P Michel

Université Victor Segalen Bordeaux 2, Burdeos, Aquitaine, France

Are you P Michel?

Claim your profile

Publications (59)122.5 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objectives Consensus methods are increasingly used as alternatives to traditional assessment methods, because of their low cost and high efficiency. The objective of our study was to investigate whether the assessment of occupational hazards linked to the use of carcinogenic, mutagenic and reprotoxic chemicals differed when done by two consensus methods, the nominal group method (based on a face-to-face group meeting) and the Delphi method (a questionnaire-based method) in comparison to direct observation. The CMR’s have been chosen due to the specificity of substances used in the laboratories and due also to the lack of prevention practices. Methods 119 professionals from 13 French research laboratories were randomly allocated to use either the Delphi or nominal group methods. Direct observation of the presence and use of chemicals was done by an external occupational hygienist who used a standardized protocol. After data collection, chemicals identified by consensus methods but not by observation were checked by local hygiene and safety correspondents. The final combined list of the present and used chemicals was defined as the reference. Sensitivities (Se) and specificities (Sp) were estimated to assess the performance of the three methods to identify the presence, and the actual use of chemicals. Characteristics associated with performance were assessed using logistic regression models. Results The total number of chemicals listed in the initial lists was 360. Observation identified 50 additional chemicals, and consensus methods another two, which were neither on the lists nor observed. Performance of the nominal group (Se presence 0.57; Se use 0.86; Sp presence 0.65; Sp use 0.74) and Delphi method (Se presence 0.59; Se use 0.83; Sp presence 0.57; Sp use 0.57) was similar. Higher seniority of the participants was the main characteristic related to better performance. Conclusions Performance of both consensus methods was low. Because of their advantages over observation (local collective involvement and lower workload), these methods might be useful before and after a valid assessment based on observation, therefore contributing at presumably affordable cost to maintain accuracy of the list, as well as team awareness and prevention commitment. Even if the observations are more burdensome to carry out, they make it possible to understand the complexity of the compromises made by operators when they face risks. In that perspective they can unearth accounts of incidents and strategies that would be otherwise difficult to verbalise through other methods. What is more, such observation methods can also help involve workers in a bottom-up approach and turn them into active stakeholders in the prevention process. It may thus be possible and relevant to develop an articulation between consensus methods and those centred on ergonomics observations.
    Safety Science 10/2014; 68:324–330. · 1.67 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose of the study Consensus methods meet the requirements and expectancies of global occupational risk prevention guidelines, in particular because they strongly involve field professionals. In addition, the nominal group technique may be easily implemented alone. The objective of our study was to compare three methods, among them two consensus methods, for identifying psychosocio-organizational (PSO) hazards in French research laboratories. Sample and methods Professionals were selected from 13 French research laboratories. The consensus methods were the Delphi and the nominal group techniques; the third technique was based on observations and interviews. Methods were independently implemented. The questionnaire was standardized and similar for the three methods. A descriptive presentation of PSO factors was performed. The concordance of the results across the three methods was analysed with Kappa coefficients. Polytomous logistic regression allowed to assess method performance adjusted to laboratory characteristics and operational modalities. Results The 176 professionals belonged to all professional categories. The observation method identified 48 PSO factors, Delphi 74 and nominal group 102, in the 13 laboratories. Observation method provided 13% of undefined answers, Delphi 8% and nominal group 7%. The most frequent PSO categories were related to the individual's role within the organization, career plan definition and work time organization. The most frequent PSO factors were lack of workstation sheet, cases of career stagnation or uncertainty, insufficient promotion system, existence of interpersonal conflicts and achieving long working hours or working outside normal hours. Concordance between the methods was low: the PSO factors identified by each method were variable. The probability to identify PSO factors by the nominal group was significantly higher (OR = 2.4) than the one by the observation method. The academics/researcher identified significantly more PSO factors than the other professionals (OR = 1.57). Discussion–conclusion The nominal group technique might be a relevant tool for PSO factor assessment for French research laboratories. The results might also be used at an individual level during occupational medical visits, for focusing the interview and prioritizing preventive actions.
    Archives Des Maladies Professionnelles Et De L Environnement - ARCH MAL PROF ENVIRON. 01/2009; 70(5):516-524.
  • F Bourrée, P Michel, L R Salmi
    [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: Consensus-based studies are increasingly used as decision-making methods, for they have a lower production cost than other methods (observation, experimentation, modeling) and provide results more rapidly. The objective of this paper is to describe the principles and methods of the four main methods - Delphi, nominal group, consensus development conference and RAND/UCLA - their use as reported in peer-reviewed publications and validation studies published in the healthcare literature. METHODS: A bibliographic search was performed in PubMed/MEDLINE, banque de données santé publique (BDSP), The Cochrane Library, Pascal and Francis. Keywords, headings and qualifiers corresponding to a list of terms and expressions related to the consensus methods were searched for in the thesauri and used in the literature search. A search with the same terms and expressions was performed on Internet using the website Google Scholar. RESULTS: All methods, precisely described in the literature, are based on common basic principles such as definition of the subject, selection of experts and direct or remote interaction processes. They sometimes use quantitative assessment for ranking items. Numerous variants of these methods have been described. Few validation studies have been implemented. Not implementing these basic principles and failing to describe the methods used to reach the consensus were both frequent reasons raising suspicion regarding the validity of consensus methods. CONCLUSION: When it is applied to a new domain with important consequences in terms of decision-making, a consensus method should first be validated.
    Revue d Épidémiologie et de Santé Publique 12/2008; · 0.69 Impact Factor
  • F. Bourrée, P. Michel, L. R. Salmi
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Consensus-based studies are increasingly used as decision making methods, for they have lower production cost than other methods (observation, experimentation, modelling) and provide results more rapidly. The objective of this paper is to describe the principles and methods of the four main methods, Delphi, nominal group, consensus development conference and RAND/UCLA, their use as it appears in peer-reviewed publications and validation studies published in the healthcare literature.
    Revue D Epidemiologie Et De Sante Publique - REV EPIDEMIOL SANTE PUBL. 01/2008; 56(6):415-423.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Evans and Bernardis proposed the 'PNB classification', by which a fingertip injury is classified according to each structure: pulp P, nail N, bone B. The objective of this study was to assess the inter-observer reliability, repeatability and accuracy of PNB. One hundred patients presenting with a fingertip injury were included prospectively, photographed, then classified in randomly chosen orders by nine independent observers. A third were drawn randomly and classified a second time to measure repeatability. A reference classification was also provided by one of the authors of the PNB system. Classifications agreed with the reference in 59% of injuries for P, 55% for N and 54% for B. The Kappa values for inter-observer agreement were 0.520 for P, 0.512 for N, and 0.504 for B; for intra-observer agreement, they were 0.616 for P, 0.658 for N, and 0.577 for B. Although levels of agreement are comparable with results found for other classifications, they are insufficient for use of the PNB classification without improvement.
    Journal of Hand Surgery (European Volume) 05/2007; 32(2):188-92. · 1.22 Impact Factor
  • V Daucourt, S Domecq, P Michel
    Revue d Épidémiologie et de Santé Publique 11/2006; 54(5):463-8. · 0.69 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The objective of our study was to perform a cost-minimization analysis of a wide-area teleradiology network. A prospective analysis of all transmissions over 1 year (data transmitted at the time of the remote consultation, and health outcomes of patients from medical record). The inter-hospital teleradiology network of the Aquitaine area (RIHRA) is a telemedicine system enabling the management of remote emergencies and elective radiology consultations. A cost-minimization study enabled a comparison of care procedures following the use of the network with those which would have been implemented without the network. The outcome measures of effectiveness were the transfers, hospitalizations, and consultations avoided or added. Fixed and variable costs were estimated. Among the 664 transmissions included in the study, 562 (85%) were performed in emergency and 102 (15%) for elective (non-emergency) cases. In emergency, 48% of transfers were avoided. For elective teleconsultations, a transfer was avoided for 37% of the patients and hospitalization for 12%. An extra consultation occurred after remote consultation for 2% of the patients. Annual saving can be estimated at 102,779 EUR for the Aquitaine area. This study underlines the efficiency of an inter-hospital teleradiology network. A qualitative evaluation of the impact of the use of the system should be carried out to improve technical and organizational operations.
    International Journal for Quality in Health Care 09/2006; 18(4):287-93. · 1.79 Impact Factor
  • V. Daucourt, S. Domecq, P. Michel
    Revue D Epidemiologie Et De Sante Publique - REV EPIDEMIOL SANTE PUBL. 01/2006; 54(5):463-468.
  • [Show abstract] [Hide abstract]
    ABSTRACT: To select a set of quality indicators (QI) in order to test them in a panel of 36 French hospitals The COMPAQH (COordination for Measuring Performance and Assuring Quality in Hospitals) project is coordinated by the French National Institute for Medical Research and supported by the French Ministry of Health and the French National Evaluation and Accreditation Agency. This project has four objectives: (1) to select a set of QI -2003- (2) to implement them in 2004-2005 in a volunteer panel of hospitals (3) to compare the hospitals anonymously (4) to explore quality management implications. QI were selected with a four-step process: (1) Establishment of a list of national priorities for Quality Improvement in relation with the Ministry of Health. (2) setting up a potential list of QI regarding these priorities. The COMPAQH staff determined a preliminary set of 81 QI, based on data in the literature and evidence about the scientific soundness of quality measures and the effectiveness of methods for improving quality. (3) Evaluation of the preliminary list. Each QI was presented in a pamphlet describing its operational definition, rationale, methodology, workload and responsibility of data collection The hospital panel (representatives) ranked the 81 QI with a validated evaluation tool which contained four dimensions: Importance, Scientific acceptability, Feasibility, and Usability. (4) Development of a consensus on a final selection. Based on a structured voting process (Delphi method, two rounds), the hospital panel selected a comprehensive set of 42 QI among the 81. (1) Eight national priorities were defined: pain management, continuity of care, management of nutritional disorders, Iatrogenic risks (including nosocomial infections), patient satisfaction, follow-up of practice guidelines, management of human resources, accessibility. (2) A set of 42 QI were selected: a set of 6 core QI and 7 to 18 specific QI according to the hospital type. Such a set of QI provides a foundation for developing a quality measurement system in French hospitals. It requires a pragmatic view for implementing them and a coherence between the different objectives of use (internal and external use).
    Revue d Épidémiologie et de Santé Publique 10/2005; 53 Spec No 1:1S22-30. · 0.69 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To compare independent and combined effectiveness and cost-effectiveness of two implementation interventions of guidelines for ordering thyroid function tests. The two implementation interventions were a Memorandum Pocket Card (MPC) and a Test Request Form (TRF). Intervention groups were wards. The study used an experimental 2*2 factorial design with matching hospitals according to size and activity and wards according to pre-intervention appropriateness for test ordering. Four ward groups were established: the dual intervention group, the order form group, the pocket card group and the control group. Physicians in all groups received guidelines and were invited to a local information meeting. The main outcome measure of effectiveness was the Guideline Conformity Rate (GCR). The cost-effectiveness ratio was the cost difference between the tested intervention and the control intervention upon effectiveness difference between the tested intervention and the control intervention. Six hospitals participated in the study (two middle-sized hospitals, two small-sized hospitals and two psychiatric hospitals). A total of 1412 orders for thyroid function tests were collected. GCR was 78% in the dual intervention group, 83% in the order form group, 73% in the pocket card group and 62% in the control group. The interaction between TRF and MPC was not significant (B=-0.70, p=0.21). Compared to simple information, TRF was effective in increasing GCR (OR=2.65, 95% Confidence Interval [CI]: 1.52-4.62), unlike MPC (OR=1.28, CI: 0.75-2.19). TRF was the less expensive and the most effective intervention. Using a robust design, our study shows a greater effectiveness of TRF than MPC and their association in implementing thyroid function test guidelines. The development of clinical practice improvement projects through the second procedure of accreditation in France is a good opportunity to develop a guidelines implementation research project.
    Revue d Épidémiologie et de Santé Publique 10/2005; 53 Spec No 1:1S79-88. · 0.69 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We conducted a prospective study of the teleradiology network which connects 15 hospitals in the Aquitaine area. All transmissions sent over a one-year period were examined (data transmitted at the time of the remote consultation and health outcomes of patients from their medical records). For emergency cases, the main outcome measure of effectiveness was the proportion of avoided transfers. For non-emergency cases, the main outcome measure of effectiveness was the proportion of transfers, hospitalizations and consultations avoided. There were 737 transmissions, of which 664 (90%) met the inclusion criteria. Of these, 562 (85%) were for emergency care and 102 (15%) for non-emergency care. In emergency care, the pathologies most often associated with a remote consultation were cerebral pathologies (88%) and traumatic spinal pathologies (8%); the proportion of avoided transfers was 48%. In non-emergency care, the specialties most often concerned with remote consultations were neurology/neurosurgery (36%), cardiology and pulmonary diseases (17%) and gastroenterology (14%). Transfer was avoided for 37% of the patients and hospitalization for 12%. An additional consultation occurred after remote consultation for 2% of the patients. The results confirm the effectiveness of an inter-hospital teleradiology network.
    Journal of Telemedicine and Telecare 02/2005; 11(4):178-84. · 1.47 Impact Factor
  • European Journal of Cardio-thoracic Surgery - EUR J CARDIO-THORAC SURG. 01/2005; 27(6):1129-1132.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Position du problèmeMesurer et comparer l’efficacité et le rapport coût-efficacité de deux interventions isolées de mise en œuvre de recommandations des examens biologiques explorant la fonction thyroïdienne (bilan thyroïdien) et de leur association. Ces deux interventions étaient : la diffusion d’une Plaquette Aide-Mémoire (PAM) et la mise en place d’une Fiche de Prescription Conseil (FPC). Elles ont toutes les deux été associées à une réunion d’information (sensibilisation).MéthodesLe schéma d’étude était un essai contrôlé randomisé, de type plan factoriel 2*2 avec appariement des établissements et des services et mesure « avant ». Les données d’efficacité (conformité aux recommandations) ont été comparées entre quatre groupes : groupe PAM seule, groupe FPC seule, groupe PAM + FPC et groupe contrôle (simple sensibilisation). Une régression logistique, ajustée sur la spécialité du service et le statut du médecin, a été utilisée pour cette comparaison. Ce modèle permettait de prendre en compte la randomisation par grappe et la stratification par paire d’hôpital. Une analyse coût-efficacité a été menée ; les coûts ont été calculés à partir d’une reconstitution du coût réel (coût de reprographie, coût des réactifs).RésultatsSix centres hospitaliers ont participé au projet, générant 1412 prescriptions. La conformité globale aux recommandations était de 74 %. L’interaction des modalités de diffusion n’était pas statistiquement significative (ß = - 0,70, p = 0,21). Comparée au groupe contrôle, la seule intervention statistiquement plus efficace était FPC seule (OR = 2,65 [IC à 95 % : 1,52-4,62). Dans chaque paire d’hôpital, la FPC était l’intervention la moins coûteuse et la plus efficace.ConclusionCette étude a permis de montrer la supériorité d’une Fiche de Prescription Conseil seule en terme d’efficacité et d’efficience pour la mise en œuvre des recommandations sur le bilan thyroïdien. De tels projets de recherche dans le domaine de la mise en œuvre des recommandations professionnelles pourraient trouver un terrain favorable dans le cadre du développement de l’évaluation des pratiques professionnelles au travers de la deuxième procédure de l’accréditation.Background To compare independent and combined effectiveness and cost-effectiveness of two implementation interventions of guidelines for ordering thyroid function tests. The two implementation interventions were a Memorandum Pocket Card (MPC) and a Test Request Form (TRF). Intervention groups were wards.Methods The study used an experimental 2*2 factorial design with matching hospitals according to size and activity and wards according to pre-intervention appropriateness for test ordering. Four ward groups were established: the dual intervention group, the order form group, the pocket card group and the control group. Physicians in all groups received guidelines and were invited to a local information meeting. The main outcome measure of effectiveness was the Guideline Conformity Rate (GCR). The cost-effectiveness ratio was the cost difference between the tested intervention and the control intervention upon effectiveness difference between the tested intervention and the control intervention.ResultsSix hospitals participated in the study (two middle-sized hospitals, two small-sized hospitals and two psychiatric hospitals). A total of 1412 orders for thyroid function tests were collected. GCR was 78% in the dual intervention group, 83% in the order form group, 73% in the pocket card group and 62% in the control group. The interaction between TRF and MPC was not significant (ß=-0.70, p=0.21). Compared to simple information, TRF was effective in increasing GCR (OR=2.65, 95% Confidence Interval [CI]: 1.52-4.62), unlike MPC (OR =1.28, CI: 0.75-2.19). TRF was the less expensive and the most effective intervention.Conclusion Using a robust design, our study shows a greater effectiveness of TRF than MPC and their association in implementing thyroid function test guidelines. The development of clinical practice improvement projects through the second procedure of accreditation in France is a good opportunity to develop a guidelines implementation research project.
    Revue d'Épidémiologie et de Santé Publique. 01/2005; 53:79-88.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Position du problèmeLe projet COMPAQH, coordonné par l’Institut national de la santé et de recherche médicale, a pour objectif de sélectionner une batterie d’indicateurs de la qualité de la prise en charge du patient (IQ) par les Établissements de Santé (ES) afin de les tester et de les valider en 2004-2005, au sein d’un panel de 36 ES volontaires, de comparer anonymement les ES et d’en étudier les implications managériales.MéthodesEn 2003, la sélection des indicateurs a été réalisée en quatre étapes : (1) Définition d’objectifs prioritaires (OP) de l’amélioration de la qualité. Elle a été élaborée par les promoteurs du projet (Direction de l’hospitalisation et de l’organisation des soins, Agence nationale d’accréditation et d’évaluation en santé et Fédérations hospitalières) et les ES participants selon cinq critères : importance du thème (notamment cohérence avec les priorités nationales de santé publique), variabilité inter-établissements, mesures existantes, marge d’amélioration, cohérence avec le manuel d’accréditation. (2) Définition de 81 IQ illustrant ces OP à partir d’une analyse de la littérature ; chaque IQ est décrit par un « cahier des charges » (justification de l’intérêt, modalités de construction, modes de recueil de données). (3) Évaluation des 81 IQ selon quatre critères: importance du phénomène étudié, qualités métrologiques, faisabilité et cohérence avec les démarches existantes. (4) Sélection définitive des IQ à tester selon une méthode Delphi.Résultats(1) Huit OP ont été définis : lutter contre la douleur, assurer la continuité de la prise en charge, lutter contre les troubles nutritionnels, lutter contre le risque iatrogène (dont les infections nosocomiales), améliorer la satisfaction du patient, respecter les bonnes pratiques cliniques, accroître la motivation, la responsabilisation et l’évaluation des compétences des professionnels, garantir l’accessibilité. (2) 42 IQ ont été sélectionnés : 6 IQ communs à tous les ES et 7 à 18 IQ spécifiques selon les secteurs d’activité des ES participants.ConclusionLes IQ sélectionnés associent majoritairement des mesures de processus à des mesures de résultats. Le principal critère de sélection retenu par les ES est la faisabilité du recueil des données. L’utilisation des IQ se conçoit selon trois objectifs : amélioration interne, diffusion publique, régulation des financements hospitaliers.Background To select a set of quality indicators (QI) in order to test them in a panel of 36 French hospitalsMethods The COMPAQH (COordination for Measuring Performance and Assuring Quality in Hospitals) project is coordinated by the French National Institute for Medical Research and supported by the French Ministry of Health and the French National Evaluation and Accreditation Agency. This project has four objectives: (1) to select a set of QI -2003- (2) to implement them in 2004-2005 in a volunteer panel of hospitals (3) to compare the hospitals anonymously (4) to explore quality management implications.QI were selected with a four–step process: (1) Establishment of a list of national priorities for Quality Improvement in relation with the Ministry of Health. (2) setting up a potential list of QI regarding these priorities. The COMPAQH staff determined a preliminary set of 81 QI, based on data in the literature and evidence about the scientific soundness of quality measures and the effectiveness of methods for improving quality. (3) Evaluatation of the preliminary list. Each QI was presented in a pamphlet describing its operational definition, rationale, methodology, workload and responsibility of data collection The hospital panel (representatives) ranked the 81 QI with a validated evaluation tool which contained four dimensions : Importance, Scientific acceptability, Feasibility, and Usability. (4) Development of a consensus on a final selection. Based on a structured voting process (Delphi method, two rounds), the hospital panel selected a comprehensive set of 42 QI among the 81.Results(1) Eight national priorities were defined: pain management, continuity of care, management of nutritional disorders, Iatrogenic risks (including nosocomial infections), patient satisfaction, follow-up of practice guidelines, management of human resources, accessibility. (2) A set of 42 QI were selected: a set of 6 core QI and 7 to 18 specific QI according to the hospital type.Conclusion Such a set of QI provides a foundation for developing a quality measurement system in French hospitals. It requires a pragmatic view for implementing them and a coherence between the different objectives of use (internal and external use).
    Revue d'Épidémiologie et de Santé Publique. 01/2005; 53:22-30.
  • [Show abstract] [Hide abstract]
    ABSTRACT: In the year 1996, accreditation process started in France by the mean of a special law, which creates for any health structure an obligation to be assessed within 5 years and a specific organism to set up the assessment process and train a network of assessors which have to be professionals still in exercise (medical, nursing and administrative staff). Since the process has to be started from scratch it was decided to set up a unique accreditation manual, which has to be applied to any setting and speciality. In this context psychiatric resources fear to be penalised because of their specificities specially their extramural activities. This paper is using the first 120 accreditation processes completed by ANAES (the French accreditation organisation) in order to compare psychiatric and no psychiatric resource results to accreditation. The process was based on assessment on how the diverse quality criteria described in the manual were completed; they were 10 main chapters concerning patients rights and information, patients charts, care organisation, management, logistic, risk management and quality control. Each of the chapter has been divided into up to 10 criteria, which set up rules for quality and propose references which have to be first self assessed by the team; the assessors will then start from this self assessment to make their own and propose a note. The notes are compiled to decide if the reference has been fully completed, partially completed or not completed at all resulting in recommendations or reserve. At the end of the process all chapter results allow to pronounce the accreditation status of the structure: no reserve no recommendation, recommendation, reserve, major reserve. This last result means no accreditation and is attributed when a major risk has been assessed; in this case the structure will be given a short delay to modify the problem and is reassessed. This paper is based on analyses of 65 no psychiatric and 41 psychiatric hospitals private or public; the 14 hospitals, which include psychiatric units, were excluded since it was impossible to attribute results to a specific unit. In comparing the global accreditation results, the psychiatric one have better results: no one got a major reserve and 22% vs. 46.6% for the no psychiatric got reserve, on the other hand the psychiatric got more recommendation 58.5% vs. 35.4% but more no reserve no recommendation 19.5% vs. 13.8%. The comparison of the average number of either reserve or recommendation is also in the favour of the psychiatric ones. In order to look at the diverse domain we compare percentages of no psychiatric and psychiatric having at least a reserve or recommendation in one of the diverse domain covered by the manual: no psychiatric have better results in all domain especially patients charts, care organisation which were significantly worse in the no psychiatric. However, when the profiles are compared (relative percentage of either reserve or recommendation) psychiatric got lower results for patient’s rights and information and better results concerning patients charts and care organisation. Since this quantitative comparison was back up by a qualitative evaluation based on interview on diverse professionals, which participated to the accreditation process, some specific theme emerged which could be check by looking in detail to the corresponding criteria. The first theme is dealing with infectious risks: sanitarians risk controls and prevention mechanisms assessments were more frequently deficient in the psychiatric hospitals than in the control group, however they were not difference for setting up these controls and mechanisms. The second theme concerns patient’s rights and information: when results are looked at globally no difference appears however this theme appears higher if the psychiatric accreditation profiles; when looking at the specific reference concerning the intimacy and dignity patient’s respect psychiatric results seem lower: 9.4% of the psychiatric hospitals reserve or recommendation are concerned by this reference vs. 2.5% in the no psychiatric. Some limitation of these comparisons should be mentioned; size effects was controlled for and no difference was found but all results presented here concern the first accreditation procedures where the most concerned and performing structures were candidates. We assume that this obvious bias affects both type of structure at a similar degree, which may be wrong. To conclude the accreditation process could be applied to psychiatric as well as no psychiatric structures and it appears that psychiatric resources obtain globally better results than no psychiatric; however two conflicting areas appear for the psychiatric resources: risk control procedures and intimacy and freedom patients rights completion.
    Annales Medico-psychologiques - ANN MEDICO-PSYCHOL. 01/2004; 162(4):255-261.
  • Revue D Epidemiologie Et De Sante Publique - REV EPIDEMIOL SANTE PUBL. 01/2004; 52:89-89.
  • Revue D Epidemiologie Et De Sante Publique - REV EPIDEMIOL SANTE PUBL. 01/2004; 52:91-91.
  • Source
    Valentin Daucourt, Philippe Michel
    [Show abstract] [Hide abstract]
    ABSTRACT: To identify the areas of needed improvement that were most frequently identified in the first 100 accredited hospitals by the French Accreditation College (FAC) according to the standards manual. To compare the outcomes of accreditation procedures according to the status and size of the accredited hospitals. We analyzed the first 100 summaries of accreditation reports available on the website of the Agence Nationale d'Accréditation et d'Evaluation en Santé (ANAES). Data were collected on hospitals, accreditation processes, and outcomes (decisions of the FAC). For each decision, we assessed the relationship with accreditation manual criteria, and analyzed their distribution by chapter. of data synthesis. Among the 100 accredited hospitals (40 public, 43 private, 17 mixed), nine were accredited without recommendations for improvement, 47 with recommendations, 40 with reservations, and four with major reservations. All of them received requests for improvement. No significant difference was found concerning the FAC decisions according to status and size of hospitals, although there was a trend that the larger the hospital, the more numerous and more serious the decisions of the FAC. The main topics addressed by decisions were those given high priority by the FAC (information given to patients and its traceability on patient records, and signing of prescriptions for medication). Despite wide heterogeneity in the summaries on accreditation and in FAC decision-making, this study provides an initial insight into common quality defects and ANAES priorities for hospitals in France.
    International Journal for Quality in Health Care 01/2004; 15(6):463-71. · 1.79 Impact Factor
  • Philippe Michel, Francois Roques, Samer A M Nashef
    [Show abstract] [Hide abstract]
    ABSTRACT: To assess whether the use of the full logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE) is superior to the standard additive EuroSCORE in predicting mortality in high-risk cardiac surgical patients. Both the simple additive EuroSCORE and the full logistic EuroSCORE were applied to 14,799 cardiac surgical patients from across Europe, of whom there were 4293 high-risk patients (additive EuroSCORE of 6 or more). The systems were compared for absolute prediction and discrimination (area under the receiver operating characteristic (ROC) curve). Actual mortality was 4.72%. The logistic model was closer to this than the additive model (4.84% (4.72-4.94) versus 4.21 (4.21-4.26)). Most of this difference was due to high-risk patients where actual mortality was 11.18% and predicted was 7.83% (additive) and 11.23% (logistic). Discrimination was similar in both systems as measured by the area under the ROC curve (additive 0.783, logistic 0.785). The additive EuroSCORE model remains a simple "gold standard" for risk assessment in European cardiac surgery, usable at the bedside without complex calculations or information technology. The logistic model is a better risk predictor especially in high-risk patients and may be of interest to institutions engaged in the study and development of risk stratification.
    European Journal of Cardio-Thoracic Surgery 06/2003; 23(5):684-7; discussion 687. · 2.67 Impact Factor
  • Source
    European Heart Journal 06/2003; 24(9):881-2. · 14.72 Impact Factor

Publication Stats

4k Citations
122.50 Total Impact Points

Institutions

  • 1998–2014
    • Université Victor Segalen Bordeaux 2
      Burdeos, Aquitaine, France
  • 1997–2003
    • University of Bordeaux
      Burdeos, Aquitaine, France
  • 1999
    • Papworth Hospital NHS Foundation Trust
      Papworth, England, United Kingdom
  • 1993
    • Unité Inserm U1077
      Caen, Lower Normandy, France
  • 1992
    • Université Bordeaux 1
      • Département Informatique
      Talence, Aquitaine, France