060.276$5,1380.167$41,908/QALY
       Third-party payer perspective$14,3070.276$9,9750.167$39,743/QALY
       0% discounting$16,4140.290$11,8290.178$40,938/QALY
       10% discounting$15,6470.263$10,7860.157$45,858/QALY
      LNF, laparoscopic Nissen fundoplication; PPI, proton pump inhibitor; ICUR, incremental cost-utility ratio; QALY, quality-adjusted life year; SF6D, Short Form 6 Dimensions; EQ5D, EuroQOL 5 Dimensions.
      Base case uses Health Utilities Index version 3 utility weights, uses total costs from a societal perspective, and discounts costs/effects at 5%.

      4. Discussion

      In light of the large expenditures devoted annually to medical management of patients with GERD and the large affect of symptoms from this prevalent condition on patient QOL, it is surprising there is such a paucity of RCT data comparing LNF and PPI for the management of patients with chronic GERD. The results from our Canadian study found no difference in symptoms at 3 years, which is consistent with the LOTUS trial [
      • Lundell L.
      • Attwood S.
      • Ell C.
      • et al.
      Comparing laparoscopic antireflux surgery with esomeprazole in the management of patients with chronic gastro-oesophageal reflux disease: a 3-year interim analysis of the LOTUS trial.
      ] but in contrast to the REFLUX [
      • Grant A.M.
      • Wileman S.M.
      • Ramsay C.R.
      • et al.
      Minimal access surgery compared with medical management for chronic gastro-oesophageal reflux disease: UK collaborative randomised trial.
      ,
      • Mahon D.
      • Rhodes M.
      • Decadt B.
      • et al.
      Randomized clinical trial of laparoscopic Nissen fundoplication compared with proton-pump inhibitors for treatment of chronic gastro-oesophageal reflux.
      ] and Nordic GORD trials [
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • et al.
      Continued (5-year) follow-up of a randomized clinical study comparing antireflux surgery and omeprazole in gastroesophageal reflux disease.
      ,
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • et al.
      Long-term management of gastro-oesophageal reflux disease with omeprazole or open antireflux surgery: results of a prospective, randomized clinical trial The Nordic GORD Study Group.
      ,
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • Hatlebakk J.G.
      • Wallin L.
      • Engstrom C.
      • et al.
      Comparison of outcomes twelve years after antireflux surgery or omeprazole maintenance therapy for reflux esophagitis.
      ] (the latter is a trial comparing open surgery and not laparoscopic surgery). Our results also found an increase in QOL for LNF-treated patients, which is consistent with REFLUX [
      • Grant A.M.
      • Wileman S.M.
      • Ramsay C.R.
      • et al.
      Minimal access surgery compared with medical management for chronic gastro-oesophageal reflux disease: UK collaborative randomised trial.
      ,
      • Mahon D.
      • Rhodes M.
      • Decadt B.
      • et al.
      Randomized clinical trial of laparoscopic Nissen fundoplication compared with proton-pump inhibitors for treatment of chronic gastro-oesophageal reflux.
      ] and LOTUS trials [
      • Lundell L.
      • Attwood S.
      • Ell C.
      • et al.
      Comparing laparoscopic antireflux surgery with esomeprazole in the management of patients with chronic gastro-oesophageal reflux disease: a 3-year interim analysis of the LOTUS trial.
      ], but is in contrast to the Nordic GORD trial [
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • et al.
      Continued (5-year) follow-up of a randomized clinical study comparing antireflux surgery and omeprazole in gastroesophageal reflux disease.
      ,
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • et al.
      Long-term management of gastro-oesophageal reflux disease with omeprazole or open antireflux surgery: results of a prospective, randomized clinical trial The Nordic GORD Study Group.
      ,
      • Lundell L.
      • Miettinen P.
      • Myrvold H.E.
      • Hatlebakk J.G.
      • Wallin L.
      • Engstrom C.
      • et al.
      Comparison of outcomes twelve years after antireflux surgery or omeprazole maintenance therapy for reflux esophagitis.
      ]. Because there is no plausible explanation for these diverse study findings across jurisdictions, it is clear that more prospective trial-based and long-term clinical studies are needed to build upon the evidentiary base of these two methods of managing gastroesophageal disease in patients with chronic GERD.
      It is similarly difficult to compare the results from this economic evaluation to other published health economic studies due largely to differences in relative prices, differences in study methods, and differences in treatment effect assumptions. Our results are broadly consistent with the REFLUX trial-based economic evaluation [
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ] and two of the modeling-based evaluations that found LNF to be cost-effective [
      • Bojke L.
      • Hornby E.
      • Sculpher M.
      A comparison of the cost-effectiveness of pharmacotherapy or surgery (laparoscopic fundoplication) in the treatment of GORD.
      ,
      • Epstein D.
      • Bojke L.
      • Sculpher M.J.
      Laparoscopic fundoplication compared with medical management for gastro-oesophageal reflux disease: cost-effectiveness study.
      ]. However, our results are in contrast to some of the other modeling-based evaluations that varied from PPIs being more effective and more costly than LNF [
      • Romagnuolo J.
      • Meier M.A.
      • Sadowski D.C.
      Medical or surgical therapy for erosive reflux esophagitis: cost-utility analysis using a Markov model.
      ], to PPIs dominating LNF [
      • Arguedas M.R.
      • Heudebert G.R.
      • Klapow J.C.
      • et al.
      Re-examination of the cost-effectiveness of surgical versus medical therapy in patients with gastroesophageal reflux disease: the value of long-term data collection.
      ], and to LNF being more effective and more costly but not cost-effective [
      • Comay D.
      • Adam V.
      • da Silveira E.B.
      • et al.
      The Stretta procedure versus proton pump inhibitors and laparoscopic Nissen fundoplication in the management of gastroesophageal reflux disease: a cost-effectiveness analysis.
      ,
      • Heudebert G.R.
      • Marks R.
      • Wilcox C.M.
      • Centor R.M.
      Choice of long-term strategy for the management of patients with severe esophagitis: a cost-utility analysis.
      ]. Although some of the differences in results can be explained by differences in relative prices across jurisdictions or differences in study methodology (e.g., time horizon), for other studies the differences are largely due to assumptions about the relative effectiveness of PPI versus LNF. For example, in two of the modeling-based economic evaluations, the authors assumed PPIs would increase QOL compared to LNF [
      • Arguedas M.R.
      • Heudebert G.R.
      • Klapow J.C.
      • et al.
      Re-examination of the cost-effectiveness of surgical versus medical therapy in patients with gastroesophageal reflux disease: the value of long-term data collection.
      ,
      • Romagnuolo J.
      • Meier M.A.
      • Sadowski D.C.
      Medical or surgical therapy for erosive reflux esophagitis: cost-utility analysis using a Markov model.
      ]. Because this assumption is contrary to the findings from our study and the other real world trial-based economic evaluation [
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ] based on actual observed patient follow-up and measurement, it is not surprising the economic evaluation results from these studies are quite different.
      Our QOL results are strikingly similar to the findings from the 1-year REFLUX trial-based economic evaluation [
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ]. Based on using the EQ5D, Grant et al. [
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ] found baseline utility values for their GERD patients to be 0.722, which is identical to our adjusted EQ5D baseline scores of 0.72. The REFLUX trial reported a difference in QALYs at year 1 of 0.088, which is very similar to our 1-year difference results and slightly smaller than our 3-year difference in QALYs of 0.109. This difference at 3 years is to be expected given the slight difference in the QOL curves that exists over years 2 and 3 before the QOL curves converge at the end of year 3.
      Most of the costing studies and economic evaluations that report longitudinal costs have noted that although LNF is generally more expensive up front, surgery becomes cost saving over time as annual costs for surgically managed patients decrease whereas costs for medical management remain relatively constant over time. The break-even point in costs has been found or estimated in some studies to occur around the second or third year of follow-up [
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ,
      • Romagnuolo J.
      • Meier M.A.
      • Sadowski D.C.
      Medical or surgical therapy for erosive reflux esophagitis: cost-utility analysis using a Markov model.
      ,
      • Nessen S.C.
      • Holcomb J.
      • Tonkinson B.
      • et al.
      Early laparoscopic Nissen fundoplication for recurrent reflux esophagitis: a cost-effective alternative to omeprazole.
      ]; however, this break-even cost has also been estimated to be 8 years [
      • Cookson R.
      • Flood C.
      • Koo B.
      • et al.
      Short-term cost effectiveness and long-term cost analysis comparing laparoscopic Nissen fundoplication with proton-pump inhibitor maintenance for gastro-oesophageal reflux disease.
      ] and 11 years [
      • Heudebert G.R.
      • Marks R.
      • Wilcox C.M.
      • Centor R.M.
      Choice of long-term strategy for the management of patients with severe esophagitis: a cost-utility analysis.
      ], or even as high as 27 years [
      • Arguedas M.R.
      • Heudebert G.R.
      • Klapow J.C.
      • et al.
      Re-examination of the cost-effectiveness of surgical versus medical therapy in patients with gastroesophageal reflux disease: the value of long-term data collection.
      ]. A simple extrapolation of the cost results from our study suggests that the break-even cost point for LNF over PPI might be between years 4 and 5 of study follow-up and this is broadly consistent with most other study findings. Given this catching up of cumulative costs of PPIs, it is not surprising that some economic evaluations have also reported a decrease in costs per QALY gained for LNF with longer time horizons. For example, based on the REFLUX trial, the cost-effectiveness of LNF was estimated at £23,284 per QALY gained at 1 year [
      • Grant A.
      • Wileman S.
      • Ramsay C.
      • et al.
      The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study The REFLUX trial.
      ] but £2,648 per QALY gained with a lifetime perspective [
      • Epstein D.
      • Bojke L.
      • Sculpher M.J.
      Laparoscopic fundoplication compared with medical management for gastro-oesophageal reflux disease: cost-effectiveness study.
      ]. Similarly in the United States, the cost-effectiveness of LNF was estimated at $1.65 million per QALY gained at 5 years but $300,000 per QALY gained at 10 years of follow-up [
      • Heudebert G.R.
      • Marks R.
      • Wilcox C.M.
      • Centor R.M.
      Choice of long-term strategy for the management of patients with severe esophagitis: a cost-utility analysis.
      ]. It is noteworthy that although these studies did find a similar pattern in lower incremental cost-effectiveness ratios with a longer time horizon, the results from these studies were qualitatively very different varying from LNF being cost-effective to not being cost-effective.
      There are a number of strengths and limitations of this study that are worthy of mention. One of the major strengths of this study is that it is a well-conducted prospective RCT in an experienced center based on actual patient measurement and outcomes as opposed to assumptions used in a model. Other strengths include a relatively long-term study follow-up period (i.e., 3 years), frequent resource use and QOL assessment (i.e., every 3 months), and QOL corroboration using three separate utility-based QOL instruments. One of the limitations of this study is the limited number of patients in the trial (n=52 in each arm) as patient recruitment into surgical trials is notoriously problematic. Another limitation of our study is the length of assessment for costs and outcomes because GERD is a chronic condition. Although 3 years is a relatively long follow-up period for a clinical trial, one of the advantages of a modeling-based study is the ability to estimate longer-term cost-effectiveness. Consistent with other longitudinal cost assessments, our results suggest that the total cost of LNF may be lower than PPI by about 4 to 5 years of treatment and follow-up and therefore LNF may dominate PPI at that time. However, only a longer-term modeling-based study based on these trial costs and outcomes would confirm these predictions.

      5. Conclusions

      We conducted a prospective concurrent RCT and economic evaluation in 104 patients in an experienced Canadian center to determine the cost utility of LNF compared to PPI for the treatment of patients with chronic GERD across a 3-year period. Although there was no difference in the primary clinical GERD symptom score (i.e., GERSS) over 3 years, LNF did result in fewer HB-free days and improved QOL. At an incremental cost of $3,205 per patient, LNF cost $29,404 per QALY gained compared to PPIs and would be considered cost-effective at WTP thresholds above $30,000 per QALY. LNF was generally found to be a cost-effective treatment for GERD patients with the cost-utility results and conclusion somewhat sensitive to the instrument used to measure QOL (i.e., EQ5D) and the unit prices for PPIs. LNF was generally found to be a cost-effective treatment for patients with symptomatic controlled GERD requiring long-term management.

      References

        • Dent J.
        • El-Serag H.B.
        • Wallander M.A.
        • Johansson S.
        Epidemiology of gastro-oesophageal reflux disease: a systematic review.
        Gut. 2005; 54: 710-717
        • Grant A.
        • Wileman S.
        • Ramsay C.
        • et al.
        The effectiveness and cost-effectiveness of minimal access surgery amongst people with gastro-oesophageal reflux disease - a UK collaborative study.
        Health Technol Assess. 2008; 12 (1–iv)
        • Shaheen N.J.
        • Hansen R.A.
        • Morgan D.R.
        • et al.
        The burden of gastrointestinal and liver diseases, 2006.
        Am J Gastroenterol. 2006; 101: 2128-2138
        • Sandler R.S.
        • Everhart J.E.
        • Donowitz M.
        • et al.
        The burden of selected digestive diseases in the United States.
        Gastroenterology. 2002; 122: 1500-1511
        • Armstrong D.
        • Marshall J.K.
        • Chiba N.
        • et al.
        Canadian Consensus Conference on the management of gastroesophageal reflux disease in adults - update 2004.
        Can J Gastroenterol. 2005; 19: 15-35
        • Eslick G.D.
        • Talley N.J.
        Gastroesophageal reflux disease (GERD): risk factors, and impact on quality of life-a population-based study.
        J Clin Gastroenterol. 2009; 43: 111-117
        • Revicki D.A.
        • Wood M.
        • Maton P.N.
        • Sorensen S.
        The impact of gastroesophageal reflux disease on health-related quality of life.
        Am J Med. 1998; 104: 252-258
        • Tessier D.
        Medical, surgical, and endoscopic management of gastroesophageal reflux disease.
        Perm J. 2009; 143: 30-36
        • Chiba N.
        • De Gara C.J.
        • Wilkinson J.M.
        • Hunt R.H.
        Speed of healing and symptom relief in grade II to IV gastroesophageal reflux disease: a meta-analysis.
        Gastroenterology. 1997; 112: 1798-1810
        • Ofman J.J.
        The economic and quality-of-life impact of symptomatic gastroesophageal reflux disease.
        Am J Gastroenterol. 2003; 98: S8-S14
        • Gerson L.B.
        • Boparai V.
        • Ullah N.
        • Triadafilopoulos G.
        Oesophageal and gastric pH profiles in patients with gastro-oesophageal reflux disease and Barrett's oesophagus treated with proton pump inhibitors.
        Aliment Pharmacol Ther. 2004; 20: 637-643
        • Lundell L.
        • Miettinen P.
        • Myrvold H.E.
        • et al.
        Continued (5-year) follow-up of a randomized clinical study comparing antireflux surgery and omeprazole in gastroesophageal reflux disease.
        J Am Coll Surg. 2001; 192: 172-179
        • Chrysos E.
        • Tsiaoussis J.
        • Athanasakis E.
        • et al.
        Laparoscopic vs open approach for Nissen fundoplication.
        Surg Endosc. 2002; 16: 1679-1684
        • Coster D.D.
        • Bower W.H.
        • Wilson V.T.
        • et al.
        Laparoscopic Nissen fundoplication--a curative, safe, and cost-effective procedure for complicated gastroesophageal reflux disease.
        Surg Laparosc Endosc. 1995; 5: 111-117
        • Velanovich V.
        Comparison of symptomatic and quality of life outcomes of laparoscopic versus open antireflux surgery.
        Surgery. 1999; 126: 782-788
        • Gould J.C.
        • Melvin W.S.
        Computer-assisted robotic antireflux surgery.
        Surg Laparosc Endosc Percutan Tech. 2002; 12: 26-29
        • Hartmann J.
        • Jacobi C.A.
        • Menenakos C.
        • et al.
        Surgical treatment of gastroesophageal reflux disease and upside-down stomach using the Da Vinci robotic system.
        J Gastrointest Surg. 2008; 12: 504-509
        • Swanstrom L.L.
        Motion-laparoscopic Nissen fundoplication is more cost effective than oral PPI administration: arguments for the motion.
        Can J Gastroenterol. 2002; 16: 621-623
        • Ip S.
        • Bonis P.
        • Tatsioni A.
        • et al.
        Comparative Effectiveness of Management Strategies for Gastroesophageal Reflux Disease.
        Agency for Healthcare Research and Quality, Rockville, MD2005
        • Grant A.M.
        • Wileman S.M.
        • Ramsay C.R.
        • et al.
        Minimal access surgery compared with medical management for chronic gastro-oesophageal reflux disease: UK collaborative randomised trial.
        BMJ. 2008; 337: a2664
        • Mahon D.
        • Rhodes M.
        • Decadt B.
        • et al.
        Randomized clinical trial of laparoscopic Nissen fundoplication compared with proton-pump inhibitors for treatment of chronic gastro-oesophageal reflux.
        Br J Surg. 2005; 92: 695-699
        • Lundell L.
        • Miettinen P.
        • Myrvold H.E.
        • et al.
        Long-term management of gastro-oesophageal reflux disease with omeprazole or open antireflux surgery: results of a prospective, randomized clinical trial.
        Eur J Gastroenterol Hepatol. 2000; 12: 879-887
        • Lundell L.
        • Miettinen P.
        • Myrvold H.E.
        • Hatlebakk J.G.
        • Wallin L.
        • Engstrom C.
        • et al.
        Comparison of outcomes twelve years after antireflux surgery or omeprazole maintenance therapy for reflux esophagitis.
        Clin Gastroenterol Hepatol. 2009; 7: 1292-1298
        • Lundell L.
        • Attwood S.
        • Ell C.
        • et al.
        Comparing laparoscopic antireflux surgery with esomeprazole in the management of patients with chronic gastro-oesophageal reflux disease: a 3-year interim analysis of the LOTUS trial.
        Gut. 2008; 57: 1207-1213
        • Arguedas M.R.
        • Heudebert G.R.
        • Klapow J.C.
        • et al.
        Re-examination of the cost-effectiveness of surgical versus medical therapy in patients with gastroesophageal reflux disease: the value of long-term data collection.
        Am J Gastroenterol. 2004; 99: 1023-1028
        • Romagnuolo J.
        • Meier M.A.
        • Sadowski D.C.
        Medical or surgical therapy for erosive reflux esophagitis: cost-utility analysis using a Markov model.
        Ann Surg. 2002; 236: 191-202
        • Comay D.
        • Adam V.
        • da Silveira E.B.
        • et al.
        The Stretta procedure versus proton pump inhibitors and laparoscopic Nissen fundoplication in the management of gastroesophageal reflux disease: a cost-effectiveness analysis.
        Can J Gastroenterol. 2008; 22: 552-558
        • Heudebert G.R.
        • Marks R.
        • Wilcox C.M.
        • Centor R.M.
        Choice of long-term strategy for the management of patients with severe esophagitis: a cost-utility analysis.
        Gastroenterology. 1997; 112: 1078-1086
        • Bojke L.
        • Hornby E.
        • Sculpher M.
        A comparison of the cost-effectiveness of pharmacotherapy or surgery (laparoscopic fundoplication) in the treatment of GORD.
        Pharmacoeconomics. 2007; 25: 829-841
        • Epstein D.
        • Bojke L.
        • Sculpher M.J.
        Laparoscopic fundoplication compared with medical management for gastro-oesophageal reflux disease: cost-effectiveness study.
        BMJ. 2009; 339: b2576
        • Anvari M.
        • Allen C.
        • Marshall J.
        • et al.
        A randomized controlled trial of laparoscopic Nissen fundoplication versus proton pump inhibitors for treatment of patients with chronic gastroesophageal reflux disease: one-year follow-up.
        Surg Innov. 2006; 13: 238-249
        • Allen C.J.
        • Parameswaran K.
        • Belda J.
        • Anvari M.
        Reproducibility, validity, and responsiveness of a disease-specific symptom questionnaire for gastroesophageal reflux disease.
        Dis Esophagus. 2000; 13: 265-270
      1. Schedule of Benefits for Physicians Services under the Health Insurance Act.
        ([Accessed March 24, 2010 2010])
        • Ministry of Health and Long-Term Care
        Ontario Case Costing Initiative (OCCI).
        ([Accessed March 24, 2010])
        • Ministry of Health and Long-Term Care
        Ontario Drug Programs Formulary.
        ([Accessed March 24, 2010])
        • Ministry of Health and Long-Term Care
        Public Information: Notice to Pharmacies regarding regulation changes made to the Ontario Drug Benefit Act (ODBA) and the Drug Interchange and Dispensing Fee Act (DIDFA) (Bill 102, Transparent Drug System for Patients Act.
        ([Accessed March 24, 2010])
      2. Community Pharmacy in Canada Executive Summary: Trends and Insights 2007 Survey of Pharmacists.
        ([Accessed March 24, 2010])
      3. Statistics Canada average hourly wages of employees by selected characteristics and profession, unadjusted data, by province (monthly) (Ontario).
        ([Accessed March 24, 2010])
        • Manca A.
        • Hawkins N.
        • Sculpher M.J.
        Estimating mean QALYs in trial-based cost-effectiveness analysis: the importance of controlling baseline utility.
        Health Econ. 2005; 14: 487-496
        • Royston P.
        Multiple imputation of missing values.
        Stata J. 2004; 4: 227-241
        • StataCorp
        Stata Statistical Software: Release 9.
        StataCorp LP, College Station, TX2005
        • Nessen S.C.
        • Holcomb J.
        • Tonkinson B.
        • et al.
        Early laparoscopic Nissen fundoplication for recurrent reflux esophagitis: a cost-effective alternative to omeprazole.
        JSLS. 1999; 3: 103-106
        • Cookson R.
        • Flood C.
        • Koo B.
        • et al.
        Short-term cost effectiveness and long-term cost analysis comparing laparoscopic Nissen fundoplication with proton-pump inhibitor maintenance for gastro-oesophageal reflux disease.
        Br J Surg. 2005; 92: 700-706