Cost-effectiveness analyses are integrated in evidence-based recommendations for market access and reimbursement of new products, provided by health technology assessment (HTA) bodies, in several countries. The objective was to compare economic protocols submitted for a same product to three HTA bodies, assessing if the same cost-effectiveness analysis and methodology was implemented and if the results provided consistent ICERs. We focused on the 3 following HTA bodies (Haute Autorité de Santé (HAS), National Institute for Health and care Excellence (NICE) and Canada's Drug and Health Technology Agency (CADTH)). The thirteen HAS opinions assessed from September 2021 to February 2022 were eligible for the analysis and selected if the health technology appraisal was available for the three HTA bodies. Comparison of economic protocols were performed using the published opinions for HAS, the Company evidence submission for NICE and the pharmacoeconomic reviews for CADTH. Thirteen methodological items were reviewed and compared. Among the thirteen opinions, four were available for the three HTA bodies: EVRYSDI® (spinal muscular atrophy), VENCLYXTO® (acute myeloid leukemia), KEYTRUDA® (metastatic colorectal cancer (MCC) and esophageal cancer (OC)). Eight items were common between submissions for KEYTRUDA® (MCC), seven for EVRYSDI®, two for KEYTRUDA® (OC) and one for VENCLYXTO®. Objective, simulated population, and treatment duration were the most common items, whereas time horizon and utilities were systematically different. These differences conduct to important range in ICER estimations (for instance, between £43,225 for NICE, €107,407 for HAS, $142,861 for CADTH for KEYTRUDA® (OC)). Heterogeneity is observed in the methodology of cost-effectiveness analyses submitted for a same product to the three considered HTA bodies. The differences in some key structural choices and hypotheses lead to heterogenous ICERs estimations, key driver for evidence-based guidance and reimbursement process.

For healthcare products liable to offer moderate to major Clinical Added Value (CAV), the CEESP issues an economic opinion in parallel of the TC which grants the level of CAV. The CAV is based on demonstrated therapeutic progress – in terms of efficacy or safety – compared to existing alternatives. The objective was to compare CEESP and TC conclusions for products with a CAV of level V (no therapeutic progress). All published CEESP opinions from February 2014 to May 2022 were eligible. Opinions on drugs or vaccine, with a CAV V granted by the TC, were selected. An extraction grid was developed with the following items from the TC and CEESP appraisals: CAV V justification, indication, efficacy comparison method, data source, methodological reservation. On the 165 CEESP opinions identified, 27 were granted a CAV V by the TC. The main reasons were the absence of relevant comparators, indirect treatment comparison method, safety concerns or limited effect size. Less than 50% (12 out of the 27) of the opinions were issued at least one major reservation or major uncertainty by the CEESP. Most of opinions with a reservation (9/12) were consistent with the TC concerns (especially clinical data and relative efficacy estimation). No pattern was observed in therapeutic areas (oncology, neurology, cardiology), or ICER result. When requested, the parallel medico-technic and economic appraisals provided by the TC and CEESP, respectively, can lead to misleading conclusions. Indeed, when a CAV V is issued by the TC, introduction of the new product on the French market must show cost savings, resulting in an irrelevant assessment of the cost-effectiveness analysis by the CEESP. The relevance of a joint evaluation between the two committees could be considered to avoid this inconsistency.

In France, the CEESP assesses the cost-effectiveness of products claiming a moderate to major therapeutic progress. A “major reservation” is issued by CEESP if the economic analysis doesn’t comply with the published methodological guidelines. The aim of this project is to assess which methodological items correlate with a risk of major reservation. We extracted all pertinent variables (n=24) with less than 20% missing data from the base case analysis of 144 Economic Opinions published for new innovative drug between October 2013 and April 2022. A logistic regression, where the dependent variable corresponds to the dichotomic event “major reservation”, was fitted. Variables with a p-value below 0.20 in a univariate regression were included in a stepwise multivariate regression (p < 0.05). From the 24 variables tested, six were associated with a p-value below 0.20 indicating a possible correlation with a “major reservation” and their inclusion in the stepwise multivariate regression. The variables were the therapeutic area, the context of the submission (1st evaluation, line extension, reassessment), the presence of an external competitors included in the dossier, utility estimation method (direct, indirect, mapping) and the tariff of the utility. Drugs submitted in virology, the presence of an external competitor, an ad-hoc study to estimate the utility and utility estimated in UK tariff were associated with a higher probability of “major reservation”. From all the information gathered in the 144 Economic Opinions published, it sems that the anticipation of a “major reservation” based on specific items isn’t straightforward. This result is consistent with the lack of predictability or standardization observed in the conclusions of CEESP opinions. Future work is needed to investigate the causality and identify clear items to predict the risk of “major reservation”.

On March 5th, 2021, a new LEEM-CEPS framework agreement was signed for 3 years, setting notably the conditions under which drugs with an ASMR IV can claim a list price consistent with the prices in 4 European markets (Germany, Italy, Spain, United Kingdom). This study aimed at analyzing the impact of this agreement on ASMR IV price, and more particularly whether drugs eligible for the European Price Guarantee have benefited from a consistent French price. Medicines with an ASMR IV granted between January 1st, 2020 and December 31st, 2021 were identified from the IQVIA Transparency Commission database, only first-time assessment and extension of indication requests which prices were published 1 month after the agreement signature were eligible. French list prices were extracted from the "Journal Officiel", and list prices in the EU-3 + UK were extracted from PricingInsight IQVIA database. Finally, we reviewed if eligible drugs met the criteria for the European Price Guarantee and we calculated the duration of price negotiation. 83 products obtained an ASMR IV between 2020 and 2021, and 16 were eligible for this analysis. At least one of the criteria for the European Price Guarantee was met by 10 drugs (62.5%), and 8 (50%) drugs had available prices in Europe. Compared to the lowest European price, 4 (25%) drugs received an equivalent list price, 2 (12.5%) drugs obtained a higher French price, and 2 (12.5%) drugs obtained a lower price, with an average discount of 1.2%. Moreover, price publication delays more than doubled for the 10 eligible drugs after the agreement (414 days on average versus 161 days in 2019). This study shows that the framework agreement is applied for most eligible products, but in return we observe an increase in negotiation time, due to potential more challenging net price negotiation.

To estimate the cost of healthcare resource utilization (HCRU) in relapsed or refractory (r/r) diffuse large B-cell lymphoma (DLBCL) patients treated with tisagenlecleucel. An economic model was developed to assess the total HCRU costs in patients with r/r DLBCL receiving tisagenlecleucel from the healthcare payer perspective in the UK and France. The HCRU costs were estimated from the time of leukapheresis to 2 months post-infusion, which is the timeframe safety monitoring and HCRU are likely occur. The model was developed using a micro-costing approach. The HCRU and safety data were from JULIET, the pivotal, global trial of tisagenlecleucel in r/r DLBCL. The model considered costs of leukapheresis, bridging and lymphodepleting chemotherapy, inpatient and intensive care unit (ICU) admission, medical professional visits, lab tests and procedures, and management of adverse events (AE). Drug costs were obtained from MIMS and eMIT for UK and ATIH for France. Unit costs for HCRU and adverse events were from public databases (NHS and PSSRU in UK; Ameli, ATIH, and GHM in France) and literature. Total HCRU costs were calculated in 2018 GBP/Euro. One-way sensitivity analysis was performed. The model estimated total HCRU costs to be £24,009 in the UK and €30,172 in France. In the UK, the costs of bridging and lymphodepleting regimens, medical professional visits, lab tests and procedures including leukapheresis, inpatient and ICU and AE management costs accounted for 7.0%, 14.6%, 11.7%, 51.4% and 15.3% of total costs, respectively. In France, the corresponding percentage was estimated to be 5.4%, 5.5%, 4.9%, 53.6% and 30.6%, respectively. The model results remained robust in the sensitivity analyses. The HCRU costs within 2 months of tisagenlecleucel administration equated to £24,009 in UK and €30,172 in France. Further research with estimates based on real-world clinical use of tisagenlecleucel is warranted.

Limited information is available on cost associated with AHSCT and in particular those detailing the cost of pre-transplant chemotherapy and post-transplant follow-up care. The aim of this study was to estimate the average hospital cost per patient for AHSCT pre- AHSCT chemotherapy cost and the 6-month transplant follow-up care cost. A retrospective study of patients with RR DLBCL (diffuse large b-cell lymphoma) at ages 18 and more was performed from French hospital database (PMSI) 2013 to 2016. Hospital stays were selected with the ICD10 code CIM C833 "Diffuse large B-cell lymphoma” in position of: Principal Diagnosis (PD) or Related Diagnosis (RD) or Associated Diagnosis between 01/01/2013 and 01/01/2014. Refractory patients could not be identified in the database. Relapsed patients were defined as patients by 2 criteria: (a) at least one complete cycle of rituximab chemotherapy (b) bone marrow biopsy followed by hospitalization. For each allograft identified with CCAM (Classification Commune des Actes Médicaux) procedure FEFFL009 "intravenous injection of a cell therapy for transplant", a cost evaluation per patient (€2017) was calculated according to the French health service perspective by detailing: the cost of pre-transplant chemotherapy, cost of transplant and the cost of follow-up. The prevalence of AHSCT refractory DLBCL was 6 patients in the PMSI. The median transplant cost per patient was € 85,038 (€79,216 - €121,199), the pre AHSCT chemotherapy median cost was € 13,940 (€ 2,521 -€33,246) and the follow-up median cost was € 11,398 (€3,157 -€51,094). The mean transplant cost per patient was € 93,233(±€14,973), the pre AHSCT chemotherapy mean cost was €14,617 (±€9,342) and the follow-up mean cost was € 16,625(±€16,378). The study showed that the economic burden of AHSCT in DLBCL patients, based on the hospital costs, is significant

L’insuffisance cardiaque (IC) est un problème majeur de santé publique dû à sa prévalence élevée en occident (2 à 3 % de la population). Afin d’améliorer la prise en charge de ces patients, le CHU de Montpellier a mis en place en 2014 un parcours de soin optimisé. Le but de notre étude était de comparer les coûts et les taux de réhospitalisation entre 2013 et 2015 et d’évaluer l’impact de ce nouveau parcours. Une étude rétrospective, observationnelle et comparative de type avant-après a été conduite. Le critère de jugement principal était le taux de réhospitalisation un an après une première hospitalisation pour IC, c’est-à-dire sans antécédents hospitaliers au cours des trois années précédentes. La base PMSI du CHU a été mobilisée pour sélectionner les séjours et les patients d’intérêt (526 et 514 patients en 2013 et 2015 respectivement), et pour recueillir les informations médicales et démographiques nécessaires. La valorisation des séjours a été réalisée via la comptabilité analytique du CHU. Les groupes ont été harmonisés via appariement sur score de propension. Les taux de réhospitalisation un an sont comparables : 18,06 % en 2013 versus 18,48 % en 2015, mais on observe un écart important 7 jours après la sortie (1,56 % en 2015 versus 3,04 % en 2013), même si cette différence n’est pas significative (p = 0,11). Sur l’année complète, les dépenses moyennes par patient s’élèvent à 8011,3 € en 2015 contre 7367,6 € en 2013 (n.s.). Le même écart a été observé pour les recettes. L’analyse par poste de dépense a révélé des différences significatives concernant les dispositifs médicaux implantables (DMI). Si ce n’est la tendance observée à j7, aucun impact sur les taux de réadmission n’a été détecté. La hausse des dépenses en 2015 pourrait en partie être expliquée par des changements dans la pratique clinique, comme l’augmentation de l’utilisation des DMI. L’étude des changements organisationnels survenus entre 2013 et 2015 doit être approfondie.

Limited information is available on cost associated with AHSCT and in particular those detailing pre-transplant chemotherapy and post-transplant follow-up care. The aim of this study was to estimate the average hospital cost per patient of AHSCT, pre- AHSCT chemotherapy cost and the 6-month transplant follow-up care cost. A retrospective study of patients with RR B-cell ALL (acute lymphoblastic leukemia) at ages 3 and 25 was performed from French hospital database (PMSI) 2013 to 2016. Hospital stays were selected with the ICD10 code CIM C910 "Acute lymphoblastic leukemia” in position of: Principal Diagnosis (PD) or Related Diagnosis (RD) or Associated Diagnosis between 01/01/2013 and 01/01/2014 excluding stays with a Nelarabine administration. Refractory patients could not be identified in the database. Relapsed patients were defined as patients by 2 criteria: (a) at least one complete cycle of chemotherapy prior to transplant (b) bone marrow biopsy followed by hospitalization. For each allograft identified with CCAM (Classification Commune des Actes Médicaux) procedure FEFFL009 "intravenous injection of a cell therapy for transplant", a cost evaluation per stay and patient (€, 2017) was calculated according to the French health service perspective by detailing: the cost of pre- AHSCT chemotherapy, cost of transplant and the cost of follow-up. The prevalence of AHSCT refractory LAL was 126 patients in the PMSI. The median transplant cost per patient was €92,740 (€38,746 - €283,596), the pre-AHSCT chemotherapy median cost was €122,266 (€415 -€344,116) and the follow-up median cost was € 8,467 (€ 628 - €74,081). The mean transplant cost per patient was € 96,738(±€33,758), the pre AHSCT chemotherapy mean cost was €133,319 (±€67,659) and the follow-up mean cost was € 15,676(±€16,940). The study showed that the economic burden of AHSCT in LAL patients, based on the hospital costs, is significant.

The objective of this study was to assess the number of patients hospitalized for sickle-cell disease (SCD) in France and for each region, and to estimate the proportion of patients experiencing Vaso-Occlusive Crisis (VOC) and Blood Transfusions (BT). A cross sectional study of all 2016 hospital stays related to SCD, using ICD10 codes as D57.(0/1/2 or 8) as PD/RD (principal/related diagnosis) or SAD (significantly associated diagnosis), were extracted from the French national hospital discharge database (PMSI). Thanks to their unique ID, patients have been tracked over hospital stays. VOC has been defined by the ICD10 code D57.0 as PD/RD, and BT has been defined by the ICD 10 code Z513 and/or the French CCAM codes (FELF00(4/5/7/11/12 or FEPF003)). Annual rates of hospitalized patient were standardized on age and sex using population residing in France on 1 January 2015 as reference. In 2016, 38,550 SCD stays for 13,059 patients (2.95 hospitalizations by patient in average) were included. Among these patients, 5.6% received only BT, 38.4% developed a VOC, and 10.8% had both. The situation was more complex at a regional level with 52.2 % of patients hospitalized in "Ile de France" region, compared to the others which oscillated between 1.0 % to 5.1%. The French 5 overseas departments collectively managed 15.9 % of these patients. The standardized rate of hospitalizations is higher in overseas departments compared to the other regions in France, respectively 97.80 vs 16.98 per 100,000 inhabitants. The same difference is showed in metropolitan France, with a higher standardized rate in "Ile de France" compared to the other regions, respectively 51.95 vs 4.20 to 14.55 per 100,000 inhabitants. This descriptive study showed real differences in SCD hospital management according to regions which could help health public policy to better manage this disease.

The PMSI (medical information system database) is the French national administrative database containing medical information and patients' features for millions of hospital stays. Most medical studies only use descriptive methods to analyze patient cohorts. The objective of this study is to assess the capability of an innovative Machine Learning algorithm to discover hidden values in health data and to bring a better understanding of diseases. We applied this approach to predict the severity of Vaso-Occlusive Crisis (VOC) for patients suffering from Sickle-Cell Disease (SCD). The severity is expressed by the frequency of VOC and the average time between two VOC. We only selected hospital stays with an SCD related ICD-10 code. Patients hospitalized with a VOC code in 2014 were extracted and followed up for two years. 20 groups of comorbidities, 7 features related to the patient's medical history and 8 features about the patient's life have been identified. Data were analyzed with an Enhanced Decision Tree technique, to explain the severity of VOC with non-linear combinations of these 35 variables. 5 630 patients were included in the study. The enhanced decision tree was able to gather patients into distinct groups based on patient and disease features. 20 distinct groups were identified. The most discriminating variables for the severity of VOC were the features related to patients' history and the number of received transfusions during the follow-up, whereas the comorbidities, the age or the gender were not discriminating. This exploratory study shows that Enhanced Decision Trees are relevant to identify patient profiles from healthcare databases and have predictive capabilities. It helps identifying leverages to prevent the severity of VOC, and so to adapt the treatment given to patients according to the risk of having a VOC in the next months.

The study objective was to estimate the budget impact related to sacubitril/valsartan introduction in treatment strategy for patients with symptomatic chronic heart failure (HF) and reduced ejection fraction. The ESC 2012 HF guidelines recommend treatment with an ACE inhibitor or angiotensin II receptor blocker. Sacubitril/valsartan demonstrated a significant risk reduction of cardiovascular deaths by 20% and hospitalizations for HF by 21% versus enalapril, an ACE inhibitor.

Heart failure (HF) is a major public health issue due to its prevalence in the western world: two to three percent of the european population have HF. To improve patient care, the Montpellier university hospital (CHU) implemented in 2014 a new organisation of HF management called “optimised pathway”. The aim of our study was to compare costs and readmissions rates between 2013 and 2015, and to assess the new pathway impact. A retrospective observationnal and comparative before-after study was conducted. The clinical endpoint was the readmission rate a year after first hospitalisation for HF. CHU databases were used to characterise the patients of interest (526 and 514 for 2013 and 2015 respectively) and their stays: administrative and medico-economic database to collect relevant demographic and medical data, and cost accounting data to value the stays. Clinical and economic outcomes were measured over a one year time horizon. No significant differences were found between 2013 and 2015 regarding the one year readmission rate (18.06% in 2013, 18.48% in 2015, p = 0.86). Time intervals between first hospitalisation and readmission weren’t significantly different (p = 0.18), even if a trend could be observed (114 days in 2013 vs. 142 in 2015). No significant difference were found in terms of total length of stay (p = 0.193). Total patient care cost more in 2015 (8468€ vs. 7223€, p < 0.01), but incomes were also higher (6871€ vs. 6368€, p = 0.088), whether not enough to compensate (difference between costs and incomes for 2013 and 2015 were -855€ and -1598€ respectively, p < 0.01). There is no evidence to say that the optimised pathway reduced costs and readmissions rates, but the increasing costs can’t be assigned to it either. Researches must go further to investigate the organisational changes that occurred between 2013 and 2015.