Cost-Effectiveness Analysis of Antihypertensive Combination Therapy among Patients with Moderate to Severe Hypertension

VA L U E I N H E A LT H 1 9 ( 2 0 1 6 ) A 1 - A 3 1 8

of the same age, race, and gender was created. The index date for the control cohort was randomly assigned during the study period to minimize selection bias. Patients in both cohorts were required to have continuous health plan enrollment for 1-year pre- and post-index date. Propensity score matching (PSM) on baseline demographic characteristics and the Charlson Comorbidity Index (CCI) score were used to compare health care costs and utilizations during the follow-up period. Results: Before 1:1 PSM, CAD patients (n= 6,558) had higher CCI scores (2.91 vs. 1.29) than control patients and were more likely to have comorbidities, including myocardial infarction (7.87% vs. 0.41%), congestive heart failure (17.34% vs. 2.23%), diabetes (39.01% vs. 17.67%), hypertension (59.80% vs. 30.36%), and depression (17.44% vs. 10.75%, all p< 0.0001). After 1:1 PSM on baseline demographic characteristics and CCI score, 1,928 patients, with balanced baseline characteristics, were matched from each cohort. Patients in the CAD cohort had higher proportions of inpatient stays (48.39% vs. 17.12%), and outpatient (98.91% vs. 74.74%), emergency room (ER) (53.79% vs. 28.27%), ambulatory (97.30% vs. 72.72%), and pharmacy visits (95.54% vs. 73.65%, all p< 0.0001). Higher health care utilization translated to higher costs for CAD patients, including inpatient ($7,932 vs. $1,791), outpatient ($4,940 vs. $2,888), ER ($453 vs. $162), ambulatory ($4,488 vs. $2,725), pharmacy ($2,867 vs. $1,898), and total costs ($15,739 vs. $6,577, all p< 0.0001).  Conclusions: Patients with CAD had significantly higher health care utilization and economic burden compared to those without CAD. PCV44 Operating Room Variable Cost of Transcatheter Aortic Valve Replacement Clancy S1, Womeldorff J2, Ryan MP2 1Edwards Lifesciences, Irvine, CA, USA, 2CTI Clinical Trial and Consulting Services, Cincinnati, OH, USA

Objectives: Heart valve disease is very common, with approximately 5 million people diagnosed annually in the United States. Transcatheter aortic valve replacement (TAVR) has become the preferred therapy in high-risk and inoperable patients with aortic stenosis. However, there is little information on the operating room (OR) costs for this procedure in a “real-world” setting from the hospital perspective. The objective of this study was to estimate the variable cost per OR minute in endovascular TAVR procedures.  Methods: The Premier database was queried from 2011 to 2014 for patients undergoing primary endovascular TAVR procedures. Patients were identified using the International Classification of Diseases, 9th Revision (ICD-9) procedure codes. OR cost per minute was calculated as the total OR hospital cost divided by the number of minutes of OR time. Only procedures that included enough detail in the billing file to quantify OR time were included. Distributions of OR cost per minute were trimmed by 0.5% at the upper and lower extremes to remove recording anomalies.  Results: There were 1,447 endovascular TAVR procedures that met the inclusion criteria. The mean age for the procedure cohort was 81.6 (SD 8.51) years. 77% of the procedures were performed in teaching hospitals, 95% in an urban setting. Approximately 40% were performed in the South region, and another 40% in the Northeast. The mean OR cost per minute was $41.55 (SD $1.69).  Conclusions: Quantifying the variable cost of OR time from a multiinstitution database provides researchers with an important benchmark to use in economic evaluations of TAVR procedures. PCV45 Cost Effectiveness of Simvastatin Plus Ezetimibe for Cardiovascular Prevention in United States Population: Results of the Improve-It Trial Almalki ZS, Guo JJ University of Cincinnati, Cincinnati, OH, USA

Objectives: Simvastatin, 40 mg, plus ezetimibe, 10 mg, daily, reduced the risk of cardiovascular events in stable patients with a history of acute coronary syndrome (ACS) in the Study of Vytorin Efficacy International Trial (IMPROVE-IT). Before the widespread use of Co-therapy with simvastatin and ezetimibe in such patients, its cost effectiveness should be demonstrated. Therefore, we investigated the cost effectiveness of adding ezetimibe to simvastatin treatment for patients with ACS in the United States.  Methods: A Markov state-transition model was developed to estimate the lifetime incremental cost effectiveness (in dollars per quality-adjusted years of life) associated with simvastatin compare with simvastatin plus ezetimibe co-therapy in patients with ACS. The model takes into account the increased risk of cardiovascular events due to increased age. The analysis was performed using a societal perspective in 2016. Direct medical cost, treatment costs and the costs of cardiovascular events, were included in the analysis. The robustness of the model to various assumptions was tested in a sensitivity analysis. Costs ($US, 2016 values) and outcomes were discounted at 3% per annum.  Results: In the IMPROVE-IT scenario, simvastatin resulted in lower QALYs with a value of 0.806, while Simvastatin plus ezetimibe resulted in 0.813 QALYs. Total costs were $8,483 for simvastatin and $9,459 for Simvastatin plus ezetimibe. Therefore, combination resulted in a gain of 0.007 QALYS at an additional cost of $975. The ICERs for combination compared with simvastatin alone was $30,855 per QALY. Based on willingness to pay threshold of $50,000 per QALY, the ICER for the combined therapy is not considered to be cost effective. Sensitivity analyses found that the model was sensitive to cost of ezetimibe.  Conclusions: Co-therapy with simvastatin plus ezetimibe prevented cardiovascular risk in IMPROVE-IT; however, it appears to be not a cost-effective treatment compared with the simvastatin monotherapy in patients with histories of ACS in U. S. PCV46 Cost Consequence Analysis of Two Different Active Flowable Hemostatic Matrices in Cardiac Surgery Patients (Us Hospital Provider Perspective) Makhija D1, Rock M1, Xiong Y2, Ikeme S1, Rubinstein M3, Arnold MR3

1Baxter 3Baxter

A47

Healthcare Corporation, Deerfield, IL, USA, 2Stratevi Consulting, Santa Monica, CA, USA, Healthcare Corporation, Westlake Village, CA, USA

Objectives: The model objective is to estimate and compare the cost consequences to hospitals, using two different active flowable hemostatic matrices (i.e. Floseal and Surgiflo) in cardiac surgery.  Methods: Using clinical outcomes from a published retrospective (US Premier Hospital Database) comparative effectiveness analysis of two hemostatic matrices in cardiac surgery patients[1], a cost-consequence framework was utilized to model the economic impact to hospitals of their use. The average annual number of cardiac surgeries and costing of major and minor complications, blood transfusions, and surgical revisions was analyzed using the 2012 Healthcare Cost and Utilization Project’s (HCUP’s) National Inpatient Sample (NIS) database (represents 95% of all U.S. community hospital discharges). Cost of hemostatic matrices and operating room (OR) was obtained from RED BOOK™ and published literature. All costs were adjusted to 2015 dollars using medical care consumer price index (CPI-medical care). One-way (OWSA) and probabilistic sensitivity analyses were performed by varying all variables within the 95% confidence interval (CI) of the point estimate or ±20% when CI was not available. [1] Tackett et al., 2014, Real-World Outcomes of Hemostatic Matrices in Cardiac Surgery, Journal of Cardiothoracic and Vascular Anesthesia doi:10.1053/j.jvca.2014.05.010.  Results: Results suggest that by using Floseal (rather than Surgiflo), a hospital performing 245 cardiac surgeries annually could avoid 11 major complications, 31 minor complications, 9 surgical revisions, 79 blood transfusions, and 260 hrs. of OR time. Avoiding these outcomes corresponds to net annualized cost savings of $1.43 million, with complication avoidance as the largest contributor to this cost saving. The risk of surgical revisions, minor and major complications had the largest impact on savings in the OWSA. Monte-Carlo simulations showed that annual savings were greater than $1.43 million, in 56% of iterations.  Conclusions: This analysis supports that using Floseal instead of Surgiflo in cardiac surgery could potentially lead to sizable cost savings to hospitals. PCV47 Cost Effectiveness of Apixaban (Novel Oral Anti-Coagulant) Compared with Conventional Therapy for Stroke Prophylaxis among Renal Impaired Patients with Atrial-Fibrillation from Perspective of Us Third Party Payer Kumar A1, Wu W2, Nili M1 of Pharmacy and Health Sciences, St. John’s University, New York, NY, USA, 2College of Pharmacy and Health Sciences, St Johns University, New York, NY, USA

1College

Objectives: Atrial Fibrillation (AF) has been a major risk factor for the development of stroke making anticoagulation therapy necessary. While on prophylaxis treatment, renal impaired patients on conventional anticoagulation (Warfarin) are at an increased risk for major bleeding. Apixaban has been shown to have better safety profile than Warfarin in face to face clinical trial. However, no study were found to have compared the cost effectiveness of Apixaban with Warfarin among renal impaired patients. We attempted to estimate the short term (one year) cost effectiveness of Apixaban compared to Warfarin for stroke prophylaxis among renal impaired patients.  Methods: We developed a decision analytic model to compare Apixaban 2.5mg/5.0 mg twice daily with dose adjusted Warfarin (INR: 2.0-3.0). Cost effectiveness was calculated separately among the patients with Severe to Moderate (25< CrCl≤ 50 ml/min) and Mild (50< CrCl≤ 80 ml/min) renal impairment. Only primary efficacy (event rate of stroke) and safety (event rate of major bleeding) were taken into consideration. Probabilities, effectiveness and safety were calculated using the clinical data from ARISTOTLE trial (NCT00412984). Inputs for cost were sourced from published US based studies.  Results: One year cost for Apixaban and warfarin was calculated as $3,670 and $2,075 respectively. Incremental cost to avert an additional event of stroke and major bleeding using Apixaban were $290,926 and $63,245 among patients with severe to moderate and $412,665 and $330,132 for mild renal function respectively as compared with warfarin. Among various sensitivity analyses, cost of Apixaban was found to have highest impact.  Conclusions: This decision analysis suggested that anticoagulation with Apixaban is a cost effective therapy as compared to dose adjusted warfarin for stroke prophylaxis among renal impaired patients. This study also indicated that severe to moderate renal impaired patients benefitted more with Apixaban compared to Warfarin. PCV48 Cost-Effectiveness Analysis of Antihypertensive Combination Therapy among Patients with Moderate to Severe Hypertension Wang X, Abughosh S University of Houston, Houston, TX, USA

Objectives: To compare the cost-effectiveness of single-pill triple antihypertensive combination therapy vs double-pill triple combination therapy (fixed-dose dual combination plus free drug class) vs three-pill triple combination therapy (free drug combination) by accounting for medication adherence to each treatment regimen in US.  Methods: A decision analytical model was constructed with one year time horizon from the patient’s perspective. Effectiveness was defined as the percentage (%) attainment of target systolic blood pressure (SBP) levels (< 140 mm Hg) based on adherence to each treatment strategy. Adherence was defined as having proportion of days covered (PDC)> = 80%. The probability of being adherent and achieving target SBP goal based on the adherence were obtained from the published literature. Direct medical cost (annual treatment cost for each strategy) was evaluated based on the adherence levels and adjusted to the year of 2015 by using consumer price index (CPI). The average cost effectiveness ratio (ACER) was calculated as the cost per % attainment of the target SBP level.  Results: The ACER for single-pill triple combination therapy was lowest ($1,566.82) followed by twopill ($1,856.32) and three-pill triple combination therapy ($2,210.42). The annual treatment cost was $964.85, $1133.72 and $1344.12 for single-pill, two-pill and three-pill combination therapy, respectively. The probability of being adherent was 55.3%, 40.40% and 32.60% to single-pill, two-pill and three-pill regimen. The probability of achieving target SBP level was 63.1% for adherent patients and 59.7% for

A48

VA L U E I N H E A LT H 1 9 ( 2 0 1 6 ) A 1 - A 3 1 8

non-adherent patients.  Conclusions: Compared to two-pill and three-pill triple combination therapy, single-pill triple antihypertensive therapy was more cost effective by considering SBP as the effectiveness outcome in the short term. Further analysis accounting for indirect medical costs from the societal perspective and using other outcome measures like quality adjusted life years (QALYs) in the long term needs to be further investigated. KEYWORDS: decision analytical model, cost effectiveness, hypertension, combination therapy. PCV49 The Real-World Cost-Effectiveness of Coronary Artery Bypass Grafting Versus Stenting in High-Risk Patients: Propensity Score Analysis of a Single Centre Experience Ariyaratne T1, Ademi Z2, Huq M3, Duffy S4, Billah B1, Rosenfeldt F4, Parkinson B5, Yap C6, Yan B7, Smith J8, Brennan A1, Tran L1, Reid C1 1Monash University, Melbourne, Australia, 2University of Basel, Basel, Australia, 3University of Melbourne, Melbourne, Australia, 4Alfred Hospital, Melbourne, Australia, 5Macquarie University, Sydney, Australia, 6Geelong Hospital, Geelong, Australia, 7Chinese University of Hong Kong, Shatin, Hong Kong, 8Monash Medical Centre, Melbourne, Australia

Objectives: To-date, several studies have evaluated the cost-effectiveness of coronary artery bypass grafting surgery (CABG) surgery versus percutaneous coronary intervention (PCI) using data from randomised controlled trials. This study will investigate the real-world cost-effectiveness of CABG compared with PCI with stents in high-risk patients with multi-vessel coronary artery disease (MVCAD). An Australian public hospital payer perspective will be adopted.  Methods: Data for 3508 patients (CABG: N= 1,440; PCI:N= 2,068 ) admitted to a major metropolitan hospital was obtained from two clinical registries, the Melbourne Interventional Group (MIG) and the Australian & New Zealand Society of Cardiac & Thoracic Surgeons (ANZSCTS). Hospital readmissions and related patient-level costs were obtained for the period of June 2009 to December 2014, from the same institution. The maximum follow-up period was five years. Adjustments for inflation and discounting will be performed over this period. Propensity score matching through the Nearest Neighbour technique will be used to balance the characteristics between the treatment (CABG) and comparator (PCI) groups. The primary and secondary measures of effectiveness will be major adverse cerebrovascular and cardiac events (MACCE), and mortality, respectively. The incremental cost-effectiveness ratios (ICERs) per MACCE avoided, and life years gained will be evaluated. Propensity score bin bootstrapping (PSBB) will be performed to further validate the results.  Results: Although several limitations apply to this analysis, we expect results to be similar to existing literature, which favours CABG compared with PCI with stents in the medium to long term. We expect highly favourable ICERs for CABG in sub-groups of patients at highest risk of complications.  Conclusions: This study will reveal the costeffectiveness of CABG compared with PCI using real-world data and new propensity score techniques. PCV50 Reevaluating the Value of Ezetimibe in the United States For Patients with History of Cvd Based on the Improve-It Result Davies GM1, Baxter C2, Vyas A3 and Co., Inc., North Wales, PA, USA, 2Merck and Co., Inc., Hoddeston, UK, 3Merck and Co., Inc., Rahway, NJ, USA

years (QALYs) of apixaban (2.5mg BID) in comparison of enoxaparin (40mg/day). CCSS used only enoxaparin as VTEp. A cohort of 1000 hypothetical patients with initial age for THR and TKR respectively are: 65 years and 71 years average[4].The distribution of surgery is 51% THR and 49% TKR[4]. The patients were introduced and distributed each therapy to analyze the probability of suffering clinical events. The horizontal time is 90 days (short-term) and 5 years (long-term) before surgery. The clinical events evaluated are: total VTE events (PE, distal and proximal DVT) and total bleedings events (Intracranial Hemorrhage, non-major clinically relevant, Major and Minor bleeds) in the short-term; VTE event and post thrombotic syndrome (PTS) in long-term. Costs were obtained from databases of CCSS[5,6]. Sensitivity analysis was also performed and 5% discount rate.  Results: The events present for 1000 hypothetical patients (51%THR/49%TKR) short-term are: 68 vs 118 total VTE and 72 vs 83 total bleeds (apixaban vs enoxaparin); in large term are: 20.6 vs 35.8 total VTE and 18.8 vs 32.6 PTS (apixaban vs enoxaparin). Overall costs and QALY per patient are: US$542.53 and 4.9587 QALY for apixaban; US$740.82 and 4.9479 QALY for enoxaparin.  Conclusions: Apixaban is cost-saving therapy for VTE prevention compared to enoxaparin after THR or TKR in the social security of Costa Rica. PCV52 Clinical Events Avoided with Apixaban Compared to Edoxaban for the Initial Treatment and Prevention of Venous Thromboembolism Lanitis T1, Hamilton M2, Quon P3, Browne C1, Cohen A4 1Evidera, London, UK, 2BMS, Princeton, NJ, USA, 3Evidera, Bethesda, MD, USA, 4King’s College Hospital, London, UK

Objectives: Prior analyses beyond clinical trials have evaluated the projected lifetime benefit of apixaban treatment compared to LMWH/VKA, rivaroxaban, and LMWH/dabigatran for initial treatment and prevention of venous thromboembolism (VTE). However, similar evaluation of apixaban compared to edoxoban has not yet been performed. Therefore, the analysis reported here evaluated the potential clinical implications of initial treatment with apixaban compared to LMWH/edoxaban over patient lifetime.  Methods: A Markov model was developed to project the lifetime clinical impact of six-month treatment of patients with either apixaban or LMWH/edoxaban following a VTE event. Clinical event rates were derived from the AMPLIFY trial and network meta-analysis of LMWH/VKA, apixaban, LMWH/ dabigatran, rivaroxaban, and LMWH/edoxaban. Background mortality rates were based on US life tables and published literature. Outcomes evaluated were lifeyears gained and the number of recurrent VTE and bleeding events avoided in a 1,000-patient cohort.  Results: Over lifetime, initial treatment for six months with apixaban versus LMWH/edoxaban was projected to result in 16 patient life-years gained in a cohort of 1,000 patients. The gain in life-years was primarily driven by fewer major bleeds and therefore fewer bleed-related case-fatalities. Apixaban was projected to result in 10 fewer major bleeds, 24 fewer clinically relevant nonmajor bleeds and two more recurrent VTE events. The increase in recurrent VTE events was attributed to longer life-expectancy for apixaban treated patients, as a result of a favorable bleeding profile, therefore a longer exposure to risk of recurrent VTE.  Conclusions: Six months of apixaban for initial treatment and prevention of VTE appears to offer a more favorable risk-benefit profile as compared to LMWH/ edoxaban leading to a projected increase in life-expectancy.

1Merck

Objectives: The IMPROVE-IT study has shown that ezetimibe (EZ) added to statin therapy provides additional benefit with respect to CVD outcomes. The objective of this study is to assess the economic value of EZ in CVD patients in the US healthcare system accounting for the change in price due to patent expiry.  Methods: We developed a model to project the long term cost and benefits of EZ added to statin therapy in patients with CVD and LDL-C values > = 70 mg/dl. Risk reduction in CVD events were based upon the relationship between LDL changes and reduction in CV risk from CTT meta-analysis. Cost and utility values were taken from recent assessments of statins in the US and Non-CVD death rates were based on US mortality statistics. An cohort of statin patients was identified from the IMS Pharmetrics and EMR databases. We conducted an evaluation where the price of EZ was fixed at the wholesale acquisition cost (WAC) for the first year and the price of EZ was reduced by 90% after one year of therapy.  Results: We identified 548 patients in the IMS database between the ages of 35-75 with a history of CVD and LDL-C > = 70 mg/dl. Patients were of 58 years age, with baseline LDL-C of 94.6 mg/dl, 55.5% were male and 35.6% of had diabetes. The reduction in current WAC price of 90% after 1 year resulted in an additional $1,363 in cost and a gain of 0.17 in QALYs, for an additional $8,150/QALY gained. Reduction in event costs due to the addition of EZ offset 80% of the incremental total drug cost of statin plus EZ.  Conclusions: With the results of IMPROVE-IT and impending patent expiry, these results suggest that initiating therapy with EZ is a clinical and cost-effective option for CVD patients treated with statins. PCV51 Cost Effectiveness Analysis Apixaban in the Prevention of Venous Thromboembolism after Total Hip or Knee Replacement in Adults in Costa Rica in 2015 Peralta-Acon M1, Poveda-Fernandez J2 1Pfizer Central America and Caribbean, San José, Costa Rica, 2Caja Costarricense de Seguro Social, San Jose, Costa Rica

Background: Patients to require total hip replacement (THR) or total knee replacement (TKR) are at high risk of developing post-operative deep vein thrombosis (DVT) and pulmonary embolism (PE), known as venous thromboembolism (VTE) [1,2]. An anticoagulant prophylaxis is recommended, low molecular-weight heparin is mostly used[3].  Objectives: Determinate cost-effectiveness of apixaban in VTE prevention (VTEp) after THR or TKR in adults from perspective of Costa Rica’s Public Health System (CCSS).  Methods: A tree-decision-model (short-term) and markov-decision-model (long-term) were designed using clinical trials data (effectiveness and clinical events) to evaluate lifetime costs and quality-adjusted-life-

PCV53 Cost Effectiveness of Intensive Lowering of Ldl Cholesterol with Pcsk9 Inhibitors as an Adjunct to High-Intensity Statin Therapy for Prevention of Major Vascular Events Lee H1, Watkins J2, Danielson D2 of Washington School of Pharmacy, Seattle, WA, USA, 2Premera Blue Cross, Mountlake Terrace, WA, USA

1University

Objectives: The focus of this assessment is to elucidate the incremental clinical effectiveness and value of PCSK9 inhibitors as adjunctive therapy to high intensity statins, compared to statin monotherapy to determine their therapeutic role in cardiovascular risk management. Methods: A Markov model was used to evaluate the long-term costs and outcomes of adjunct PCSK9 inhibitor therapy to background statin therapy compared to high intensity statin monotherapy in multiple risk groups. Model parameters such as treatment efficacy, LDL Cholesterol (LDL-C) achievement rates, cardiovascular risk, survival, utilities, and costs were derived from large clinical trials and meta-analyses, and other published sources. Cardiovascular event rates were adjusted based on absolute LDL-C reductions and applied to each respective risk category. Costs were valued in USD 2015, and costs and benefits were discounted at a rate of 3% per annum.  Results: Over a time horizon of 30 years, the costs of managing cardiovascular risk with PCSK9 inhibitors in addition to high-intensity statin therapy were substantially higher than high-intensity statin monotherapy across all risk groups. The incremental costs per QALY of adjunct PCSK9 inhibitor therapy were $515,568.05, $770,436.62, and $385,168.83 for Coronary Heart Disease (CHD) equivalent, medium cardiovascular risk, and clinical CHD cohorts, respectively. The results of this model were most sensitive to changes in PCSK9 inhibitor cost, statin LDL-C goal achievement rates, and relative risks for CHD Death and non-fatal myocardial infarction.  Conclusions: More intensive lowering of circulating LDL-C to levels below 70mg/dL does not produce sufficient incremental benefit to justify the current cost of PCSK9 inhibitors. Despite inferior target LDL-C achievement rates, high-intensity statin monotherapy was just as effective in preventing major vascular events across all risk groups at a substantially lower cost. However, the PCSK9 inhibitors may demonstrate value in high risk statin intolerant patients, but their current therapeutic role as an adjunct therapy limits this potential. PCV54 Development of a Pharmacoeconomic Model to Assess the Cost Effectiveness of Pharmacogenomics Tests in Chronic Heart Failure: The Case of Ivabradine Iliza AC1, Fanton Aita F1, Mitchell D1, Guertin JR2, Matteau A3, Lelorier J1