Expert group and authors

Pierre-Yves ancel, Epidemiologist in perinatal and women's health, INSERM IFR 69, Villejuif

Christine bonnier, Paediatric neurologist, Clinique Saint-Luc, UCL, Brussels, Belgium

Antoine burguet, Infantile resuscitation, University hospital, Poitiers

Evelyne combier, Health-related economy and management, INSERM U537, CNRS UMR 8052, Le Kremlin-Bicêtre; ENSP, Rennes

Brigitte estournet-mathiaud, Paediatrician and infantile resuscitation, Raymond Poincaré Hospital, Garches

Vincent gautheron, Physical and rehabilitation medication, University hospital, Saint-Etienne

Dominique mahieu-caputo, Obstetrics, Xavier-Bichat Hospital, Paris

Stéphane marret, Neonatal paediatrics and resuscitation, University hospital, Rouen

Françoise molenat, Child psychiatry, University hospital, Montpellier

Eric plaisance, Sociology of education, CERLIS (Research centre on social links), CNRS UMR 8070, Paris 5 university

Jean-Christophe roze, Neonatology, University hospital, Nantes

Annie triomphe, Health economist, CNRS UMR 8595, Paris 1 University

Danielle valeur-masson, Paediatrics, EMSA, Nursery nursing and perinatology institute, Paris

Véronique zupan-simunek, Neonatal resuscitation, Antoine Béclère Hospital, Clamart

The following presented communications

Alain beucher, Multidisciplinary and departmental EMSA, Robert Debré University hospital, Angers

Christine cans, Child handicaps register and perinatal observatory, Isère, Grenoble

Scientific and editorial coordination

Fabienne bonnin, scientific assistant, INSERM collective expertise centre, Xavier-Bichat Faculty of medicine, Paris

Catherine chenu, scientific assistant, INSERM collective expertise centre, Xavier-Bichat Faculty of medicine, Paris

Jeanne etiemble, director, INSERM collective expertise centre, Xavier-Bichat Faculty of medicine, Paris

Catherine pouzat, scientific assistant, INSERM collective expertise centre, Xavier-Bichat Faculty of medicine, Paris

Marie-Josèphe saurel-cubizolles, in charge of expertise, INSERM collective expertise centre, Xavier-Bichat Faculty of medicine, Paris

Bibliographical assistance

Chantal rondet-grellier, documentalist, INSERM collective expertise centre, Xavier-Bichat Faculty of medicine, Paris

The International Classification of Diseases (ICD) and the International Classification of Function, handicap and health (ICF) are the two reference classifications

The International Classification of Diseases (ICD), which was created in 1948 under the direction of the WHO, is the principal official, terminological reference in medicine and epidemiology on an international level. It classifies health problems into diseases, disorders, lesions and traumas. This classification, however, based on a biomedical model, does not allow assessment of the chronic consequences of a trauma or a disease. The International Classification of deficiencies, Incapacities and Handicaps (ICH1) was established in 1980 with an experimental classification statute. This classification considers a handicap as an individual phenomenon and describes the cause-effect relationship in a linear fashion: the disease triggers an organic and functional deficiency—a lesion-related aspect—which leads to an incapacity in terms of a person's behaviour and activities—a functional aspect—that, in turn, leads to a disadvantage (handicap)—the situational aspect.

The context of the subject's life is, however, not apparent to the ICH1. Between 1996 and 2000, six temporary classifications were co-ordinated by the WHO, and the social handicap model was extensively imposed.

In 2001, the WHO General Assembly adopted the International Classification of Function, handicap and health (ICF) as a new tool for defining the state of health of the population and, in particular, the prevalence and significance of debilitating situations.

The International Classification of Function refers to three aspects of the handicap: the deficiency, the limitations in terms of activities and participation restriction. Deficiency affects an organ with regard to its function or structure. The limitation in terms of activities and the participation restrictions refer to the difficulties that a person may encounter when carrying out an activity or adopting a social role. The ICF is thus organised into two sections, each comprising two components: the first deals with function and the handicap in question (organic functions and anatomical structures, activities and participation), whereas the second concerns contextual factors (environmental and personal factors). Each component can be expressed either positively or negatively. These components are subdivided into fields within which are categories (classification units) at two, three, or four levels.

This is a framework classification that allocates a qualitative code defining the importance of the level of health or the severity of the attack (e.g., D166 expresses the reading performance). In the "activities and participation" component, the diagrams distinguish between two qualitative codes: the capacity code, which characterises a person's aptitude to carry out a task or action in a standard environment, and the performance code, which describes that the person in question is capable of dealing with the context of everyday life. The environmental framework can thus help to limit a person's activity and his/her participation. The discrepancy between capacity and performance reflects the difference in impact between a standard environment and a uniform environment. The ICF proposes a multi-dimensional vision in which environmental factors play a key role.

This classification can be used at different levels. Everyone, depending on his/her profession, can fit such or such a component. The "organic functions and anatomical structures" component is more specifically informed by the health care professionals, whereas the "activities and participation", "environmental factors", and "personal factors" components are informed by educational and social professionals.

In children, the deficiency component is the most used because it is known by care teams, and it is well described in the literature and can be measured more easily than limitations in terms of activities or participation restriction, for which the measuring tools are rare or even inexistent in children.

The literature refers to other approaches or handicap classifications relating to infant/child handicap. Generally developed in the United States, these classifications are essentially pragmatic. They are aimed at identifying the requirements of handicapped children and their families to assess the cost and generate assistance. They are not necessarily appropriate for identifying the perinatal origin of the handicap or aetiological approaches. Some authors emphasise the interest of non-category methods that analyse the consequences of various disorders of a medical, behavioural, or intellectual performance nature, considering that these methods are better suited to provide indications to the departments and organisations in charge of handicapped children.

The cause of approximately 50% of deficiencies and handicaps could occur during the perinatal period

The definition of a handicap of perinatal origin is not unique. For some authors, the handicap of perinatal origin may be defined as a handicap, the causal event of which occurs during life in the uterus after 22 weeks of amenorrhoea (according to the WHO) and up to 8 days or 28 days after birth, depending on whether the authors count the peri- and neonatal periods. More often than not, it is only following the elimination of other causes that handicaps are listed, the perinatal origin of which is feasible.

Approximately 20–35% of severe infant/childhood deficiencies are of antenatal origin (abnormal morphogenesis, chromosome aberrations, and genetic abnormalities) and between 5 and 10% of severe deficiencies are post-neonatal associated with trauma and, more rarely, an infection or a tumour. Thus, using a process of elimination, it is estimated that 55–75% of deficiencies could originate during the perinatal period. A cautious approach must, however, be adopted when interpreting this aspect because a perinatal risk factor is involved (prematurity, hypotrophy, and asphyxia at birth) or a cerebral abnormality of peri- or neonatal origin (anoxo-ischaemic encephalopathy, periventricular leucomalacia, and intraventricular haemorrhage) occurs in only 15 to 45% of cases.

These facts highlight the extent of the difficulties involved in trying to establish the origin of handicaps. This particularly affects children in whom no causal or risk factor is likely to account for the onset of the handicap, as well as those presenting with a risk factor (prematurity, hypertrophy, and asphyxia at birth) without it being formally possible to establish a cause-effect relationship.

The part played by perinatal causes varies depending on the type of deficiencies: motor, sensory, intellectual, and psychiatric

According to studies, motor deficiencies include motor infirmity of cerebral origin (MICO), which refers to a motor disorder, the origin of which can be traced back to the perinatal period or cerebral palsy, which signifies permanent motor disorders linked with a non-advancing, cerebral lesion, without pre-judging its origin. These also include deformities affecting the central nervous or musculo-skeletal systems, as well as degenerative or hereditary diseases.

Intellectual deficiencies are defined by an intellectual quotient (IQ) of less than 70. According to the WHO agreement, children with an IQ of between 70 and 85 are in a limit zone, whereas those with an IQ of 50 to 70 present with moderate mental retardation, and children with an IQ of less than 50 are classed as severely or deeply mentally retarded. Intellectual deficiencies also include trisomy 21, regardless of the severity of the mental retardation (but that is not considered as being of perinatal origin).

Auditory deficiencies refer to a bilateral loss exceeding 70 decibels (dB) before correction and visual deficiencies to a visual acuity of less than 3/10 in the best eye (after correction).

Mental disorders occurring between 0 and 3 years of age are mainly autism and other invasive disorders affecting development.

Finally, numerous handicaps associate a motor deficiency with a severe or profound mental deficiency, resulting in extremely restricted autonomy and the possibilities of perception, expression, and relations.

The role of perinatal causes varies according to the type of deficiencies. For instance, it seems higher in the case of motor deficiencies than for intellectual and sensory deficiencies.

In France, according to the data listed in the Register of infant/child handicaps and the perinatal research institute (RHEOP), the role of risk factors of perinatal origin in motor deficiencies is twice that for other deficiencies.

The prevalence of severe child deficiencies in the general population is approximately 1%

Regardless of the definitions considered and the methodologies employed, most of the epidemiological surveys conclude that the global prevalence of severe deficiencies is approximately 1% in children. It may reach 2% if moderately severe deficiencies are taken into account. However, data relating to the prevalence of handicap depends on the age of the child in question. On average, it takes 1 or 2 years to identify a motor or neuro-sensorial deficiency or incapacity, 3 or 4 years to detect a fine motor disorder and often longer than that to identify a disorder involving cognitive functions. The prevalence will therefore be higher in an older population. Prevalence data are generally set for children in the 5- to 7-year age bracket.

The prevalence of motor handicap (cerebral palsy) is estimated in the population registers of western countries to be between 1.5 and 3 per 1,000 infants born alive.

According to a recent review of the international literature, almost 3% of children will be moderately mentally retarded (IQ ranging from 50 to 70), but substantial variations can be seen between the studies (0.5% to 8%). As regards severe mental retardation (IQ <50), the average incidence is estimated to be 3.8 per 1,000 children based on the overall findings of the studies. The heterogeneity between studies is less significant than for moderate mental retardation.

As regards sensorial deficiencies, the prevalence of visual deficiencies ranges from 0.20 and 1.81 per 1,000 live births in Europe and the United States. In the case of blindness, the prevalence varies from 0.10 to 0.50 per 1,000 live births in Europe and the United States. Finally, the prevalence of auditory deficiencies ranges from 1.1 and 1.3 per 1,000 live births for a threshold set at 40 dB between 0.4 and 0.7 per 1,000 in the case of the most serious forms (auditory loss >70 dB).

On the basis of numerous international studies (including three French investigations), the average prevalence can be estimated at 1 per 1,000 children from 0 to 19 years of age. Taking all of the forms of invasive developmental disorders into consideration, the average prevalence is close to 3 per 1,000.

The fluctuations in terms of handicap prevalence observed in the literature reflect a better recording of the handicap, changes in medical practices and handicap definition. Furthermore, comparisons of care strategies adopted by teams in the different countries show that the number of children to be looked after, their survival and neurological fate can, to a large extent, be modified by ethical, cultural, social, and organisational choices (access to care).

The changing trends observed in recent years in several countries are, nevertheless, convergent. Taking all births into account, the prevalence of deficiencies has stagnated or even increased between 1970–1975 and from 1985–1990.

Half of the severe childhood deficiencies or handicaps affect children born full-term or close to term

The presentation of prevalence rates of handicaps of perinatal origin and changes in these rates can generally be clearly distinguished between infants born full-term and premature infants. Studies focusing on the fate of infants born full term mainly come from the population registers, whereas the studies concentrating on the fate of premature infants or infants with a low birth weight arise especially from cohort studies conducted in reference hospital centres (United States, Canada, and Japan) or are based on regional surveys (Europe and Australia).

In premature infants, the deficiency rate is far higher than in that observed in infants born full-term, namely of the order of 3 to 10% for the cohort of very premature babies following less than 33 weeks of amenorrhoea (AW) or even higher in the lower gestational age brackets.

The probability of seeing a deficiency or handicap develop is, overall, inversely proportional to the gestation period or birth weight. The prevalence of handicaps must, therefore, be considered per week of gestational age.

A marked increase in the prevalence of cerebral palsies was observed between 1975 and 1985 in very premature infants, especially for the more immature among them (born before 28 AW). Since then, prevalence has stagnated or even decreased in these infants. In infants born between 31 and 36 AW, the prevalence of cerebral palsy fell or stagnated between 1980–1985 and 1990–1995. Finally, no changes were observed in infants born full-term during the same period.

French data on the prevalence of severe deficiencies are similar to international data

The information relating to infant/child handicaps in France is either taken from a permanent and continuous recording within the register framework, or ad hoc transversal or cohort surveys. The principal deficiencies investigated are motor, intellectual, sensorial, and psychiatric.

Two registers currently exist in France, one in the Isère region and the other in the Haute-Garonne region. The Isère register (Register of infant/child handicaps and perinatal research institute, RHEOP) was created in 1991. To be included in this register, a child must have at least one severe, principal deficiency (defined above) and be a resident in the Isère region at 7 years of age. The main source of information is the Commission départementale de l'éducation spéciale (Special education departmental commission – SEDC). The other possible sources are the centres of early medico-social action (CAMSP) and hospital or specialist departments. The Haute-Garonne register is more recent since this was set up in 1999.

Three ad hoc or transversal surveys were carried out within INSERM unit 149. The first two surveys, which were conducted in 1985–1986 and 1989, focused on all children presenting with severe deficiencies and born in 1972, 1976, and 1981, and whose parents were residing in one of the 14 regions covered by the survey (a total of 6,013 children between 9 and 14 years of age). The third study involved all deficient children born between 1976 and 1985 whose parents were living in three regions (Isère, Haute-Garonne, and Saône-et-Loire) between 1992 and 1993 (6,174 children between 8 and 17 years of age). The SEDC proved to be the main source of information for these surveys. A survey was also carried out involving all handicapped children born in 1984 and residing in the Lorraine region (357 children). Finally, more recently, the Handicaps, Incapacities and Dependence survey (HID) carried out by Insee focused on a representative sample of persons living in specialist institutions in 1998 and on a sample of persons residing in these establishments in 1999. The main aim of the survey was to describe the incapacities and their consequences. More than 1,000 children at least 15 years of age were included in each sample.

The cohort surveys include populations of children at high risk of handicap, especially the very premature infants. The first was carried out in 1985 in children born between 25 and 32 AW in the Paris region, who were monitored up until 2 years of age. The second survey was carried out on an identical population of children born in Franche-Comté in 1990–1992 and followed up until 5 years of age. The third involved all of the very premature infants managed in the Fort-de-France neonatal resuscitation department that treated all children born live at less than 33 AW in Martinique between 1992 and 1995. Lastly, in 1997, all of the very premature infants born between 22 and 32 AW in 9 regions of France were included in the EPIPAGE study and followed up until 5 years of age. It was at this specific age that a full medical and psychological evaluation was organised in the centres set up for this purpose.

Currently, 9 in 1,000 infants born in the early 1990s suffer from a severe deficiency. The prevalence of motor deficiencies affects more than 3 in 1,000 children and almost 2 in 1,000 in the case of cerebral palsy. The prevalence rates for visual and auditory deficiencies range from 0.2 to 1.3 per 1,000. Finally, depending on the extent of the mental retardation, estimations for multi-handicapped children vary from 0.7 to 1.3 in every 1,000.

In the very premature infants, the first studies reported between 5 and 13% of cases of cerebral motor infirmity and almost 15% of cases of moderate or severe mental retardation. More recently and focusing on a larger sample, the prevalence of cerebral palsy reached 8% at 2 years of age (EPIPAGE study).

The Isère handicaps register shows that the number of children presenting with a severe deficiency has increased from 6.85 per 1,000 in 1980–1982 to 8.85 per 1,000 in 1989–1991. This change reflects the increase in the prevalence of motor deficiencies (+26–42% for cerebral palsies) and psychiatric disorders (+90%) between 1980–1982 and 1989–1991. The marked increase in the latter could reflect changes in the way in which disorder classifications are recorded rather than a real increase in the prevalence rate. No change has been reported for intellectual, auditory, and visual deficiencies.

In the very premature infants, a comparison of the results of the 1985 survey carried out in the Paris region with those of the EPIPAGE survey, which was conducted in the same region in 1997, does not reveal any change in the prevalence of motor deficiencies. None of the data allow changes in other, sensory or intellectual deficiencies to be followed up in these children.

Numerous factors are likely to influence the prevalence estimations provided by these studies. One of the crucial elements for these studies is the recording quality. As regards this particular point, the results depend on the way in which the information is collated. In the cohort studies, patients lost to follow-up were the main cause for concern. They represent between 7% and 20% of children taking part in the French studies, but the published data show that they constitute a population at high risk of handicap. In the transversal surveys and the registers, the problem has been one of exhaustivity. The exhaustivity rate is 86% in the Isère register. This is difficult to evaluate in transversal surveys. These facts suggest a slight underestimation in the prevalence of deficiencies in the results produced by registered as those arising from cohort surveys.

The terms used to define a deficiency may differ, depending on the studies in question. Such is the case for motor deficiencies. Cerebral motor infirmity (CMI, a French term) and cerebral palsy are two entities that match up but which cannot be superimposed on each other. However, studies recently carried out in France increasingly refer to cerebral palsy. Other phenomena may trigger differences. In particular, the degree of severity of the deficiency may vary between studies. The choice of the age at which data are registered can also affect the prevalence of handicaps. If the information is recorded at too early an age, then not all of the deficiencies, especially those of an intellectual nature, can be identified. Lastly, the quality of the data collated will play an important role. In transversal and cohort studies as well as registers, the sources of information are numerous as indeed are the number of people involved. This cannot guarantee the homogeneity of the data collection.

As regards changes over the coming years in France, two phenomena could help to reverse the trend observed to date. The first concerns the improvement in the antenatal screening of congenital deformities and a constant increase in medical involvement in pregnancies since the end of the 1980s. The second refers to the improvement in the neurological prognosis of populations at high risk of handicap such as very premature infants. This is, in fact, suggested in foreign data that confirm a stabilisation in the prevalence rate of cerebral palsy since 1990. Other phenomena could, however, counter-balance the effects of deformity screening and the management of very premature infants. In particular, if the age at maternity increases, then the number of infants born with deformities will also increase. Similarly, if prematurity continues to increase, the population of children at high risk of sequelae will also rise.

On the basis of the data presented above, it is interesting to estimate the annual number of children with a deficiency of perinatal origin among the population of children born full term and the number of children in the premature population (also distinguishing those born very prematurely). If 1% of children between 7 and 8 years of age is considered to present with a severe deficiency, the number of these children in France can be estimated to be 7,500 (750,000 births each year) for one generation. According to the published data, the proportion of handicaps of perinatal origin varies from 15% (low estimation) to 65% (high estimation). An average estimation of 50% seems reasonable. Moreover, premature infants are considered to represent approximately 50% of children suffering from a severe deficiency.

Thus, in one generation, 1,120 to 4,870 children will develop a severe deficiency of perinatal origin. Half of them were born prematurely (before 37 AW), including 280 to 1,120 very premature infants (born before 32–33 AW).

Perinatal events can be implicated in the onset of deficiencies or handicaps

The various perinatal diseases having a significant impact on the onset of handicaps or deficiencies can be classed in three major groups: prematurity and especially very early prematurity; neurological accidents in babies born full-term, mainly perinatal asphyxia; the other diseases threaten brain development.

Prematurity and very early prematurity in particular (birth before a gestational age of 32 weeks) has a major impact on the risk of all types of handicaps.

Evidence relating to various sequelae depends on the child's age. As a general rule, motor sequelae are diagnosed before 2 years of age and remain relatively fixed. Other disorders (learning or behavioural disorders) are detected much later at a pre-school or school age. The latter are progressive, and some of them are compensated for or regress over time such as, for instance, language disorders or social behaviour disorders. However, in the case of premature infants presenting with motor sequelae, the gap in non-motor skills persists and even tends to become accentuated over time.

Different factors influence the outcome of premature infants: intrinsic factors such as gestational age, birth weight, sex, twin births or neonatal diseases (cerebral lesions and bronchopulmonary dysplasia, etc.) and environmental factors such as pre- and post-natal care or family and social environment.

In the case of very premature infants, the early use of neurological exploration techniques improves the quality of the prognosis. This is especially evident with cerebral imaging—cerebral echography, magnetic resonance imaging (MRI)—and electroencephalograms (EEGs) during the neonatal period. The clinical neuromotor examinations carried out in the first few months post-term are useful, although their limitations in terms of predictive value must be remembered. These are sometimes difficult to interpret and their accessibility or quality vary from one neonatology unit to the next.

In full-term newborn infants in the event of neurological accident (perinatal anoxia, cerebro-vascular accident, spontaneous intracranial or traumatic haemorrhaging, and neonatal meningitis, etc.), the prognosis varies considerably depending on the nature of the accident. The best-established prognosis concerns neonatal encephalopathies, which are most generally assumed to be post-anoxic. These are manifested by a change in neurological behaviour during the first week of life. Minor encephalopathies primarily have no effect upon the child's outcome. Conversely, however, moderate and severe encephalopathies have a very important impact. Thus, in the case of moderate encephalopathy, neurological sequelae, half of which are cerebral palsies, are observed in 25% of newborn infants. The other sequelae are essentially cognitive (intellectual deficiency or language disorders). In cases of severe encephalopathies, 50% of surviving children present with neurological sequelae. Convulsions multiply the risk of motor infirmity 3-fold.

In full-term babies suffering from neonatal encephalopathy, the prognosis is based on three types of early examinations: a clinical examination during the first days of life; imaging, mainly MRI and electrophysiological examinations (EEG and/or evoked potentials). The predictive nature of these examinations, especially MRI, is described as very good in the literature (under optimal conditions). However, the experience of practitioners, who are confronted with various cases of falsely reassuring imaging, must be taken into account.

There are other perinatal risk situations, the impact of which on the onset of handicaps is more difficult to evaluate, either because they are rare or because they are combined with another risk situation (e.g. prematurity). Regardless of the prematurity and hypotrophy factors, twins, and monochorial twins in particular, may present with clastic cerebral lesions in utero or during the post-natal period, and deformities. The risk of lesions is all the greater when a transfuser-transfused syndrome is present.

There is a link between the presence of intra-uterine growth retardation (IUGR) and the occurrence of handicaps, regardless of the term of pregnancy. This risk is enhanced in premature cases. The prognosis appears to be chiefly conditioned by the severity of the hypotrophy. Prognostic factors in the event of IUGR are: perinatal asphyxia, gestational age, masculine sex, and a small head circumference (especially the failure of the head circumference to catch up). All of these factors potentiate the risk of intellectual deficiencies in particular.

Antenatal exposure to toxic substances (alcohol) or post-natal exposure (neurotoxic and ototoxic medicinal products), post-natal malnutrition, perinatal infections, severe jaundice, and severe neonatal diseases warranting a prolonged stay in an intensive care unit are also likely to promote the onset of sequelae, especially developmental problems and deafness.

All of these neonatal diseases demand prolonged follow-up, but this approach is not straightforward to arrange. A considerable difficulty for paediatric follow-up must be taken into account. Neuro-developmental sequelae are observed with higher frequence among families who fail to comply with follow-up schedules. At the present time in France, extended follow-up is proposed for children who are most at risk (early premature infants and neonatal encephalopathy in full-term babies). Even if it was applied, this follow-up policy would overlook more than half of the handicap-deficiencies of perinatal origin.

Different pathological situations can be seen at the origin of perinatal events involved in the onset of deficiencies or handicaps

Different pathological situations can affect events that impact upon the onset of a deficiency or handicap such as prematurity, extensive prematurity, encephalopathy in full-term infants, and other risk events.

Approximately 1/3 of prematurity is induced by medical consent and 2/3 are spontaneous. Within the scope of spontaneous prematurity, a distinction is made between premature labour on intact membranes (1/3 of premature labours) and premature labour on ruptured membranes (1/3). Variations, however, exist according to gestational age because the work on intact membranes affects 25% of premature labours between 27 and 32 AW, 44% between 24 and 26 AW, and 64% before 24 AW.

The risk factors exposing children to moderate prematurity and extreme prematurity are the same. The link with prematurity is, nevertheless, greater for certain risk factors including the socio-economic level, living conditions (single mother, etc.), the age of the mother, the body mass index, the obstetric medical history and smoking. Intra-uterine growth retardation is also a risk factor in prematurity.

Premature labour is often associated with another complication of pregnancy: arterial hypertension (2%), placenta praevia (4%), retroplacental haematoma (7%), and haemorrhagic disease (16%). Placenta anomalies (placenta praevia and retroplacental haematoma) are frequently associated with extensive prematurity (50%), as well as infections (38%) and immunological factors (30%). Cervical incompetence (16%), maternal diseases (10%), trauma and major surgery (8%), and foetal anomalies (6%) are also related factors. These causes are interlinked in 58% of cases and isolated in 38% of cases. No factor has been identified in conjunction with prematurity in 4% of cases.

Prenatal infection is also a significant cause of preterm delivery with intact membranes. Asymptomatic, intra-amniotic infection is observed in 13% of cases involving premature investigation of intact membranes. Chorio-amniotitis will develop secondarily in 37.5% of cases. These infections are also directly involved in the onset of cerebral lesions observed in premature infants.

Twins present with an extremely high risk of prematurity and low birth weight. The risk of prematurity is increased 10-fold and the risk of weight below 2,500 grams by 11. The excess risk is observed for both moderate and extreme prematurity. Thus, 10% of twins are born before 32 weeks compared with 1% of single infants. Similarly, the proportion of infants weighing below 1,000 g is 4% in twins and 0.4% in single infants. The condition of infants at birth is characterised by a higher number of stillborn infants and babies having an Apgar score of less than 8 at one minute or five minutes in the case of twins compared with single infants.

The threat of premature labour (TPL) is the primary cause of hospital admissions during pregnancy in France (38% of the 20% of pregnant women admitted to hospital).

Neonatal encephalopathy is another risk factor affecting the onset of deficiencies and handicaps at birth. Neonatal asphyxia is, in approximately 50% of cases, a consequence of labour, otherwise it can be prevented or screened. In fact, certain cases of per-partum asphyxia may be a consequence of antepartum pathological conditions. The other diseases screened by neonatal asphyxia indicators could be exclusively pre-natal pathologies.

The entire problem stems from the major difficulty involved in identifying per-partum asphyxia in situations where the indicators are hardly specific and common to the entire perinatal period.

It is difficult to specify the relative role of various per-partum anoxia risk factors because this varies depending on the populations studied. In favoured countries, the risk situations are placental insertion anomalies, primiparity, alcohol abuse, pre-eclampsia, the masculine sex of the fetus, and intra-uterine growth retardation.

As regards the impact on the approach adopted during labour, study results tend to diverge. Generally speaking there does not appear to be a routine Caesarean policy that would reduce the number of children suffering from cerebral palsy.

The risk situations for intra-uterine growth retardation are as follows: embryopathies of chromosomal, infectious, genetic, and toxic nature including the "foetal alcoholisation" syndrome and foetal malnutrition. Vasculo-renal syndromes remain the main cause of hypotrophy at birth (approximately 50% of cases). Risks are also heightened by multiple pregnancies, smoking, chronic vascular diseases, and certain maternal diseases. Echography plays a crucial role in the screening and diagnosis of intra-uterine growth retardation. Its sensitivity varies, depending on the term. It appears to be optimal at 34 AW with sensitivity on the order of 60% for a specificity of 85%. Some authors recommend combining clinical and echographical criteria to improve screening and evaluation of the severity of intra-uterine growth retardation, and propose predictive scores.

Early screening of cerebral lesions is essential but not life-threatening

Early screening of cerebral lesions or identification of the risks of impaired cerebral development are essential in a child. In fact, early intervention now holds an important place among the possibilities currently available for action.

Cerebral lesions acquired at the end of pregnancy or during the perinatal period represent the majority of lesions responsible for paediatric neurodevelopmental handicaps. Early screening can be carried out:

• before birth based on current echographist performances that reveal porencephalopathies, ischaemo-haemorrhagic strokes, microcephalies, calcifications of the cortical grey matter or basal ganglia, cytomegalovirus-induced foetal diseases in particular and ventricular dilatations;
• at birth in the presence of revealing clinical symptoms in a full-term newborn infant: early neonatal encephalopathy or isolated convulsions or hypotension;
• at birth by systematic cerebral imaging monitoring (transfontanellar ultrasonographic study or nuclear magnetic resonance imaging) of certain populations at risk: extremely early prematurity, twin pregnancies, even intra-uterine growth retardation or infants with an excessive birth weight;
• during the months following birth in the presence of early neuromotor abnormalities sometimes combined with growth retardation of the cranial perimeter during monitoring procedures in certain populations at risk during mandatory examinations performed at 9 months and 2 years in the general population.

The circumstances under which cerebral lesions are discovered vary, depending on the extent of the lesions and their topography. Their clinical expression during the immediate neonatal period is often poor when accidents have occurred at the end of pregnancy or in a premature, newborn infant. Routine cerebral imaging among populations at risk (twins, infants presenting with intra-uterine growth retardation and premature babies) will reveal cerebral lesion of ischaemic and/or inflammatory origin. Sometimes, the clinical picture will be more extensive, especially in full-term infants subjected to diffuse anoxic accidents affecting not only the basal ganglia but also the grey matter. Their clinical detection will often be delayed in relation to birth because of secondary neuromotor abnormalities.

Lesion topography is, in itself, closely related to the stage of cerebral development. Thus, given the vulnerability of its white matter, very premature infants often present with a disease affecting the white matter accompanied in particular by periventricular leucomalacias or intraventricular/intra-parenchymatous haemorrhages. In full-term infants, given the functional maturity of the neurones, grey matter and grey nuclei, early neonatal encephalopathy is evident from early neonatal encephalopathy by diffuse or localised neuronal necrosis.

The prognosis of various cerebral lesions and the importance of sequelae are directly related to the aetiology of the lesions but also to their extent and the stage of cerebral development at the time of the accident. Severe, secondary, intellectual deficiencies will rather be specific to the full-term newborn infant subjected to diffuse anoxia whereas more moderate deficiencies with praxic, memory, and educational learning disorders will be observed in more premature infants. Determination of a specific prognosis is still difficult, even impossible, for several reasons. The concept of prognosis remains highly subjective. Perinatal lesions occur in a fully developing brain. Abnormal images do not inform us about subsequent cerebral functions and the brain's capacity for reorganisation. In most cases, only a risk situation, the relevance of which will vary from one individual to the next, will be apparent.

However, electroencephalograms conducted during the first week and nuclear magnetic resonance are still currently the best tools in trained hands. Anomalies in the basic tracing in full-term newborn infants and premature infants alike help with determining the prognosis. Similarly, the presence of central grey nuclei involvement on MRI will be linked with a greater risk of motor sequelae in the event of cerebro-vascular accident or diffuse anoxic attacks. The use of these tools nevertheless presents acute problems: the reduced number of EEG experts as in neonatal imaging, discordance between the abnormalities observed with transfontanel echography and MRI, and a lack of MRI accessibility in certain regions.

Most handicaps or deficiencies of perinatal origin can be screened in the first 3 years

Three types of handicap can be identified in the first three years: motor handicap, neuro-sensorial deficiencies, and disorders affecting development (autism), intellectual deficiencies being identified at a later stage. Age correction in the event of prematurity must be carried out up until 2 years of age. There is frequently an additional delay with language in premature infants.

In the first months (before 6 months), severe motor handicaps of a tetraparesis nature, frequently associated with sensorial and intellectual deficiencies are highlighted by neuromotor examination, which confirms the lack of head posture, global hypotension, impaired continued vision, eating difficulties, and no spontaneous motility. These children must be assessed swiftly in specialised centres to complete the motor, visual, and auditory evaluation.

During the second 6 months of life, slow progress and delayed acquisition of head and seated posture draw attention. This is precisely the context in which diagnosis is often delayed and, in the case of non-specialist doctors, warrants referral for a specialist opinion. Development consultations in the neonatology or neuropaediatric departments or even in early medico-social action centres (EMSAC) should be used as referral options in such situations. The points noted in the medical records with the examinations performed at the 4th and 9ths month allow diagnosis to be approached in a precise manner.

At the end of the first year, serious and severe developmental abnormalities must be highlighted. These are mainly numerous handicaps, tetrapareses, severe diplegias and haemiplegias. If a family is given the diagnosis, the physician in question assumes that the child will be seen again on several occasions because it is essential to assess the slow advance to be certain of motor involvement. In the case of intellectual deficiencies, diagnosis is very rarely made before 3 years of age.

As soon as abnormalities are observed, the children have to be referred to development specialists who will conduct a more precise semiological analysis and investigate concomitant, particularly visual disorders for motor cerebral infirmities. Other abnormalities may develop, such as flexion spasms, respiratory disorders secondary to swallowing disorders, and orthopaedic abnormalities.

During the second year of life, a management approach must be structured for severe motor handicaps. Other deficiencies, especially orofacial involvement with its effect on verbal language, must be evaluated. Lastly, for some children, moderate motor disorders will disappear virtually completely, allowing the diagnosis of IMC to be diagnosed a minima or even transient neuromotor disorders.

Visual handicaps mainly affect the very premature infants who presented with retinopathy during the first weeks of life and for whom screening and follow-up must be rigorously organised as soon as they are admitted to hospital, especially if a therapeutic approach is adopted in hospital. In the advanced stages of retinopathy, 50% of infants are at risk of blindness. For others, an ophthalmological examination is urgent in the event of strabismus (squint) or nystagmus.

Auditory deficiency must be highlighted as soon as possible. The recommended age for identifying a major deficiency is 3 months in Anglo-Saxon countries. This seems early because, in some children, early auditory abnormalities can revert to normal in the second trimester of life. Babbling during the first few months does not mean that there is no deafness. In France, the diagnosis is usually made at the age of around 9 months.

Invasive developmental disorders (autism) are investigated during several consultations. Warning signs must lead to a specialist referral.

In the general population, medical records are inspected and, if examined thoroughly, should allow most deficiencies or handicaps to be highlighted. Certain items can be added: assessment of cranial sutures, opening of thumb, test for parachute reactions, analysis of the failure to adopt a seated position.

In the populations at risk, a full diagnosis is required. This involves a motility study focusing on passive, comparative, dorsal and ventral tone, active tone, anti-gravity suspension reactions, an assessment of joint amplitudes, especially in the feet, anterior and lateral drawn-seated manoeuvres as described by the French School of developmental neurology. These specific neurological examinations must be carried out by doctors trained in the neuromotor examination of young children in hospital consultations specialising in high-risk children or in EMSAC possessing a screening activity. They can also be conducted by a functional rehabilitation doctor.

A simple grid together with a few diagrams should enable paediatricians to highlight more development-related disorders. The generalisation of acoustic, provoked, oto-emissions (APOE) (plus the use of ringing toys) as well as an ophthalmological examination with "baby vision" should give a more precise interpretation of sensorial deficiency.

Chronologically, at the end of the first 6 months of corrected age, severe and moderate deficiencies can be highlighted but still without a specific approach. The second 6-month period should allow the disorders to be identified more precisely. Severe or moderate disorders will persist during the second year and may trigger the severity of certain handicaps and test for concomitant handicaps: orthopaedic, convulsions, involuntary, parasitic movements, sensory disorders, and the absence of language. Follow-up for the risk population of extremely premature infants must be extended beyond 3 years as specific, cognitive difficulties are apparent only in children around 4 years of age. This follow-up procedure calls for highly specific screening techniques that are practised by neuropsychologists.

Early intervention is based on the concept of cerebral plasticity

Early interventions can be defined as all of the intended prevention and treatment strategies in the first, even the second, year of life in order to promote cerebral development and optimal structuring. They are based primarily on two major axes: conventional, educational intervention strategies (neuromotor education, educational actions involving parents, etc.) and various types of neuroprotection strategies.

The early educational intervention strategies are based on the cerebral plasticity concept. These various plasticity processes are mobilised by re-educational approaches in a fully developing brain. Recent understanding of a few of the basic mechanisms of development, especially the importance of the phenomena of cell death and synaptic stabilisation and post-natal, neuronal circuits make up the theoretical and clinical basics in favour of early intervention. The capacities for motor compensation of the immature central nervous system secondary to lesions in the pre-central area are greater than those of the adult central nervous system in monkeys (Kennard principle). In children, destruction of the nervous tissue occurs in an environment that is still only slightly or not functionally specialised, unlike in the case of adults. Furthermore, the molecular and cellular response to the lesions varies with the extent of cell maturation. Under certain conditions when feasible, neural plasticity may be beneficial even if the ensuing development trajectory differs from normal. However, it can also be unsuitable if the structural connections do not permit an appropriate environmental response.

The concept of "neuroprotection", initially reserved for molecules protecting or preventing cell death phenomena, can now be extended to all actions promoting harmonious cerebral development and preventing the onset of specific disorders affecting the latter. A distinction is made between organisational, therapeutic, and environmental strategies.

On an organisational level, in utero transport to allow children at risk to be born in specialist, level III establishments, has shown its impact on the reduction in early cerebral lesions. On the other hand, the effect of labour methods (pelvic or Caesarean route, before or after work) and the beneficial role of the Caesarean section have not been fully consolidated.

From a therapeutic perspective, recent animal modelling tools for paediatric cerebral lesions have led to advances in the protection of the developing brain and even in the curative treatment of acquired cerebral lesions. Numerous molecules with protective effects have thus been used against the biological phenomena involved in cell death: pro-inflammatory cytokine antagonists, membrane stabilisers, inhibitors of free radical formation, inhibitors of glutamate release, calcium channel antagonists and the NMDA glutamate receptor as well as anti-apoptotic molecules, etc. However, the difficulty in envisaging neuroprotective treatments preventing or reducing lesions in premature or full-term infants is attributed to: the vast array of risk factors and mechanisms responsible for cerebral lesions; with no knowledge of the time (most usually antenatal) at which the deleterious cascade resulting in the cerebral lesion begins; the poor specificity of foetal stress markers (meconial fluid, anomalies in foetal heart rate) at the start of the deleterious cascade and abnormalities in adapting to extra-uterine life (Apgar score, metabolic acidosis); and finally, the difficulty in finding therapeutic agents that can be tested in clinical trials because they do not have any harmful side effects. Transfer to the clinical trial stage in human pathology is, therefore, problematic. Only some molecules or therapeutic means have been studied to date and found to offer a certain advantage: hypothermia in anoxia in full-term new-born infants, magnesium sulfate and fluoride steroids administered at the ante-natal stage and even nitric oxide at the post-natal stage in premature infants. Extensive multicentre trials using "cocktails" having several points of impact on the harmful cascade are both necessary and urgent.

Finally, the awareness of paediatricians of the role of the environment in a child's mental and cognitive development (acquired) has allowed preventive measures to be introduced in several areas: preservation of the mother-child relationship and the processes of attachment (bonding), stress-reduction programmes in a hospital environment, "care geared towards the individual development of every child" in the neonatal intensive care unit, consideration of pain and costly education programmes for small children and taught at home in the United States.

All of these aspects confirm the belief that early intervention, just like neuroprotection strategies, are an essential element and must be evaluated more effectively. Furthermore, early involvement is totally justified from an ethical standpoint.

The highlighting of a deficiency or handicap must be accompanied by the implementation of a suitable therapeutic project

Motor cerebral infirmity (MCI) or cerebral palsy is caused by definitive cerebral lesions that develop very early in life and are responsible for problems in the neuromotor field and language and praxis domains with deviations in neuropsychological and mental-emotional development that can also alter both actual and potential capacities in children. According to a French study, the distribution of different clinical profiles in cerebral palsy is as follows: 40% in quadriplegia, 17% in diplegia, and 21% in haemiplegia.

It is important to consider these disorders in an early, continuous, and multidisciplinary context. Consistent with the care and neuro-orthopaedic prevention carried out in neonatology, children must benefit from motor rehabilitation provided by a physiotherapist or psychomotor expert trained in techniques specific to young children. Depending on their wishes and potential for involvement, the parents receive practical recommendations to give their infant a proper introduction into daily life.

Consequently, child management involves a list of his/her capacities and focuses in particular on the motor, sensorial, and relational potential, which will be checked and highlighted throughout growth. The therapeutic project is based on a global vision of childhood development with reference to the growth schedule.

Non-pharmacological treatment involves physiotherapy (including electrostimulation), ergotherapy and orthophony, orthetic equipment, plastered postures, technical aids, and a combination of these different techniques. The scientific proof of the interest shown in physiotherapy, ergotherapy and orthophony is missing, despite the fact that these are always indicated and represent an essential complement to surgery.

Conventional physiotherapy and ergotherapy are sometimes supplemented by other techniques, which are recognised to varying degree as "patterning", conductive therapy, equitherapy, balneotherapy, the Votja technique, the Polish programme, and segmental exclusion by induced constraint. This last technique is undergoing evaluation but initial results have already proved encouraging. Methods such as hyperbaric oxygenotherapy are still at the highly experimental stage.

For 20 years or so, the treatment of spasticity has benefited from numerous methods that are difficult to apply to children: administration of medicinal products, muscular denervation agents (botulin A toxin), intrathecal injections (baclofen), or surgery (neurectomy, rhizothomy).

The child's progress and even difficulties are highlighted by regular clinical evaluation with reference to neuro-sensorial and psycho-behavioural development scales. The results of these evaluations are periodically presented to parents in a verbal and/or written format in an easily understandable language so as to provide the answers to the numerous questions that are asked, particularly with regard to the future.

The project, which involves medical follow-up, rehabilitation care, often devices, psychological support and rehabilitation sessions is co-ordinated by a referred physician, who possesses the necessary paediatric and/or neurology and/or rehabilitation skills. A strong family involvement is desirable, especially for the project to be set up in the first place, its practical execution and regular evaluation. Information concerning environmental adaptation and preparations for the future are provided over time with anticipation: games, home, district, school, work, leisure activities and transport. In the longer term, this support can continue with the educational project, either ordinary, managed, or even specialised. Autonomy is developed for as long as possible using local educational or social equipment.

An early visual function deficiency may interfere with a child's development and impact upon all of his/her skills, whether of a motor, cognitive or emotional nature, and thus have repercussions on school performances and both social and professional insertion. The existence of a critical visual development period occurring in the first few months of life points in favour of early intervention in visual deficiencies in young children. In most cases, these are situations involving a risk of amblyopia, which is treatable. Uncorrected ametropies are the primary cause of strabismus and functional amblyopia. They must be controlled as quickly as possible, preferably by intervening during the sensitive period, i.e., before 1 year of age. Any anisometropia >1 dioptre, any astigmatism >1.5 dioptre, any hyperopia ≥3.5 dioptres and any myopia ≥3.5 dioptres ¹ must be managed by optical correction involving the constant wearing of glasses and, in some cases (myopias, marked anisometropias, and unilateral aphakia), contact lenses. This corrective measure must be accompanied, where necessary, with functional amblyopia treatment.

Similarly, the early intervention in strabismus and other oculomotor disorders as well as potentially related functional amblyopia, conditions the quality of the result obtained. Oculomotor disorders must be treated before a child turns 3 years of age, studies having confirmed that the earlier the alignment of both eyes is obtained (around 2 years of age), the more satisfactory the long-term binocular co-operation.

The treatment of congenital glaucoma and congenital cataract is surgical and represents a therapeutic emergency to limit the risks of blindness and to restore the transparency of the media to allow normal vision to develop. At the same time, the retinoblastoma must be medically controlled without delay if the child's life is to be saved.

The rehabilitation of low vision in paediatric ophthalmology must allow the good social integration of the child. This rehabilitation involves the participation of professionals who have benefited from specific training (specialist rehabilitation, locomotion instructors, specialists in providing assistance with living from day to day, etc.). As for the management approach adopted, it must be organised as closely as possible to suit the child's location because many towns do not have the adequate structures. Low vision rehabilitation must be structured in the form of management networks or co-ordinated care networks under medical supervision.

The early support given to a very young child (under 3 years of age) presenting with severe, auditory deficiency is based on three complementary and intrinsic principles: assistance with auditory perception (early prosthetic insertion, cochlear implants, etc.); learning to "hear" using auditory aids (early auditory rehabilitation) and assistance with implementing communication allowing the child to grasp a better understanding of the world around him/her and the demands made by others (parents, siblings, family, professionals at the establishments frequented by the child), but also to express himself/herself; the involvement of parents and other family members to help them permanently adapt to the child by introducing a suitable relational balance and rediscovering spontaneous communication. This involvement will help to prepare very young children to adapt to nursery school by proposing extended personal and individual assistance.

As regards children presenting with cognitive function deficiency (all of the mental processes involved in processing information), the aetiologies are diverse albeit unknown in a large number of cases. Former terms such as "backward" or "frail" are now firmly in the past, primarily because they posed insurmountable obstacles for any form of education or rehabilitation. However, numerous terms are currently used and refer to different, theoretical frameworks, which can have important practical consequences in terms of institution and treatment. These include: intellectual deficiency, predominantly intellectual deficiency associated with neuropsychic disorders, mental retardation (slight, moderate, severe), mental handicap, progressive disharmony and cognitive functional deficiency, etc. In France, autism still poses classification problems because the frameworks used (concept of psychosis in the French classification system or invasive developmental disorders in the international classification system).

Raising and educating children suffering from invasive developmental disorders (including autism) requires special know-how, which is not part of the normal list of parental skills and which even goes beyond the techniques regularly taught in traditional professional training courses. The efficacy of the interventions used will depend on the availability and adequate training of competent professionals with a sound knowledge of the services that can be accessed.

Treatments for autism are largely based on the behavioural principal of operating conditioning (Lovaas method) as well as on psycho-educational and behavioural approaches centered on the acquisition of cognitive and developmental skills (TEACCH method). Both approaches are used within the perspective of educational or social normalisation. Studies applying both these approaches highlight substantial gains in the cognitive development (IQ) and speech of children suffering from autism or other invasive developmental disorders (IDD). The institution of treatment at an early age seems to be an essential condition to guarantee successful intervention. Regardless of the location where these programmes are carried out (either at home or in specialist centres), close co-operation between parents and professionals over a long period of time is fundamental for success. The gains obtained are usually maintained once treatment has ended. The children who make the most progress are those who have the better cognitive skills to start with.

Motor, neuro-sensorial, intellectual, and psychiatric deficiencies call for an early management approach that is suitable for each situation and which is based on validated methods where possible. The various techniques have still only been assessed to a minor extent in France.

Early stimulation programmes have demonstrated a certain efficacy for children who are at risk of developing a deficiency

Knowledge of the various mechanisms involved in cerebral development confirms the crucial importance of the first years of life for acquiring cognitive skills. These concepts have, however, been modulated by highlighting the process of delayed plasticity and the cognitive compensation of early deficiencies by a highly stimulating life style.

For several years now, early stimulation programmes have been devised for children who are at risk of developing a deficiency: vulnerable treatment due to the fact that they belong to an unfavourable and hardly stimulating socio-economic category; premature infants or infants with a low birth weight.

On the basis of studies showing that the socio-economic category was a crucial factor in child development for the first 5 years, programmes (Head start programs) were developed in the United States during the 1960s for children belonging to an unfavourable and hardly stimulating socio-economic category. Positive effects were recorded in school failure rates and, at a social level, by a reduction in the number of crimes committed. Since 1990, support has been available to parents, but the additional positive effects are still unknown.

Programmes have also been launched in many countries for premature infants and infants with a low birth weight, who represent a population exposed to a high risk of developing these deficiencies. These programmes are targeted either to the child himself/herself or even to the family or combined to provide assistance for both the child and his/her family. They are initiated on leaving hospital in the transition period between being discharged and arriving home or else affect post-hospital follow-up.

The first programmes targeted towards the period of neonatal hospitalisation were based on the idea that premature infants lack stimulation and babies benefit from various forms of stimulation (multi-sensorial, tactile and physiotherapeutic) provided by nurses or ergotherapists. The results evaluated within the scope of these controlled studies appear to be positive. Other programmes aimed, on the contrary, to reducing stress and boosting children's capacities for autoregulation. These were carried out by nursing staff and gradually by parents. The results were also positive with regard to the short-term development scores. The programmes given the best evaluation are those involving both parents and children and are known as NIDCAP programmes (Newborn individualised developmental care and assessment programmes). These are individualised programmes intended for babies with a very low birth weight (less than 1,500 g) in neonatal intensive care units. The short-term results are positive (weight gain, duration of hospital stay) but are more difficult to evaluate in terms of long-term development.

Other programmes carried out during the hospital-home transition period are aimed at training parents to respond appropriately to the signs given by their child. Training is given over 7 days before discharge and then for 3 months after discharge. The beneficial effects are evident at 3 years of age and continue for up to 9 years (one single study). As for post-hospitalisation programmes, they confirm the positive effects on cognitive development and on parent-child interactions in the first two years of life, especially in cumulative risk situations (prematurity and poverty).

In the early 1990s, a longitudinal, multi-site programme on early stimulation over 3 years, known as the Infant health and development programme (IHDP), was launched in the United States for infants with a gestational age of less than or equal to 37 weeks and a birth weight of less than or equal to 2,500 g. The results confirm positive cognitive and behavioural effects at 24 and 36 months, which are more marked for children weighing between 2,000 and 2,500 g and for slightly more educated mothers.

Generally speaking, it is a well-known fact that combined parent-child programmes are the most useful. The effects on parent-child relationships and on the child's cognitive development are positive if stimulation is maintained. Much better results are seen at a cognitive level as opposed to a motor level. These programmes are especially effective when mothers have a low level of education. They therefore compensate in part for what should have come from the family. The results show that the children benefiting from these programmes experience display less of a cognitive decline than children in control groups. This constitutes a genuine preventive action.

It remains to ascertain how risk groups should be selected considering studies where programme efficacy has been proven. A combination of prematurity and/or a low birth weight with an unfavourable socio-economic background appears to correspond to a combination of risks that clearly respond to the programmes proposed.

An economical screening and management approach is still barely used in decision-making

Examination of the economic literature relating to the screening and the management of handicaps of perinatal origin is highly deceiving in more than one respect.

The published studies date back some years and essentially concern the evaluation of prematurity costs. Such paucity in the recourse to economic evaluation tools and, in particular, the total lack of any recent French study can be explained by the fact that economic analytical methodologies, if they existed, are sometimes difficult to instigate, especially due to the rate of pitfalls inherent in the cost inventory.

Cost studies provide information on the nature of financial flows used when carrying out a prevention or patient management programme. This expression of cost is generally advanced to plead in favour of the allocation of financial resources to a category of actions or health system agents. The numbering of cost can also be of a macro-economic nature and aimed at answering the following question, for example: "What is the cost to the society for prematurity and related handicaps?" A recent study carried out on this topic in California is based on a cohort of children with low birth weight admitted to the neonatal resuscitation centre. The results clearly show the interest in preventing early prematurity, not only to prevent the onset of handicaps but also to avoid the very high expenditure concerning both living and deceased children.

The cost-efficacy analyses are aimed at comparing the "profitability" of various preventive or sanitary actions from the point of view of a collective interest. They allow various strategies to be classified per order of the decreasing cost-efficacy ratio, provided that the real value of the consumed resources is measured. In the cost-efficacy analyses, consequences are determined by a "physical efficacy indicator", which can be, for instance, a biological indicator or autonomy, the number of abnormalities screened, lives saved, or deaths avoided. One of the difficulties in analysing cost-efficacy is the choice of a relevant efficacy indicator to measure the performance of evaluated actions. Medical innovations increasingly interfere not only with the "quantity of life" that they can obtain but also the quality of life for the subject who will benefit from this. Certain actions can extend the life-span but at the cost of a certain morbidity or deterioration in the quality of life of these subjects during the health procedure.

The economic evaluation can thus take the form of a cost-utility approach developed essentially in Anglo-Saxon contexts. This approach weighs the physical efficacy indicator by assessing the quality of life (QALY: Quality adjusted life year). This assessment is made on a subjective basis by interviewing medical experts or target individuals of the action or general population. This is a compared evaluation of two types of treatment of retinopathy in the premature infant (cryotherapy and laser), which links to each treatment the number of years of life in good health that are saved.

The cost-benefit analysis is the fullest form of the economic evaluation but also the most delicate because it attempts to link an economic or monetary value to the sanitary benefits of health care actions. This approach poses methodological problems because it is necessary to monetarise the consequences of changes in health conditions such as production losses. A British study relates the additional expenditure of screening and retinopathy treatment of the premature retinopathy to costs saved by reducing the number of visual handicapped subjects whose condition must be controlled.

Nevertheless, extreme caution must be exercised when interpreting the results of all the economical studies. A comparison of the results obtained in different countries is always risky because of the differences in social protection and sickness insurance systems.

Certain countries have more experience than others in organising management approaches

In Europe there is considerable heterogeneity concerning the organisation of neonatal care for extremely premature infants. In the Netherlands, Portugal, and the Czech Republic, newborn infants of less than 32 weeks are managed in level III units. On the other hand, in Denmark, Northern Ireland, Germany, the United Kingdom, and France, these children are also admitted to hospital in level II units. Despite a consensus of opinion regarding the need to transfer pregnant women at risk of premature labour to maternity wings linked to resuscitation and neonatal care services, almost have of the very premature French infants were outborn in 1996. There is still no comparative evaluation of these different organisational decisions.

NIDCAP programmes were initially developed in the United States and then in Sweden to respond to the needs of children born very prematurely in terms of stimulation and care during a critical period of development of their central nervous system. The pain and discomfort caused by care practices during hospital admission were taken into account. These programmes require changes in the organisation of care in neonatology departments and considerable rehabilitation efforts. Rehabilitation specialists going on to become development specialists are portrayed as parents' allies, bringing them confidence and skills to take care of their child. These specialists are starting to take part in research into neonatology care as members of multidisciplinary teams.

Thirty years or so ago, the United States introduced a policy to assist children with physical, mental, or behavioural handicaps. Since 1990, the programme launched by the Individuals with Disabilities Education Act (IDEA) developed a decentralised, family-orientated system of multidisciplinary care. At the heart of the system, the "medical homes" represent the ideal approach to regionalised management and partly correspond to the vocation of Early French Medico-Social Action Centres (EMSACs). They ensure continuity of neonatal care to adolescence, identification of requirements in medical, specialised consultations and their co-ordination, interaction with school, social life and the centralisation of information on medical care and the hospital admittance of each child. The type and extent of management approaches are based on the Individualised Family Service Plan, an individualised, family-orientated programme, which defines the most suitable form of education for each child in line with his/her development. The paediatrician at the medical device centre discusses the child's progress, treatment, and clinical course with the family and helps to locate additional services for secondary prevention purposes.

A recent study that evaluated access to the early management programme (IDEA) at the national level indicates that the average age at which parents report an abnormality is 7.4 months. The diagnosis is carried out on average 1.4 months later, and admission to the programme occurs on average 5.2 months after diagnosis. The individualised management plan is initiated 1.7 months later at an average age of 15.7 months. Most families are satisfied with their involvement in an early intervention programme. They report having had an opportunity to discuss matters with a health care professional and have benefited from straightforward access to services corresponding to their requirements. They positively assess the work conducted by the professionals and felt that they played a role in the decisions taken regarding their child.

In Europe, the disappearance of specialist institutions in various countries (Sweden and Italy, etc.) led to accentuating the decentralisation of skills in terms of handicap to promote a local management approach. In Sweden, committees have adaptation centres for handicapped children combining screening and handicap treatment resources (physiotherapy, ergotherapy and orthophony, etc.). These centres are also responsible for the pre-school insertion of handicapped children. Where possible, handicapped children with normal intellectual capacities go to a mainstream school and are assisted, if required, by personal assistance and suitable equipment. Children suffering from an intellectual handicap and possibly other functional limitations go to a "separate" school. In Italy, the law of 1977 abolished the differentiated (streamed) classes and special classes by defining flexible forms of integration in favour of teaching handicapped children at a mainstream school with specialist teaching support. The right to integration is recognised throughout the mandatory school period but also in crèches and secondary schools.

The United Kingdom comes mid-way between countries that have opted to close specialist establishments and France or Belgium where there is still an important specialist education sector. The policy is directed at having a maximum number of children in ordinary establishments with the assistance, where feasible, of external specialists, but this does not imply the closure of specialist establishments, which are viewed as a complement and not as an alternative to "mainstreaming". In the United Kingdom, a large number of autistic children are taught in an ordinary environment with specialist support.

Comparison of the different ways of organising management approaches for handicapped children carried out by the European Agency for developing the education of persons with particular requirements in various EC states is interesting from several angles. It shows that some countries have more experience than others in organising management approaches. Common points in terms of actions and programmes nevertheless warrant emphasis. Generally speaking, early intervention is recognised as essential in the same way as a complete evaluation of the child's potential is needed to determine the assistance required. The teams cover four key areas of early intervention: medical, psychological, educational, and social. These teams consist of additional professionals, depending on requirements.

For several years now, we have witnessed a tendency to promote "developmental care" in deficient or handicapped children in an attempt to integrate them in an ordinary environment. The services in charge of "developmental care" are most often under the responsibility of the public health or social affairs "authorities", but educational skills (schooling) are more actively needed in the programmes than in the past. Amongst the points emphasised as necessitating improvement, there are practical difficulties in organising the co-ordination of services and professionals, given the lack of organisational measures and strategies.

The involvement of families in organising developmental care is essential

The role played by parents in the care system introduced for a very young child can be analysed at three levels:

• their involvement in the management of a child suffering from a proven or potential handicap;
• the type of support experienced by the parents, from pregnancy and subsequently, when a risk of disease is highlighted;
• the highlighting and management of their vulnerability factors in preventing certain obstetric-paediatric complications (e.g., prematurity or growth retardation) as well as the effects of perinatal measurement approaches on the forging of parent-child links.

We can rely on a simple diagram combining all of the interactive systems surrounding the newborn or unborn infant to integrate these three levels of reflection. These systems take into account the biological, emotional and inter-relational elements that interfere in a child's somato-psychic development.

This diagram enables a link to be made between pregnancy, the post-partum period, early infancy and also between the medical, social, and psychiatric fields. The follow-up given to an extremely premature child will depend on obstetrico-paediatric co-ordination, which will have consolidated or broached parents' confidence with regard to the care system.

The interest in involving parents in the event of a handicap is aimed primarily at preventing an "additional handicap", which would come from a distorted link and poor resolution of the trauma resulting from announcing the handicap. Prospective clinical trials describing the bonding process between a mother and her premature child have shown that, in the case of a similar disease in neonatal resuscitation, the neuropsychic fate of children was strongly correlated with the quality of the emotional environment.

Figure

Interactive systems in a child's somato-psychic structure .

At the same time, copious amounts of published data attempt to describe the expectations of families when faced with the handicap of their child. The conditions and content of parent information become a research topic. Parents want recognition for their role and skills while indicating the type of support required (counselling): coming out of isolation, being specifically informed of the handicap and its consequences, being helped by talking to people in similar situations, being guided in finding out and accessing the various services available in administrative burdens.

The effects of parent integration in management programmes have been assessed by reducing stress, better adjustment to their child's requirements, and conjugal balance. The effect of involving the father in subsequent cognitive development has been observed. These results seem to be evidence-based, but this evidence was denied during the technological and institutional development period to the extent that a new type of professional has emerged: the "insertion technician" became the parents' mouthpiece and support when confronted with a series of services difficult to identify and rejoin.

Parental involvement is very important in certain programmes concerning autism, a condition that, in the past, was overlooked and which generated a feeling of guilt against which families strongly reacted, demanding information and respect for their role. Forms of intensive intervention integrated them as "co-educators" with positive results. Identical effects can therefore be seen between direct stimulation programmes with the child and programmes to support parent-child interaction.

Action programmes (such as Head Start) were launched in the United States in an attempt to reduce the dependency with regard to social assistance for families existing in the precarious situation chiefly present in intensive home visits. Their efficacy is improved when the visits begin before the birth and continue for a sufficiently long period. Antenatal visits have resulted in a reduced incidence of prematurity, growth retardation, and hypertension with variants depending on the population categories, particularly ethnic origin. Better integration of the mothers has been achieved. The criteria for success in children concern the prevention of cognitive deficiencies and behavioural disorders with limited efficacy.

Figure

Different environmental registers. EMSAC: early medico-social action centres; SEHCS: special education and home care services; MIP: maternal and infantile protection

A sociological analysis has attempted to exempt the efficacy criteria from programmes that have "worked". These are, in fact, conditions governing participants, which appear as the most relevant indicators: autonomy, flexibility, proximity to families, a holistic approach, a feeling of competence—all of those factors offering a type of model (restoration of confidence in oneself and in others) for mothers experiencing social and emotional insecurity. The authors emphasise the interest of minor teams exempt from heavy public or private administrative burdens that divide up the action deviating from the real expectations of families, highlighting a lack of accessibility and delayed or confused responses.

Parental associations play an important role as intermediaries capable of helping to highlight the numerous specialist services, to improve the adjustment of responses and, at the same time, sharing the emotion of trauma inherent in the presence of a handicap.

In this context, clinical trials are beginning accurately to describe the interactions between professionals and parents but also within participation networks—mutual respect and complementarity seem to be decisive factors. The notion of "alliance" between families and professionals is added to the provision of routine approaches. Thus the interesting experiences of treating and managing "in networks" drug-addicted pregnant women are manifested by a significant reduction in the incidence of prematurity, parental involvement in weaning a baby, the disappearance of authoritarian emergency placements, and a decrease in frequent or repeated pregnancies. These experiences define a new clinic of "networking", which has been made possible by the gradual decompartmentalisation of public and private, obstetric and paediatric, medical, social, and psychiatric services in France in particular.

The harmonisation of interventions in the fields concerned and the early consideration of emotional vulnerability factors are important. Reflection must focus on the initial training of carers working closest to families (obstetricians, general practitioners, and paediatricians) and on teaching methods that would facilitate a better mutual understanding to offer children and families at risk a flexible, consistent "human cover" that would continue for as long as required, regardless of the origin of the established or suspected deficiency.

Network organisation is essential for the success of a perinatal health policy in France

The organisation of network care is currently presented as a model because this type of organisation is conducive to co-ordinating an action in which the potential of all those involved would be utilised: users, health care professionals, institutions and people in charge of decision-making. Various types of network already exist, some of which are formalised while others are not, in the field of perinatology. These networks are based on subjects, which vary in terms of specificity, interlinking various key players in the perinatology field.

The town-hospital network for following up pregnancies in the general population links general practitioners interested in pregnancy follow-up with freelance midwives, Maternal and Infant Protection protocol (MIP), and both public and private hospital structures. These networks are directed toward more specific populations such as those that find themselves in a precarious situation. Combined town-hospital networks and inter-hospital co-ordination arrange antenatal diagnosis at the regional level. The inter-hospital co-operation networks allow women to be managed in an establishment that is suitable for their medical situation. Town-hospital networks allow a return to home visits after childbirth to be organised.

Co-ordination within a "perinatal health network" of all these networks is desirable to maintain consistency and to allow actions of promoting the awareness of and providing information for the population to be incorporated. The aim of this type of network is to create a common culture among all of the participants, to make common tools available to them, and to motivate them to meet the mutual shared objectives in order to improve the status of perinatal health.

Better care co-ordination will improve the health of the population. This fact has been studied chiefly in perinatology from the angle of care regionalisation, which is one of the main objectives of inter-hospital co-operation networks. Regionalisation improves perinatal health indicators in terms of perinatal mortality and neonatal morbidity. A fall in the perinatal mortality rate is especially evident from the foetal aspect. This gain is not achieved at the cost of a deterioration in neonatal morbidity rates but, on the contrary, is accompanied by a reduction in cerebral lesions or an improvement in the development score. It is, therefore, logical to reach favourable progression for health indicators with better co-ordinated perinatal care. Care organisation upstream from birth and surrounding birth thus helps to reduce perinatal mortality. The effect of improved co-ordination of perinatal care on changes in the total number of surviving children at risk of developing a handicap is more difficult to determine. Although improved care helps to reduce the number of children at risk of developing a handicap among previously surviving children, it mathematically increases the total number of children at risk due to the increase in the overall number of surviving newborn infants belonging to risk populations.

A perinatal care network must involve the organisation of follow-up appointments until children at risk of developing a handicap reach school age. The parents of extremely premature infants complain of being abandoned after their discharge from the neonatology department. They are confronted by a series of professionals, who are informed to varying degree and who are competent in following up their children. The absence of a consensus opinion and references contributes to this fact. Similarly, the lack of structural co-ordination relating to specialist care in the event of single and multi deficiencies adds to misinterpretation of the health policy in the field of handicapped infancy in the opinion of the population and health care professionals. Conversely, the success of introducing perinatal care in certain regions has promoted the launch of follow-up networks known as downstream networks of perinatal care networks. The aim of such downstream networks is two-fold. The principal objective is to organise post-natal follow-up in a consistent matter to boost access to early management of motor, sensorial, or developmental incapacities or limitations. This follow-up must allow the consequences of deficiencies to be reduced, secondary handicaps to be limited and medical error often accompanied by a feeling of parental abandonment to be avoided, while promoting involvement and effective, relevant assistance. The secondary aim is to carry out a long-term evaluation of the regional perinatal policy and thus of the perinatal care network. This evaluation is essential. Care practices change very quickly without having been assessed in the long term.

Figure

Implementation of downstream network guiding high-risk patients towards suitable structures for early management .

The organisation of a downstream network is based essentially on the principle of improved co-ordination of professionals and care structures. The setting up of a co-ordination cell, the creation of a common follow-up tool in the form of a common dossier, paper and computerised on a regional server, the organisation of training to use this tool and recourse to specialist consultations are the principle means required to create such a network. The evaluation must be two-fold: quantitative and qualitative. The quantitative evaluation is relatively easy: percentage of children included and followed up until 5 years of age. The qualitative evaluation is trickier to carry out: parent satisfaction, number of children highlighted by the follow-up network, and improvement of the quality of life for children receiving early management. The evaluation of networks implemented should lead to the conclusion as to whether there is a need to generalise the experiments under way.

Clarification and the co-ordination of institutional responsibilities that would facilitate improved management

Although the decentralisation laws have allocated in their principles clear-cut fields of competence for each major hospital/institution, the State and general councils retain decisive skills when it comes to managing and developing the procedure for the management of handicapped people. Although this overlapping of skills is well known for the management of handicapped adults, it is less so for children under 6 years of age.

Early infancy is the period during which handicaps of perinatal origin are going to be recognised. Furthermore, some of these will not be detected before 6 years of age. Every infant is reliant on the people around him/her and it is only over time that a gap in their learning ability in relation to that of the majority of children of the same age will appear. This delay in the time taken for clinical signs to become apparent will lead to the diagnosis and justify the 20 mandatory examinations carried out in children within the MIP.

During this period, the parents of children in whom a deficiency is expected have a choice in the approach and practitioners that they are going to use and consult, as with any other disease. EMSACs, which have a screening and family guidance mission, have entrusted the rehabilitation and care section covered by sickness insurance to general councils such as the MIP.

The State, which guarantees national solidarity, has retained the responsibility for recognising the handicap and allocating financial aid. It provides the supervision for all medico-social establishments and guides children through the various structures and establishments, apart from the EMSACs.

As regards the local management of the dossiers of handicapped children under 20 years of age, the State has entrusted the exercising of its skills to departmental commissions, the Special Education Departmental Commissions (SEDCs). These commissions recognise the handicap and assess the level of incapacity entitling suffers to the Allocation of Special Education (ASE) and its complements, allocate invalidity cards, direct children over 3 years of age to nurseries or medico-social structures, and authorise, from birth if necessary, the intervention of care at home (special education services and care at home, SESCH). Whereas MIP regulations and actions are directed toward all children, without exception, the SEDCs must be contacted to be able to intervene, and they are generally approved by the parents.

All the methods for welcoming young infants, such as crèches, day nurseries or nursery assistants are placed under the responsibility of general councils, even when places specifically reserved for handicapped children actually exist. Admissions are carried out according to the methods specific to each administrative body without any need for approval by the SEDCs. The decree dated August 1, 2000, which gives parents the option to keep a child in a crèche up until 6 years of age, and the Family Allocations Fund (FAF) circular, which excludes contracts for investing in establishments that would not be able to receive handicapped children, must facilitate the admission of the latter and keep them in infant care establishments until they begin mainstream school or enter a socio-educational establishment coming under State jurisdiction.

Even if these transfers of competences follow a relatively clear logic, perception of the boundary marking out the respective fields of the two institutions often proves to be unclear to the extent that Social Security (sickness insurance and family allocations) is one of the main financers regardless of the institution concerned. Numerous reports, including that of the Accounts Court, have stressed the confusion that arises in the role of institutional participants involved in handicap policies since the decentralisation laws came into force. Some authors conclude that even if these fragmented responsibilities, which generate joint financing, can lead certain players to deny their specific skills or not to recognise those of another.

It is a well-recognised saying that, "any handicap before 20 years of age is up to the State", forgetting that the prevention, screening of handicaps before 6 years of age, their subsequent management by EMSACs, and the methods of hosting young infants are departmental skills. Given this fact, all the readily accessible statistical data concerning the management of childhood handicaps are limited to SEDCs and to Specialist Education (SE) enquiries, which ignore the EMSACs located outside the State's area of competence. To use these statistical data for children of pre-school age (under 6 years of age) comes back to ignoring those for whom the handicap has not been the subject of administrative recognition.

The lack of statistical French data concerning handicaps in very young infants has led the legislator to introduce in 1973, during the first perinatality plan, infant/child health certificates to organise the collection and processing of data and to provide finance.

These certificates must be completed by the doctor who examined the child during the first 8 days of life and at both 9 months and 24 months. The same doctor is legally obliged to send this certificate within a period of 8 days and in a confidential envelope to the doctor in charge of the MIP service in the child's home department. These mandatory investigations are carried out in full and, to examine the child and complete and send the certificate, general practitioners, like paediatricians, are given a paediatric fixed price, which amounts to 5 € per assessment. This fixed fee is also integrally reimbursed by sickness insurance.

These certificates are processed in each department, and the database thus composed is placed under the responsibility of the MIP doctor. The processing of health certificates, data analysis, and the annual transmission of results to the Ministry for Health fall within the realm of the general councils. The regrouping of data transmitted by the departments to the Ministry for Health must result in the creation of a national database.

More than 30 years after the introduction of health certificates, although the procedures are ironed out and the finances exist, and the CNIL is never opposed to the computer processing of data for statistical purposes, the weaknesses of the system cannot be overlooked. There are still no reliable data regarding the perinatal period and infant/child handicaps. The quality procedure for completing these certificates is substandard and the quantity received by doctors responsible for MIP services is low (approximately 70% for those conducted at 9 months and 24 months), which does not mean that these assessments have not been carried out or indeed that they have been carried out incorrectly. One of the causes of this situation is that certain doctors refuse to give basic medical data to medico-social services that use such information for the purposes of individual prevention by selecting so-called "at risk" children based on medical or social criteria that do not always match.

Furthermore, there is no listing such as a SE enquiry that would identify the number of handicapped children admitted to establishments that fall under the jurisdiction of the general councils (EMSAC methods of receiving children).

The decentralisation laws have blurred the vision regarding regulations underlying the management policy for taking very young, handicapped children into care. This sharing of skills has led to the increase in a standard phenomenon in France, which, regardless of the underlying disease, involves emphasising the "care" section to the detriment of prevention and screening. This vision of things is hampered by the fact that the only national statistics that can be readily accessed concern establishments and services that fall under the jurisdiction of the State. Furthermore, the lack of reliable epidemiological data both at a national and a departmental level makes it difficult to estimate requirements. The same applies when it comes to assessing the offer available for the management of handicaps of perinatal origin in children of pre-school age.

Boosting awareness and training

Preventing risk situations

Early screening

Management

Promoting research