[Avoidable hospitalizations: the sole responsibility of primary care?]. / Hospitalisations potentiellement évitables: une responsabilité des seuls soins de premier recours?

BACKGROUND: Avoidable hospitalizations are used as a performance indicator of primary care in many countries. We investigate here the validity and usefulness of this measure both at a global scale and for the French healthcare system. METHODS: A scoping study was performed to take a critical look at this concept. The different uses of avoidable hospitalizations as an indicator have already been reported in two recent systematic literature reviews. RESULTS: Rates of avoidable hospitalizations seem to be far more correlated with the socioeconomic attributes of patients than with primary care supply. The few studies conducted in France confirm this international trend. Several weaknesses have been spotted in the building of this indicator: the choice of conditions that can be considered as sources of avoidable hospitalizations, their identification among hospitalization disease codes, the quality of hospital coding procedures, the ecological bias in the data collection of illustrative variables. CONCLUSION: Guidelines for improvement of this indicator are provided. In particular, we discuss the possibility of its use at the scale of the whole healthcare system.

[How to build the legitimacy of patient and consumer participation in health issues?]. / Comment construire la légitimité de la participation des usagers a des problématiques de santé?

Initially introduced by Juppé in 1996, the legislative reforms of January 2nd and March 4th 2002 legally enacted new forms of consumer representation and participation in the development of the health system. However, it appears that while this new role which was created to ensure legitimate participation has been recognised by law in theory, it has not necessarily received the same recognition and incorporation in practice at the grass roots level. As a result, it is now essential to think about practical methods of representation in order to sustain local legitimacy of consumers and patients on the ground and construct it from the bottom-up. The goal of this work was to understand how and under what conditions local legitimacy for health care system consumers, as a particular group of actors, can be effectively built, independently and irrespective of the specific question of elective democratic processes. The foundation of this work is based on material which resides in the collection of data from various local participation experiments that we or other researchers have contributed to establishing in a select group of health care settings. The results of this analysis serve to update a list of principle factors through which the legitimacy of the health care system's users is constructed. Such factors include the following: the promoting agents' expectations vis-à-vis the system's users and the a priori status which is given to them; the identification and selection methods used for choosing users, and the link to the types of users in terms of representation; the nature of the "generalisation" process for decision-making, understood as the process which transforms individuals' words and perspectives into collective ones; and the conditions for and modes of interaction between laypersons and professional experts. Finally, the paper presents the potential conflictive relationship or tension which may exist between representation and legitimacy with regard to the process for building legitimacy. The authors discuss the links between local, direct and elective democracy. The paper describes how effective democratic conditions can be built on the ground (from self-administered legitimacy to externally acknowledged legitimacy). In conclusion the paper succinctly distinguishes the articulation between representative democracy and participatory democracy, and how local processes relying upon direct democracy may be properly linked to the more "classical representative model of democracy".

[Benchmarks for evaluating health care networks]. / Repères pour l'évaluation des réseaux de soin.

The implementation of health care networks is accompanied by a certain number of changes in professional practices. At the heart of the economic, organisational, and political stakes, new questions are posed, especially in terms of evaluation. It is true that no matter what the level of evaluation, it is essential today to rely on the support of existing information systems to develop methodologies and to better describe the produced effects, in economic terms and health outcomes, but also from a more qualitative point of view, in terms of public satisfaction and benefits to clients. Laying down some reference points for evaluating health care networks can only contribute to developing new strategies for showing their effectiveness and efficiency and learning from this experimentation.

[Risk factors, diseases and health care acceptance in perinatology: a predictive model of hospital use]. / Facteurs de risque, pathologies et recours au soin en périnatologie: un modèle de prédiction du recours à l'hôpital.

BACKGROUND: The goal of our study was to develop a predictive model of resource use for pregnancy and perinatal care based on the knowledge of the distribution of risk factors in a given population of pregnant women. METHODS: Data recorded in Outcome of Pregnancy Certificates (CIG) from 11 voluntary maternities of the district of Seine-Saint-Denis allowed us to identify those pathologies that were predictive of premature births and prenatal hospitalization of mothers. We built a classification of disease states and of risk level. A logistic regression using disease states as dependent variables and risk levels as independent variables allowed us to compute expected rates with their confidence intervals. RESULTS: Among singletons, malformations, diabetes, toxemia, intra-uterin growth retardation, premature rupture of membranes covered 25% of all pregnancies but explained 64% of maternal hospitalizations; 90% of all mothers hospitalized and with delivery before 37 weeks gestation had at least one of these disease states. But 85% of the women who did not belong to disease classes had a normal pregnancy and delivery. CONCLUSIONS: In a given population, the distribution of risk levels is predictive of the incidence of disease per class. Then, given the length of stay of mothers per class, the rate of transfer of babies and the length of stay in postnatal care, we can simulate bed occupancy and compute bed capacities. The precision of the model is globally good, despite the relatively modest size of our initial data base: it will improve with the use of the model and the expected more widespread availability of data in France.

-Anesthesia and analgesia practice patterns in French obstetrical patients-. / Anesthésies et analgésies pratiquées dans les maternités françaises.

OBJECTIVE: To assess the rate of epidural analgesia (EA) for parturition and the techniques of anaesthesia for Caesarean section (CS). STUDY DESIGN: Retrospective study. PATIENTS: A series of 84,235 deliveries. METHODS: The series was extracted from a total of 770,054 deliveries carried out in 1991, according to the number of births in each hospital (1/1 if the births were < or = 100 per year, up to 1/25 if they were > or = 2,000 per year). The data analyzed included: anaesthesia technique, whether or not there was an anaesthetist on night duty at the hospital, birth rate in the hospital, type of hospital: university (UH), general (GH) or private (PH). For vaginal deliveries, the mode of labour commencement (spontaneous or induced), the multiplicity of pregnancies and a history of past CS were also noted. RESULTS: Vaginal deliveries: the overall rate of EA was 37.2%. EA were not carried out in 5% of maternity hospitals. In cases of spontaneous labour, the average rate was 32.1%, significantly less than for induced labour (59.6%, P < 0.0001) and in cases of previous CS (39%, P < 0.05). There was no statistical difference in cases of multiple pregnancies (35.7%). The average rate of EA was correlated to the number of annual births (P < 0.001) and was increased when the anaesthetist was present in hospital at night (P < 0.001). It was also significantly lower in GH (P < 0.001) than in UH or PH, which were equivalent. Scheduled CS: general anaesthesia (GA) was carried out at a significantly higher rate than regional anaesthesia (RA) (49.7% vs 48.4%, P < 0.05). In 15.1% of hospitals, RA was not available. The incidence of RA was influenced neither by the rate of annual births nor by the presence of the anaesthetist in the hospital during night. However, RA was significantly less frequent in GH (46.3%, P < 0.001) than in UH (48.6%) and in PH (53.6%) which were equivalent. CS during labour: the incidence of RA was significantly higher than GA (53.2% vs 44.1%, P < 0.001). In 17.1% of hospitals, RA was never carried out. The rate of RA was correlated to the size of the maternity hospital, and significantly higher (P < 0.001) when the anaesthetist was present in hospital during night. The differences between UH, GH and EP were the same than for scheduled CS. CONCLUSION: In France in 1991, the average rate of 37.2% for EA for obstetrics was high when compared to the rate in United Kingdom. It was equivalent to those in United States and Ontario, Canada. The discrepancies between hospitals were mainly related to structural and organizational factors. The influence of the size of the maternity hospital, the 24-hour service of EA was also shown in other studies. However, the difference between GA and UH and PH is a French particularity. The high rate of GA for CS differs largely with those in the UK or the USA. The time saving aspect of GA was probably an important factor for the choice of this technique. This study must be reactualized and enlarged to determine the demand of EA for labour by parturients and obstetricians.

[Type of birth center and conditions of transfer of neonates under 1500 g or gestational age under 33 weeks]. / Lieux de naissance et conditions de transferts des enfants de moins de 1500 g ou d'âge gestationnel strictement inférieur à 33 semaines.

BACKGROUND: Perinatal care's organization has been widely discussed in France during this last decade. Until now, transfer of high-risk neonates from their birth maternity to a pediatric unit using mobile vehicles led by specialized teams is encouraged in this country. POPULATION AND METHODS: Retrospective analysis of the type of maternities of birth for a population of 717 newborns, weighing less than 1,500 g and/or of gestational age under 33 weeks, extracted from a sample of 84,279 births in 1991. RESULTS: Only 15.6% of studied births took place in a maternity including a special intensive care pediatric unit (international level 3); 58.7% of those newborns where transferred outborn. There was a significant difference between the immediate access of newborns to a level 3 pediatric unit according to the location-of birth: significantly fewer newborns were directly transferred to a level 3 unit when born in a facility that included a level 2 pediatric unit, compared with those born in facilities that included a level 1 or 3 pediatric unit. CONCLUSION: Strong efforts should be made to identify mothers at high risk of giving birth to extremely prematured babies or babies with a very low birthweight so that births could take place in maternities properly equipped for their care. Perinatal care's organization should be built on a hierarchical network of maternities and pediatric services related to the risk of the population. Accreditation of maternities and pediatric services could help moving towards this kind of organization.

[Analysis of delivery facilities and conditions in 1991 in mothers of infants with birth weights below 1500 grams and/or gestational age under 33 weeks]. / Analyse des lieux et des conditions d'accouchement en 1991 des mères d'enfants dont le poids de naissance était inférieur à 1500 g et/ou l'âge gestationnel inférieur strictement à 33 sem.

AIMS: To identify maternities (in terms of level of activity and linkage with pediatric services) where, in 1991, mothers gave birth to infants of gestational age less than 33 weeks and/or with birth weight under 1500 g. To analyse factors linked to the probability of choosing a particular maternity as place of delivery for this population and measure the impact of maternal transferts on the rates of deliveries. To estimate the number of neonate transferts which could be avoided with simple recommendations. To propose policies that would allow France to come closer to the results of other reference countries in term of management of obstetrical and neonate care. MATERIAL AND METHOD: We extracted from a retrospective sample of 84,279 births (out of a total of 770,148), 717 infants meeting previously defined criteria and related to the population described above, using univariate and multivariate analysis and logistic regression. RESULTS: The Odds-ratio for a delivery to take place in a maternity with a volume of more than 2000 deliveries a year, compared with those doing less than 300 deliveries, is 4, 12. Only 15.8% of those deliveries took place in maternities linked to a level 3 pediatric unit (i.e where an intensive care neonates unit was located in the same building). 39.5% of births took place in level 1 maternities where no required pediatric service existed. Logistic regression techniques showed that the choice of a maternity for mother referral was more linked to the number of deliveries than to its level of pediatric services. In the studied population, 46% of the difference between the observed number of births in high volume maternities (compared to the expected number) could be explained by a maternal referral. The analysis of deliveries showed that for 34% of mothers who gave birth to a baby in a level 1 or 2 maternity, there was a possibility of being referred easily in a level 3 maternity. CONCLUSION: This study shows that the level of care of mothers at high risk of delivering a very premature and/or hypotrophic infant is far from international standards. Simple actions could double the number of births taking place in adapted maternities. We propose to both obstetricians and pediatrists, a common program to enhance the level of care.

[Cesarean sections in France: impact of organizational factors on different utilization rates]. / Césariennes en France: impact des facteurs organisationnels dans les variations de pratique.

In this study, we analysed the potential impact of organizational factors to explain the variation of cesarean sections' rates. We used a retrospective sample of 84,372 deliveries and two subsamples of low risk deliveries for cesarean sections. We determined different organisational factors that included: juridical and financial status of maternities, their architecture, the type of on-call for obstetricians, pediatrists and anesthetists, the annual number of deliveries and the level of pediatric staff and equipments of the maternities. We used multiple regression techniques to study the specific effect of each parameter, while controlling effects of age and parity of the mothers. We have found that even on the low risk samples, variation of rates were important. The type of on-call, the level of pediatric services and the architecture of maternities exerted a strong and significant effect on the rate of cesarean sections compared to the absence of impact of the number of deliveries. We discuss the reasons why, explaining the occurrence of those factors and then, stress the need to take into account the relevant factors for organizational audits. It appears that, in the context of the new regulation of the health system, these results should give obstetricians reasons to enhance their efforts to correct inefficient practices and to respect consensual guidelines and joint accreditation of obstetric and pediatric units.

[Coding problems in medical information in the framework of the medicalization of the hospital information system]. / Problèmes posés par le codage de l'information médicale dans le cadre du programme de médicalisation du système d'information hospitalier.

The authors review the difficulties presented by the description of medical data, on the basis of the french experience with the programme to medicalize the hospital information system. They explain the different steps in preparing a hospital discharge abstract, and the difficulties presented by each one. They particularly stress the phase of hierarchization in choosing the principal diagnosis. They propose some solutions to improve data quality and an approach based on medical practice patterns.

Measles-specific lymphocyte reactivity and serum antibody in subjects with different measles histories.

Measles hemagglutination-inhibiting (HAI) antibody titers and measles-specific lymphocyte reactivity were studied in subjects who had previously received killed measles vaccine and had been recently reimmunized with live vaccine--persons who previously had atypical measles, subjects with multiple immunizations with live vaccine, adults with a history of measles, and persons with primary measles immunization. Twelve to 18 months after revaccination of former recipients of killed vaccine, all 31 subjects had measles HAI antibody titers greater than or equal to 10 and only three had lymphocyte stimulation indices (LSI) greater than 2.8. The mean LSI and geometric mean titer of antibody were similar to those of persons who had had natural measles many years ago. One subject who previously had atypical measles had an HAI titer of less than 5 and an LSI of 0.5; another had an LSI of 6.0 and an HAI titer of 80. Only one of seven subjects who were immunized more than once with live vaccine had an LSI of greater than 2.8 one year later. The finding of measles-specific lymphocyte reactivity and antibody levels in revaccinated former recipients of killed vaccine that are similar to those in persons who had natural measles suggests that the risk of future atypical measles in this group is slight.

Revaccination of previous recipients of killed measles vaccine: clinical and immunologic studies.

Clinical and immunologic studies were performed in association with revaccination with live measles vaccine in 75 adolescents 11 to 14 years after immunization with killed measles vaccine. Ten subjects had local pain with swelling or erythema or both at the injection site; in three, the local reactions were severe and disturbing systemic complaints were also noted. These marked reactions were more common in subjects in whom the interval between the last dose of killed vaccine and the dose of live vaccine of the primary immunization series was less than or equal to 2 months, in subjects with prevaccination serum HAI antibody titers of less than or equal to 5, and in subjects with high measles antigen specific lymphocyte stimulation ratios. Serum complement levels could not be correlated with clinical reactions. Measles specific lymphocyte stimulation ratios were significantly higher in recipients of killed vaccine than in three compara,ive groups, in subjects with a killed-live interval in the initial vaccine series of less than or equal to 2 months as compared with greater than or equal to 3 months, and in subjects with prevaccination HAI antibody titers of less than or equal to 5 as compared with titers greater than or equal to 10. Although both low serum antibody and high measles specific lymphocyte reactivity were associated with marked local reactions, and probably indicative of susceptibility to atypical measles, our findings suggest that exaggerated lymphocyte reactivity is of greater importance in the adverse clinical response.