EU Pharmaceutical Expenditure Forecast

A254

VALUE IN HEALTH 16 (2013) A1-A298

have been taken so far to assess the importance associated by the health care decision-makers to incorporate ethical principles into practice. The review serves as an initial basis for a survey to study US decision-makers’ perspectives about the integration of evidence, economics and ethics in health care decision making. PHP59 MARKET ACCESS LESSONS FROM EXISTING PRODUCT DEVELOPMENT PARTNERSHIPS (PDPS) Ramadan F, Marinoni G, Ando G IHS, London, UK

OBJECTIVES: To give an overview of recent PDPs between the pharmaceutical industry, the public sector, International Health Organizations (IHOs) and academia and draw best practices from two case studies, the Medicines for Malaria Venture (adaptive PDP) and the Malaria Vaccine Initiative (innovative PDP). METHODS: We surveyed the ‘Health Partnerships Database’ (HPD) to list PDPs to date. We searched for partnerships involving product development of drugs, vaccines and other pharmaceutical and diagnostics products. For the case studies, we analysed websites of the specific PDPs. RESULTS: The HPD lists 19 PDPs, which can be classified into two categories: adaptive PDPs, i.e. research and development (R&D) initiatives focused on tailoring existing products to developing countries’ needs, and innovative PDPs, i.e. R&D initiatives focused on developing new products to address developing countries’ diseases. Existing PDPs focus on R&D of drugs (8 PDPs), vaccines (7 PDPs) and preventive and diagnostic techniques (4 PDPs) for diseases such as tuberculosis (TB, 5 PDPs), malaria (6 PDPs), meningitis (1 PDP) and HIV/AIDS (6 PDPs). Meanwhile, recently developed PDPs are tailored as an access-to-market strategy for essential and neglected disease treatments in developing countries (4 PDPs). In-depth analysis of two PDPs showed that 1) assessment of specific health care needs in developing countries prior to partnership initiation; 2) local government engagement in partnerships; 3) sustainability of funding flow accompanied by national expertise building; and 4) monitoring, reporting and transparent sharing of results are crucial for the success of PDPs. CONCLUSIONS: Multinational pharmaceutical companies are increasingly dedicating part of their development pipeline to developing-country diseases and turning to IHOs as strategic partners to secure market knowledge and access in these countries. However, most collaborations still happen around the “big three” diseases, Malaria, HIV/AIDS and TB. HEALTH CARE USE & POLICY STUDIES – Formulary Development PHP60 FORECASTING THE HEALTH AND HEALTH SPENDING CONSEQUENCES OF VALUE-BASED DRUG INSURANCE DESIGN FOR INSUREDS WITH FIVE HIGHCOST CONDITIONS Olchanski N1, Milstein AS2, Chambers JD1, Neumann PJ1 1 Tufts Medical Center, Boston, MA, USA, 2Stanford University School of Medicine, Stanford, CA, USA

OBJECTIVES: The goal of value-based insurance design (VBID) is to set encourage use of high-value therapies and discourage use of low-value ones by appealing to evidence of value, rather than a drug’s price, when tiering drugs and setting copayment levels. The aim of this study was to simulate the impact of an illustrative VBID approach on savings and health gains. METHODS: We used simulation to model an impact of a value-based formulary where drugs with less favorable cost-effectiveness (CE) had higher co-pays, and those with more favorable clinical effectiveness and CE were tiered with lower co-pays. Model inputs included drug utilization and cost data from the Medical Expenditure Panel Survey (MEPS) for 2008 and CE and health gain data from the published literature (Tufts CEA Registry). Estimates of behavioral effects of changing co-pay amounts on drug utilization were based on published literature. Modeled conditions included lipid disorders, hypertension, diabetes, esophageal disorders, and depression. In sensitivity analyses we ran the model using a large health plan’s 2010 drug cost and utilization data. RESULTS: In the esophageal and lipid-lowering scenarios, the total annual prescription costs decreased (39%), in the hypertension scenario costs remained roughly the same, and in depression and diabetes costs increased slightly (2-4%). Total costs of care decreased in all but one scenario by 1-2%, and remained steady in the hypertension scenario. Health benefits also showed a positive impact across all scenarios, increasing total quality-adjusted life years gained by 118,000 to 254,000, about 1%. The results were similar in scenarios using health plan data. CONCLUSIONS: The simulation illustrates the importance of evolving from the prevalent sub-optimal practice of tiering therapies based largely on drug costs to tiering based on overall value. PHP61 COST-SAVING MEASURES IN REIMBURSEMENT OF ORPHAN DRUGS UNDER MEDICARE PRESCRIPTION DRUG PLANS Sepulveda B, Doyle JJ Quintiles, Hawthorne, NY, USA

OBJECTIVES: The increase in premium-priced orphan drugs coupled with health care budget constraints will pressure managed care plans to consider restricting market access. Coverage and reimbursement of ten FDA-designated orphan drugs (alpha-glucosidase, bexarotene, bosentan, galsulfase, idursulfase, iloprost, laronidase, mecasermin, nitisinone, plerixafor) were analyzed for seven popular Medicare PDP (BCBS Rx, AARP Medicare Rx Preferred, Cigna Medicare Rx Plan One, Humana Enhanced, Aetna/CVS, EmblemHealth, UnitedHealthcare). METHODS: Formulary tier structure, monthly retail costs and utilization restrictions (UR)—pre-authorization (PA), quantity limits (QL) and step therapy

(ST)—were obtained from CMS (www.medicare.gov). UR were assigned point values reflecting most to least restrictive—QL, 1; ST, 2; PA, 3; exclusion from formulary, 6. Each drug-plan combination can be assigned a maximum of 6 points. Disease prevalences were obtained from a variety of sources. RESULTS: Monthly retail prices ranged from $782.45 (laronidase) to $11,845.25 (mecasermin). Iloprost was most frequently excluded from formularies (2) and was subject to PA from the remaining five plans. The most frequently PA drug was mecasermin (6). There was slight negative correlation between price and prevalence of the disease for which the drugs were indicated (r2=0.080). There was a slight positive correlation between price and the number of UR points (r2= 0.046). There was virtually no correlation between UR points and disease prevalence (r2=0.004). However, there is a moderate correlation between the number of plans that implement PA for a drug and that drug’s price (r2=0.322). CONCLUSIONS: With a wide range of drug prices, there is virtually no correlation between the number of UR points, drug price and disease prevalence. However, it is clear that PA is perceived to be the most effective cost-saving measure as it is used most for the more expensive drugs. HEALTH CARE USE & POLICY STUDIES – Health Care Costs & Management PHP62 IMPACT OF HEALTH, SOCIAL, LIFESTYLE, AND ECONOMIC FACTORS ON LIFE EXPECTANCY IN OECD COUNTRIES Augustine JM, Skrepnek GH The University of Arizona, Tucson, AZ, USA

OBJECTIVES: The purpose of the study was to assess the relationship of pharmaceutical expenditures (PE) and other social and economic factors on life expectancy for different age groups in Organisation for Economic Co-operation and Development (OECD) countries. METHODS: This retrospective study analyzed data from the OECD database on the association between life expectancy in 2010 and PE, social/lifestyle (tobacco, alcohol, fruits, vegetables, and sugar consumption), and economic (health care expenditures and gross domestic product (GDP)) factors from a sample of 27 developed countries. Data were pooled according to gender at 40, 60, and 65 years while controlling for the lag variables of alcohol and tobacco (lagged 20 years), food intake variables (7 years), and economic factors (12 years). All variables were transformed via logathrimic form to yield full elasticities. A bootstrapped random intercept regression model with sandwich estimators was conducted to determine which social and health factors were associated with a change in life expectancy. RESULTS: The mean life expectancy for males was 39.02±2.30 years for age 40, 21.44±1.65 years for age 60, and 17.58±1.39 years for age 65. The mean life expectancy for females was 43.22±2.09 years for age 40, 25.21±1.68 years for age 60, and 20.90±1.57 years for age 65. Results of the regression of life expectancy indicated significant associations (p≤0.05) of: male gender (coefficient=−0.146. CI: −0.157-0.135); age group of 60 years (−0.570, CI: −0.583-0.558); age group of 65 years (−0.763, CI: −0.777-0.750); calorie intake (−0.144, CI: −0.246-0.041); fruit and vegetable intake (0.06, CI:0.036-0.084); overall health care expenditures (−0.047, CI: −0.092-0.001); and GDP (0.252, CI:0.189-0.314). Pharmaceutical expenditures were not associated with change in life expectancy for any age group. CONCLUSIONS: Across 27 OCED countries, specific food intake, GDP, and health care expenditures were significantly associated with altering life expectancy. Increased PE was not significantly associated with life expectancy. PHP63 HOMELESS PATIENTS USE OF URBAN EMERGENCY DEPARTMENTS IN THE UNITED STATES Coe AB, Moczygemba LR, Harpe SE, Gatewood SBS Virginia Commonwealth University, Richmond, VA, USA

OBJECTIVES: This study compares homeless patients’ utilization of the urban emergency department (ED) in the United States (US) with non-homeless patients and examines the relationship between homelessness and frequency of ED use, arrival to the ED by ambulance, waiting time to be seen, length and type of ED visit. METHODS: The emergency department component of the 2009 National Hospital Ambulatory Care Survey database (NHAMCS-ED) was used for this cross-sectional, descriptive analysis. Patients were included if their patient residence was homeless or private residence and if they visited an urban ED. Descriptive statistics, bivariate analyses, and logistic regression were used to examine the relationship between demographics and ED use variables with homelessness. Nationally representative weights were applied to the estimates. RESULTS: The weighted sample size was 100,847,969 patient visits. A total of n=636,399 (0.63%) of the patients were homeless. The mean age of homeless patients was 44.0 years (95% CI: 41.7-46.3) compared to 34.7 years for nonhomeless patients (95% CI: 33.5-35.9, p<0.0001). The majority of homeless patients were male (75.9%) versus only 44.9% of non-homeless patients (p<0.0001). More homeless patients arrived to the ED via ambulance (45.2 vs. 14.9%, p<0.0001). Homeless patients had a significantly different length of ED visit and number of ED visits in the past 12 months (both p=0.007) than nonhomeless patients. Homeless patients were more likely to be older (OR=1.02, 95% CI: 1.003-1.040), male gender (OR=3.34, 95% CI: 1.191-9.349), arrive to the ED via ambulance (OR=7.56, 95% CI: 4.045-14.137), have a longer length of ED visit (OR=1.002, 95%CI: 1.001-1.003), and have a past visit to the ED in the last 12 months (OR=1.06, 95%CI: 1.033-1.080) than non-homeless patients. CONCLUSIONS: In 2009, homeless patients who visited US urban emergency departments were significantly different than non-homeless patients in their demographics, frequency, access, and pattern of ED use. PHP64 EU PHARMACEUTICAL EXPENDITURE FORECAST

VALUE IN HEALTH 16 (2013) A1-A298

Rémuzat C1, Toumi M2, Vataire AL3, Cetinsoy L3 1 Creativ-Ceutical France, Paris, France, 2University Claude Bernard Lyon 1, Lyon, France, 3 Creativ-Ceutical, PARIS, France

OBJECTIVES: With constant incentives for health care payers to contain their pharmaceutical budgets forecasting has become critically important. Some countries even developed pharmaceutical horizon scanning units. The objective of this project was to build a model to assess net effect of the entrance of new patented medicinal products versus medicinal products going off-patent, with a forecast horizon until 2016, on seven selected EU Member States’ pharmaceutical budgets: France, the UK, Germany, Poland, Portugal, Greece and Hungary. This model should take into account the ageing population, as well as current and future country-specific pricing, reimbursement and market access policies. (performed for the EU Commission -http://ec.europa.eu/health/healthcare /key_documents/index_en.htm) METHODS: A model was developed for generics and biosimilars for each country. This model estimated a separate and combined effect of the direct and indirect impact of patent cliff. A model was also developed for new entrants, which estimated the sales development and risk of development failure. New entrants were reviewed individually to assess their clinical potential and translate into commercial potential. Forecast was performed according to three perspectives (health care public payer, society and manufacturer), several types of distribution chain (retail, hospital, combined retail and hospital). Probabilistic and deterministic sensitivity analyses were carried out. RESULTS: All countries would experience drug budget reduction except Poland (€+41 million). Savings are expected to be the highest in the UK (€-9367 million), France (€-5589 million), and far behind, followed by Germany (€-831 million), Greece (€-808 million), Portugal (€243 million) and Hungary (€-84 million). Cardiovascular, central nervous system , respiratory areas and biosimilar entry will be the source of savings, while oncology, immunology and inflammation will lead for additional expenditure. The model is very sensitive to time to market branded products, generic prices, geneneric penetration, distribution of biosimilars. CONCLUSIONS: Pharmaceutical expenditures are expected to decrease in the analyzed period. The model is sensitive to policy decisions. PHP65 PHARMACEUTICAL SPENDING LEVELS ACROSS EUROPE – CONSEQUENCES OF CURRENT INTERNATIONAL PRICING POLICIES ON LOWER INCOME COUNTRIES Boehler CEH1, Merész G1, Kalo Z2 1 Syreon Research Institute, Budapest, Hungary, 2Eötvös Loránd University, Budapest, Hungary

OBJECTIVES: Current mechanisms of pharmaceutical price control, as for instance international price referencing, may result in a convergence of pharmaceutical prices across countries at different levels of economic attainment. Within the EU, this process is intensified through the single European Market, which fosters parallel trade if price discrepancies for pharmaceuticals are sufficiently large. A narrow pharmaceutical price corridor within the EU should lead to disproportionally high percentages of pharmaceutical spending on total health care spending in lower income EU countries. This paper explores whether such a discrepancy exists between European countries after controlling for other factors potentially influencing the level of pharmaceutical spending. METHODS: We analysed OECD and EUROSTAT data from 21 European countries between 2001 and 2010. A two-level hierarchical growth curve model was used to assess the relationship between pharmaceutical expenditure as percentage of total health expenditure and covariates both on measurement and country-level, including GDP, pharmaceutical and total health expenditure per capita, the cost of labour in EU member countries and life expectancy at birth. RESULTS: Initial results show an inverse relationship between GDP per capita and pharmaceutical expenditure as a percentage of total health expenditure. This relationship remains highly significant even after controlling for life expectancy at birth and differences in labour cost across EU member states. CONCLUSIONS: Pharmaceutical pricing policies within the EU put disproportionally high pressure on countries with more restricted health care budgets, which further limits access to health care - and thereby the potential to achieve health benefits - in lower income EU countries. Future analysis should use more disaggregated data and link pharmaceutical prices to health status and data on actual pharmaceutical consumption. Ultimately, new strategies of establishing differential pharmaceutical pricing across countries at different levels of economic attainment are needed to ensure universal and equitable access to innovative health care technologies. PHP66 PERSISTENCE IN OUT-OF-POCKET HEALTH CARE EXPENDITURES AMONG MEDICARE BENEFICIARIES Yoo M1, Biskupiak J2 1 University of Utah, Salt lake City, UT, USA, 2University of Utah, Salt Lake City, UT, USA

OBJECTIVES: This study examines the dynamics of out-of-pocket health care spending by looking at the persistence of such spending among Medicare beneficiaries. It provides insight into how long high health care out-of-pocket spending persists, its association with certain health conditions and health shocks, and identify the determinants of the likelihood that an individual will be a persistent high health care spender. METHODS: This study uses the Health and Retirement Study (HRS) from 2001 to 2010. Descriptive statistics show concentration and persistence of out-of-pocket health care spending. Logistic model is used to study the likelihood of persistently being in the top percentiles of out-of-pocket health care spending among elderly adults who were initially in the top percentiles and to identify determinants of such probabilities. RESULTS: High out-of-pocket spending persisted for about six years and then remained fairly stable afterwards. Having a certain health condition such as diabetes, strokes or emotional problems increased the probability of persistence by up to 9 percentage points. Compared to elderly adults who were covered by only Medicare, people having other government health insurance such as Medicaid,

A255

CHAMPUS, VA, or other military programs were about 19 percentage points less likely to be a persistently high out-of-pocket spender. By contrast, people who were covered by Medicare and any supplementary health insurance were more likely to be a persistently high spender. CONCLUSIONS: Non-trivial number of elderly adults remained in the same upper percentiles of the distribution for multiple years. The crucial factors that triggered high out-of-pocket spending on health care for several years were having certain chronic conditions or experiencing the onset of new health condition, even after controlling individual characteristics. Additionally, having an existing health insurance that supplements Medicare coverage or the acquisition of a new supplementary health insurance had a significant impact on the probability of persistence. PHP67 PUBLISHED BUDGET IMPACT ANALYSES IN THE UNITED STATES: ARE THEY (MIS)INFORMING DECISION MAKERS? Botteman MF, Li C, Corman S, Gao X Pharmerit International, Bethesda, MD, USA

OBJECTIVES: Budget impact analysis (BIA) should be used alongside costeffectiveness analysis (CEA) to inform formulary decision-making. In the present analysis, US-published BIAs of pharmacological interventions were analyzed to assess trends in reported outcomes and conclusions. METHODS: Systematic searches (01/2002-12/2012) of PubMed and other sources (i.e., ISPOR/AMCP proceedings, manual reference search) were conducted to identify published BIAs reporting per-member per-month costs (PMPMC or equivalent) associated with the introduction of new pharmacotherapies. The outcome measures (PMPMC and BIA-author judgments of their own results) were analyzed descriptively. Associations between these two measures and other study characteristics (publication year and type, analytical horizon, and patient population size) were evaluated. RESULTS: Fifty BIAs were identified (40% via PubMed), reporting a total of 82 PMPMC outcome estimates (as some studies reported multiple populations/indications). Half were published in/after 2009. The median (min; max) PMPMC across all 82 estimates was 2 (-66; 894) cents (US$). Thirteen percent of all estimates were cost-saving while 75% and 95% were ≤5 and ≤90 cents, respectively. 65% of studies reported results between -5 and +5 cents PMPM. No study characteristics were associated with PMPMC. In 84% of studies the authors provided qualitative judgments of the acceptability of their BIA results. In 6% of studies and estimates, the PMPMC was reported as unacceptable. CONCLUSIONS: BIA results cluster within a narrow range (-5 to +5 cents) and are nearly always reported as acceptable. These findings raise questions regarding the relevance and validity of published BIAs. In part, BIAs expressed in PMPMC (particularly in cents on the dollar) may easily provide a false impression of minimal impact. Whether BIA authors should self-judge the acceptability of PMPMC estimates is questionable. Without pre-defined benchmarks, judgments regarding BIA acceptability may be best left to decisionmakers and within the context of more comprehensive assessment (including notably CEA results). PHP68 PHARMACEUTICAL EXPENDITURE FORECAST MODEL TO SUPPORT HEALTH POLICY DECISION MAKING Rémuzat C1, Toumi M2, Vataire AL3, Cetinsoy L3, Aballea S4 1 Creativ-Ceutical France, Paris, France, 2University Claude Bernard Lyon 1, Lyon, France, 3 Creativ-Ceutical, PARIS, France, 4Creativ-Ceutical, Paris, France

OBJECTIVES: With constant incentives for health care payers to contain their pharmaceutical budgets, modelling policy decision impact become critical. The objective of this project was to test various policy decisions on pharmaceutical budget (developed for the European Commission for the project “EU Pharmaceutical expenditure forecast” -http://ec.europa.eu/health/healthcare/ key_documents/index_en.htm). METHODS: We built a model to assess policy scenarios impact on pharmaceuticals reference forecast for seven EU Member States (France, the UK, Germany, Poland, Portugal, Greece and Hungary). The following scenarios were tested: expanding UK policies to EU, time to market access, reimbursement rate , generic price and penetration, distribution chain of biosimilars. RESULTS: Applying the UK policy would incur dramatic savings for Germany (10 times the reference forecast). France and Portugal would also enjoy substantial additional savings under this scenario (2 and 4 times the reference forecast respectively). Delaying time to market is very powerful to reduce the pharmaceuticals expenditure. While applying the EU transparency directive (6 month-process for pricing and reimbursement), it would increase budget for all countries (from 1.1 to 4 times the reference forecast), except in Germany (additional savings). Decreasing price of generic and boosting the penetration rate, associated to distribution of biosimilar through hospital channel is a powerful way to reduce pharmaceuticals expenditure.In comparison, reimbursement rate has a modest impact. CONCLUSIONS: The most important leverages that were identified by the model on the pharmaceutical budget are driven by generic and biosimilar prices penetration rate and distribution. Reducing, even slightly, the prices of generics will have a major impact while reimbursement rate has less impact. Time to access for innovative products is an effective leverage used by some countries for delaying the bill. PHP69 ECONOMIC ANALYSIS OF THE MEXICAN'S INSTITUTE FOR SOCIAL SECURITY (IMSS) EXPENDITURE Huicochea JL Universidad Iberoamericana, Distrito Federal, Mexico

BACKGROUND: The Mexican institute for social security IMSS represents alone 44% of the total health care market in the country. Selling to this major client