La maladie de Fabry (MF) est une maladie lysosomale rare causée par un déficit en alpha-galactosidase. Les objectifs de l’étude étaient de décrire les caractéristiques des patients, l’utilisation des ressources de santé (URS) et les coûts chez les patients MF traités par les thérapies de substitution enzymatique (TSE) disponibles en France (agalsidase alfa et/ou agalsidase bêta). Cette cohorte rétrospective utilise la base de données nationale de santé (SNDS) pour les patients ayant reçu au moins une TSE (2009-2016). Les patients nouvellement et précédemment traités ont été suivis de l’initiation du traitement (date index) jusqu’à la fin de la période d’étude/décès. Des statistiques descriptives sur les caractéristiques sociodémographiques et cliniques ont été évaluées à la date index. Les URS/coûts (taux annuels moyens) ont été quantifiés pendant l’exposition au traitement. Au total, 453 patients traités (94 % d’adultes) ont été inclus dans l’analyse, avec un suivi moyen de 5,8 ans (± 2,4). L’âge moyen à l’inclusion était de 46,4 [ET ± 14,2] ans (59 % d’hommes) chez les adultes et de 12,7 [ET ± 4,7] ans (83 % d’hommes) chez les enfants. Chez les adultes, les lésions organiques les plus fréquentes étaient les atteintes rénales (32 %) et cardiovasculaires (30 %). Soixante-trois pour cent des patients ont été traités à la fois à l’hôpital et à domicile, 23 % uniquement à l’hôpital et 14 % uniquement à domicile. Le coût moyen annuel total (hospitalisation, consultations, médicaments) a été estimé à 175 486 €. Cette étude est la première à fournir des données de vie réelle sur l’ensemble des patients MF traités en France. Elle contribue ainsi à définir le profil clinique de ces patients, permettant de suivre leur prise en charge médicale. L’étude a également identifié les limites du système de codage actuel, l’absence de données cliniques détaillées limitant l’interprétation des données, entravant ainsi leur exploitation par des approches innovantes comme l’intelligence artificielle.

L’Angioplastie pulmonaire (AAPB) est une option dans la prise en charge des patients atteints d’hypertension pulmonaire thromboembolique chronique (HTP-TEC) inopérable ou récurrente après traitement chirurgical. Cette technique est réalisée depuis 2014 par 2 centres en France, Paris-Sud (Marie Lannelongue, Kremlin-Bicêtre) et le CHU de Grenoble. L’intervention nécessite des séances de dilatations réalisées au cours de plusieurs séjours hospitaliers. Nous avons évalué le coût hospitalier de cette technique et de ses complications en France. Tous les patients adultes avec ≥1 séjour hospitalier pour HTP associée à un acte d’AAPB entre le 01/01/2014 et le 30/06/2016 ont été inclus, à partir des données du PMSI pour les 2 centres réalisant cet acte. Les patients ont été suivis durant 6 mois, période jugée pertinente pour estimer les coûts attribuables à l’acte d’AAPB. Durant le suivi, les séjours liés à la prise en charge des complications de l’AAPB ont été distingués de ceux liés à la prise en charge de l’HTP-TEC, à partir d’une lecture médicale considérant le diagnostic du séjour, sa durée et le délai par rapport au dernier acte d’AAPB (< 30 jours). Les coûts hospitaliers (y compris les transports médicaux) ont été estimés à partir des tarifs publics pour les actes médicaux de 2014 à 2016, mis à jour en euros 2017, et valorisés selon la perspective de l’assurance maladie. 191 patients ont été inclus: 53 % d’hommes, d’âge moyen 64,3 ans et en majorité sans antécédent de traitement chirurgical (3,1 %). Le premier séjour d’AAPB était réalisé 1,1 an (0.0–10.0), en médiane, après la première hospitalisation pour une HTP. En moyenne, 2,8 séjours avec au moins un acte d’APB ont été enregistrés par patient de durée moyenne de 7,8 jours pour le premier séjour et 5,8 jours pour les séjours ultérieurs. Le coût hospitalier attribuable aux séjours d’AABP et ses complications a été estimé en moyenne (écart-type) à 8 764€ (± 3 435) par séjour, 21 245€ (±12 843) par patient etreprésentait un total de 4 057 825€ au cours de la période d’étude. Le coût variait selon la classe d’âge, la densité de la commune de résidence et la présence de comorbidités. Il diminuait avec le temps, probablement en lien avec l’expérience des opérateurs. Ces données fournissent une estimation en vie réelle du coût hospitalier de l’AAPB chez les patients atteints d’HTP-TEC en France; particulièrement pertinente alors que sa position dans l’algorithme thérapeutique est régulièrement affinée.

Background The prognosis of patients hospitalized with community-acquired pneumonia (CAP) with regards to intensive care unit (ICU) admission, short- and long-term mortality is correlated with patient’s comorbidities. For patients hospitalized for CAP, including P-CAP, we assessed the prognostic impact of comorbidities known as at-risk (AR) or high-risk (HR) of pneumococcal CAP (P-CAP), and of the number of combined comorbidities. Methods Data on hospitalizations for CAP among the French 50+ population were extracted from the 2014 French Information Systems Medicalization Program (PMSI), an exhaustive national hospital discharge database maintained by the French Technical Agency of Information on Hospitalization (ATIH). Their admission diagnosis, comorbidities (nature, risk type and number), other characteristics, and their subsequent hospital stays within the year following their hospitalization for CAP were analyzed. Logistic regression models were used to assess the associations between ICU transfer, short- and 1-year in-hospital mortality and all covariates. Results From 182,858 patients, 149,555 patients aged ≥ 50 years (nonagenarians 17.8%) were hospitalized for CAP in 2014, including 8270 with P-CAP. Overall, 33.8% and 90.5% had ≥ 1 HR and ≥ 1 AR comorbidity, respectively. Cardiac diseases were the most frequent AR comorbidity (all CAP: 77.4%). Transfer in ICU occurred for 5.4% of CAP patients and 19.4% for P-CAP. Short-term and 1-year in-hospital mortality rates were 10.9% and 23% of CAP patients, respectively, significantly lower for P-CAP patients: 9.2% and 19.8% (HR 0.88 [95% CI 0.84–0.93], p < .0001). Both terms of mortality increased mostly with age, and with the number of comorbidities and combination of AR and HR comorbidities, in addition of specific comorbidities. Conclusions Not only specific comorbidities, but also the number of combined comorbidities and the combination of AR and HR comorbidities may impact the outcome of hospitalized CAP and P-CAP patients.

Biosimilars drugs represent significant savings for the French National Insurance. The biologic drugs environment is challenging in France and health authorities create stronger incentives to encourage biosimilar prescription and the loss of patents has accelerated. The aim of the study is to analyszed the economic burden and the market share evolution of biosimilar drugs based on claims data "permanent sample of national health insurance" between 2007 (first biosimilar launch) and 2017. Retrospective observational study of biosimilars and reference products consumptions. We have extracted the health insurance reimbursement for all biosimilar and reference medicinal products commercialized in France during the period. Market shares were calculated by dividing the number of biosimilars packages delivered to all drug packages (biosimilar and references). Costs (in Euro) were extrapolated to national level by using coefficient published by the national health insurance office (CNAM). Consumptions of 11 biosimilars and 6 reference drugs were analyzed. Biosimilar expenditures were €0,7 billion compared to €6,5 billion for reference drugs. Biosimilar market shares increased from 4% to 19% between 2010 and 2017. Biosimilar drugs like filgrastim and epoetin had the most important expenditures (respectively €286 million and €186 million). In 2017, these same drugs had the most important market shares with respectively 80% and 63%. On the other side, biosimilar drugs like etanercept and insulin glargin had the smallest expenditures during the period (respectively €7,3 million and €7,2 million) and the smallest market share (4% each) in 2017. This study demonstrate that biosimilar drug market increased during the period. The "Health National Strategy » biosimilar objective for 2022: 80% of market share is achievable if the market keeps the actual trend. Patient's characteristics description could be complementary to this study, to give more information to the biosimilar landscape.

Le suivi de l’épidémiologie hospitalière des infections bactériennes permet d’observer l’émergence de souches résistantes, devenues un enjeu de santé publique critique. La qualité et l’exhaustivité de description de ces situations est cruciale et conditionne directement la pertinence des usages des données ainsi recueillies, en particulier pour le suivi de l’écologie bactérienne hospitalière. C’est dans cet objectif que notre étude propose d’examiner le contenu des bases de données hospitalières issues du Programme de médicalisation des systèmes d’information (PMSI), afin d’analyser leur cohérence concernant les différents types de résistances aux antibiotiques en fonction de la localisation des infections, des agents bactériens en cause. Les codes utilisés pour identifier les types d’infections bactériennes, les agents bactériens incriminés et les éventuelles résistances associées ont été définis à partir de la CIM-10 (Classification internationale des maladies–10e édition). Les séjours hospitaliers comportant au moins l’un de ces codes ont été extraits de la base nationale du PMSI 2014. Au total, 1 617 893 séjours correspondant à un total de 1 258 462 patients ont été extraits de la base PMSI 2014. Parmi ces séjours, 46 % comportaient un code de bactérie et 7 % de résistance. Les infections respiratoires basses étaient les plus nombreuses (32 % des séjours ; pneumonie présente dans 95 % des cas), suivies par les infections génito-urinaires (26 %), les infections intra-abdominales et diarrhées (24 %) et les infections de la peau et des tissus mous (15 %). L’analyse a mis en évidence des incohérences entre types d’infections et agents bactériens associés, ainsi qu’entre agents bactériens et résistances associées. Ces incohérences sont liées à de probables défauts de codage à la source. Dans l’état actuel du codage du PMSI, la réalisation d’une cartographie nationale de l’écologie bactérienne par type d’infection via cette base de données ne serait pas pertinente. Ces résultats soulignent néanmoins l’intérêt d’améliorer le codage des données du PMSI pour en faire un outil additionnel de surveillance épidémiologique des infections bactériennes.

Objective. - A preliminary analysis of data consistency on different types of bacterial resistance by infection site and causative agents was conducted using the French hospital discharge database (French acronym PMSI) to assess the use of the database in a national cartography tool. Material and methods. - Hospital stays in medical, surgical, and obstetrical units were extracted from the 2014 PMSI database using the ICD-10 diagnosis codes. Bacterial infections, causative agents, and resistance corresponding to these stays were also identified. Results. - Data from 1258462 patients, corresponding to a total of 1617893 stays, was extracted. Among these stays, 46% were associated with a bacteria code and 7% with a resistance code. Lower respiratory tract infections were the most frequent infections (32% of stays; pneumonia in 95% of cases), followed by genitourinary infections (26%), intra-abdominal infections and diarrhoeas (24%), and skin and soft tissue infections (15%). Inconsistencies were observed between the types of infection and associated bacteria and between bacteria and associated resistance. These inconsistencies are likely due to initial coding errors. Conclusion. - The cartography of bacterial infections cannot be developed using the data of the current PMSI coding. These results underline the need to improve the coding of PMSI data for its use as a complementary tool of epidemiological surveillance of bacterial infections. (C) 2018 Elsevier Masson SAS. All rights reserved.

Functional menorrhagia is responsive of significant economic burden, as its initial management is based on surgical strategies and implies hospitalization in case of failure or severe complications. The objective of this study is to compare 4 surgical strategies used for the treatment of functional menorrhagia in terms of costs and failure or complication avoided. A retrospective database analysis was performed using the French exhaustive national hospital discharge database (PMSI). All hospital stays from 2009 to 2015 with 4 types of menorrhagia surgery identified by CCAM codes associated with ICD-10 codes were extracted: 2nd generation (2G), 1st generation (1G), curettage, hysterectomy. Only incident 35-55 year-old women were analyzed (no surgery since 2006). Rehospitalizations related to surgery failure or severe complication were followed during at least 18 months. Hospital costs associated with these patients were estimated using the French official tariffs expressed in 2017 Euro. A cost-effectiveness analysis was performed comparing each surgical procedure to 2G, in terms of cost and rate of failure or severe complication avoided. 7,863 patients with 2G (7%), 39,935 with 1G (36%), 38,923 with curettage (35%), 23,163 with hysterectomy (21%) were included. Mean cost per patient was respectively €4,285 for curettage, €6,064 for hysterectomy, €4,182 for 2G and €3,765 for 1G. Failure or complication occurred in respectively in 17.9%, 30.6%, 10.1% and 21.5% of patients treated by 2G, curettage, hysterectomy and 1G. As compared to 2G, curettage was dominated (less effective and more expensive), hysterectomy was more expensive and more effective (ICER = €24,128 per % of patient with failure or complication avoided) and 1G was less effective and less expensive (ICER = €11,583 per % patient with failure or complication avoided). This study shows 1G and 2G techniques are cost-effective, in line with their recommended use at first stage in France.

To assess the current hospital burden of functional menorrhagia surgically treated in France. A retrospective database analysis was performed using the French exhaustive national hospital discharge database (PMSI). All hospital stays from 2009 to 2015 with 4 types of menorrhagia surgery identified by CCAM codes associated with ICD-10 codes were extracted: 2nd generation (2G), 1st generation (1G), curettage, hysterectomy. Only incident 35-55 year-old women were analyzed (no surgery since 2006). An algorithm was completed with the medical input of experts in order to exclude any patient identified as presenting comorbidities that would introduce bias in the results (breast or colorectal cancer, myoma, endometriosis…). Patients operated on before 2014/06/30 were followed at least 18 months from their surgery. Another algorithm and a medical review identified rehospitalizations related to surgery failure or complication. Hospital costs associated with these patients were estimated using the French official tariffs expressed in 2017 Euro. 7,863 patients with 2G (7%), 39,935 with 1G (36%), 38,923 with curettage (35%), 23,163 with hysterectomy (21%) were included. Whereas the global evolution was -13% from 2009 to 2015, it was +80% for 2G and +5% for 1G, and -37% for curettage and -15% for hysterectomy. The 18-month failure rate ran from 2.8% for hysterectomy to 9.9% for 2G, 12.7% for 1G and 20.6% for curettage, whereas the 18-month complication rate was 1.4% for curettage, 1.5% for 1G, 1.9% for 2G and 5.3% for hysterectomy. The trend was similar at 24 and 60 months. The 18-month median cost per patient varied from €782 [Q1 741-Q3 2,732] for curettage to €1,059 [913-2,075] for 1G, €1,173 [1,002-2,231] for 2G, €3,090 [2,909-4,189] for hysterectomy. This study shows that mini-invasive 1G and 2G techniques low complication rates and costs are in line with their recommended use at first stage.

To assess hospitalization costs associated with HIV in France in 2013/2014. The study was realized using the French Medical Information System database (PMSI, “Programme de Médicalisation des Systèmes d’Information”). All hospital stays in 2013 with HIV ICD-10 (International Classification of Diseases) codes (B20*, B21*, B22*, B23*, B24*) were extracted from this comprehensive database. Through an algorithm and a medical review, stays were classified as directly associated to HIV (low hypothesis, LH) or probably associated to HIV (High hypothesis, HH). HH included directly and probably associated to HIV stays. Each patient was followed for one year (e.g. from March 2013 to March 2014) from its first stay in hospital. Only hospitalizations excluding day care inpatient stays contributed to this evaluation. The total annual cost of hospitalization and the median annual cost per patient were estimated. Valuation was performed considering French official tariffs for 2013 and 2014 and expressed in 2016 euros. In this database, 4,095 patients with at least one hospitalisation were identified in the LH and 10,236 in the HH; the overall number of stays were 6,855 and 19,910, corresponding to 1.7 and 1.9 stays per patient, respectively. Overall, the economic annual cost of hospitalizations associated with HIV was estimated at €45 million in the LH, with a median cost per patient of €5,271 (€113 million in the HH, median cost of €5,226 per patient). 49.8% of hospitalised HIV patients in LH (33.4%, HH) had at least one opportunistic infection, representing 68.9% of the total annual cost (55.9%, HH). Hospital annual related costs associated with HIV represent at least €45 million. Hospitalizations with opportunistic infections represent half this cost and could be reduced by improving the HIV care patients cascade: earlier HIV diagnosis, earlier linkage to care, and better retention in care to avoid opportunistic occurrences.

Atrial fibrillation (AF) is associated with numerous cardiovascular (CV) complications. The objective of this study was to estimate the national annual burden of CV complications in patients with AF in hospitals. All patients hospitalized in 2012 with a diagnosis of AF were identified from the French National public/private hospital database (PMSI). Comorbidities and medical data were collected during a 5-year look-back period and used to calculate stroke risk score (CHA2DS2-VASc). Reasons for CV-related hospitalization, emergency admission, rehabilitation and death at discharge were described. Costs of acute care were determined using Diagnosis Related Groups and corresponding tariffs (2012 Euros). In total, 533,044 AF patients were hospitalized for any reason. Mean age was 78.0 (±11.4) years, 53% were males and mean CHA2DS2-VASc score was 4.0 (±1.8). CV-related hospitalizations occurred in 267,681 patients: 34% for AF management care, 28% for heart failure, 8% for strokes, 7% for other ischemic heart diseases, 5% for vascular diseases, 4% for transient ischemic attacks and systemic embolisms, 2% for bleedings and 12% for other CV reasons. In CV-related hospitalizations, 45% of patients required emergency admission, especially patients hospitalized for strokes (75%) and for bleedings (67%). Death at hospital occurred in 6% of patients with a CV-related hospitalization: 17% in patients with strokes and 10% in patients with bleedings. Rehabilitation was needed for 34% patients with non-fatal strokes with a mean length of 48 days. The annual total cost (acute care and rehabilitation) for all hospitalized CV events during the year 2012 was €1.94 billion. Among this, heart failure represented €518m, AF management €306m, strokes €291m and hemorrhages €48m. A half-million AF patients were hospitalized in 2012 in France. CV-related hospitalizations involved over a quarter-million AF patients for a global burden of almost €2 billion.

Complications are common following kidney transplantation. The French Hospital National Database (PMSI) allows patients follow-up through their hospitalization reports. This study assessed 4-year incidences of complications after kidney transplantation and estimated their economic impact, in particular renal complications. From the years 2006-2010, hospitalization data for all patients with kidney transplantation in year 2006 were extracted. Patients' hospital stays were followed during a period of 4 years. Major hospitalizations of interests were pooled in renal, cardiovascular and infectious complications. Renal complications were detailed in graft loss, acute renal insufficiency (ARI), acute rejection (AR), recurrence of initial nephropathy (RIN), chirurgical complication and, others. Incremental costs were estimated by subtracting transplantation-year mean cost of patients without any renal complications to the one of those with renal complications the same year. Mean annual costs of patients with graft loss were assessed separately for each year. Costs were calculated according to the 2010 National Hospital Tariff and National Scale. A total of 2392 patients with kidney transplantation in 2006 were identified. A total of 61.5% were males and mean age was 45.0(±14.9) years old. 4-year cumulative incidences of renal, cardiovascular and infectious complications were 54.8% (n=1,311), 20.8% (n=498) and 19.8% (n=474). Incidences of ARI, AR, RNI, chirurgical complication and, others renal complications were 24.5%, 21.1%, 4.1%, 3.1% and 22.4%, respectively. Compared to other patients without any renal complication (First-year mean costs: €25,170; Q1:€17,341–Q3:€27,649), the corresponding incremental costs for renal complications were €7,046, €10,376, €10,238, €7,874 and, €5,668. During the 4-year period, graft loss occurred in 4.4% patients. Annual mean costs of graft loss for the first, second, third and fourth years were €32,159 (Q1:€11 723–Q3:€41,890) (n=105), €19,085 (n=64), €25,269 (n=52) and, €20,780 (n=41), respectively. After transplantation, short-term renal complications are frequent and expensive. Intensive interventions on renal preservation and graft loss prevention are needed.

The prevalence of atrial fibrillation (AF) in France approaches one million people. The major complication associated with AF is stroke. Current anticoagulation options for stroke prevention increase the risk of hemorrhages. Objectives were to estimate the 2-year cumulative incidence and costs of hospitalizations for strokes and hemorrhages in adults hospitalized for AF and eligible for stroke prevention. Data for patients with an AF-related hospitalization in 2008 were extracted from the French Hospital National Database (PMSI). Risk scores (i.e.CHADS2; range:0–6) were calculated from 2006–2008 data. Patient eligible for stroke prevention with anti-coagulants (i.e.CHADS2≥1) were selected for the follow-up analysis. Strokes and hemorrhages hospitalizations were identified according to ICD-10 codes. Strokes severity was based on rehabilitation length and death. Cumulative incidence was calculated by the number of new hospitalizations during the 2-year period divided by the number of patients. Mean hospital costs were calculated from to the 2011 National Hospital Tariff for acute and rehabilitation care. A total of 61,582 AF patients were identified. Mean age was 75.0(±11.0) years old and mean CHADS2 was 1.90(±0.99). 2-year cumulative incidences of any strokes and hemorrhages were 3.21% (ischemic/60%; hemorrhagic/24%; unspecified/16%) and 5.31% (gastro-intestinal/26%; intracranial/5%; others/69%), respectively. Mean costs of ischemic and hemorrhagic strokes were €4,848 and €7,183 (mild), €10,909 and €14,298 (moderate), €29,065 and €29,701 (severe) and, €6,035 and €4,590 (fatal), respectively. Mean costs of hemorrhages were €3,601 and €7,331 for gastro-intestinal and intracranial localizations and, €3,941 and €2,552 for others major and non-major hospitalized bleeds. Frequencies and cost of hospitalized hemorrhages appear important to be taking into account in the global burden of AF. This data should be useful for future French pharmacoeconomic evaluations of new oral anti-coagulants. Thus, this real world data study may be helpful to assess consistency of patients' features within recent published clinical trials.