Search

Objectives: We investigated whether the risk of death among noncoronavirus disease 2019 critically ill patients increased when numerous coronavirus disease 2019 cases were admitted concomitantly to the same hospital units. Design: We performed a nationwide observational study based on the medical information system from all public and private hospitals in France. Setting: Information pertaining to every adult admitted to ICUs or intermediate care units from 641 hospitals between January 1, 2020, and June 30, 2020 was analyzed. Patients: A total of 454,502 patients (428,687 noncoronavirus disease 2019 and 25,815 coronavirus disease 2019 patients) were included. Interventions: For each noncoronavirus disease 2019 patient, pandemic exposure during their stay was calculated per day using the proportion of coronavirus disease 2019 patients among all patients treated in ICU. Measurements and Main Results: We computed a multivariable logistic regression model to estimate the influence of pandemic exposure (low, moderate, and high exposure) on noncoronavirus disease 2019 patient mortality during ICU stay. We adjusted on patient and hospital confounders. The risk of death among noncoronavirus disease 2019 critically ill patients increased in case of moderate (adjusted odds ratio, 1.12; 95% CI, 1.05–1.19; p < 0.001) and high pandemic exposures (1.52; 95% CI, 1.33–1.74; p < 0.001). Conclusions: In hospital units with moderate or high levels of coronavirus disease 2019 critically ill patients, noncoronavirus disease deaths were at higher levels.

RATIONALE: Nurse-to-nurse familiarity at work should strengthen the components of team working and enhance its efficiency. However, its impact on patient outcomes in critical care remains poorly investigated. OBJECTIVES: To explore the role of nurse-to-nurse familiarity on inpatient deaths during intensive care unit stay. METHODS: Retrospective observational study in eight adult academic intensive care units between 01/01/2011 and 31/12/2016. MEASUREMENTS AND MAIN RESULTS: Nurse-to-nurse familiarity was measured across day and night 12-hour daily shifts as the mean number of previous collaborations between each nursing team member during previous shifts within the given Intensive Care Unit (suboptimal if\textless50). Primary outcome was a shift with at least one inpatient death, excluding death of patients with a decision to forego life-sustaining therapy. A multiple modified Poisson regression was computed to identify the determinants of mortality per shift, taking into account intensive care unit, patients' characteristics, patient-to-nurse and patient-to-assistant nurse ratios, nurse experience length and workload. A total of 43,479 patients were admitted of whom 3,311 (8%) died. Adjusted model showed a lower risk of a shift with mortality when nurse-to-nurse familiarity increased in the shift (relative risk 0.90 [0.82-0.98] 95%confidence intervals per 10 shifts, p=0.012). Low nurse-to-nurse familiarity during the shift combined with suboptimal patient-to-nurse and assistant-nurse ratios (suboptimal if \textgreater2.5 and \textgreater4, respectively) were associated with increased risk of shift with mortality (1.84 [1.15-2.96], p\textless0.001). CONCLUSIONS: Shifts with low nurse-to-nurse familiarity were associated with an increased risk of patient deaths.

The annual prevalence of metastatic neuroendocrine tumours (mNETs) is rising, leading to significant healthcare costs. The present study aimed to describe healthcare resource use (HRU) and the corresponding costs among patients with mNETs, according to primary tumour location, functioning status and type of treatments. The LyREMeNET study included consecutive mNET patients with a diagnosis performed between January 2010 and December 2017, who were seen at least once in the ENETS center of excellence in Lyon. The median HRU and costs per patient were estimated, up to 3 years before and after the diagnosis. The Cancer database of the center was linked to the French national health data system. HRU and related costs were described per person per month (PPPM). Among 316 patients presenting with a mNET, 48.4% had a small-intestinal mNET, 32.3% had a pancreatic mNET and 39.2% had carcinoid syndrome. The mean overall cost increased from €615 to €2875 PPPM between the years preceding and following the diagnosis, and remained above €2500 in the two subsequent years. The two main cost drivers of total healthcare expenditure were drugs (€1161) and hospital stay (€662). Median costs of mNETs arising from pancreas and small intestine were €2325 and €2540 PPPM, respectively. Costs were higher in patients with a functional mNET (€2807 PPPM for carcinoid syndrome) and during peptide receptor radionuclide therapy (PRRT) (€8835 PPPM). The highest overall cost was found during the first year following the diagnosis. Cost of care was higher for small intestine mNETs, for functional mNETs and during peptide receptor radionuclide therapy.

Hospital beds management is critical for the quality of patient care, while length of inpatient stay is often estimated empirically by physicians or chief nurses of medical wards. Providing an efficient method for forecasting the length of stay (LOS) is expected to improve resources and discharges planning. Predictions should be accurate and work for as many patients as possible, despite their heterogeneous profiles. In this work, a LOS prediction method based on deep learning and embeddings is developed by using generic hospital administrative data from a French national hospital discharge database, as well as emergency care. Data concerned 497 626 stays of 304 931 patients from 6 hospitals in Lyon, France, from 2011 to 2019. Results of a 5-fold cross-validation showed an accuracy of 0.73 and a kappa score of 0.67 for the embeddings method. This outperformed the baseline which used the raw input features directly.

Data about incidence and severity of reoperations up to 6 months after bariatric surgery are currently limited. The aim of this cohort study was to evaluate the incidence and severity of reoperations after initial bariatric surgical procedures and to compare this between the 3 most frequent current surgical procedures (sleeve, gastric bypass, gastric banding). Nationwide observational cohort study using data from French Hospital Information System (2013–2015) to evaluate incidence and severity of reoperations within 6 months after bariatric surgery. Hazard ratios (HR) of longitudinal comparison between historical propensity-matched cohorts were estimated from a Fine and Gray’s model using competing risk of death. Cumulative reoperation rates increased from postoperative day-30 to day-180. Consequently, 31.1 to 90.0% of procedures would have been missed if the reoperation rate was based solely on a 30-day follow-up. Reoperation rate at 6 months was significantly higher after gastric bypass than after sleeve (HR 0.64; IC 95% [0.53–0.77]) and corresponded to moderate-risk reoperations (HR 0.65; IC 95% [0.53–0.78]). Reoperation rate at 6 months was significantly higher after gastric banding than after sleeve (HR 0.08; IC 95% [0.07–0.09]) and corresponded to moderate-risk reoperations (HR 0.08; IC 95% [0.07–0.10]). Cumulative incidence of reoperations increased from 30 days to 6 months after sleeve, gastric bypass, or gastric banding and corresponded to moderate-risk surgical procedures. Consequently, 30-day reoperation rate should no longer be considered when evaluating complications and surgical performance after bariatric surgery.

International audience; OBJECTIVES: The objective of the study is to evaluate the performance of high-dimensional propensity scores (hdPSs) for controlling indication bias as compared with propensity scores (PSs) in surgical comparative effectiveness studies. STUDY DESIGN AND SETTING: Patients who underwent interventional transcatheter aortic valve implantation (TAVI) or surgical aortic valve replacement (SAVR) between 2013 and 2017 were included from the French nationwide hospitals. At each hospital level, matched pairs of patients who underwent TAVI and SAVR were formed using PSs, considering 20 patient baseline characteristics, and hdPSs, considering the same patient characteristics and 300 additional variables from procedure and diagnosis codes the year before surgery. We compared death, reoperation, and stroke up to 3 years between TAVI and SAVR using Cox or Fine and Gray models. RESULTS: Before matching, 12 of 20 patient characteristics were imbalanced between the included patients who underwent TAVI and SAVR. No significant imbalance persisted after matching with both methods. Hazard ratio of 1-year death, reoperation, and stroke was 1.3 [1.1; 1.4], 1.6 [1.1; 2.4], and 1.4 [1.2; 1.7] for TAVI relative to SAVR with PSs (n = 9,498 pairs) and 1.1 [1.0; 1.3], 1.3 [0.8; 2.0], and 1.3 [1.0; 1.6] with hdPSs (n = 7,157). CONCLUSION: HdPS estimations were more consistent with results seen in randomized controlled trials. The HdPS is a promising alternative for the PS to control indication bias in comparative studies of surgical procedures.

Background France is among the countries showing fastest growth of thyroid cancer (TC) incidence and highest incidence rates in Europe. This study aimed to clarify the temporal and geographical variations of TC in France and to quantify the impact of overdiagnosis. Methods We obtained TC incidence data in 1986–2015, and mortality data in 1976–2015, for eight French departments covering 8% of the national population, and calculated the age-standardised rates (ASR). We estimated the average annual percent changes (AAPC) of TC incidence, overall and by department and histological subtype. Numbers and proportions of TC cases attributable to overdiagnosis were estimated by department and period, based on the comparison between the shape of the age-specific curves with that observed prior to changes in diagnostic practice. Results During 1986–2015, there were 13,557 TC cases aged 15–84 years. Large variations of TC incidence were observed across departments, with the highest ASR and the fastest increase in Isere. Papillary subtype accounted for 82.8% of the cases, and presented an AAPC of 7.0% and 7.6% in women and men, respectively. Anaplastic TC incidence decreased annually 3.0% in women and 0.8% in men. Mortality rates declined consistently for all departments. The absolute number (and proportion) of TC cases attributable to overdiagnosis grew from 1074 (66%) in 1986–1995 to 3830 (72%) in 2006–2015 in women, and varied substantially across departments. Conclusions Overdiagnosis plays an important role in the temporal and regional variations of TC incidence in France. Monitoring the time trends and regulating the regional healthcare practice are needed to reduce its impact.

International audience; PURPOSE: The volume of bariatric surgery has significantly increased over the past decade with concomitant postoperative outcomes improvement. The goal of this nationwide study was to estimate the volume-outcome relationship in bariatric surgery at the hospital level. MATERIALS AND METHODS: A cross-sectional analysis of all patients who underwent bariatric surgery procedure in France from January 2011 to December 2014 was designed. Volume-outcome relationship was analyzed using generalized estimating equations. RESULTS: We identified 184,332 inpatient stays for bariatric surgical procedures performed in 606 hospitals. Health care institutions performing more than 200 bariatric cases per year were significantly associated with shorter average length of stay (p~\textless~0.001) and less frequent need for intensive or critical care unit (p~=~0.003) during the index stay in comparison with lower volume institutions. Reoperations rate increased from 3.1% [95% CI, 2.8-3.3] (n~=~5627) at 1~month to 4.9% [4.6-5.2] at 3~months and 8.2% [7.8-8.7] at 6~months. The risk of reoperation after gastric bypass was 1.37 times less frequent in higher volume institutions (>=~200 inpatient stays per year, p~=~0.003), while it was 1.26 times more frequent after gastric banding in higher volume institutions (p~=~0.057) and was unaltered regarding sleeve gastrectomy (p~=~0.819). CONCLUSION: This study showed for the first time in bariatric surgery that reoperation rate after gastric bypass or sleeve significantly increased at 3 and 6~months postoperatively. Health care institutions performing more than 200 bariatric cases per year were significantly associated with improved postoperative outcomes and less frequent need for reoperation.

Introduction Le Tour de France (TDF) rassemble chaque annee en juillet plus de 10 millions de spectateurs le long des routes exposes a de fortes chaleurs. L’objectif de cette etude etait de determiner si le passage du TDF etait associe a une augmentation du nombre de patients admis pour des pathologies liees a la chaleur dans les services d’urgences situes a proximite du trace du TDF. Methodes Nous avons conduit une etude retrospective a partir des donnees nationales extraites des resumes de passages aux urgences des etablissements situes a proximite des etapes du TDF en 2016 et 2017. Pour chacun des services d’urgences concernes, nous avons compare le nombre de patients admis pour pathologies liees a la chaleur le jour de passage du TDF avec un jour controle dans le meme service, apparie sur le « Heat Index » (indice de chaleur ressentie) et la saisonnalite (mois de juillet et aout). Resultats Au total, 34 etapes du TDF correspondant a 6041 kilometres de routes et 30 343 patients admis dans 240 services d’urgences ont ete analyses. Il n’y avait pas de difference concernant le nombre d’admissions par jour aux urgences les jours de passage du TDF pour des pathologies liees a la chaleur par rapport aux jours controles, que ce soit parmi tous les patients (3,2 versus 3,4, p-value = 0,640) ou uniquement ceux residant a proximite des urgences (2,7 versus 2,7, p-value = 0,671). De meme, aucune difference n’etait observee pour l’analyse des etapes avec une temperature maximale superieure a 30°C Celsius et pour l’analyse des admissions toutes pathologies confondues. Discussion/Conclusion L’exposition a la chaleur durant le TDF n’augmentait pas le risque de pathologie liee a la chaleur necessitant un recours aux soins d’urgence. Parmi les explications possibles a ce resultat, nous pouvons citer le role des services de premiers secours lors de la course et la bonne adherence des spectateurs aux mesures de prevention.

OBJECTIVE The aim of the study was to investigate whether patients who undergo surgery in hospitals experiencing significant length of stay (LOS) reductions over time are exposed to a higher risk of severe adverse events in the postoperative period. SUMMARY BACKGROUND DATA Surgical care innovation has encouraged hospitals to shorten LOS under financial pressures with uncertain impact on patient outcomes. METHODS We selected all patients who underwent elective colectomy or urgent hip fracture repair in French hospitals between 2013 and 2016. For each procedure, hospitals were categorized into 3 groups according to variations in their median LOS as follows: major decrease, moderate decrease, and no decrease. These groups were matched using propensity scores based on patients' and hospitals' potential confounders. Potentially avoidable readmission for severe adverse events and death at 6 months were compared between groups using Cox regressions. RESULTS We considered 98,713 patients in 540 hospitals for colectomy and 206,812 patients in 414 hospitals for hip fracture repair before matching. After colectomy, patient outcomes were not negatively impacted when hospitals reduced their LOS [hazard ratio (95% confidence interval): 0.93 (0.78-1.10)]. After hip fracture repair, patients in hospitals with major decreases in LOS had a higher risk of severe adverse events [1.22 (1.11-1.34)] and death [1.17 (1.04-1.32)]. CONCLUSIONS Patients who underwent surgical procedures in hospitals experiencing major decreases in LOS were demonstrated worse postoperative outcomes after urgent hip fracture repair and not after elective colectomy. Development of care bundles to enhance recovery after emergency surgeries may allow better control of LOS reduction and patient outcomes.

Introduction Il existe un exces de chirurgie thyroidienne pour nodule benin. La prise en charge des nodules est heterogene malgre des recommandations qui placent les resultats de la cytoponction a l’aiguille fine (CAF) au centre de la decision chirurgicale. Notre objectif etait d’evaluer si le respect du parcours de soins recommande avait une influence sur la pertinence de la realisation d’une chirurgie de nodule thyroidien pour cancer. Methodes Nous avons selectionne, dans les donnees de l’echantillon general des beneficiaires, les patients adultes operes d’un nodule thyroidien entre 2012 et 2015. L’exposition etait le parcours de soins dans l’annee precedant la chirurgie. Trois parcours ont ete consideres : le parcours 1, contenant une CAF et une consultation d’endocrinologie, et le 2, contenant une CAF, etaient consideres optimaux. Le parcours 3, sans CAF, etait non optimal. Le critere de jugement principal etait la nature maligne du nodule. L’association entre les parcours et la probabilite de cancer thyroidien a ete analysee par regression logistique multivariee. Resultats Parmi les 1080 patients inclus dans l’etude, 404 (37,4 %) patients ont presente un parcours de soins optimal, avec 197 (18,2 %) parcours 1 et 207 (19,2 %) parcours 2, alors que le parcours etait non optimal pour 676 (62,6 %) patients. Un total de 257 (23,8 %) cancers thyroidiens ont ete diagnostiques, avec respectivement 72 (36,5 %), 66 (31,9 %) et 119 (17,6 %) cas parmi les parcours 1, 2 et 3. Les odds de cancers thyroidiens etaient superieurs dans le parcours 1 (OR : 2,67 ; IC95 % : 1,88–3,81 ; p-value Discussion/Conclusion Moins d’un tiers des chirurgies etaient associees a un diagnostic de cancer. La realisation d’une CAF etait associee a des indications chirurgicales plus pertinentes mais cet examen etait insuffisamment realise. Une tendance a l’amelioration des pratiques a cependant ete observee au cours des annees.

Introduction En pharmaco-epidemiologie, les scores de propension a haute dimension (hdPS) ont prouve leur utilite pour tenir compte de facteurs confondants non mesures. Notre objectif etait d’evaluer la performance des hdPS par rapport aux scores de propension classiques (PS) pour controler le biais d’indication lors de la comparaison de resultats entre des patients beneficiant de l’implantation d’une valve aortique par voie percutanee (TAVI) et des patients beneficiant d’un remplacement chirurgical de valve aortique (SAVR). Methodes A partir de la base PMSI nationale MCO, nous avons selectionne les patients ayant beneficie de l’une de ces deux procedures entre 2013 et 2017. Au sein de chaque hopital, les sejours realises pour TAVI ou SAVR ont ete apparies par : –PS considerant 20 caracteristiques patients ; –hdPS considerant 20 caracteristiques patients et 300 variables issues des codes actes et diagnostiques de l’annee precedant l’hospitalisation. En utilisant chacune de ces methodes, les risques observes de deces, reprise chirurgicale et accident vasculaire cerebral de un mois a trois ans ont ete compares entre TAVI et SAVR avec des modeles de Cox ou Fine et Gray. Resultats Avant appariement, 12 des 20 caracteristiques patients presentaient un desequilibre entre les groupes. Apres appariement, les groupes etaient equilibres sur toutes les caracteristiques des patients pour les deux methodes. Avec la methode PS, les risques d’evenements etaient plus eleves parmi les patients TAVI par rapport aux SAVR alors qu’avec la methode hdPS, les ecarts de risque entre TAVI et SAVR etaient moins marques (ex : deces a 1 an HR 1,3, IC95 % [1,1, 1,4] avec la methode PS et 1,1 [1,0, 1,3] avec hdPS ; reprise chirurgicale a 1 an 1,6 [1,1, 2,4] avec PS et 1,3 [0,8, 2,0] avec hdPS), se rapprochant davantage des resultats obtenus experimentalement dans le cadre d’essais cliniques. Discussion/Conclusion La methode hdPS semble une alternative prometteuse au PS pour controler le biais d’indication dans les etudes de comparaison d’efficacite menees a partir des bases hospitalieres nationales.

Introduction Les systemes de soins actuels demeurent soumis a des risques d’erreurs qui peuvent avoir des consequences dramatiques pour les patients. Comprendre la variation des resultats chirurgicaux et comment dispenser des soins surs est imperatif pour leur amelioration. Methodes Nous avons mene un essai randomise en cluster dans 40 hopitaux francais pour evaluer l’impact de la mise en place d’un suivi prospectif des resultats chirurgicaux avec retour regulier aux equipes de bloc. Suite a un etat des lieux (2014–2015), les hopitaux ont ete repartis de maniere aleatoire pour beneficier ou non des cartes de controle (2017–2018). Le critere principal etait un composite d’evenements indesirables graves dans les 30 jours suivant la chirurgie extrait du PMSI. Nous avons estime le ratio des odds ratios (ROR) pour comparer l’evolution des resultats chirurgicaux entre les hopitaux des groupes intervention et controle en ajustant sur les caracteristiques des patients. Resultats Parmi 159 688 patients operes, 155 362 (97,3 %) ont ete inclus dans l’analyse. Une reduction significative d’evenements indesirables graves a suivi la mise en place des cartes de controle dans les hopitaux intervention versus controle (ROR : 0,89 ; IC95 % : 0,83 a 0,96 ; p = 0,001). Les memes tendances etaient observees concernant les deces des patients (0,84 ; 0,71 a 0,99 ; p = 0,042), leur passage en reanimation (0,85 ; 0,76 a 0,94 ; p = 0,001) et leur reoperation (0,91 ; 0,82 a 1,00 ; p = 0,056). Discussion/Conclusion La mise en œuvre de cartes de controle avec retour regulier aux equipes de bloc sur leurs resultats etait associee a une reduction de la morbi-mortalite des patients en chirurgie digestive.

Background: Idiopathic pulmonary fibrosis (IPF) and combined pulmonary fibrosis with emphysema (CPFE) are associated with an increased risk of pulmonary hypertension (PH). Few data are available regarding PH severity and correlation with respiratory physiology. Aim: To evaluate baseline respiratory and hemodynamic characteristics of patients with IPF-PH and CPFE-PH. Methods: Patients with a diagnosis of IPF or CPFE were analysed from a prospective multicenter observational study (HYPID) conducted in French expert centers for rare pulmonary diseases and/or PH (NCT01443598). Severe PH was defined as mean pulmonary artery pressure (mPAP) > 35 mmHg or cardiac index (CI) Results: Out of 220 patients included in the cohort, 61 had IPF-PH and 75 CPFE-PH. Patients were 74% males, with a mean age of 69 ± 9 years. The mean FVC at baseline was 75 ± 25%, and mean DLco was 25 ± 10%. NYHA class was I-II in 18% of patients, III in 56%, IV in 24%, N/A in 2%. Six-min walk distance was 276 ± 145 m. Hemodynamic characteristics were mPAP 38 ± 9 mmHg, CI 2.5 ± 0.6 L/min/m², and pulmonary vascular resistance (PVR) 8.7 ± 3 WU. Kco was lower in CPFE-PH than in IPF-PH (table); other parameters were comparable between groups. DLco at baseline significantly correlated with mPAP (p=0.01 in IPF and 0.04 in CPFE group). DLco also correlated with PVR and CI in IPF-PH (p=0.001 and 0.01 respectively), but not in CPFE-PH. Conclusion: Hemodynamic severity is comparable in patients with IPF-PH and CPFE-PH, and correlates with DLco. Further study is needed to evaluate the prognostic significance of hemodynamic severity.

cited By 3; International audience; Objective: To compare the clinical outcomes and direct costs at 5 years between transcatheter aortic valve implantation (TAVI) and surgical aortic valve replacement (SAVR) using real-world evidence. Methods: We performed a nationwide longitudinal study using data from the French Hospital Information System from 2009 to 2015. We matched, inside hospitals, 2 cohorts of adults who underwent TAVI or SAVR during 2010 on propensity score based on patient characteristics. Outcomes analysis included mortality, morbidity, and total costs and with a maximum 60-month follow-up. Clinical outcomes were compared between cohorts using hazard ratios (HRs) estimated from a Cox proportional hazards model for all-cause death, and from Fine and Gray's competing risk model for morbidity. Results: Based on a cohort of 1598 patients (799 in each group) from 27 centers, a higher risk of death was observed after 1 year with TAVI compared with SAVR (16.8% vs 12.8%, respectively; HR, 1.33; 95% confidence interval [CI], 1.02-1.72) and was sustained up to 5 years (52.4% vs 37.2%; HR, 1.56; 95% CI, 1.33-1.84). At 5 years, the risk of stroke was increased (HR, 1.64; 95% CI, 1.07-2.54) as was myocardial infarction (HR, 2.30; 95% CI, 1.12-4.69) and pacemaker implantation (HR, 2.40; 95% CI, 1.81-3.17) after TAVI. The hospitalization costs per patient at 5 years were €69,083 after TAVI and €55,687 after SAVR (P

BackgroundQuality improvement and epidemiology studies often rely on database codes to measure performance or impact of adjusted risk factors, but how validity issues can bias those estimates is seldom quantified.ObjectivesTo evaluate whether and how much interhospital administrative coding variations influence a typical performance measure (adjusted mortality) and potential incentives based on it.DesignNational cross-sectional study comparing hospital mortality ranking and simulated pay-for-performance incentives before/after recoding discharge abstracts using medical records.SettingTwenty-four public and private hospitals located in FranceParticipantsAll inpatient stays from the 78 deadliest diagnosis-related groups over 1 year.InterventionsElixhauser and Charlson comorbidities were derived, and mortality ratios were computed for each hospital. Thirty random stays per hospital were then recoded by two central reviewers and used in a Bayesian hierarchical model to estimate hospital-specific and comorbidity-specific predictive values. Simulations then estimated shifts in adjusted mortality and proportion of incentives that would be unfairly distributed by a typical pay-for-performance programme in this situation.Main outcome measuresPositive and negative predictive values of routine coding of comorbidities in hospital databases, variations in hospitals’ mortality league table and proportion of unfair incentives.ResultsA total of 70 402 hospital discharge abstracts were analysed, of which 715 were recoded from full medical records. Hospital comorbidity-level positive predictive values ranged from 64.4% to 96.4% and negative ones from 88.0% to 99.9%. Using Elixhauser comorbidities for adjustment, 70.3% of hospitals changed position in the mortality league table after correction, which added up to a mean 6.5% (SD 3.6) of a total pay-for-performance budget being allocated to the wrong hospitals. Using Charlson, 61.5% of hospitals changed position, with 7.3% (SD 4.0) budget misallocation.ConclusionsVariations in administrative data coding can bias mortality comparisons and budget allocation across hospitals. Such heterogeneity in data validity may be corrected using a centralised coding strategy from a random sample of observations.

Introduction Les avancees medicales et les incitations financieres ont encourage les hopitaux a reduire leur Duree de sejour (DS), avec un impact incertain sur la securite des patients. Notre objectif etait d’investiguer si les patients operes dans des hopitaux reduisant significativement leur DS au cours du temps etaient exposes a un risque plus eleve d’evenements indesirables graves apres l’operation. Methodes Nous avons selectionne dans la base PMSI nationale MCO les patients operes dans un etablissement francais entre 2013 et 2016 pour une colectomie programmee ou une chirurgie en urgence pour fracture de l’extremite superieure du femur (FESF). Pour chacune des deux chirurgies, les hopitaux ont ete categorises en trois groupes selon l’evolution de leur DS mediane (diminution majeure, diminution moderee, pas de diminution). Ces groupes ont ete apparies par scores de propension considerant les caracteristiques des patients et des hopitaux. La re-hospitalisation potentiellement evitable du patient pour evenement indesirable grave a six mois a ete comparee entre les groupes avec des regressions de Cox. Resultats Avant appariement, 98 713 patients dans 540 hopitaux ont ete consideres pour la colectomie et 206 812 patients dans 414 hopitaux pour la chirurgie pour FESF. Apres la colectomie, les resultats chirurgicaux des patients n’etaient pas negativement impactes quand les hopitaux reduisaient leur DS (Hazard Ratio [Intervalle de Confiance a 95 %]: 0,93 [0,78–1,10]). Apres la chirurgie pour FESF, les patients operes dans les hopitaux avec diminution majeure de leur DS avaient un risque plus eleve d’evenements indesirables graves (1,22 [1,11–1,34]) et de deces (1,17 [1,04–1,32]). Discussion/Conclusion Les patients operes dans les hopitaux avec reduction majeure de leur DS presentaient de plus mauvais resultats apres la chirurgie en urgence pour FESF et pas apres la colectomie programmee. Le developpement de « care bundles » pour ameliorer la recuperation apres des chirurgies en urgence pourrait permettre un meilleur controle des resultats chirurgicaux des patients en cas de reduction des DS.

Introduction Dans le contexte de la sclerodermie systemique, les patients porteurs a la fois d’une hypertension pulmonaire et d’une pneumopathie interstitielle diffuse (SSc-HTP-PID) ont un pronostic plus defavorable que ceux porteurs d’une hypertension arterielle pulmonaire (HTAP) isolee, sans pneumopathie interstitielle (SSc-HTAP). Il n’y a pas a ce jour de preuve en faveur d’une efficacite des traitements specifiques de l’HTAP chez les patients SSc-HTP-PID. L’objectif de notre etude etait de comparer la survie et la reponse au traitement entre SSc-HTAP et SSc-HTP-PID. Methodes Nous avons mene une analyse retrospective de 128 patients (66 SSc-HTP-PID, 62 SSc-HTAP), issus de 15 centres, dont l’HTAP a ete diagnostiquee par catheterisme cardiaque droit entre 2001 et 2016. Tous les patients ont recu un traitement specifique de l’HTAP. Le diagnostic de pneumopathie interstitielle diffuse (PID) etait confirme ou infirme par la tomodensitometrie. Resultats Au diagnostic, l’HTAP etait plus severe dans le groupe SSc-HTAP resistances vasculaires pulmonaires moyennes de 8,7 contre 5,7 UW (p = 0,0005). Les patients porteurs d’une PID avaient un trouble ventilatoire restrictif et une hypoxemie plus marques que dans le groupe SSc-HTAP (capacite vitale forcee moyenne de 68 % contre 100 %, p Conclusion Ce travail confirme le pronostic plus severe de la SSc-HTP-PID par rapport a la SSc-HTAP. La reponse aux traitements specifiques de l’HTAP serait decevante dans la SSc-HTP-PID en ce qui concerne la dyspnee, mais semble equivalente sur le plan hemodynamique a ce qui est observe dans la SSc-HTAP.

The recommended treatment for acute calculous cholecystitis combines antibiotics and cholecystectomy. To reduce morbidity and mortality, guidelines recommend early cholecystectomy. However, the optimal timing for surgery on first admission remains controversial. This study aims to determine the best timing for cholecystectomy in patients presenting with acute calculous cholecystitis. The French national health-care database was analyzed to identify all patients undergoing cholecystectomy for acute cholecystitis during the same hospital stay between January 2010 and December 2013. Data regarding patients, procedures, and hospitals characteristics were collected. The relationship between surgery’s timing and clinical outcome was evaluated by multiple logistic regressions. Overall, 42,452 patients from 507 hospitals were included in the study. Postoperative complications requiring invasive treatment occurred in 961 patients (2.3 %), and the mortality rate was 1.1 %. Adverse postoperative outcomes—intensive care admission, reoperation, and postoperative sepsis—were significantly lower when surgery was performed between days 1 and 3 (3–3.3, 0.5–0.6, and 3.8–4.1 %, respectively) when compared to surgery performed on the day of admission (5.6, 1.2, and 5.2 %, p

Objectives To establish the pattern of change in individual scientific production over the career of medical researchers. Design Retrospective cohort based on prospectively collected data in a hospital information system. Setting Multicentre university hospital in France. Participants Two distinct populations of 1835 researchers (full professors vs non-academic physicians) having produced 44 723 publications between 1995 and 2014. Main outcome measures Annual number of publications referenced in Medline/PubMed with a sensitivity analysis based on publications as first/last author and in high impact journals. The individual volume of publications was modelled by age using generalised estimating equations adjusted for birth cohort, biomedical discipline and academic position of researchers. Results Averaged over the whole career, the annual number of publications was 5.28 (95% CI 4.90 to 5.69) among professors compared to 0.82 (95% CI 0.76 to 0.89) among non-academic physicians (p

Introduction Medical and surgical innovations have allowed hospitals to shorter length of stay (LOS) under financial constraints. It remains unclear whether hospitals with uncontrolled LOS reduction may expose patients to unsafe care. We aimed to determine if there was an association between trend in hospital LOS and patient's risk of readmission over time. Methods We selected all patients who underwent elective colectomy or urgent hip fracture repair in French hospitals between 2013 and 2016. For each procedure, hospitals were categorized in three groups according to potential variations in their median length of index stays over study period as follows: major decrease, moderate decrease, or no decrease. Three cohorts of patients were then designed by matching these groups of hospitals two-by-two, based on a propensity score approach tacking into account potential confounding factors related to patients and hospital characteristics. In each cohort, outcomes were compared between hospitals groups at 1, 3 and 6 months. Potentially avoidable readmission for severe adverse event was modelled using Fine and Gray's approach with competing risk of death, while number of days in all acute care hospitalizations over study period was analyzed using generalized estimating equations (GEE) Poisson regression. Clustering of patients in matched pairs was taken into account. Results We considered 98,713 patients in 540 hospitals for colectomy and 206,812 patients in 414 hospitals for hip fracture repair before matching. After colectomy, trend in hospital LOS over time was not associated with risk of readmission, except for patients operated in hospitals with a moderate decrease in LOS whose experienced a lower risk at 3 months compared to those operated in hospitals with no decrease (hazard ratio [95% CI]: 0.85 [0.78–0.93]). After hip fracture repair, patients operated in hospitals with major decrease in LOS compared to those operated in hospitals with no decrease had a higher risk of readmission at 1, 3 and 6 months (1.37 [1.18–1.59], 1.23 [1.10–1.37], 1.22 [1.11–1.33], respectively). For the two procedures, major or moderate decrease in hospital LOS over time compared to hospitals with no decrease were associated with fewer number of days in acute care hospitalizations at 1, 3 and 6 months (P Conclusion Reduction in LOS over time in French hospitals was not associated with worse patient outcomes after colectomy, whereas a higher risk of severe adverse event was observed after hip fracture repair in hospitals with major LOS decrease. For both procedures, clear relationship was found between reduction in LOS and lower consumption of hospitalization days in acute care.

Introduction La sleeve, le bypass et l’anneau gastrique sont les trois principales chirurgies bariatriques realisees en France. Leurs complications postoperatoires etant principalement evaluees au cours du premier mois postoperatoire, notre objectif etait de comparer entre ces trois chirurgies, le risque de reprise chirurgicale et sa severite, dans les six mois postoperatoires. Methodes Nous avons selectionne dans la base PMSI nationale MCO, les patients operes entre 2013 et 2015 pour l’une de ces trois chirurgies. Au sein de chaque hopital, les patients operes d’une sleeve ont ete apparies aux patients operes d’un bypass (cohorte A) et d’un anneau (cohorte B), par des scores de propension considerant les caracteristiques des patients. Les reprises chirurgicales ont ete classees selon leur severite (faible, moderee, elevee), a l’aveugle, par trois chirurgiens sur la base des actes CCAM. Le risque de survenue d’une reprise chirurgicale et sa severite a ete compare dans les deux cohortes avec des modeles de survie tenant compte du risque competitif de deces, dans les six mois postoperatoires. Resultats La cohorte A etait constituee de 34252 patients operes dans 304 hopitaux. Le risque de reprise chirurgicale etait plus faible pour la sleeve que pour le bypass lors de la premiere semaine suivant l’intervention (HR 0,64 IC95 % [0,56–0,73]) ainsi qu’apres deux mois postoperatoires (0,64 [0,53–0,77]). La cohorte B etait constituee de 25206 patients operes dans 378 hopitaux. Le risque de reprise etait plus eleve pour la sleeve que pour l’anneau la premiere semaine suivant l’intervention (1,58 [1,29–1,94]) alors qu’il etait plus faible apres deux mois postoperatoires (0,08 [0,07–0,09]). Dans les deux cohortes, les resultats etaient similaires pour les reprises de severites moderees. Discussion/Conclusion Au-dela de deux mois postoperatoires, la sleeve est la procedure presentant le moins de risque de reprises chirurgicales y compris pour les severites moderees.

Introduction La fibrose pulmonaire idiopathique (FPI) et le syndrome emphyseme fibrose (SEF) sont associes a un risque accru d’hypertension pulmonaire (HTP). Peu de donnees sont disponibles sur la relation entre hemodynamique et physiologie respiratoire, et l’effet du traitement de l’HTP. Les objectifs de notre etude etaient d’evaluer les caracteristiques respiratoires et hemodynamiques des patients atteints de FPI ou SEF et d’HTP ; de rechercher des correlations entre elles ; et d’evaluer et de comparer la reponse au traitement de l’HTP et le pronostic. Methodes Parmi 220 patients avec HTP pre-capillaire inclus dans l’etude observationnelle multicentrique prospective HYPID (NCT01443598), 61 patients avec FPI et 75 avec SEF ont ete inclus dans l’analyse. Resultats Le 89 % des patients etaient hommes avec un âge moyen de 70 ± 8,5 ans. La CVF moyenne au diagnostic d’HTP etait de 76 ± 24,5 % et la DLco de 24,7 ± 10,9 %. La classe NYHA etait I-II (18 %), III (58 %), IV (24 %). La distance parcourue en 6 min (DM6) etait de 276 ± 145,4 m. La PAPm etait 37 ± 9 mmHg, l’index cardiaque (IC) 2,6 ± 0,6 l/min/m2 et les resistances vasculaires pulmonaires (RVP) 6,6 ± 2,9 UW. Le KCO etait plus bas chez les SEF que chez les FPI (35 % vs 49 % ; p = 0,0005), ainsi que la SpO2 (93 % vs 94,6 %, p = 0,006). La DLco etait significativement correlee avec la PAPm (r = −0,55, p = 0,0002) et les RVP (r = −0,53, p = 0,005) chez les FPI mais pas chez les SEF. La SpO2 etait significativement correlee avec tous les parametres hemodynamiques dans le groupe SEF, mais avec aucun d’entre eux chez les FPI. L’HTP etait severe chez 73,3 % des patients SEF contre 78,7 % des FPI (p = 0,47). Le traitement de l’HTP a entraine une diminution significative de la PAPm et des RVP et une augmentation de l’IC chez les patients SEF, mais pas chez les FPI. En considerant la totalite des patients traites, un traitement combine avec un antagoniste des recepteurs de l’endotheline et un inhibiteur de la phosphodiesterase − 5 a montre une amelioration significative des RVP (p = 0,05). La survie mediane etait de 20,2 mois, sans difference significative de survie entre les deux groupes. La DM6 etait le seul facteur predictif independant de la survie globale (HR 0,997, IC 0,995–0,998, p Conclusion L’hemodynamique de l’HTP est comparable au cours de FPI et SEF, mais la relation avec la fonction pulmonaire et la reponse au traitement different, suggerant des differences physiopathologiques. La distance parcourue au test de marche est le seul facteur independant de la survie.

Background As the medicine practiced in hospital settings has become more specialized, training in primary care is becoming increasingly essential for medical students, especially for future general practitioners (GPs). Only a few limited studies have investigated the representativeness of medical practices delivering this training. The aim of this study was to assess the representativeness of French GP trainers in terms of socio-demographics, patients and activities. Methods We conducted a cross-sectional study covering all private GPs practicing in the Rhône-Alpes region of France in 2011. This population consisted of 4992 GPs, including 623 trainers and 4369 non-trainers, managing 8,198,684 individual patients. Data from 2011 to 2012 were provided by the Regional Health Care Insurance (RHCI). We compared GP trainers with non-trainers using the Pearson chi-square test for qualitative variables and the Student t-test for quantitative variables Results GP trainers do not differ from non-trainers for gender, but they tend to be younger, more frequently in mid-career, and more likely to practice in a rural area. Their patients are broadly representative of patients attending general practice for age (with the exception of a higher consultation rate for infants), but patients with medical fee exemption status relating to low income are underrepresented. GP trainers have a heavier workload in terms of office visits and on-call duties. They prescribe a higher proportion of generic drugs, perform more electrocardiograms and cervical smears, and fewer plaster casts. GP trainers show better performance in diabetes follow-up, and to a lesser extent for seasonal flu vaccination and mammograms. Conclusions GPs and patients of training practices are globally representative, which is particularly critical in countries such as France, where the length of specialty training in a general practice setting is still limited to a few months. In addition, GP trainers tend to have better clinical performance, which conforms to their teaching modelling role and may encourage other GPs to become trainers.

International audience; OBJECTIVES: To describe the clinical outcomes of patients undergoing transcatheter aortic valve implantation (TAVI) and to determine the direct costs before and after TAVI. METHODS: A nationwide longitudinal study using data extracted from the French Hospital Information System. Selection criteria: all patients who underwent TAVI between 1 January 2010 and 31 December 2010. Period of follow-up: 12 months preceding TAVI to 36 months after. End-points: mortality, morbidity and total costs of acute and rehabilitation care from the perspective of the hospital. RESULTS: A total of 1332 patients (mean age: 82.0 +/- 7.2 years; 50.2% men) were identified. The mean hospitalization length of stay was 13.5 +/- 9.3 days. The intrahospital mortality from any cause was 7.9% during the index hospitalization, 8.8% at 30 days, 14.8% at 6 months, 18.4% at 1 year, 24.8% at 2 years and 32.3% at 3 years. The mean number of hospital stays per patient was 4.79 the year preceding TAVI and 4.11 the year after. The cumulated number of hospital stays at 2 and 3 years post-TAVI was 6.88 and 9.69, respectively. The mean hospitalization costs were 14 665(sic) the year preceding, 26 575(sic) for the index procedure and 12 308(sic) the year after TAVI. The cumulated hospitalization costs per patient at 3 years after TAVI were 22 110(sic) for acute hospitalizations and 5689(sic) for rehabilitation. CONCLUSIONS: Mortality at 3 years is consistent with other published studies. After TAVI, hospitalization stays in both acute and rehabilitation settings, and the associated costs do not appear to be reduced compared with the year preceding TAVI. The total cost for patients undergoing TAVI is substantial at 3 years.

Background Use of French Diagnosis Related Groups (DRGs) program databases, apart from financial purposes, has recently been improved since a unique anonymous patient identification number has been created for each inpatient in administrative case mix database. Based on the work of the group for cancer epidemiological observation in the Rhone-Alpes area, (ONC-EPI group), we review the remaining difficulties in the use of DRG data for epidemiological purposes and we consider a longitudinal approach based on analysis of database over several years. We also discuss limitations of this approach. Difficulties The main problems are related to a lack of quality of administrative data, especially coding of diagnoses. These errors come from missing or inappropriate codes, or not being in accordance with prioritization rules (causing an over- or under-reporting or inconsistencies in coding over time). One difficulty, partly due to the hierarchy of coding and the type of cancer, is the choice of an extraction algorithm. In two studies designed to estimate the incidence of cancer cared in hospitals (breast, colon-rectum, kidney, ovaries), a first algorithm, including a code of cancer as principal diagnosis with a selection of surgical procedures less performed than the second one including a code of cancer as principal diagnosis only, for which the number of hospitalizations per patient ratio was stable across time and space. The chaining over several years allows, by tracing the trajectory of the patient, to detect and correct inaccuracies, errors and missing values, and for incidence studies, to correct incident cases by removing prevalent cases. Discussion However, linkage, complete only since 2007, does not correct data in all cases. Ways of future improvement certainly pass through improved algorithms for case identification and especially by linking DRG data with other databases.

Objective. Use of cancer cases from registries and PMSI claims database to estimate Département-specific incidence of four major cancers.Methods. Case extraction used principal diagnosis then surgery codes. PMSI cases/registry cases ratios for 2004 were modelled then Département-specific incidence for 2007 estimated using these ratios and 2007 PMSI cases.Results. For 2007, only colon-rectum and breast cancer estimations were satisfactorily validated for infranational incidence not ovary and kidney cancers. For breast, the estimated national incidence was 50,578 cases and the incidence rate 98.6 cases per 100,000 person per year. For colon-rectum, incidence was 21,172 in men versus 18,327 in women and the incidence rate 38 per 100,000 versus 24.8. For ovary, the estimated incidence was 4,637 and the rate 8.6 per 100,000. For kidney, incidence was 6,775 in men versus 3,273 in women and the rate 13.3 per 100.000 versus 5.2.Conclusion. Incidence estimation using PMSI patient identifiers proved encouraging though still dependent on the assumption of uniform cancer treatments and coding.

Introduction It is commonly accepted that the more a hospital perform a procedure the better the procedure is done, even if explanatory mechanisms of this dogma have not been deeply elucidated so far. We aimed to identify if surgical outcomes were altered when patients were admitted to hospitals experiencing significant volume changes over time and if a learning effect existed at the individual hospital level. Methods We selected all patients who underwent one of ten specific surgeries in French hospitals from a nationwide database, between 2010 and 2014. For each surgery, we ascertained three groups of hospitals according to their volume trend toward increase, decrease or no change over study period. In-hospital mortality, reoperation, unplanned hospital readmission and a composite outcome of these complications within 30 days were compared between groups using Cox regressions, taking into account clustering of patients within hospitals and potential confounding factors. Interaction between the group of trend in hospital volume and year of surgery was also introduced to study learning effect. Results Over 5 years, 759,518 patients from 694 hospitals were analyzed. Compared to hospitals with unchanged volume of procedures, patients’ composite outcome was not different in hospitals experiencing volume increase (HR: 0.99, 95% CI [0.92–1.06]) or decrease over time (1.01 [0.95–1.08]) and did not improve or deteriorate during study period when hospitals increased (1.01 [0.99–1.03]) or decreased their volume (0.99 [0.97–1.01]). Similar results were found when analyzing outcomes and procedures separately. Conclusions Trend in hospital volume over time does not appear to impact patient outcomes, which questions about the explanatory mechanism of “practices makes perfect” dogma at the hospital level.

Introduction L’utilisation de la mortalite intra-hospitaliere pour comparer la performance des hopitaux requiert son ajustement valide sur le « case-mix » des patients pris en charge. Un codage heterogene de leurs comorbidites pourrait influencer le classement des hopitaux et leur paiement a la performance. Methodes Dans chacun des 24 etablissements francais participants, nous avons tire au sort 30 sejours MCO realises en 2010 parmi les 78 GHM presentant la mortalite la plus elevee au niveau national. Ces sejours ont ete recodes de maniere homogene par un binome d’information medicale a partir du dossier patient. Un modele hierarchique Bayesien a estime la performance du codage initial pour chaque couple hopital/comorbidite (index d’Elixhauser/Charlson). A partir de ces estimations, nous avons re-impute la presence de comorbidites sur l’ensemble des patients pris en charge en 2010 dans ces etablissements pour corriger leur mortalite ajustee. Les changements de classement des hopitaux avant-apres correction ont ete utilises pour simuler l’influence des heterogeneites de codage sur des modeles incitatifs de paiement a la performance. Resultats Parmi les 70 402 sejours realises dans les hopitaux participants, 715 ont ete recodes. Les valeurs predictives positives medianes allaient de 56,5 a 97,5 % pour les 31 comorbidites de l’Elixhauser et de 52,2 a 92,9 % pour les 17 comorbidites du Charlson ; pour la sensibilite, 33,6 a 89,7 % pour l’Elixhauser et 38,7 a 90,2 % pour le Charlson. Pour l’Elixhauser, 70,3 % des hopitaux changeaient de position dans le classement par mortalite apres correction de l’ajustement, revelant que 6,53 % (± 3,6 %) d’un budget incitatif typique serait incorrectement attribue a des hopitaux tendant a surcoder les comorbidites de leurs patients. Pour le Charlson, 61,5 % des hopitaux changeaient de position avec 7,32 % (± 4 %) de mauvaise attribution du budget. Discussion/Conclusion Une meilleure validite des comorbidites renseignees par les hopitaux est necessaire pour permettre leur utilisation pour un paiement a la performance non biaise. Il est possible de corriger ces heterogeneites de codage par une relecture centralisee partielle des dossiers.

Introduction De nombreuses etudes se sont interessees a la relation entre le volume de chirurgies d’un etablissement et la survenue de complications. Cependant, les mecanismes explicatifs de cette relation ne sont pas elucides. Pour determiner s’il existe un effet d’apprentissage, nous avons etudie la relation entre l’evolution au cours du temps du volume de chirurgies realisees et la survenue de complications. Methodes Les patients operes dans un etablissement francais entre 2010 et 2014, pour l’une des dix chirurgies etudiees, ont ete identifies dans la base PMSI. Pour chaque chirurgie, les hopitaux ont ete repartis en trois groupes selon l’evolution de leur volume de procedures (augmentation, stabilite, baisse) sur cinq ans. Des modeles de survie ont estime la mortalite, la reprise chirurgicale, la rehospitalisation potentiellement evitable et le composite de ces trois complications a 30 jours, en considerant le regroupement des patients dans les hopitaux et divers facteurs confondants lies aux caracteristiques des patients et des hopitaux. Resultats En cinq ans, 610 180 patients ont ete consideres dans 694 etablissements. Le risque de reprise chirurgicale diminuait dans le groupe augmentant son volume de prothese de hanche (HR 0,79 IC 95 % [0,67–0,94]) et de pontage aorto-coronarien (0,38 [0,15–0,97]) mais aussi dans celui diminuant son volume de pontage aorto-coronarien (0,23 [0,09–0,62]). Il augmentait dans le groupe diminuant son volume de reperfusion d’un anevrisme de l’aorte abdominale (AAA) (1,41 [1,04–1,91]) et d’endarteriectomie (1,28 [1,01–1,64]). Le risque de rehospitalisation augmentait dans le groupe diminuant son volume d’endarteriectomie (1,29 [1,07–1,55]) et augmentant son volume d’œsophagectomie (1,73 [1,36–2,20]). Le risque d’avoir au moins une complication augmentait dans le groupe augmentant son volume de reperfusion d’un AAA (1,41 [1,04–1,91]). Discussion/Conclusion Le manque de coherence observe entre les resultats obtenus pour les differentes chirurgies ne permet pas de degager un lien clair entre l’evolution du volume et la survenue de complication operatoires. Davantage de preuves sont necessaires pour utiliser le volume comme reflet de la qualite chirurgicale.

International audience; OBJECTIVE: To estimate breast cancer incidence in the general population using a method that corrects for lack of sensitivity and specificity in the identification of incident breast cancer in inpatient claims data. STUDY DESIGN AND SETTINGS: Two-phase study: phase 1 to identify incident cases in claims data, and phase 2 to estimate sensitivity and specificity in a subset of the population. Two algorithms (1: principal diagnosis; 2: principal diagnosis+specific surgery procedures) were used to identify incident cases in claims of women aged 20 years or older, living in a French district covered by a cancer registry. Sensitivity and specificity were estimated in one district and used to correct incident cases identified. RESULTS: The sensitivity and specificity for algorithms 1 and 2 were 69.0% and 99.89%, and 64.4% and 99.93%, respectively. In contrast to specificity, the sensitivity for both algorithms was lower for women younger than 40 years and older than 65 years. Cases reported by cancer registries were closer to cases identified with algorithm 2 (-3.2% to +20.1%) and to corrected numbers with algorithm 1 (-1% to +15%). CONCLUSION: To obtain reliable estimates of breast cancer incidence in the general population, sensitivity and specificity, which reflect medical and coding practice variations, are necessary.

International audience; One approach to estimate cancer incidence in the French D?rtements is to quantify the relationship between data in cancer registries and data obtained from the PMSI (Programme de M?calisation des Syst?s d'Information M?cale). This relationship may then be used in D?rtements without registries to infer the incidence from local PMSI data. We present here some methodological solutions to apply this approach. Data on invasive breast cancer for 2002 were obtained from 12 D?rtemental registries. The number of hospital stays was obtained from the National PMSI using two different algorithms based on the main diagnosis only (Algorithm 1) or on that diagnosis associated to a mention of "resection" (Algorithm 2). Considering registry data as gold standard, a calibration approach was used to model the ratio of the number of hospital stays to the number of incident cases. In D?rtements with registries, validation of the predictions was done through cross-validation. In D?rtements without registries, validation was done through a study of homogeneity of the mean number of hospital stays per patient. Cross-validation showed that the estimates predicted by the model were true with data extracted by Algorithm 1 but not by Algorithm 2. However, with Algorithm 1, there was an important heterogeneity between French D?rtements as to the mean number of hospital stays per patient, which had an important impact on the estimations. In the near future, the method will allow using medico-administrative data (after calibration with registry data) to estimate D?rtemental incidence of selected cancers.

Objective: Matching healthcare staff resources to patient needs in the ICU is a key factor for quality of care. We aimed to assess the impact of the staffing-to-patient ratio and workload on ICU mortality. Design: We performed a multicenter longitudinal study using routinely collected hospital data. Setting: Information pertaining to every patient in eight ICUs from four university hospitals from January to December 2013 was analyzed. Patients: A total of 5,718 inpatient stays were included. Interventions: None. Measurements and Main Results: We used a shift-by-shift varying measure of the patient-to-caregiver ratio in combination with workload to establish their relationships with ICU mortality over time, excluding patients with decision to forego life-sustaining therapy. Using a multilevel Poisson regression, we quantified ICU mortality-relative risk, adjusted for patient turnover, severity, and staffing levels. The risk of death was increased by 3.5 (95% CI, 1.3–9.1) when the patient-to-nurse ratio was greater than 2.5, and it was increased by 2.0 (95% CI, 1.3–3.2) when the patientto-physician ratio exceeded 14. The highest ratios occurred more frequently during the weekend for nurse staffing and during the night for physicians (p < 0.001). High patient turnover (adjusted relative risk, 5.6 [2.0–15.0]) and the volume of life-sustaining procedures performed by staff (adjusted relative risk, 5.9 [4.3–7.9]) were also associated with increased mortality. Conclusions: This study proposes evidence-based thresholds for patient-to-caregiver ratios, above which patient safety may be endangered in the ICU. Real-time monitoring of staffing levels and workload is feasible for adjusting caregivers’ resources to patients’ needs. (Crit Care Med 2015; 43:1587–1594)

Introduction Les index de comorbidite de Charlson et d’Elixhauser sont classiquement utilises pour categoriser les comorbidites de patients basees sur les codes de diagnostic de la classification internationale des maladies. Avant de recommander leur usage a partir des donnees hospitalieres francaises, il parait souhaitable de verifier que la qualite de documentation des sejours dans le PMSI est suffisante pour garantir la validite de ces index. Methodes Une etude observationnelle multicentrique retrospective a ete conduite pour comparer, avant et apres recodage, la documentation des diagnostics servant au calcul de ces index de comorbidite. Un echantillon de sejours d’hospitalisation complete realises en 2010 dans 24 etablissements francais a ainsi ete audite. Trente Resumes de sortie anonymes ont ete tires au sort par etablissement afin de permettre l’analyse du dossier medical des patients et le recodage du sejour par un meme binome de professionnels de l’information medicale. Les deux index ont ete calcules avant et apres recodage afin d’estimer leur validite en considerant le recodage standardise comme gold standard. Resultats Parmi 720 sejours tires au sort, 715 disposaient d’un dossier et ont pu etre analyses. Globalement, 40,8 % [IC95 %, 37,2–44,5 %] n’ont pas necessite le recodage d’une categorie de comorbidite, que ce soit pour l’index de Charlson ou pour celui d’Elixhauser. Pour les 17 comorbidites comprises dans l’index de Charlson, la sensibilite mediane etait de 71 % (etendue allant de 43 a 100 %) et la valeur predictive positive mediane de 79 % (etendue allant de 44 a 100 %). Pour les 31 comorbidites comprises dans l’index d’Elixhauser, la sensibilite mediane etait de 71 % (etendue allant de 48 a 100 %) et la valeur predictive positive mediane de 85 % (etendue allant de 50 a 100 %). Discussion/conclusion La mesure des deux index a partir des donnees hospitalieres francaises presente une validite satisfaisante en depit de la variabilite observee d’une comorbidite a l’autre.

Introduction Une mauvaise gestion des lits peut se repercuter sur l’organisation des soins, l’acces des patients aux prises en charge dont ils ont besoin et les recettes de l’hopital. Mieux gerer les flux de patients necessite de predire leur duree de sejour pour anticiper leur sortie des que possible. Notre objectif etait de developper un algorithme informatise de prediction des durees de sejour des patients hospitalises puis de comparer l’exactitude des durees predites par l’algorithme a celles predites par l’Homme. Methodes L’algorithme a ete developpe a partir des donnees de resumes d’unite medicale collectees en 2013 et 2014 aux Hospices Civils de Lyon (HCL). Les durees de sejours ont ete predites en testant differentes approches statistiques (regression lineaire log a effets mixtes, modele de Cox et « machine learning »). La validite des predictions a ete estimee selon un schema de type train-test en mesurant l’erreur quadratique moyenne (RMSE) et la difference entre durees de sejour predites et observees. Les variables explicatives finalement retenues dans l’algorithme comprenaient l’unite medicale d’hospitalisation du patient, son âge, sexe et diagnostic principal. Une enquete prospective a ensuite ete menee sur le terrain afin de comparer les capacites de prediction de l’algorithme a celles de l’Homme. Vingt-cinq medecins repartis dans six services des HCL ont ainsi predit la duree de 187 sejours entrants consecutifs de mars a mai 2015. Resultats La difference moyenne entre la duree predite par l’algorithme et celle observee etait de 2,6 jours (ET = 6,6 et mediane = 0,8). Cette difference etait de 1 jour ou moins dans 73 % des cas et tendait a augmenter en cas de sejour long. Les capacites de prediction de l’algorithme (RMSE = 8 [IC95 % 7–9]) semblaient plus elevees que celles de l’Homme (RMSE = 14 [13-15]). Discussion/conclusion La mise a disposition d’un outil de prediction des durees de sejour a partir des informations recueillies en routine dans le systeme d’information hospitalier pourrait s’averer utile pour les equipes de soins.

BACKGROUND: Mass tourism during winter in mountain areas may cause significant clustering of body injuries leading to increasing emergency admissions at hospital. We aimed at assessing if surgical safety and efficiency was maintained in this particular context. METHODS: We selected all emergency admissions of open surgery performed in French hospitals between 2010 and 2012. After identifying mountain areas with increasing volume of surgical stays during winter, we considered seasonal variations in surgical outcomes using a difference-in-differences study design. We computed multilevel regressions to evaluate whether significant increase in emergency cases had an effect on surgical mortality, complications and length of stay. Clustering effect of patients within hospitals was integrated in analysis and surgical outcomes were adjusted for both patient and hospital characteristics. RESULTS: A total of 381 hospitals had 559,052 inpatient stays related to emergency open surgery over 3 years. Compared to other geographical areas, a significant peak of activity was noted during winter in mountainous hospitals (Alps, Pyrenees, Vosges), ranging 6-77% volume increase. Peak was mainly explained by tourists' influx (+124.5%, 4,351/3,496) and increased need for orthopaedic procedures (+36.8%, 4,731/12,873). After controlling for potential confounders, patients did not experience increased risk for postoperative death (ratio of OR 1.01, 95%CI 0.89-1.14, p = 0.891), thromboembolism (0.95, 0.77-1.17, p = 0.621) or sepsis (0.98, 0.85-1.12, p = 0.748). Length of stay was unaltered (1.00, 0.99-1.02, p = 0.716). CONCLUSION: Surgical outcomes are not compromised during winter in French mountain areas despite a substantial influx of major emergencies. Language: en