Search

Resume Notre hypothese de travail est qu’une meilleure connaissance du devenir des patientes souffrant d’anorexie mentale devrait contribuer a la prevention des rechutes et des complications et permettrait d’evaluer l’efficacite des traitements. L’evaluation actuelle du devenir de l’anorexie mentale s’appuie classiquement sur l’etude de la survenue d’evenements objectifs, qui ne refletent qu’imparfaitement la realite du devenir des patientes en termes de vecu subjectif. Nous proposons ici la construction d’un outil d’evaluation du devenir de l’anorexie mentale base essentiellement sur la perception des patientes. La methodologie utilisee a repose sur : (1) la conduite d’entretiens sur le mode de l’association libre aupres de 30 patientes anorexiques mentales ; (2) l’analyse de leurs contenus etayee par les hypotheses preexistantes et les hypotheses nouvelles soulevees par l’expression du vecu des patientes pour degager des themes a inventorier ; (3) la construction d’un autoquestionnaire a partir du developpement de chaque theme en plusieurs questions reprenant les mots des patientes. L’utilisation de ce nouvel outil est proposee dans des etudes epidemiologiques prospectives ulterieures pour une evaluation scientifique du devenir des patientes, afin de comparer l’efficacite des differents traitements et de mieux prevenir les complications futures.

Resume Objectifs. – Decrire la politique obstetricale concernant la gestion de l'accouchement en cas de grossesse gemellaire biamniotique, a terme. Patientes et methode. – Une enquete a ete realisee aupres des responsables des maternites appartenant au reseau sentinelle Audipog ( n = 170). Un questionnaire a ete ainsi envoye par la poste. Resultats. – Le taux de participation a ete de 72,35 %. L'analyse porte sur 124 questionnaires. Une cesarienne prophylactique est realisee par 0,8 % des repondants en cas de grossesse biamniotique et par 57 % des repondants lors d'une grossesse monoamniotique. Une cesarienne prophylactique est realisee respectivement pour 74 % des repondants si le premier et le deuxieme jumeaux sont en siege (J1 et J2), pour 81 % des participants si J1 est en siege et J2 en cephalique et pour 68 % des repondants si J1 est en siege et J2 en transverse. 86,7 % des repondants considerent que l'accouchement d'une grossesse gemellaire avec J1 en siege et J2 en cephalique est ressenti comme plus risque que celui d'un siege a terme, en cas de grossesse unique. Si J1 et J2 sont en siege, 73 % des participants considerent l'accouchement comme etant egalement plus a risque neonatal que celui d'un siege unique. En revanche, si J1 est en cephalique et J2 en siege, ils ne sont plus que 17,5 % des repondants a considerer cet accouchement comme etant plus a risque neonatal que celui d'un siege a terme. Discussion et conclusion . – A l'exception de la radiopelvimetrie, de la presence du pediatre et d'un anesthesiste pour l'accouchement, les politiques de service apparaissent assez homogenes. Une politique de cesarienne prophylactique systematique en cas de grossesse gemellaire biamniotique est peu repandue.

Identification of newborn babies with fetal growth restriction remains a problem both from the multi-factorial aspect of fetal growth and from statistical definition. Besides the usual terms: ‘Small for gestational age’ (SGA) and intrauterine growth retardation (IUGR), often used synonymously, the term ‘fetal growth retardation’ was recently introduced in reference to the genetic growth potential of infants. From a sample of 72,000 births, we have designed a statistical model in order to estimate the expected birth weight taking into account gestational age, sex, birth rank, maternal age, height and pregravid weight, we then calculated an individual limit of birth weight under that a newborn must be considered as having a ‘fetal growth restriction’ quoted FwGR. This new approach allowed us to identify 2 groups of newborns with FGR, one classically identified as SGA (noted FwGRI) (3.9%), and the other newly identified as FGR (noted FwGRII) (1.4%). In contrast, this approach allowed us to identify a group of ‘constitutionally small’ infants according to their constitutional growth potential (1.1%). In other words, among infants usually defined as SGA (5%), 22% appeared in fact to be ‘constitutionally small’ and therefore misclassified. As an initial validation, we observed that the proportion of maternal hypertension during pregnancy was low in the ‘constitutionally small’ infants (close to that of the normal group), and three times higher in the newly identified group FwGRII than in the normal group. Following these results, it seems to be meaningful to follow-up this new group of FwGRII infants, in terms of catch-up growth and neurodevelopmental outcome.

Objectives: After a previous study had shown the existence of psychological risk factors of pre-term delivery, we designed a study aimed at assessing the effect of psychotherapeutic support of pregnant women hospitalised with pre-term labour, followed by a second multicentric study aimed at demonstrating the reliability of such an intervention. Methods: Both studies were conducted in two successive cohorts of patients hospitalised with pre-term labour at 18–35 weeks of gestation. The initial study comprised 157 patients in each group, whereas the reliability study comprised 191 patients in the experimental group versus 202 in the control group. In each experimental group, the patients were offered psychotherapeutic support in addition to the usual clinical management. The psychological support included interviews with a psychologist and a collaborative work plan implemented with the nursing staff. Results: The analysis, conducted in the ‘intention to treat’ manner, shows a significant decrease in the early pre-term birth rate (Conclusion: This study offers new and major results related to the prevention of delivery before 35 weeks of gestation, both in the initial study as well as in the reliability study. Thus, providing this type of psychological support to women hospitalised for pre-term labour, in the context of antenatal care, can help to avoid early pre-term births and their complications in terms of brain damage and neuropsychological development.

Resume Un programme d'education pour la sante a l'ecole maternelle a ete mis en œuvre afin d'ameliorer le sommeil des jeunes enfants. La methode retenue est celle d'une intervention avec randomisation en deux groupes (groupe intervention et groupe temoin), des enfants de 3 ans inscrits a l'ecole maternelle dans le departement du Rhone en 1992 (140 classes et 1 500 enfants par groupe). L'intervention, d'une duree de 2 ans, a ete realisee par les medecins de protection maternelle et infantile et de sante scolaire. Leur role, lors des bilans de sante, etait de mobiliser les equipes pedagogiques, de diffuser des outils pedagogiques specifiques, et de sensibiliser les familles au respect des rythmes de sommeil des enfants. L'evaluation de ce programme a ete faite en recourant a une analyse statistique utilisant un modele de regression logistique avec prise en compte des facteurs de confusion potentiels, completee par une analyse en sous-groupes. Cette evaluation a porte sur les modifications des connaissances des familles et de la duree de sommeil des enfants. Les resultats montrent la faisabilite d'une telle intervention sur une large population d'enfants scolarises en ecole maternelle, ainsi que l'efficacite de l'intervention, d'une part au niveau des connaissances des families avec une diminution significative du risque de « mauvaises connaissances(OR = 0,76), notamment en zone urbaine, et, d'autre part, chez les enfants avec reduction significative du risque de « nuit courte(OR = 0,50) dans les sous-groupe des enfants « petits dormeurs a 3 ansdes familles de bas niveau social.

Our working hypothesis is that a better insight into the outcome of patients suffering from anorexia nervosa should contribute to preventing relapses and further complications and assessing treatment efficiency. Through anorexia nervosa, the patients express the difficulty they have to view themselves as specific subjects.The current classic outcome evaluation is based on the study of objective events, which only partially reflect the reality of the patients' outcome at a subjective level. The objective of this study was to set up a new assessing instrument of the outcome of patients suffering from anorexia nervosa, essentially based on the patients' perception of their experience.The methodology used has been based on: (1) the conduct by the main investigator of unstructured interviews using "free association", with the help of an interview guide. The anorexia nervosa patients were recruited among those who were hospitalized on an isolation contract, or among outpatients under a psychiatrist/psychoanalyst's supervision, aged over 25 years old so that they may have started their reproductive life. The study included 30 patients; (2) the analysis of the interview contents backed by preexisting hypotheses and by new ones suggested by the expression of the patients' perception, so as to set up an inventory of new themes; (3) the construction of a self-administered questionnaire starting from the development of each theme into several questions taking up the patients' own words and offering 4 possible answers (disagree completely, disagree, agree, quite agree).The analysis of the interviews contents has led to the development of 11 themes. The self-administered questionnaire includes a total of 124 items stemming from the development of each theme into between 9 and 16 items that were mixed in the version submitted to patients.This original interpretation of the outcome of the patients through their experience provides a better understanding of their relation to desire and pleasure, and consequently of the evolution of their subjectivity. By integrating several aspects of the disease expression, our instrument constitutes an alternative to the combination of several non-specific tools in anorexia nervosa. It thus avoids the atomization of the pathology and respects the specificity of its structure. The analysis of the disease function in the emergence of their subjectivity rather than the static observation of its symptoms has led to the development of new themes.The validation of this new methodological approach of the follow-up of anorexia nervosa based on the patients' perception of the evolution of their disease, aside from anthropometrical or physiological parameters, will have to be tested on a new population of patients. A quantitative score will be developed in association with the self-administered questionnaire. Its use in further epidemiological studies will enable a scientific assessment of the patients' outcome, and better prevent further complications and relapses, by screening patients with a pejorative risk. The ultimate aim is to improve these patients' care.

Early postpartum discharge is a recent practice in France and for which there are few national data.The Sentinel AUDIPOG network was used to describe the practice of postpartum early discharge (3 days in case of vaginal delivery and5 days in case of caesarean section) from 1994 till 2002, after standardization on the distribution according to the type of hospital and the region (n = 128232). Secondarily, the prognostic medical factors of an early postpartum hospital discharge were researched, in univariate analysis and logistic regression analysis, of the births of 2001-2002.Early postpartum discharges concerned 3% of the deliveries in 1997 and 7% in 2002. Eearly postpartum discharge was more frequent in the level II and III obstetric facilities, in non- university hospitals, in facilities with more than 1 500 deliveries a year, in urban hospitals and in Paris and the surrounding area. The women leaving early in the postpartum were more often multiparas, with no pregnancy pathology, with a single pregnancy, without postpartum hemorrhage, and a child2500 g without risk of infection. At present, 40% of vaginal delivery and 25% of caesarean section primiparas and 55% of vaginal delivery and 30% of caesarean section multiparas could be discharged early.The reduction of the postpartum hospital stay is inevitable but it is advisable to take care that the women who are discharged are medically fit for discharge.

This study was carried out to build statistical models for defining FGR (Fetal Growth Restriction) in weight and/or length after taking growth potential of an infant into account. From a cohort of pregnant women having given birth to 47,733 infants in 141 French maternity units, two statistical models gave individualized limits of birth weight and birth length (based on the 5th centile) below which, after adjustment for its individual growth potential, a newborn must be considered as FGR in weight and/or in length. A sample of 906 infants had measures taken of cord blood growth factors (IGF1, IGFBP3). The FGR(W) definition (weight

The AUDIPOG association (Association of Users of Computerized Files in Perinatalogy, Obstetrics and Gynaecology) proposed in 1994 the setting up of the Sentinel Network, including voluntary public and private maternity units from all areas in France. Each year, every member sends individual data related to a sample of births to the AUDIPOG Network. Today, the common database comprises 95,000 births collected during one month by all members of the network. Thanks to a standardization procedure, the perinatal health indicators, obtained from the common database, are valid estimations at the national level. The results of a given year are published at the beginning of the following year, providing a continual survey of perinatal indicators in France. Since the regionalization of perinatal services, a common perinatal information system is of great importance for a continual survey of perinatal health indicators and assessment of medical practices. Among the most important results of this perinatal network, we observed an increase in the rate of multiple pregnancies from 3.5% of births in 1998 to 4.2% in 2001. The increase of multiple births seems to be related to a lower maternal age in case of procreative medical assistance. As a consequence, the rate of preterm births was also increased, resulting in 7.2% of all births. In another section a regular increase in the rate of cesarean sections, from 16.3% in 1999 to 17.6% was observed in 2001. Cesarean sections before labour are chiefly concerned. From 1999, each member of the AUDIPOG network is given access to the entire common database, on the Internet AUDIPOG site, in order to describe medical practices when faced with a particular clinical situation of interest. This access to the common database aids self-assessment of medical practices and quality improvement (http://audipog.inserm.fr).

1- In order to define a "consensual attitude" in case of breech presentation, we queried the AUDIPOG network on obstetrical practice in search for the more frequently accepted criteria for normal delivery. 2- For consensual practices that could be identified, early neonatal complications were compared between women cared for in maternity wards applying these practices and those cared for in maternity wards not applying these practices.1- A confidential survey was sent to participating obstetrics units to determine their practical attitude in case of breech presentation. A consensual attitude was established on the basis of their responses. 2- Crude and adjusted comparisons were made concerning neonatal results between the consensual and non consensual groups of obstetrics units.Response rate was 85% for 175 obstetrics departments consulted. Six criteria were identified leading to classification of 42% of the units as consensual obstetrics units. The risk of major neonatal complications was lower in the consensual group than in the non-consensual group: adjusted OR=0.27 (95% CI: 0.09-0.85).A questionnaire on obstetrical practice can help define consensual attitudes associated with lower risk of neonatal complications.

To describe the evolution of medical practices in breech presentation at term and to compare early neonatal complications according to initial obstetrical decision on the type of delivery.71919 pregnancies are included in the AUDIPOG Database from 1994 to 2000. Among this cohort, pregnant women with a singleton in breech, at term were selected which represented 2136 women after the exclusion of in utero deaths and medical abortions. The first outcome was global criteria of severe early neonatal complications ("death during per or immediate post partum or transfer of the newborns to an intensive care or surgery unit"). The size of the sample authorized a power of 90%.The rate of cesarean section before labor was 40%. We found 2.3% of early neonatal complications in the group in which labor was accepted vs. 1.9% in the group with cesarean section performed before labor (p0.05). After having taken into account prognostic factors, we do not see a significant difference for the risk of neonatal complications between the two groups (OR=1.33; 95% CI: 0.63-2.80).The analysis of the AUDIPOG Database describes the French obstetrical practice in breech presentation at term. We do not find a different risk in morbi-mortality as regards to the initial obstetrical choice regarding the type of delivery, but the absence of randomization in our study does not authorize a strong medical evidence to guide national recommendations.

The aim of this study was to examine the relation between cervical length and the presence of funneling and the risk of preterm delivery.This prospective blind cohort involved 200 hospitalized women with preterm labor in a tertiary care hospital. Women were recruited for a single transvaginal ultrasonography to assess cervical length and presence of funneling. The main outcome measures were: (1) relative risks (RR) and adjusted odds ratios of preterm delivery (37 weeks' gestation); (2) time interval between the cervical ultrasonography date to 37 weeks' gestation or to-preterm birth.The RR of preterm delivery according to the cervical length (cut-off of30 mm) was 2.79 (95% CI 1.70--4.59). The RR according to the presence of funneling (cut-off of5 mm) was 1.39 (95% CI 0.99--1.95). The adjusted odds ratio was 3.92 (95% IC 1.75--8.75) for cervical length and 0.77 (95% CI 0.35--1.67) for funneling. Women with a cervical length of30 mm had a significantly shorter interval from ultrasonography date up to 37 weeks' gestation than did women with a cervical length of30 mm (P0.003).Ultrasonographic mensuration of the cervix provides predictive information on the risk of preterm delivery.

The purpose of this study was to describe the care level, legal status and size of the maternity units where deliveries take place in France according to risk level for mother and infant. We analyzed the 1997-1998 data to better implement the network's perinatal policy.and methods. Our standardized sample included 4200 single births in 1997 and 3650 in 1998 collected by the French Sentinel Network after applying a sample rectification technique to offset the methodological problems created by the volunteer nature of the sample population. Distribution of care level, legal status and size of the maternity unit where deliveries took place were recorded according to the risk level of the patients.In 1997-1998, 22% of pregnant women delivered in level III maternity units and 33% in level II units. Twenty percent of the deliveries took place in level II maternity units with less than 1500 deliveries per year. During this period, women with diabetes or hypertension delivered more often in level III units (31% and 27% respectively) than women in the general population. This was not true for women with a previous perinatal death (23%). Inversely, births of infants before 33 weeks gestation or weighing less than 1500 g occurred more often in level III maternity units (55% and 59% respectively) than in the general population. Twenty-seven percent of the infants requiring neonatal transportation were born in level I maternity units. For 'low risk' mothers, delivery occurred more often in level I maternity units (more than 50%) or in small maternity units with less than 1000 deliveries per year (45%) than for the general population (45% and 36% respectively).These data obtained from the Sentinel Network provide precise information on where deliveries occur in France. These data will be useful for implementing the network's perinatal policy. They will also provide a means of following referral practices in the future.

The growth supervision of children using growth curves is a widespread and useful tool in general pediatric practice. In France the latest reference curves are rather ancient, therefore it seemed to us interesting to re-examine some growth parameters and to compare them to the current reference data.The studied sample was composed of 7,000 children from the Rhône-Alpes region in France who were seen for a school health check-up. Anthropometric measurements (35,000) related to weight, height and cranial circumference of these children aged from 0 to 6 were selected from their health booklets. Centile curves for these three variables were drawn from these measurements using the LMS method, which is specifically suited to these types of data.If height and cranial circumference can be considered as normally distributed, weight is markedly skewed to the right, reflecting a high prevalence of children with heavy weight. Comparison with current references data shows more or less similar results for weight, height and cranial circumference: the medians and the 2.5 centile are constantly higher in our sample than those of the reference data (the discrepancy increasing with age). The differences are more important regarding the part of the distribution which concerns the highest values: overall, the 90th centile of our distribution corresponds rather closely to the 97.5th one of the current reference data. Thus, at 6 years of age, the 97.5th centile of the reference weight distribution for girls is 23,400 kg, while it is estimated at 27,770 kg in our sample. One should take in account the different characteristics of the two studies to interpret these differences.The aim and use of such growth curves is discussed, together with recent computer applications in this field.

The establishment of the mother-baby relationship is often disturbed in case of preterm birth. Clinical observations and epidemiological studies have highlighted the existence of a relationship between psychological factors and preterm birth. Therefore, a psychological support may have a beneficial effect in order to prevent preterm delivery. The aim of this study was to assess the effectiveness of psychological support in pregnant women with symptoms of preterm labor, which is considered as a somatic expression of psychological difficulty related to pregnancy. The study, which involved two cohorts of women, hospitalized or not, identified during two successive periods in the same maternity ward, included 309 women in the experimental group and 323 in the control group. The women in the control group were followed up according to the usual therapeutic procedures, whereas additional psychological support was offered to the experimental group. 92% of the women who met the psychologist accepted the psychological support. The analysis, conducted "in intent to treat", showed a significant decrease in the preterm birth rate in the experimental group (12%) compared with the control group (26%). This study confirms the feasibility of this kind of intervention and the effectiveness of psychological support on the risk of preterm delivery. The interviews help the women regain confidence in their own capacity to fulfil their pregnancy.

The predictive value of lipoprotein(a), Lp(a), for coronary artery disease, is strongly suspected, though unproven. The normal serum level is 0.3 g/l. We searched for correlations between serum Lp(a) levels and coronary artery disease in a population of patients hospitalized in a general cardiology unit.Serum Lp(a) was assayed in all patients consecutively hospitalized during 1994 in the Valence hospital cardiology unit. Two groups were distinguished: patients with coronary artery disease (n = 444) and those presumed free of coronary artery disease (n = 555). Coronography were performed when required. Serum Lp(a) levels were compared for the following variables: age, sex, smoking habits, blood pressure, total cholesterol, HDL and LDL-cholesterol, triglycerides and apolipoproteins A1 and A2. Univariate, then multivariate analysis were performed first patients of all ages, then for those aged more and less than 60 years.Univariate analysis demonstrated that Lp(a)0.3 g/l was associated with coronary heart disease (OR = 1.33; p = 0.03), although this correlation was no longer significant after adjustment for other known risk factors (OR = 1.28; p = 0.07), except in the subgroup of patients over 60 years of age (OR = 1.37; p = 0.04).There was a non-significant trend favoring an association between serum Lp(a) level0.3 g/l and the development of coronary artery disease.

The observation of medical practices and the analysis of outcome as a function of practices are essential today for monitoring and improvement of the healthcare system. The AUDIPOG Sentinel Network provides a tool for the self-assessment of practices that enables each unit to assess its position relative to a group of maternity wards.Pooling of individual data about births that occurred during the month of January each year in about a hundred volunteer wards allows assessing perinatal health indicators and following their evolution over time. A sample rectification technique, adapted from the quota method, allows offsetting the methodological problems created by volunteering and assessing perinatal indicators at the national level.The analysis of data collected from 1994 to 1997 indicates a number of trends, even in the short run, and justifies that a real practice monitoring and assessment tool be set up in order to improve perinatal health.

An educational project in nursery schools was carried out with the aim to improve children's sleep. Its effectiveness was tested in terms of awareness of families and length of sleep of children. The method was a prevention trial based on the random division of classes of 3-year old into two groups (intervention group and control group) comprising children enrolled in nursery schools in the Rhône region (France) in 1992. The project, lasting 2 years, relied on physicians from the community (Maternal, Child Health Service and School Health Service), their role being to mobilise the teaching teams, distribute specific teaching tools and raise awareness of families to respect the sleep patterns of children, during routine medical examinations. The study was carried out in 140 nursery schools with 1,500 children in each group. The evaluation was based on the application of a logistical regression model taking potentially confounding factors into account, and an analysis in sub-groups in order to demonstrate the effectiveness of intervention on particularly exposed groups. The results show: 1) the feasibility of such an intervention among nursery school children on a large scale, 2) the effectiveness of the action, with a reduction of the risk of "poor knowledge" among parents (OR = 0.76), particularly in urban areas, and a reduction in the risk "short nights" in the sub-group of children who had " little sleep at 3 years" in low social class families (OR = 0.50).

PURPOSE To define the role of a 10-Gy boost to the primary tumor in the conservative treatment of early infiltrating breast carcinoma treated by limited surgery and radiotherapy. PATIENTS AND METHODS Between 1986 and 1992, 1,024 women with early breast carcinoma (< or = 3 cm in diameter) were treated by local excision, axillary dissection, and conventional 50-Gy irradiation given in 20 fractions over 5 weeks and then randomly assigned to receive either no further treatment or a boost of 10 Gy by electrons to the tumor bed. The median follow-up time was 3.3 years as of September 1994. The occurrence of telangiectasia was reported, and the patients were asked to evaluate the cosmetic result. RESULTS At 5 years, 10 patients of 521 who had received the boost (Kaplan-Meier estimate of local relapse rate, 3.6%) and 20 of 503 who had received no further treatment (Kaplan-Meier estimate of local relapse rate, 4.5%) had developed a local recurrence (P = .044). After adjustment for the main prognostic variables, the relative risk was still significantly lower for the boost group (0.3; range, 0.12 to 0.95). The boost group had a higher rate of grade 1 and 2 telangiectasia (12.4% v 5.9%), but no difference was seen between the two treatment arms in the self-assessment score for the cosmetic result. CONCLUSION Delivery of a boost of 10 Gy to the tumor bed after 50 Gy to the whole breast following limited surgery significantly reduces the risk of early local recurrence, with no serious deterioration in the cosmetic result. Additional follow-up evaluation will be required to assess the long-term results.

Breast cancer is the most common cancer in women worldwide. Many studies have been performed worldwide to assess the effectiveness of screening in terms of reduced mortality due to breast cancer. Since the end of 1989, 10 breast cancer mass screening programs using mammography have been carried out in France under the sponsorship of the National Fund for Health Prevention, Education, and Information (FNPEIS) from the National Health Insurance of Salaried Workers (CNAMTS). These 10 campaigns, which are on a district scale, are organized according to variable methods and are assessed using a common procedure. Four groups of criteria are measured in this procedure, which investigates the impact, quality, effectiveness, and costs of screening programs. The average and extreme values of each criterion as calculated from the campaigns are presented in this paper. In order to enlighten the judgment on the French results, a comparison with the international standards in force and with the results of foreign screening programs is proposed.

The purpose of this work was to comparatively assess the results of mass screening programs for breast cancer implemented in six French departments in 1986, within the scope of the National Fund for Health Prevention, Education and Information of the National Health Insurance Office of Salaried Workers.The data collected by the screening centres were analyzed by ten assessment teams that were independent from the program promotion staff, all using the same evaluation form. A complementary population study performed in eight French districts then, allowed assessing the frequency of self-referred screening (mammography performed out of program).The rate of participation in screening programs, in relation to the invited population, ranged from 21 to 48%, according to the district (36% in average). This low participation was probably related to the extent of self-referred screening. In fact, 19 to 40% of women, according to the district, had previously had a screening mammographic coverage: rate was around 68% in women aged 50 to 69 years. Positive findings with mammography ranged from 4.5 to 15.8% (10.1% in average), while intervention rates ranged from 0.7 to 1.6% and detection rates from 3.8 to 6.2%. The ratio between benign tumors and cancers ranged from 0.7 to 2.1 according to the district. In order to enlighten the judgement on French results, we propose a comparison with the international standards in force.The various experiences with breast cancer screening in France show that this screening is technically feasible on the basis of existing medical structures. However, some criteria are still below the expected values, especially if compared with international standards. This result is probably accounted for by the high rate self-referred screening before age 40 in France. In these conditions, the question is whether extending breast cancer screening programs in France is an appropriate course of action.

A file of 100,716 children born between 1984 and 1988 in 22 maternity wards was used to estimate for each week of gestational age and gender, the exponent β which makes an index Weight/Length

The identification of intrauterine growth retarded infants remains a problem. It is useful to distinguish between small for date which uses the reference of a statistical birthweight limit and intrauterine growth retardation. A study comprising nearly 100,000 births from many French regions allowed a definition of the limits of birthweights taking into account gestational age, sex, birth rank, maternal height and pregravid maternal weight. This new approach distinguished two groups of newborns, the "constitutionaly small" and those who are newly called "growth retarded". The validity of this new classification is discussed.

The collection of nearly 100,000 births from 22 maternity hospitals in various French regions permitted to establish foetal growth curves for weight, height and head circumference. The 50th percentiles of weight at term obtained in this sample are not different from those given by Leroy and Lefort in the Paris region in 1971. However below 38 weeks of gestation, we found that 50th percentiles are 100 or 150 g less than those given by Leroy and Lefort. The discussion from international papers show large variations of the 10th percentile for weight in according to gestational age and sex, these variations are often related to the choice of different target populations.

A first experiment aimed at collecting data from 98 maternity hospitals grouped in the AUDIPOG Sentinel Network addressed 7,837 deliveries in January, 1994. The technique used for recovering the results was adapted from the quota method and aimed at strict respect of marginal distributions of deliveries per regional area and per hospital type, and at the best the joint distribution of deliveries per regional area/hospital type, to obtain estimates of perinatal health indicators. Selection indicators (age, family status, socio-economic conditions, previous history), practising indicators (supervision of the pregnancy, entrance motive, presentation, delivery mode, anaesthetic) and result indicators (prematurity, small for date, transfer, pathology of the new-born) are given.

Knowing that 20-30% of intra-uterine growth retardation are unexplained, the authors assume a psychological component that may influence fetal growth. A first psychologic approach consisted in having interviews with about fifty women who had just given birth to small-size babies. The analysis of the contents of these interviews, performed in the free association mode, allowed screening six psychopathologic themes : denial of pregnancy, painful emotional or physical experiences, operative functioning mode, unessential role of the husband, lack of an imaginary child, and previous affective deprivation. The verification of the hypothesis arising from the psychological approach will be based on a complementary epidemiological approach.

In the framework of the National Fund for Prevention, Health Education and Information, the setting-up in France, on an experimental basis, of mass breast cancer screening programs by mammography in ten departments leads to suggest a common protocol for the evaluation of these programs, before an eventual generalization to the whole country. These programs are based on common principles: local screening based on existing medical facilities, single view mammography with double reading. However, the screening organisation is different in each department according to the target population, the screening interval, the mode of invitation, the methods for arranging double reading, etc. The purpose of this article is to report on the work of the evaluation group of the ten programs, and to propose a methodology for comparing the screening experiences in order to measure the role of the organisation methods. It gives definitions and modes of calculations for the evaluation criteria in four fields: impact, quality, efficacy and cost (with details about the feasibility of data collection) and proposes a reflection on analysis techniques in order to develop an optimum mass screening strategy.

The increasing level of childhood obesity has prompted some research into the early risk factors likely to lead to preventive measures. A case-control study was conducted of a five year-old population in the Rhône and Isère administrative "départements" of France during the children's first visit to primary school, with the participation of 327 obese children and 704 controls. Anthropometric data on the children since birth, together with data on their lifestyles, were collected in interviews with parents. "Family obesity" and "obesity at birth" were found to be closely related to the child's obesity at five years old (adjusted OR = 2.7 and 2.1 respectively). Of the environmental factors, the hypothesis tested paid specific attention to television viewing, snacks between meals and lack of sleep. These three variables were all found to be risk factors of obesity at five years old. However after allowing for parental obesity, the only remaining significant risk factor for obesity at five was lack of sleep (adjusted OR = 1.4). The pathogenic assumptions raised by these results are discussed.

The purpose of this study was to explore the eventual relationship between the working conditions, and overweight women working in hospitals. The sample comprised 594 women working in hospitals in Lyon; 34% of them were overweight which is 10% above the theoretical weight. The rate of overweight hospital staff increased with the woman's age, the number of children and when the social level decreased. The analysis showed that being overweight was more frequent in women working in cleaning staff (56%) and in nursing auxiliaries (36%) than in nurses and administrative staff (24% and 27% respectively). After taking the above personal risk factors into account, this relationship remained the same. However, the night work and the length of service in hospitals are not found to be risk factors, after adjustment for the personal risk factors and type of work. Considering the women who were not overweight at the time of commencing employment, the relationship between being overweight and the type of working conditions in the hospitals disappeared.

Tumorecomy with axillary dissection followed by locoregional irradiation is the standard treatment of T1 T2 ≤ 3 cm breast cancer. The optimum dose to the breast and the tumor bed remains controversial. To answer this question a phase III study was conducted in Lyon since 1986. Main criterias of inclusion were T1 T2 ≤ 3 cm with free margins on the pathological specimen, age no more than 70 years. Randomization was between two dose levels: Arm 1: whole breast irradiation 50 Gy/20 F/5 weeks (dose specified on the 95% ICRU isodose); Arm 2: same dose plus a boost of 10 Gy with electrons 9 or 12 Mev (isodose 90%). First end point will be local recurrence. For this criterium, on the assumption of a 1.2 annual rate of local recurrence, in patients with boost, and a 2.4 rate in those not receiving boost, with an annual censured rate of 1%, at one side sent, a type I error equal to 5% and a power of 85%, a number of 1014 patients must be included in the trial. Between 01/86 until 07/92, 1024 patients were enrolled in this trial. Randomization was performed with stratification on pTNM classification. The groups were comparable for know prognostic factors. Results The local relapse free survival after 5 years was 96.4% in the 60 Gy group (tested group) and 95.5% in the 50 Gy group (controlled group). The hazard ratio, adjusted for other prognostic factors was 0.31 (0.11 to 0.88—P = 0.028). The disease free survival after 5 years was 86%, in the 60 Gy and 82.2% in the 50 Gy group. The adjusted hazard ratio was 0.6 (0.38 to 0.96—P = 0.037) There was no significant difference for overall survival between the two groups. Telangiectasis grade I were present in 6% in the 50 Gy group and 12% in the 60 Gy group. This result suggests a trend in favor of the boost (60 Gy) in term of local control with a small and acceptable detrimental effect on cosmetic results.

The purpose of this study, which interviewed 356 couples of childbearing age in the Lyon region of France, was to analyze the different attitudes of men and women towards parenthood, and to identify the social and personal characteristics which may modify these attitudes. The 2 groups in the study, the 183 couples who wanted an additional child and the 127 who did not, were compared in relation to socioeconomic status, woman's work status, woman's age at 1st birth, and absence of a child of each sex. (46 couples with conflicting intentions were excluded.) The couples intending to have further children on average had significantly fewer children, were younger, and less frequently had a child of each sex. The other factors were statiscally insignificant. The fertility rate of couples not wanting more children was above the replacement rate of 2.6. The rate for couples intending to have another child was comparable at 2.4, except that the average 5 year age differences between the 2 groups was higher than the usual interval between 2 births, reinforcing the trend toward a model family of 2 or 3 children and greater intervals between consecutive births.

RésuméLa Hopkins Symptom Checklist-58 ou HSCL-58 (Derogatis et al., 1974) est un autoquestionnaire de 58 items largement employé au cours des essais de psychotropes et en épidémiologie psychiatrique pour l’évaluation des troubles névrotiques et affectifs. Cet instrument permet des scores sur 5 dimensions définies sur des bases théoriques et empiriques (analyses factorielles), dont la validité repose sur la stabilité de la structure factorielle au sein de différents échantillons. Or, la plupart des analyses factorielles de la HSCL-58 ont concerné des groupes de patients recrutés dans des centres de soin (groupes cliniques). Il importait donc de vérifier la stabilité factorielle sur des sujets névrotiques provenant d’une population tout-venant (groupes subcliniques).Au cours d’une enquête épidémiologique sur les facteurs de risque de prématurité (Mamelle et al., 1987), menée dans 4 maternités de Lyon, 1 643 femmes enceintes de 6 mois avaient rempli un questionnaire psychopathologique comprenant 45 items extraits de la HSCL (la plupart des items manquants par rapport à la HSCL-58 n’appartenaient à aucune des 5 dimensions). Les évaluations d’avant-grossesse ont été utilisées pour définir un groupe ≪subclinique≫, en sélectionnant les femmes qui avaient obtenu les scores totaux à la HSCL (45 items) les plus élevés, ce score procurant une estimation de gravité névrotique. Le seuil de gravité ne pouvant être qu’arbitraire, nous avons décidé de retenir un ratio nombre de sujets/nombre de variables égal à 10. Au total, 457 femmes (27,8% de l’échantillon initial) ont été considérées dans l’analyse. Pour nous assurer de la proximité de ce groupe ≪subclinique≫ d’avec les groupes ≪cliniques≫, nous avons comparé les notes moyennes des items d’anxiété et de dépression de notre échantillon avant et après sélection, avec celles de 3 groupes décrits par Derogatis et al. (1974) : 2 groupes cliniques, un de patients névrotiques anxieux, un autre de déprimés névrotiques, et un troisième groupe de sujets normaux, représentatif de la population d’Oackland. Une analyse en composantes principales avec rotation varimax assortie d’une méthode de choix du nombre de facteurs décrite par Comrey (1978) a été utilisée. Les notes moyennes des items d’anxiété et de dépression de notre groupe ≪subclinique≫ de 457 femmes se sont révélées proches de celles des névrotiques anxieux de Derogatis, mais plus faibles que celles des déprimés névrotiques, et plus élevées que celles des sujets normaux (Tableau I). Une solution à 4 facteurs est apparue la meilleure: ≪vulnérabilité≫, ≪somatisation≫, ≪tension≫, et ≪troubles cognitifs≫ (Tableau II). Deux facteurs (somatisation et troubles cognitifs) reproduisent de près la structure attendue, le facteur vulnérabilité étant proche d’un facteur dépression, mais les items de sensitivité et d’anxiété n’ont pas formé les facteurs espérés.Nos résultats confirment donc les données de la littérature concernant la stabilité relative des dimensions somatisation et obsession (ou plutôt troubles cognitifs), et l’instabilité des dimensions anxiété et sensitivité, le cas de la dépression étant intermédiaire. Il semble, en conclusion, que les scores dimensionnels de la HSCL-58 ne présentent pas une validité suffisante pour être employés dans les groupes subcliniques.

In a French Department (Rhône), the prevalence of infantile autism was found to be 10.8 per 10,000 in the 5-9 age group, using broad operational criteria. The prevalence decreases to 5.1 per 10,000 when another set of criteria, similar to DSM-III, is used. The boy/girl sex ratios for these two sets of criteria were 2.1:1 and 2.3:1 respectively. The use of DSM-III turned out to be problematic for taking into account a broad spectrum of early psychotic disorders, autistic-like, yet more polymorphous. A subdivision concerning the existence of associated major neurological anomalies has appeared promising but requires finer validation.

INSERM established a working group to evaluate the effects of VDT work on pregnancy outcome. The group examined all available scientific papers concerning radiation emitted by VDTs, animal experimentation, and epidemiological studies. An estimation of risk was made for each outcome (congenital malformations, spontaneous abortion, stillbirth, premature birth, and low birthweight) according to a decisional algorithm, a method that gives preponderance to epidemiological arguments. On this basis, the group was able to show that the use of VDTs poses no increased risk of congenital anomalies, stillbirth, or premature birth. Consensus evaluation of divergent epidemiological results concluded that VDT work is not a risk factor significantly affecting spontaneous abortions. The risk of low birthweight could not be validly assessed.

This review of the causes of recurrent spontaneous abortion is combined with a statistical evaluation of the frequency of abortion and of the probability of abortion according to the way in which previous pregnancies have ended.

The known variations in weight for gestational age as well as compounding maternal and/or fetal factors, puts into question the usual definition of small for gestational age (SGA) which is based on gestational age alone. On the basis of more than 20,000 births studied in 4 maternity hospitals from 3 separate regions in France, the authors propose a new definition for growth retardation in full-term babies, taking into account 4 factors: gestational age, sex, birth rank, height and usual weight of mothers. This definition allows constitutionally SGA newborns to be considered normal, while some babies previously classified as normal would now lead one to suspect intra-uterine growth retardation (IUGR). This approach tries to take constitutional and environmental risk factors into account in fetal growth hopefully allowing for help better detection of IUGR.

A study whose aim was to define the role of psychological factors in prematurity etiology has led us to develop a methodology for evaluation of personality which could have a connection with the occurrence of a premature birth. The proposed method--interviews based on free association followed by the construction of questionnaires serving as a base for an epidemiological study--seem to us applicable in order to show the psychological components which intervene in the occurrence of a somatic illness.