- Email: [email protected]
Reality in Market Access In Germany and France – Comparative Analysis of Added Benefit Decisions on Innovative Pharmaceutical Therapies
A440
VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6
of the HTA process in Japan, which is excluded from this analysis. NICE, PBAC and the SMC all require the number of patients with the indicated disease and a clear statement of patient numbers eligible for treatment per year for 5 years. Beyond this, the requirements of the Australian PBC and the SMC were similar, specifying prevalence, incidence and mortality data, whereas NICE requires a measure of disease burden (not clearly defined) and life expectancy among those with the disease. The epidemiology requirements did not differ by disease area. Conclusions: HTA bodies stipulate the inclusion of epidemiological data to estimate economic impact of interventions. Some requirements are common to all agencies, but there are also some important differences. Specific epidemiological data needs for individual agencies must be considered by drug developers when planning and gathering information for HTA submissions. PHP212 Inclusion and Consideration of Patient Preferences in Amnog Early Benefit Assessments Obradovic M , Rauland M GfK Market Access, Nuremberg, Germany .
.
Objectives: To explore the inclusion and evaluation of patient preference data in AMNOG early benefit dossiers. Methods: Patient perspective is becoming increasingly important in health care decision making. Health technology assessment (HTA) agencies include patient views in their appraisals in different ways and to various extents. Patient views and preferences can be captured either by engaging patient organisations, patients themselves, or informal carers in the HTA process (as is done for example in the UK or Canada). Such patient aspects can be also explored in qualitative and quantitative studies which are considered as part of evidence documentation by the assessment committees. We have reviewed value dossiers and corresponding benefit assessments in Germany from 1st January 2011 to 31th March 2014. Types of patient preference data included in the value dossiers and their consideration in the assessments were collected and summarised. Results: A total of 68 dossiers were analyzed. 18 dossiers (26%) included data on patient preferences. Those data related to: a) relevance of different treatment endpoints from the patient’s perspective in oncology, hepatitis C, diabetes, or HIV infection; b) patient preference for a specific drug administration route (e.g. oral vs. injections), administration system (e.g. different inhalation systems), or administration frequency; c) patient preference for therapy duration or type of therapy. In none of the assessment reports did the evaluating committee specifically address the evidence presented on patient preferences. A comment on patient preference data was stated in one assessment report only (for aclidiniumbromid). Conclusions: About a quarter of value dossiers referred to data on patient preference. Surprisingly, it appears that the evidence on patient preference has not been considered in the AMNOG benefit assessments despite the fact that benefit to the patient is the central criterion of the AMNOG early benefit assessment. PHP214 Rapid Relative Effectiveness Assessment of Pharmaceuticals: Transferability and Completeness of Information Derived From Global Value Dossiers To Complete A Eunethta Submission Neeser K 1, Lister J 1, Stanisic S 2, Stengel C 1, Mueller E 1 1LASER ANALYTICA, Loerrach, Germany, 2LASER ANALYTICA, Milano, Italy .
.
.
.
.
Objectives: EUnetHTA is currently evaluating the applicability of the Core Model® for Rapid Relative Effectiveness Assessment of Pharmaceuticals of selected drugs for which a market authorization is intended between 2013 and 2015. A global value dossier (GVD) represents an important tool for pharmaceutical manufacturer (PM) to consolidate information on considered disease, its management and treatment, epidemiology, as well as health economic outcomes. The current study is intended to quantify the information that may be transferable from a well prepared, comprehensive GVD to EUnetHTA submission via gap analysis. The EUnetHTA submission comprises of 4 modules: 1. Description of the health problem; 2. Technical characteristics of technology and appropriate comparator(s); 3. Clinical effectiveness, and 4. Safety of the new drug. Methods: A GVD, developed for a pharmaceutical product, was used as data source to evaluate the feasibility to address all questions in the four EUnetHTA modules. Results: In modules 1 and 2 about 60% of data that are required to complete the EUnetHTA submission are missing from the GVD. Most of this information (e.g. implementation of the new drug, current use of the drug) may be easily obtained from other documents held by PM. In modules 3 and 4 about 70% of queries cannot be answered without additional assessments (e.g. complementary evidence synthesis, study quality assessments), systematic searches (e.g. on-going unpublished studies) and additional sources. Conclusions: A GVD can be a useful pool of knowledge for a new drug in a specific indication. Despite this large body of evidence, a considerable part of the information that is required for a EUnetHTA submission may be missing from GVDs and needs to be derived from existing clinical study reports, extensive systematic literature searches and additional evaluations. Applying of validated and systematic methods during the GVD development process may reduce additional work for assessment reports. PHP215 Encepp-HTA Working Group Survey on Capacity To Conduct Research in Support Of Health Technology Assessment Toussi M 1, Lis Y 2, Qizilbash N 3, Blake K V 4, Ehrenstein V 5, Kurz X 4, Moore N 6, Sinclair M 7 Health, Paris La Défense, France, 2PAREXEL International, Uxbridge, UK, 3OXON Epidemiology Limited and Department of Primary Care and Public Health, Imperial College, London, UK, 4European Medicines Agency, London, UK, 5Aarhus University Hospital, Aarhus N, Denmark, 6INSERM CIC-P 0005, Université de Bordeaux, CHU de Bordeaux, Bordeaux, France, 7University of Ulster, Newtownabbey, UK .
.
.
.
.
.
.
.
.
1IMS
Objectives: In light of growing interest in Health Technology Assessment (HTA) related outcomes being incorporated into post-authorisation studies (PAS) of medicines, an HTA Working Group of the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) has been established.
The Working Group’s current focus is on establishing baseline data to shape the network’s activities in meeting emerging demands for evidence generation in PAS. Therefore, an exploratory survey was undertaken to map current capabilities related to research that supports HTA and to identify relevant training needs. Methods: An online survey was developed and piloted within the HTA Working Group prior to distribution to over 200 researchers on the ENCePP mailing list. The items in the survey were both qualitative and quantitative. Data will be analysed using simple descriptive statistics and thematic analysis. Results: The survey is ongoing and will close in July 2014. Preliminary results indicate that (50%) of those sampled had conducted studies involving patient reported outcomes (PROs) including quality of life data. The most frequently used data collection tools were: paper (80%), handheld devices (50%), interactive voice (30%) and webbased (30%). Clinical outcomes were the most frequent primary study endpoints (100%), although at least some data on PROs and cost outcomes were included in 86% and 71% of studies, respectively. The most common design was primary data collection prospective cohort studies. ENCePP guidelines were the most frequently referenced followed by ISPOR and ISPE methodological good practice guidelines. Full data analyses will be available in September 2014. Conclusions: The findings of the ENCePP-HTA WG survey provide baseline data on ENCePP expertise and resources in research supporting HTA that are of interest to regulatory and HTA stakeholders across Europe. PHP216 What The English Could Learn From The Irish: Making The Nice Approval Process More Cost-Effective Macaulay R HERON Commercialization, London, UK .
Objectives: The National Institute of Health and Care Excellence (NICE) is a body that has gathered worldwide respect for its fully comprehensive clinical and economic review of pharmaceutical agents. However, it has been criticized for taking too long to conduct assessments and, as a consequence of this, only being able to appraise a subset of all new agents coming to market. Other Health technology Asessment (HTA) bodies achieve a much higher output of treatment appraisals with a much shorter turnaround time (e.g. in 2013 NICE appraised 27 drugs compared to 57 by the National Centre for Pharmacoeconomics (NCPE), Ireland). Methods: A thematic analysis of the appraisal processes of the NCPE was conducted to determine if any key criteria could be extrapolated that could potentially increase the speed and output of the NICE appraisal process. Results: One particularly interesting aspect of the NCPE system was revealed – a 2 to 4 week rapid preliminary review that all drugs must undergo to determine the necessity of a full pharmaco-economic assessment (PEA). Full PEAs are reserved for products with high cost, significant budget impact, and/or questionable value for money. This enables products that are cheaper and at least as efficacious as the relevant comparator to gain rapid market access. It also enables the NCPE to focus their time and resources on the products that will have a substantial/questionable clinical and/or economic impact. Conclusions: Adoption of such a systematic rapid review procedure into the NICE appraisal process with reimbursement conditional on NICE approval could similarly enable a better focussing of NICE resources to where it is most impactful. This could also incentivise pharmaceutical companies to drop their prices to gain rapid market access rather than going through the expensive and time consuming procedure of a full NICE submission. PHP217 Going Beyond The Qaly In Assessing The Benefits of Medical Devices Wilkinson G 1, Drummond M 2 School of Economics and Political Science, London, UK, 2University of York, Heslington, York, UK ..
.
1London
Objectives: Many medical devices offer improvements over current care that may be difficult to assess using standard methods of benefit such as the qualityadjusted life-year (QALY). The objective of this research was to identify whether these benefits could be measured and valued by alternative approaches, such as contingent valuation (willingness-to-pay) or conjoint analysis (discrete choice experiments). Methods: We undertook a systematic review of the literature from 1996 to 2013 to identify empirical studies of the benefits of medical devices using the following methodologies: willingness-to-pay (WTP), discrete choice experiments (DCEs), multi-criteria decision analysis (MCDAs) and subjective well-being (SWB). We recorded the number and category of individuals surveyed, the attributes explored and the key findings of each study. Results: The search resulted in 2772 hits, of which 2016 were considered not relevant and 76 were duplicates. On further examination, 47 of the remaining 680 papers were found not to meet our inclusion criteria, 240 were methodological papers and 363 involved non-device technologies. This left 30 relevant empirical studies, of which 18 were WTP and 12 DCEs. The types of devices studied included hearing aids, hip and knee implants and colostomy bags. Comfort, convenience or ease of use was the most common attribute explored other than effectiveness and quality of the device. Where both were studied, patient preferences sometimes differed from those of physicians. Conclusions: This research demonstrates that it is feasible to measure and value the benefits of devices using alternative approaches to QALYs, but that the literature is quite small as compared with that for non-device technologies. This is surprising given the often-repeated claim that devices have benefits that are difficult to assess using standard methods. Whilst the studies demonstrate the relative importance of each attribute, thought is required on how to incorporate these estimates into comparative cost-effectiveness studies. PHP218 Reality in Market Access In Germany and France – Comparative Analysis of Added Benefit Decisions on Innovative Pharmaceutical Therapies Droeschel D 1, Chicoye A 2, Lepretre M 2 University - SRH FernHochschule Riedlingen, Riedlingen, Germany, 2ESSEC Executive Education, Paris-La Défense, France .
1Riedlingen
.
.
A441
VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6
Objectives: The German market access system for drugs have been changed significantly in the last years, by introducing a similar focus on benefit assessment as in the French system. The research question remains whether they produce consistent results in terms of additional benefit (AB) for pharmaceuticals which have passed the assessment in both systems. Methods: The G-BA and IQWiG as well as the Transparency Commission (TC) databases were searched systematically to identify those products, which have been processed in both systems between Jan 2011 and Dec 2013. For further comparison a data grid consisting of 26 items for evaluation has been developed including study comparator, primary clinical endpoints, health related quality of life inclusion. Results: Overall, 140 new therapies have been assessed in France by TC, and 80 in Germany by the G-BA. According to inclusion criteria, 44 products could be identified which have passed through both systems including 7 orphan drugs. Thirteen products (30%) had no AB granted by both Agencies, whereas 9 (20%) were in both cases granted with a minor AB, (assuming that “minor” values are equivalent between the two systems), amounting to 22/44 cases with a similar resolution. Five cases (11%) showed a discrepancy in added benefit, all times TC = no and G-BA = yes. However, varying magnitudes appeared to be the greatest difference (n = 17 (39%) remaining drugs), conditioned by lacking concordance of both scale grade systems. Conclusions: Decisions of the agencies in both countries show partial heterogeneity in driving criteria like benefit levels (ASMR and AB). Although the evidence package for initial assessment in both countries is largely similar, preliminary results suggest their contextualization and scales are different. Further analysis based on results of the grid is needed to better assess criteria leading to different benefit levels and their reimbursement impact. PHP219 Factors Influencing Dutch Drug Reimbursement Recommendations; A Database Analysis Alleman C J M , Spoorendonk J A , Charokopou M , Hensen M , Heeg B Pharmerit International, Rotterdam, The Netherlands .
.
.
.
.
.
.
.
Objectives: In the Netherlands, manufacturers need to apply for reimbursement of outpatient drugs on either list 1A (no added benefit) or 1B (added benefit). For expensive drugs, hospitals can receive additional reimbursement if the drug is included on the expensive drug list (EDL). Pharmacoeconomic evidence is only required for list 1B and EDL evaluations. The National Health Care Institute (NZi) evaluates submissions and makes (provisional) reimbursement recommendations to the Dutch government. The aim of this study was to identify explanatory variables for the recommendation by NZi. Methods: A database of published evaluations from February-2006 to March-2014 was created, consisting of the final reimbursement recommendation and a range of corresponding explanatory variables such as the therapeutic indication, clinical and economic characteristics. Univariate analyses were performed to assess the impact of the individual explanatory variables on the recommendation by means of odds ratios. Results: In total 262 applications were included; the number of positive recommendations by NZi were 121/122 (99%) for 1A, 77/107 (72%) for 1B and 19/28 (68%) for EDL. Pharmacoeconomic analysis was reported in 36/107 (34%) 1B evaluations, of which 27 (75%) were recommended. For the EDL category, pharmacoeconomic analysis was reported in 20/28 (71%) evaluations, out of which 17 (85%) received a positive recommendation. Univariate analyses for the 1B subgroup showed that NZi recommendations were significantly (α = 0.05) influenced by clinical trials with life-saving primary endpoint (positive), non-inferior trial outcomes compared to placebo (negative) and budget impact below € 2,500,000 (positive). Whereas, the univariate analyses on EDL evaluations demonstrated that ATC-code L (antineoplastic and immunomodulating agents), clinical trials with life-saving primary endpoint and reporting of economic analysis outcomes had a significant and positive impact on the final NZi recommendation. Conclusions: These univariate analyses demonstrated that for 1B and EDL evaluations indication, clinical and economic factors impact the NZi reimbursement recommendations. PHP220 Measuring Extent of Access For Nice Health Technology Assessment Decisions: Trends From 2008 to 2013 O’Neill P , Devlin N Office of Health Economics, London, UK .
.
Objectives: When assessing trends in NICE HTA decisions it would be useful to ascertain their implications on access for groups of technologies. A specific issue is to understand the degree of access associated with ‘optimised’ decisions, where usage has been restricted to a subgroup of patients relative to the scope of the appraisal. Using a previously developed method, we calculate the degree of recommended access for medicines and assess trends between 2008 and 2013 by therapeutic area and over time. Methods: In a previously published paper we developed a measure, M, to assess access associated with NICE technology optimised appraisal decisions. This was defined as M=(p/P)x100, where M is a measure of the level of patient access (0 equals no access, 100 full access), P is the set of patients considered in the guidance as potential candidates for treatment (given the scope of appraisal and license), and p is the number of patients for whom NICE did recommend. Applying measure M to NICE HTA decisions for medicines between January 2008 and December 2013 we assess trends by therapeutic area and over time. In this paper, to understand trends, we extend the analysis to include recommended and not recommended decisions. We assume a recommended decision scores 100 using measure M, a not recommended decision 0, and optimised decisions, where not possible to determine M, a score of 50. Results: For 201 decisions between 2008 and 2013, on average, M was equal to 52, ranging from 37 in 2008 to 57 in 2011. At therapy level, M scored between 38 for cancer medicines to 100 for Hepatitis C treatments. Conclusions: The results for this period suggest around half of patients have been recommended by NICE to receive treatment, relative to scope of appraisal and license. These considerations address access not implementation issues.
PHP221 A Comparison of International Health Technology Assessment Systems – Does The Perfect System Exist? Kerr A 1, Todd C 1, Hebborn A 2, Ulyate K 1 1Roche Products Pty. Ltd., DEE WHY, Australia, 2F. Hoffmann-La Roche AG, Basel, Switzerland .
.
.
.
Objectives: There are a number of common elements considered good practice in Health Technology Assessment (HTA) that have been published by organizations representing the field. These components include: clear processes and decisionmaking, including scope for pragmatic approaches and appeal; transparency in methodology, value judgments and decisions; and a facility for stakeholder involvement. The objective of this study was to compare international HTA systems to rank their performance against the ideal components of HTA. Information was also collected on emerging topics such as combined regulatory-payer scientific advice, coverage with evidence, evaluation of drug-diagnostic pairs and disinvestment. Methods: A survey was designed to collect information on the HTA systems in the United Kingdom (UK), France, Germany, Italy, The Netherlands, Sweden, Central Eastern Europe, Canada, Australia, New Zealand (NZ), Korea and Taiwan. Questions were grouped under the topics: process, methods, data, societal input and transparency. The survey was completed by Roche affiliates with first-hand experience working with the HTA system in their country. Results: The majority of countries give consideration to rare diseases and low budget impact with leniency in decision making and/or process. Transparency in decision-making is lacking in many of the countries surveyed. Whilst consumer members sit on decision-making committees in several countries, only the UK involves a group of citizens in setting the decision making criteria applied by the committee. Combined regulatory-payer scientific advice is only available in European countries. Australia is the only country to evaluate drug-diagnostic pairings for both costs and outcomes. Only the UK and NZ have routine disinvestment reviews. Conclusions: Each country is performing well in some elements of their HTA system, but none met all the requirements of an ideal system. HTA systems can learn from the experiences in other countries when considering improvements to processes and efficiency. PHP223 Trends In Early Engagement Between Industry And HTA: Analysis of Scientific Advice Service Provided By Nice Since 2009 Maignen F M 1, Osipenko L 1, Gajraj E 1, Chivers R 2 Institute for Health and Care Excellence, London, UK, 2National Institute for Health and Clinical Excellence, Manchester, UK .
.
.
.
.
1National
Objectives: Regulatory Scientific Advice (SA) provided by EMA, FDA, MHRA and other agencies is highly demanded by manufacturers but health technology assessment (HTA) scientific advice is still far from becoming a routine step in the product development cycle. NICE has been running an advisory service for 5.5 years. Methods: This work presents analysis of requests to the programme: types of advice projects, number and type of requests per company, clinical indication, stage of clinical development when the advice is sought, reason for seeking advice and current development and regulatory status of products. Results: Between 2009 and 2014 NICE conducted 109 advisory projects (107 medicinal products and two diagnostic tests). 23 of these projects were done in parallel with regulatory agencies and/or other HTA bodies. 78% of all requests were in the following four therapeutic areas: oncology, neurology, rheumatology and cardiology. Majority of products (61%) were in phase II of clinical development when advice was sought. At the time of this analysis, 71 products (66%) were still in development, 6 (5.5%) were subject of a review for a marketing authorisation (MA), 8 (7.5%) had received a MA, the authorisation was not granted to 2 products (2%) and the clinical development was discontinued in 20 cases (19%). Most products that received NICE scientific advice are yet to be referred to the technology appraisals programme. Conclusions: Over the last few years, requests for scientific advice diversified into personalised medicines, regenerative medicines and products for rare and very rare diseases. Most HTA scientific advice requests continue to come from top 20 Pharma companies, however we are starting to see an increasing number of inquiries and project bookings from small-medium size companies. PHP224 Exploring Uncertainty in Economic Evaluation Of Medicines: A Review of The First Manufacturers’ Submissions To the French National Authority For Health (Has) Ghabri S 1, Hamers F F 1, Josselin J M 2, Harousseau J L 1 1Haute Autorité de Santé (HAS), Saint-Denis La Plaine, France, 2Université de Rennes 1, Rennes, France .
.
.
.
.
.
.
Objectives: Since October 2013, HAS is required to provide the inter-ministerial Economic Committee on Health Care Products (CEPS) with an economic evaluation on innovative medicines likely to have a significant budget impact on the national health insurance scheme. HAS economic evaluations are based on critical appraisals of cost-effectiveness analyses (CEA) submitted by manufacturers. Exploration of uncertainty around incremental cost-effectiveness ratio is critical to assess the robustness of CEA. Our objective was to assess how uncertainty exploration has been undertaken by manufacturers, using HAS guidelines on economic evaluation as an analytical framework. Methods: Manufacturers’ submissions assessed by end of May 2014 (n= 13) were reviewed. Three sources of uncertainty were considered: uncertainty around model input parameters, uncertainty around model structure and methodological uncertainty. Tools to explore uncertainty included deterministic sensitivity analysis (DSA), probabilistic sensitivity analysis (PSA), as well as overall compliance with HAS guidelines. Results: Model input parameters were the most frequently explored source of uncertainty. Both DSA and PSA were systematically used. However, reporting of DSA varied substantially across submissions, with frequent lack of justification of parameters ranges. Regarding PSA, the choice of distribution was not systematically justified and lacked consistency across similar parameters. Most submissions failed to consider parameters correlations. Exploration of uncertainty around model structure was rarely presented. Where applicable, alternative methods for extrapo-
VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6
of the HTA process in Japan, which is excluded from this analysis. NICE, PBAC and the SMC all require the number of patients with the indicated disease and a clear statement of patient numbers eligible for treatment per year for 5 years. Beyond this, the requirements of the Australian PBC and the SMC were similar, specifying prevalence, incidence and mortality data, whereas NICE requires a measure of disease burden (not clearly defined) and life expectancy among those with the disease. The epidemiology requirements did not differ by disease area. Conclusions: HTA bodies stipulate the inclusion of epidemiological data to estimate economic impact of interventions. Some requirements are common to all agencies, but there are also some important differences. Specific epidemiological data needs for individual agencies must be considered by drug developers when planning and gathering information for HTA submissions. PHP212 Inclusion and Consideration of Patient Preferences in Amnog Early Benefit Assessments Obradovic M , Rauland M GfK Market Access, Nuremberg, Germany .
.
Objectives: To explore the inclusion and evaluation of patient preference data in AMNOG early benefit dossiers. Methods: Patient perspective is becoming increasingly important in health care decision making. Health technology assessment (HTA) agencies include patient views in their appraisals in different ways and to various extents. Patient views and preferences can be captured either by engaging patient organisations, patients themselves, or informal carers in the HTA process (as is done for example in the UK or Canada). Such patient aspects can be also explored in qualitative and quantitative studies which are considered as part of evidence documentation by the assessment committees. We have reviewed value dossiers and corresponding benefit assessments in Germany from 1st January 2011 to 31th March 2014. Types of patient preference data included in the value dossiers and their consideration in the assessments were collected and summarised. Results: A total of 68 dossiers were analyzed. 18 dossiers (26%) included data on patient preferences. Those data related to: a) relevance of different treatment endpoints from the patient’s perspective in oncology, hepatitis C, diabetes, or HIV infection; b) patient preference for a specific drug administration route (e.g. oral vs. injections), administration system (e.g. different inhalation systems), or administration frequency; c) patient preference for therapy duration or type of therapy. In none of the assessment reports did the evaluating committee specifically address the evidence presented on patient preferences. A comment on patient preference data was stated in one assessment report only (for aclidiniumbromid). Conclusions: About a quarter of value dossiers referred to data on patient preference. Surprisingly, it appears that the evidence on patient preference has not been considered in the AMNOG benefit assessments despite the fact that benefit to the patient is the central criterion of the AMNOG early benefit assessment. PHP214 Rapid Relative Effectiveness Assessment of Pharmaceuticals: Transferability and Completeness of Information Derived From Global Value Dossiers To Complete A Eunethta Submission Neeser K 1, Lister J 1, Stanisic S 2, Stengel C 1, Mueller E 1 1LASER ANALYTICA, Loerrach, Germany, 2LASER ANALYTICA, Milano, Italy .
.
.
.
.
Objectives: EUnetHTA is currently evaluating the applicability of the Core Model® for Rapid Relative Effectiveness Assessment of Pharmaceuticals of selected drugs for which a market authorization is intended between 2013 and 2015. A global value dossier (GVD) represents an important tool for pharmaceutical manufacturer (PM) to consolidate information on considered disease, its management and treatment, epidemiology, as well as health economic outcomes. The current study is intended to quantify the information that may be transferable from a well prepared, comprehensive GVD to EUnetHTA submission via gap analysis. The EUnetHTA submission comprises of 4 modules: 1. Description of the health problem; 2. Technical characteristics of technology and appropriate comparator(s); 3. Clinical effectiveness, and 4. Safety of the new drug. Methods: A GVD, developed for a pharmaceutical product, was used as data source to evaluate the feasibility to address all questions in the four EUnetHTA modules. Results: In modules 1 and 2 about 60% of data that are required to complete the EUnetHTA submission are missing from the GVD. Most of this information (e.g. implementation of the new drug, current use of the drug) may be easily obtained from other documents held by PM. In modules 3 and 4 about 70% of queries cannot be answered without additional assessments (e.g. complementary evidence synthesis, study quality assessments), systematic searches (e.g. on-going unpublished studies) and additional sources. Conclusions: A GVD can be a useful pool of knowledge for a new drug in a specific indication. Despite this large body of evidence, a considerable part of the information that is required for a EUnetHTA submission may be missing from GVDs and needs to be derived from existing clinical study reports, extensive systematic literature searches and additional evaluations. Applying of validated and systematic methods during the GVD development process may reduce additional work for assessment reports. PHP215 Encepp-HTA Working Group Survey on Capacity To Conduct Research in Support Of Health Technology Assessment Toussi M 1, Lis Y 2, Qizilbash N 3, Blake K V 4, Ehrenstein V 5, Kurz X 4, Moore N 6, Sinclair M 7 Health, Paris La Défense, France, 2PAREXEL International, Uxbridge, UK, 3OXON Epidemiology Limited and Department of Primary Care and Public Health, Imperial College, London, UK, 4European Medicines Agency, London, UK, 5Aarhus University Hospital, Aarhus N, Denmark, 6INSERM CIC-P 0005, Université de Bordeaux, CHU de Bordeaux, Bordeaux, France, 7University of Ulster, Newtownabbey, UK .
.
.
.
.
.
.
.
.
1IMS
Objectives: In light of growing interest in Health Technology Assessment (HTA) related outcomes being incorporated into post-authorisation studies (PAS) of medicines, an HTA Working Group of the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) has been established.
The Working Group’s current focus is on establishing baseline data to shape the network’s activities in meeting emerging demands for evidence generation in PAS. Therefore, an exploratory survey was undertaken to map current capabilities related to research that supports HTA and to identify relevant training needs. Methods: An online survey was developed and piloted within the HTA Working Group prior to distribution to over 200 researchers on the ENCePP mailing list. The items in the survey were both qualitative and quantitative. Data will be analysed using simple descriptive statistics and thematic analysis. Results: The survey is ongoing and will close in July 2014. Preliminary results indicate that (50%) of those sampled had conducted studies involving patient reported outcomes (PROs) including quality of life data. The most frequently used data collection tools were: paper (80%), handheld devices (50%), interactive voice (30%) and webbased (30%). Clinical outcomes were the most frequent primary study endpoints (100%), although at least some data on PROs and cost outcomes were included in 86% and 71% of studies, respectively. The most common design was primary data collection prospective cohort studies. ENCePP guidelines were the most frequently referenced followed by ISPOR and ISPE methodological good practice guidelines. Full data analyses will be available in September 2014. Conclusions: The findings of the ENCePP-HTA WG survey provide baseline data on ENCePP expertise and resources in research supporting HTA that are of interest to regulatory and HTA stakeholders across Europe. PHP216 What The English Could Learn From The Irish: Making The Nice Approval Process More Cost-Effective Macaulay R HERON Commercialization, London, UK .
Objectives: The National Institute of Health and Care Excellence (NICE) is a body that has gathered worldwide respect for its fully comprehensive clinical and economic review of pharmaceutical agents. However, it has been criticized for taking too long to conduct assessments and, as a consequence of this, only being able to appraise a subset of all new agents coming to market. Other Health technology Asessment (HTA) bodies achieve a much higher output of treatment appraisals with a much shorter turnaround time (e.g. in 2013 NICE appraised 27 drugs compared to 57 by the National Centre for Pharmacoeconomics (NCPE), Ireland). Methods: A thematic analysis of the appraisal processes of the NCPE was conducted to determine if any key criteria could be extrapolated that could potentially increase the speed and output of the NICE appraisal process. Results: One particularly interesting aspect of the NCPE system was revealed – a 2 to 4 week rapid preliminary review that all drugs must undergo to determine the necessity of a full pharmaco-economic assessment (PEA). Full PEAs are reserved for products with high cost, significant budget impact, and/or questionable value for money. This enables products that are cheaper and at least as efficacious as the relevant comparator to gain rapid market access. It also enables the NCPE to focus their time and resources on the products that will have a substantial/questionable clinical and/or economic impact. Conclusions: Adoption of such a systematic rapid review procedure into the NICE appraisal process with reimbursement conditional on NICE approval could similarly enable a better focussing of NICE resources to where it is most impactful. This could also incentivise pharmaceutical companies to drop their prices to gain rapid market access rather than going through the expensive and time consuming procedure of a full NICE submission. PHP217 Going Beyond The Qaly In Assessing The Benefits of Medical Devices Wilkinson G 1, Drummond M 2 School of Economics and Political Science, London, UK, 2University of York, Heslington, York, UK ..
.
1London
Objectives: Many medical devices offer improvements over current care that may be difficult to assess using standard methods of benefit such as the qualityadjusted life-year (QALY). The objective of this research was to identify whether these benefits could be measured and valued by alternative approaches, such as contingent valuation (willingness-to-pay) or conjoint analysis (discrete choice experiments). Methods: We undertook a systematic review of the literature from 1996 to 2013 to identify empirical studies of the benefits of medical devices using the following methodologies: willingness-to-pay (WTP), discrete choice experiments (DCEs), multi-criteria decision analysis (MCDAs) and subjective well-being (SWB). We recorded the number and category of individuals surveyed, the attributes explored and the key findings of each study. Results: The search resulted in 2772 hits, of which 2016 were considered not relevant and 76 were duplicates. On further examination, 47 of the remaining 680 papers were found not to meet our inclusion criteria, 240 were methodological papers and 363 involved non-device technologies. This left 30 relevant empirical studies, of which 18 were WTP and 12 DCEs. The types of devices studied included hearing aids, hip and knee implants and colostomy bags. Comfort, convenience or ease of use was the most common attribute explored other than effectiveness and quality of the device. Where both were studied, patient preferences sometimes differed from those of physicians. Conclusions: This research demonstrates that it is feasible to measure and value the benefits of devices using alternative approaches to QALYs, but that the literature is quite small as compared with that for non-device technologies. This is surprising given the often-repeated claim that devices have benefits that are difficult to assess using standard methods. Whilst the studies demonstrate the relative importance of each attribute, thought is required on how to incorporate these estimates into comparative cost-effectiveness studies. PHP218 Reality in Market Access In Germany and France – Comparative Analysis of Added Benefit Decisions on Innovative Pharmaceutical Therapies Droeschel D 1, Chicoye A 2, Lepretre M 2 University - SRH FernHochschule Riedlingen, Riedlingen, Germany, 2ESSEC Executive Education, Paris-La Défense, France .
1Riedlingen
.
.
A441
VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6
Objectives: The German market access system for drugs have been changed significantly in the last years, by introducing a similar focus on benefit assessment as in the French system. The research question remains whether they produce consistent results in terms of additional benefit (AB) for pharmaceuticals which have passed the assessment in both systems. Methods: The G-BA and IQWiG as well as the Transparency Commission (TC) databases were searched systematically to identify those products, which have been processed in both systems between Jan 2011 and Dec 2013. For further comparison a data grid consisting of 26 items for evaluation has been developed including study comparator, primary clinical endpoints, health related quality of life inclusion. Results: Overall, 140 new therapies have been assessed in France by TC, and 80 in Germany by the G-BA. According to inclusion criteria, 44 products could be identified which have passed through both systems including 7 orphan drugs. Thirteen products (30%) had no AB granted by both Agencies, whereas 9 (20%) were in both cases granted with a minor AB, (assuming that “minor” values are equivalent between the two systems), amounting to 22/44 cases with a similar resolution. Five cases (11%) showed a discrepancy in added benefit, all times TC = no and G-BA = yes. However, varying magnitudes appeared to be the greatest difference (n = 17 (39%) remaining drugs), conditioned by lacking concordance of both scale grade systems. Conclusions: Decisions of the agencies in both countries show partial heterogeneity in driving criteria like benefit levels (ASMR and AB). Although the evidence package for initial assessment in both countries is largely similar, preliminary results suggest their contextualization and scales are different. Further analysis based on results of the grid is needed to better assess criteria leading to different benefit levels and their reimbursement impact. PHP219 Factors Influencing Dutch Drug Reimbursement Recommendations; A Database Analysis Alleman C J M , Spoorendonk J A , Charokopou M , Hensen M , Heeg B Pharmerit International, Rotterdam, The Netherlands .
.
.
.
.
.
.
.
Objectives: In the Netherlands, manufacturers need to apply for reimbursement of outpatient drugs on either list 1A (no added benefit) or 1B (added benefit). For expensive drugs, hospitals can receive additional reimbursement if the drug is included on the expensive drug list (EDL). Pharmacoeconomic evidence is only required for list 1B and EDL evaluations. The National Health Care Institute (NZi) evaluates submissions and makes (provisional) reimbursement recommendations to the Dutch government. The aim of this study was to identify explanatory variables for the recommendation by NZi. Methods: A database of published evaluations from February-2006 to March-2014 was created, consisting of the final reimbursement recommendation and a range of corresponding explanatory variables such as the therapeutic indication, clinical and economic characteristics. Univariate analyses were performed to assess the impact of the individual explanatory variables on the recommendation by means of odds ratios. Results: In total 262 applications were included; the number of positive recommendations by NZi were 121/122 (99%) for 1A, 77/107 (72%) for 1B and 19/28 (68%) for EDL. Pharmacoeconomic analysis was reported in 36/107 (34%) 1B evaluations, of which 27 (75%) were recommended. For the EDL category, pharmacoeconomic analysis was reported in 20/28 (71%) evaluations, out of which 17 (85%) received a positive recommendation. Univariate analyses for the 1B subgroup showed that NZi recommendations were significantly (α = 0.05) influenced by clinical trials with life-saving primary endpoint (positive), non-inferior trial outcomes compared to placebo (negative) and budget impact below € 2,500,000 (positive). Whereas, the univariate analyses on EDL evaluations demonstrated that ATC-code L (antineoplastic and immunomodulating agents), clinical trials with life-saving primary endpoint and reporting of economic analysis outcomes had a significant and positive impact on the final NZi recommendation. Conclusions: These univariate analyses demonstrated that for 1B and EDL evaluations indication, clinical and economic factors impact the NZi reimbursement recommendations. PHP220 Measuring Extent of Access For Nice Health Technology Assessment Decisions: Trends From 2008 to 2013 O’Neill P , Devlin N Office of Health Economics, London, UK .
.
Objectives: When assessing trends in NICE HTA decisions it would be useful to ascertain their implications on access for groups of technologies. A specific issue is to understand the degree of access associated with ‘optimised’ decisions, where usage has been restricted to a subgroup of patients relative to the scope of the appraisal. Using a previously developed method, we calculate the degree of recommended access for medicines and assess trends between 2008 and 2013 by therapeutic area and over time. Methods: In a previously published paper we developed a measure, M, to assess access associated with NICE technology optimised appraisal decisions. This was defined as M=(p/P)x100, where M is a measure of the level of patient access (0 equals no access, 100 full access), P is the set of patients considered in the guidance as potential candidates for treatment (given the scope of appraisal and license), and p is the number of patients for whom NICE did recommend. Applying measure M to NICE HTA decisions for medicines between January 2008 and December 2013 we assess trends by therapeutic area and over time. In this paper, to understand trends, we extend the analysis to include recommended and not recommended decisions. We assume a recommended decision scores 100 using measure M, a not recommended decision 0, and optimised decisions, where not possible to determine M, a score of 50. Results: For 201 decisions between 2008 and 2013, on average, M was equal to 52, ranging from 37 in 2008 to 57 in 2011. At therapy level, M scored between 38 for cancer medicines to 100 for Hepatitis C treatments. Conclusions: The results for this period suggest around half of patients have been recommended by NICE to receive treatment, relative to scope of appraisal and license. These considerations address access not implementation issues.
PHP221 A Comparison of International Health Technology Assessment Systems – Does The Perfect System Exist? Kerr A 1, Todd C 1, Hebborn A 2, Ulyate K 1 1Roche Products Pty. Ltd., DEE WHY, Australia, 2F. Hoffmann-La Roche AG, Basel, Switzerland .
.
.
.
Objectives: There are a number of common elements considered good practice in Health Technology Assessment (HTA) that have been published by organizations representing the field. These components include: clear processes and decisionmaking, including scope for pragmatic approaches and appeal; transparency in methodology, value judgments and decisions; and a facility for stakeholder involvement. The objective of this study was to compare international HTA systems to rank their performance against the ideal components of HTA. Information was also collected on emerging topics such as combined regulatory-payer scientific advice, coverage with evidence, evaluation of drug-diagnostic pairs and disinvestment. Methods: A survey was designed to collect information on the HTA systems in the United Kingdom (UK), France, Germany, Italy, The Netherlands, Sweden, Central Eastern Europe, Canada, Australia, New Zealand (NZ), Korea and Taiwan. Questions were grouped under the topics: process, methods, data, societal input and transparency. The survey was completed by Roche affiliates with first-hand experience working with the HTA system in their country. Results: The majority of countries give consideration to rare diseases and low budget impact with leniency in decision making and/or process. Transparency in decision-making is lacking in many of the countries surveyed. Whilst consumer members sit on decision-making committees in several countries, only the UK involves a group of citizens in setting the decision making criteria applied by the committee. Combined regulatory-payer scientific advice is only available in European countries. Australia is the only country to evaluate drug-diagnostic pairings for both costs and outcomes. Only the UK and NZ have routine disinvestment reviews. Conclusions: Each country is performing well in some elements of their HTA system, but none met all the requirements of an ideal system. HTA systems can learn from the experiences in other countries when considering improvements to processes and efficiency. PHP223 Trends In Early Engagement Between Industry And HTA: Analysis of Scientific Advice Service Provided By Nice Since 2009 Maignen F M 1, Osipenko L 1, Gajraj E 1, Chivers R 2 Institute for Health and Care Excellence, London, UK, 2National Institute for Health and Clinical Excellence, Manchester, UK .
.
.
.
.
1National
Objectives: Regulatory Scientific Advice (SA) provided by EMA, FDA, MHRA and other agencies is highly demanded by manufacturers but health technology assessment (HTA) scientific advice is still far from becoming a routine step in the product development cycle. NICE has been running an advisory service for 5.5 years. Methods: This work presents analysis of requests to the programme: types of advice projects, number and type of requests per company, clinical indication, stage of clinical development when the advice is sought, reason for seeking advice and current development and regulatory status of products. Results: Between 2009 and 2014 NICE conducted 109 advisory projects (107 medicinal products and two diagnostic tests). 23 of these projects were done in parallel with regulatory agencies and/or other HTA bodies. 78% of all requests were in the following four therapeutic areas: oncology, neurology, rheumatology and cardiology. Majority of products (61%) were in phase II of clinical development when advice was sought. At the time of this analysis, 71 products (66%) were still in development, 6 (5.5%) were subject of a review for a marketing authorisation (MA), 8 (7.5%) had received a MA, the authorisation was not granted to 2 products (2%) and the clinical development was discontinued in 20 cases (19%). Most products that received NICE scientific advice are yet to be referred to the technology appraisals programme. Conclusions: Over the last few years, requests for scientific advice diversified into personalised medicines, regenerative medicines and products for rare and very rare diseases. Most HTA scientific advice requests continue to come from top 20 Pharma companies, however we are starting to see an increasing number of inquiries and project bookings from small-medium size companies. PHP224 Exploring Uncertainty in Economic Evaluation Of Medicines: A Review of The First Manufacturers’ Submissions To the French National Authority For Health (Has) Ghabri S 1, Hamers F F 1, Josselin J M 2, Harousseau J L 1 1Haute Autorité de Santé (HAS), Saint-Denis La Plaine, France, 2Université de Rennes 1, Rennes, France .
.
.
.
.
.
.
Objectives: Since October 2013, HAS is required to provide the inter-ministerial Economic Committee on Health Care Products (CEPS) with an economic evaluation on innovative medicines likely to have a significant budget impact on the national health insurance scheme. HAS economic evaluations are based on critical appraisals of cost-effectiveness analyses (CEA) submitted by manufacturers. Exploration of uncertainty around incremental cost-effectiveness ratio is critical to assess the robustness of CEA. Our objective was to assess how uncertainty exploration has been undertaken by manufacturers, using HAS guidelines on economic evaluation as an analytical framework. Methods: Manufacturers’ submissions assessed by end of May 2014 (n= 13) were reviewed. Three sources of uncertainty were considered: uncertainty around model input parameters, uncertainty around model structure and methodological uncertainty. Tools to explore uncertainty included deterministic sensitivity analysis (DSA), probabilistic sensitivity analysis (PSA), as well as overall compliance with HAS guidelines. Results: Model input parameters were the most frequently explored source of uncertainty. Both DSA and PSA were systematically used. However, reporting of DSA varied substantially across submissions, with frequent lack of justification of parameters ranges. Regarding PSA, the choice of distribution was not systematically justified and lacked consistency across similar parameters. Most submissions failed to consider parameters correlations. Exploration of uncertainty around model structure was rarely presented. Where applicable, alternative methods for extrapo-