contribution of to antenatal care in france: impact level ...during the following trimester or...

Journal of Epidemiology and Community Health, 1987, 41, 321-328

Contribution of specialists to antenatal care in France:impact on level of care during pregnancy and deliveryBRUNO HUBERT, BEATRICE BLONDEL, AND MONIQUE KAMINSKIFrom the Unitk de Recherches Epidkmiologiques sur la Mere et l'Enfant, INSERM, 16 Avenue PaulVaillant-Couturier, 94807 Villejuif Ctdex, France

SUMMARY This study was based on a survey of a national sample of births in France in 1981 whichincluded 5508 women. Four pathways of antenatal care were defined according to the stage ofpregnancy at first intervention of a specialist, as opposed to a general practitioner, in the care of thepregnancy. Taking into account the sociodemographic and medical characteristics of the women in alogistic regression, a large number ofantenatal visits, an ultrasound examination, and hospitalisationduring pregnancy were more frequent when the degree ofspecialisation ofthe pathway increased. Butthe influence ofpathways was less significant for deliveries. Caesarean section rates, for example, didnot vary according to pathway. However, induction of labour and intrapartum electronic fetalmonitoring were less frequent among women cared for solely by a general practitioner than amongthose who had consulted a specialist at least once during pregnancy. The increase in medical care andthe role of the specialist in antenatal care are discussed.

In several European countries, such as Belgium, WestGermany, and France, pregnant women may choosethe place where they will get their antenatal care andthe person who will be responsible for it.1 In France in1981, 38% of pregnant women were cared forexclusively by a specialist, 9% exclusively by a generalpractitioner, while, for 53%, antenatal care was sharedbetween both of them.2 The consequences of thesepatterns ofcare have been studied in several countries.In Belgium, it has been shown that the frequency ofvisits differed between the public and private sectors;3in France, the degree to which care was ensured byspecialists constituted one of the main sources ofinequality.4 Since these studies were limited to theantenatal period and, specifically, to the number ofvisits, the impact of this specialisation on other aspectsof care during pregnancy or on the conditions ofdelivery remains unknown.Other factors have an influence on antenatal care:

women under 20 years of age, women with a highparity, a low educational level or a residence in a ruralarea make fewer visits and are less frequently seen by aspecialist.5- But few studies have simultaneouslytaken into account all these maternal characteristics inorder to evaluate the role of the type of care providerson the level of care. The objective of this study hasbeen to specify the role of the degree of specialisationin antenatal care, while taking into account maternalcharacteristics and complications during pregnancy.

We have tried to determine whether the qualificationof the person responsible for care was associated withthe level of antenatal care and whether thesedifferences in antenatal care also had an influence onconditions of delivery. We used the "pathway" toexpress the process by which pregnant women gothrough the various systems of antenatal care existingin France.

Population and methods

The data were derived from a national survey carriedout in France in 198 1.2 This study included arepresentative sample of 5508 births. Collection of thedata was made: (1) by interview-of the mothers, afterdelivery, about their sociodemographic situation,antenatal care, pregnancy complications, andoutcome of previous pregnancies; and (2) byabstraction of information on the delivery andneonatal period from hospital records.The definition of the pathways was based on the

trimester of pregnancy during which a specialist firstundertook the antenatal care. The concept ofspecialistincluded the following qualifications: gynaecologist,obstetrician, midwife, and doctor in training tobecome a gynaeco-obstetrician. Use of this definitionmeant the exclusion of 588 women (11% of thesample) based on the following reasons: lack ofinformation on the person responsible for the

321

Protected by copyright.

on January 27, 2020 by guest.http://jech.bm

j.com/

J Epidem

iol Com

munity H

ealth: first published as 10.1136/jech.41.4.321 on 1 Decem

ber 1987. Dow

nloaded from

http://jech.bmj.com/

322 Eantenatal care for at least one trimester (6%); absenceof visit for at least one trimester (3%); isolatedintervention of a specialist, that is, with no continuityduring the following trimester or trimesters (2%).After these exclusions, the analysis was carriedthrough on 4920 women.Four pathways were defined:pathway I. care provided exclusively by a generalpractitioner (8% of the women);pathway 2: care by a specialist beginning in thethird trimester (I17% of the women);pathway 3. care by a specialist beginning in thesecond trimester (16% of the women);pathway 4. care by a specialist beginning in thefirst trimester (59% of the women).

Three variables were used to characterise careduring pregnancy: more than seven visits, at least oneultrasound examination, and hospitalisation.Indicators of care during labour and delivery werecaesarean section, induction of labour with oxytocinor prostaglandins, and intrapartum electronic fetalmonitoring.

Maternal sociodemographic and medicalcharacteristics were taken into consideration: (1)parity; (2) age; (3) sociocultural level, measured bydiploma (no diploma, intermediate level diplomacorresponding to 7 to 10 years of schooling, and upperlevel diploma based on at least 12 years of schooling);(4) rural or urban residence, rural environment beingdefined as towns having less than 4000 inhabitants andlocated away from an urban centre; (5) history ofadverse outcome in the previous pregnancy: perinataldeath, preterm delivery (before 37 completed weeks ofgestation), birthweight less than 2500 g, malformation,spontaneous abortion, or ectopic pregnancy.

In the analysis of antenatal care, the occurrence ofat least one of the following complications was takeninto account: hypertension (systolic blood pressure> 13 or diastolic >8), proteinuria (>01 g/l),bleeding, threatened abortion or preterm delivery. Inaddition, the trimester of occurrence of thesecomplications was considered in the study of thenumber of antenatal visits.

For the study of induction, we took into accountpost term deliveries (pregnancy duration of more than42 weeks).

For intrapartum electronic fetal monitoring, wenoted the presence of at least one of the principalindications for this examination proposed byNiswander et al:8 severe hypertension (which wejudged in our data to be present when there was aprescription of an antihypertensive duringpregnancy), diabetes mellitus, suspected intrauterinegrowth retardation, gestational age less than 37 ormore than 42 weeks, meconium-stained amniotic fluid

Bruno Hubert, Beatrice Blondel, and Monique Kaminskior induction by oxytocin or prostaglandins.

In order to study the relation between antenatal carepathways and delivery conditions, the maternity unitswere classified in three groups according to theirannual number of deliveries: less than 500, 500 to 1 99and 1200 deliveries or more yearly. This criterion is anindicator of the level of equipment available and therate of obstetric interventions practised in the units.9We first compared the characteristics of the women

in each antenatal care pathway. Then to study the careduring pregnancy and delivery we used the followingprocedure: we first examined the relation between careindicators and both pathways and characteristics ofthe women; secondly, we studied the specific effect ofthe pathways on care in a multivariate analysis, takinginto account all the maternal characteristics for whicha significant relation to care has been observed in theprevious stage of analysis. Age and parity do notappear in the tables, but they were included in themodels whenever they were significantly related tocare in the univariate analysis.

Statistical methods used were Pearson's x2 test andmultiple logistic regression; statistical significance ofthe adjusted odds ratios was tested by the maximumlikelihood ratio.'0

Results

Characteristics of the women differed according topathway (table 1); in the pathways where visits fromspecialists began early, women were more frequentlyof low parity, older age, high educational level, urbanplace of residence, with an adverse outcome of theprevious pregnancy, or early occurrence of acomplication during the index pregnancy.

ANTENATAL CAREThe proportion ofwomen having had more than sevenantenatal visits increased with the earliness of aspecialist's involvement in care (table 2). Thefrequency of visits increased with educational level,residence in an urban environment, presence ofadverse outcome of the previous pregnancy, and earlyoccurrence of complication. After adjustment formaternal characteristics in logistic regression, thepathways continued to have an important effect oncare, while these characteristics all maintained theirspecific effect on the frequency of visits.

Ultrasound examination during pregnancy wasstrongly linked to the degree of specialisation of thepathway: the frequency was only 51% in pathway Ibut increased regularly in the intermediate pathwaysto reach 90% in pathway 4 (table 3). But maternalcharacteristics also played an important role, identicaloverall to that found for the number of visits. Afteradjustment, having an ultrasound examination



j.com/

J Epidem

iol Com

munity H


ber 1987. Dow

nloaded from


Contribution of specialists to antenatal care in France: &npact on level of care during pregnancy and deliveryTable I Maternal characteristics according to antenatal care pathways

323

Pathway 1 Pathway 2 Pathway 3 Pathway 4

(No) (No) (No) (No)

Parity0 37 41 43 411-2 48 47 51 523 or more 15 12 6 7

(411) (828) (790) (2877)

Age (yr)less than 20 9 9 7 320to 34 86 86 89 9034+ 5 5 4 7

(410) (830) (792) (2859)

DiplomaNone 33 32 23 17Intermediate 57 56 58 61Upper 10 12 19 32

(384) (786) (747) (2712)

Rural environment 48 46 35 24(405) (821) (785) (2831)

Adverse outcome ofprevious pregnancy 10 10 12 17

(379) (787) (753) (2732)

Trimester of pregnancy complicationNone 73 69 67 643rd 14 17 15 152nd 5 5 10 81st 9 9 8 12

(390) (808) (774) (2803)

* p<000l

Table 2 Percentage ofpregnant women with more than seven antenatal visits according to antenatal carepathways andmaternalcharacteristics

Logistic regression

No. % S(l) S(2) OR Cl 95%

Antenatal care pathway1 414 22 12 832 35 2 0 1-5-2-73 791 31 1-4 1-1-2-04 2879 52 3-1 2-4-4-1

DiplomaNone 1007 35 1Intermediate 2462 43 1-2 10-14Upper 1156 53 14 1-2-1-8

EnvironmentRural 1527 37 1Urban 3311 46 * 1-2 1-0-1-4

Adverse outcome of previous pregnancyNo 3983 42 1Yes 664 53 1-6 1-3-1-9

Trimester of pregnancy complicationNone 3164 39 13rd 741 49 1-5 1-3-1-82nd 362 58 2-2 1-7-2-8Ist 504 58 2-0 1-6-2-5

(1) Not adjusted(2) Adjusted for all significant characteristics including age and parity

* p<005p<O400l



j.com/

J Epidem

iol Com

munity H


ber 1987. Dow

nloaded from


324 E Bruno Hubert, Beatrice Blondel, and Monique KaminskiTable 3 Frequency of ultrasound examination according to antenatal care pathway and maternal characteristics

Logistic regression

No. % S(J) S(2) OR CI 95%

Antenatal care pathway1 392 51 12 813 72 24 1 8-323 774 83 36 27-494 2829 90 66 5-1-87

DiplomaNone 986 72 1Intermediate 2416 84 1 7 1.2-2 4Upper 1138 91 2 4 14-40

EnvironmentRural 1490 77 1Urban 3245 85 NS 12 10-14

Adverse outcome of previous pregnancyNo 3912 82 1Yes 650 89 1 7 13-22

Pregnancy complicationNo 3095 80 1Yes 1591 88 17 14-21

(1) Not adjusted(2) Adjusted for all significant characteristics including age and parity*** p<O00I

Table 4 Frequency of hospitalisation during pregnancy according to antenatal care pathway and maternal characteristics

Logistic regression

No. % S(l) 5(2) OR CI 95%

Antenatal care pathway1 410 7 12 826 1 1 1 8 1 1-3 03 788 15 25 15-414 2866 18 28 1 8-45

DiplomaNone 1000 18 1Intermediate 2461 14 * 07 06-0 9Upper 1153 16 07 06-10

EnvironmentRural 1519 13 IUrban 3303 17 ** NS 12 10-1-5

Adverse outcome of previous pregnancyNo 3974 14 1Yes 663 26 *** 17 14-2 2

Pregnancy complicationNo 3164 6 1Yes 1606 33 7-4 61 9 0

(1) Not adjusted(2) Adjusted for all significant characteristics including age and parity

p<0.05* p<OOOI

remained significantly linked to the pathway; allmaternal characteristics, except place of residence,remained significant.The hospitalisation rate increased with the degree of

specialisation of the care pathway (table 4).Hospitalisation was more frequent among women

who lived in an urban environment, who had a historyof adverse outcome of the last previous pregnancy, orwho had a complication of the pregnancy; it was lessfrequent among women with a high educational level.After adjustment, only the relation betweenhospitalisation and place of residence disappeared.



j.com/

J Epidem

iol Com

munity H


ber 1987. Dow

nloaded from


Contribution of specialists to antenatal care in France: impact on level of care during pregnancy and delivery 325Table 5 Place of delivery according to antenatal care pathway

Pathway I Pathway 2 Pathway 3 Pathway 4% % % % Significance

Maternity unitNo. of deliveries/year<500 47 21 14 16500-1-199 31 43 40 41,>1200 21 35 46 43

(405) (822) (776) (2812)

*** p<O001

Table 6 Frequency ofinduction oflabour according to antenatal care pathway, place ofdelivery, and maternal characteristics

Logistic regression

No. % S(l) S(2) OR CI 95%

Antenatal care pathway1 412 3 12 832 8 ** 24 1 2-493 791 10 2-9 1-4-5-74 2882 9 2-9 1 5-55

Maternity unitNo. of deliveries/year< 500 923 6 1500-1199 1944 9 * NS 12 09-1 7> 1200 1945 10 1-3 10-1 9

DiplomaNone 996 6 1Intermediate 2434 9 15 1 1-21Upper 1144 11 2-0 1 4-2 8

EnvironmentRural 1503 8Urban 3280 9 NS

Adverse outcome of previous pregnancyNo 3269 9Yes 653 10 NS

Gestational age >42 weeksNo 4414 9 1Yes 158 17 2-2 1-43-5

(1) Not adjusted(2) Adjusted for all significant characteristics (age and parity excluded as they were not significantly associated with induction)** p<0-01

*** p<0-00

DELIVERY CONDITIONS

At delivery, the women in the four pathways weredistributed differently among the three types ofmaternity units defined according to the annualnumber of deliveries (table 5). The women cared forexclusively by a general practitioner gave birth morefrequently in small units.The caesarean section rate did not differ

significantly among the four pathways: 9%, 10%,11%, and 11% respectively, for pathways 1,2,3, and 4.

Induction of labour was less often performedamong women of pathway 1 (table 6). The rate ofinductions was greater among the women with a highlevel of education and among those with a prolonged

pregnancy. After adjustment in a model including thetype of maternity unit, the educational level, and theduration of gestation, induction of labour was almostthree times less frequent in pathway 1 than in theothers, which did not otherwise differ amongthemselves. The type of maternity unit appeared to beof no significance; on the other hand, level ofeducation and length of gestation retained specialimportance. Intrapartum electronic fetal monitoring(EFM) was also less frequent among those in pathway1 (table 7). The role ofmedical indications ofEFM wasvery weak: only 77% of the women with suchindicators were monitored with EFM versus 67% ofthe other women. After taking into account the type ofmaternity unit, parity, age, place of residence, and



j.com/

J Epidem

iol Com

munity H


ber 1987. Dow

nloaded from


Table 7 Frequency of intrapartum electronic fetal monitoring according to antenatal care pathway, place of delivery, andmaternal characteristics

Logistic regression

No. % S(1) S(2) OR Cl 95%

Antenatal care pathway1 409 55 12 823 69 1-6 1-2 2-13 785 71 1-6 1-2 2-14 2852 71 1-6 13 21

Maternity unitNo. of deliveries/year

<500 913 59 1500-1199 1922 66 1-2 1-0 1-4-1200 1929 77 2-0 1 6-24

DiplomasNone 987 69Intermediate 2416 70 NSUpper 1133 71

EnvironmentRural 1486 66 1Urban 3250 71 NS 11 0-9 1-3

Adverse outcome of previous pregnancyNo 3909 70Yes 645 70 NS

Medical indication for monitoringNo 3269 67 1Yes 1200 77 1-6 14-1-9

(1) Not adjusted(2) Adjusted for all significant characteristics including age and parity*** p<O001

medical indications, EFM remained less frequentamong the women of pathway 1. In this model, placeof residence was no longer significant, contrary towhat one observed for medical indications and thetype of maternity unit.

Discussion

This study stressed the important influence on

antenatal care of the degree of specialisation of thepathway: the amount of care increased with theearliness of the specialist's intervention. Duringdelivery, the influence ofpathways was less important.Women in the most specialised pathways gave birthmore frequently in large maternity units. However, therate of obstetric interventions was overall the same inthe various pathways, except among the women caredfor exclusively by a general practitioner.

DEFINITIONS OF PATHWAYS

In this study, the earliness of the specialist'sinvolvement in antenatal care has been used ratherthan the proportion of antenatal visits by a specialist.This definition made it possible to take into accountthe trimester in which a complication appeared in thecourse of a pregnancy, and to specify the "dose-

response" effect between the pathway's degree ofspecialisation and the amount ofantenatal care. It alsoallowed one to know the qualification of the personresponsible for care during the last trimester, who canhave an influence on the management of labour anddelivery. This definition also reflected a mechanismcommon in France-transfer of care from generalpractitioner to specialist in the second or thirdtrimester of pregnancy-which accounted for thesmall proportion (2%) of women excluded from thestudy because they had only an isolated visit from aspecialist.There were several reasons for including midwives

with the specialists: midwives have a three-yeartraining in obstetrics. When they are involved inantenatal care, they generally work within a maternityunit, along with specialists, and are seldom solelyresponsible for the supervision of a pregnancy. In ourstudy population, only 46 women were cared forexclusively by a midwife.

ACCESSIBILITY OF CAREIn France in 1981, as already observed earlier,6 thechoice of specialised antenatal care was largelydetermined by the sociodemographic characteristicsof the women. But, apart from the influence of the

326 E Bruno Hubert, Beatrice Blondel, and Monique Kaminski



j.com/

J Epidem

iol Com

munity H


ber 1987. Dow

nloaded from


Contribution of specialists to antenatal care in France: inpact on level of care during pregnancy and deliveryspecialisation of the medical pathways, importantdisparities in care have persisted; they are linked todifferences of geographic and socioculturalaccessibility.'2 Geographic accessibility, estimatedfrom the place of residence, had an influence onantenatal care, but this influence was attenuated afteradjustment for the other sociodemographic variablesand the pathways. This appears to indicate that thedifferences perceived are to a large extent influencedby the supply of services.

After medical complications and care pathways hadbeen taken into account, women with a higheducational level tended to have more examinationsand interventions. The importance of educationallevel was also reported in a study on the diffusion ofultrasound examinations in France between 1976 and1981.12 These results suggest that care was not givenstrictly for medical reasons and that the requests ofwomen or their social situation influenced medicalbehaviour.'3 On the other hand, hospitalisationduring pregnancy was more frequent among womenwith a low educational level. This finding agreed withthat of Mizrahi and Mizrahi'4 who observed that "animportant substitution takes place between outpatientvisits and hospitalisation as income increases".

MEDICAL PRACTICESThe "dose-response" relationship between the degreeof specialisation and the number of visits andultrasound examinations reflected a difference inpractices between general practitioners and specialists.By training, specialists have an approach to pregnancyfocused on diseases and they tend to have frequentrecourse to visits and procedures, while generalpractitioners tend to be more orientated to supervisionand advice adapted to a normal pregnancyI5 and, bythe same token, are less interventionist. Their workorganisation is also different: specialists areaccustomed to regular and planned appointmentswhile general practitioners exercise a form ofcare thatis first of all adapted to the four visits recommended bythe Ministry of Health. Concerning interventionsduring labour and delivery, differences dependedupon whether or not a specialist was providing the carein the third trimester. That the smallest number ofobstetric interventions occurred in pathway 1 can alsobe explained by the fact that, in this pathway,deliveries often took place in very small maternityunits (fewer than 300 deliveries annually) or werecarried out by a general practitioner.9

MODIFICATIONS IN THE PROVISION OFOBSTETRIC CAREThe results of this study must be examined in thecontext of the modifications in care during pregnancyand delivery in France. Specialisation of antenatal

care increased from 1976 to 1981; the proportion ofgeneral practitioners providing complete care topregnant women declined from 20-8% to 8-8%.2 Thenumber of gynaecologists and obstetricians in Franceincreased from 3158 in 1980 to 3994 in 1984, bringingthe rate of these specialists up from 3-9 to 5 3 per 1000births-an increase of 36% in four years.'6

In the same way, the number of procedures andinterventions increased between 1976 and 1981: theproportion of women having had more than sevenvisits rose from 25 to 42%, ultrasound examinationsfrom 11 to 82%, hospitalisation during pregnancyfrom 13 to 16%, intrapartum electronic fetalmonitoring from 31 to 71 %, induction of labour from7 to 8%, and caesarean sections from 8-5 to 11%.2These augmentations concerned to a great extentscreening in pregnancy; interventions increased moremoderately and did not attain the rates found in othercountries such as the United States where, during thesame period, caesarean section rates increased from10-4 to 17 9%.17Two factors have contributed to this increase in

medical care: (1) the specific effect of procedures andinterventions: this is the case, for example, ofintrapartum electronic fetal monitoring associatedwith an increased risk ofcaesarean section 18 19; (2) thespecific effect of specialists who play a dual function inthe increase of care: by their leading role in the use ofnew technologies and by their increasing involvementin the supervision of normal pregnancies. Thisphenomenon leads to a rapid and large-scale use ofrecent technological innovations for pregnant women,whether their pregnancy is complicated or not.However, the effectiveness of these practices has notalways been sufficiently assessed, and manyuncertainties remain as to what constitutes optimumcare for a normal pregnancy.2>25

THE ROLES OF GENERAL PRACTITIONER,MIDWIFE, AND SPECIALISTThe above observations lead one to reflect on therespective roles of general practitioner, midwife, andspecialist in antenatal care. Keirse, Parboosingh, andRobinson have shown how these three healthprofessions complement each other.'5 26 27 Specialistsare trained to care for obstetric disorders; generalpractitioners integrate antenatal care in an overallsupervision of the health of women and children andare in a better position to ensure routine care;midwives have a privileged role in assuring primarycare and offering pregnant women support and adviceadapted to their situation. This practice is wellperceived in the countries of northern Europe whereobstetricians take charge of complicated pregnanciesor assure a limited number of antenatal visits fornormal pregnancies.' The same type oforganisation is

327



j.com/

J Epidem

iol Com

munity H


ber 1987. Dow

nloaded from


E Bruno Hubert, Beatrice Blondel, and Monique Kaminskiat the base of experience in Aberdeen: here, through abetter definition of the objectives of antenatal care, ithas been possible to achieve, among other things, amore balanced share of antenatal care visits betweenspecialists and general practitioners for normalpregnancies.28 Such an approach may be moredifficult to pursue in systems of care that are morecompetitive in nature, such as the one in France.

Conclusion

Our study has illustrated the relation between thespecialisation ofantenatal care and the amount ofcareand number of interventions. The question remainswhether increased specialisation is desirable for thosewomen experiencing a normal pregnancy, as thisspecialisation favours the diffusion of procedures andtechniques, the effectiveness of which has not beensufficiently evaluated.

Correspondence should be addressed to: BeatriceBlondel, U 149 INSERM, 16, Avenue Paul Vaillant-Couturier, 94807 Villejuif cedex, France.

References

1 Blondel B, Pusch D, Schmidt E. Some characteristics ofantenatal care in 13 European countries. Br J ObstetGynaecology 1985; 92: 565-8.

2 Rumeau-Rouquette C, du Mazaubrun C, Rabarison Y(eds). Naitre en France. 10 ans d'evolution 1972-1981.Paris: INSERM-Doin 1984.

3Wollast E, Vandenbussche P, Buekens P. Evaluation de lasurveillance prenatale en Belgique et comparaison entreles secteurs medicaux publiques et prives. Rev Epidem etSante Pubi 1986; 34: 52-8.

4 Blondel B, Kaminski M, du Mazaubrun C, Rumeau-Rouquette C. Surveillance prenatale et filieres medicalespendant la grossesse. Rev Epidem et Sante Publ 1982; 30:21-34.

5 Robine JM, Maquin P, Nicaud V, Hatton F. Facteursdeterminants des pratiques de sante: exemple de lasurveillance de la grossesse en milieu rural et urbain. RevEpidem et Sante Publ 1985; 33: 203-11.

6 Blondel B, Kaminski M, Breart G. Antenatal care andmaternal demographic and social characteristics.Evolution in France between 1972 and 1976. J EpidemiolCommunity Health 1980; 34: 157-63.

Cooney JP. What determines the start of prenatal care?Prenatal care, insurance and education. Med Care 1985;23: 986-97.

8 Niswander K, Henson G, Elbourne D, et al. Adverseoutcome of pregnancy and the quality of obstetric care.Lancet 1984; ii: 827-31.

9Hubert B. Specialisation de la surveillance prenatale. effetssur les soins pendant la grossesse et sur les conditions del'accouchement. Paris: Memoire de DEA, 1986.

10 Dixon DJ, et al. BMDP statistical software. Berkeley:University of California Press, 1985.Kleinbaum DG, Kupper LL, Morganstern H.Epidemiologic research. Belmont: Lifetime LearningPublication, 1982.

Poisson-Salomon AS, Breart G, Maillard F, Rumeau-Rouquette C. Diffusion of ultrasound examinationduring pregnancy in France between 1976 and 1981. Int JEpidemiol 1987; 16: 234-8.

3 Garcia J, Blondel B, Saurel-Cubizolles MJ. The socialneeds of childbearing families. Social policies and theorganisation ofhealth care. In: Effective care inpregnancyand childbirth. Enkin M, Keirse M, Chalmers I (eds).Oxford: Oxford University Press, in press.

14 Mizrahi A, Mizrahi A. Influence des *facteurs socio-economiques sur l'hospitalisation. Paris: CREDOC, 1977.

5 Keirse MJ. Interaction between primary and secondaryantenatal care, with particular reference to theNetherlands. In: Effectiveness andsatisfaction in antenatalcare. Enkin M, Chalmers I (eds). London: SIMP/Heinemann, 1982; 222-33.

16 Caisse Nationale d'Assurance Maladie. Carnetsstatistiques, 1985; 20.

7 Placek PJ, Taffel S, Moien M. Cesarean section deliveryrates: United States, 1981. Am J Public Health 1983; 73:861-2.

18 Leveno KJ, Cunningham FG, Nelson N, et al. Aprospective comparison of selective and universalelectronic fetal monitoring in 34 995 pregnancies. N EnglJ Med 1986; 315: 615-9.

19 Breart G. Un probleme de decision en perinatalogie:1'enregistrement continu du coeur foetal pendant letravail. Rev Epidem et Sante Publ 1984; 32: 34-9.

20 Committee to study the prevention of low birthweight.Preventing low birth weight. Washington: Institute of

2Medicine, National Academy Press, 1985.

21 Hall MH, Chng PK, McGillivray 1. Is routine antenatalcare worthwhile? Lancet 1980; ii: 78-80.

22 National Institutes of Health consensus developmentconference. The use of diagnostic ultrasound imagingduring pregnancy. JAMA 1984; 252: 669--72.

23 Kaminski M, Breart G, Buekens P, et al (eds). Perinatalcare delivery sYstems, description and evaluation inEuropean Community countries. Oxford: OxfordUniversity Press, 1986.

24 Enkin M, Chalmers I. Effectiveness and satisfaction inantenatal care. London: SIMP/Heinemann, 1982.

25 Chard T, Richards M. Benefits and hazards of the newsobstetrics. London: SIMP, 1977.

26 Parboosingh J, Kerr M. Innovations in the role of obstetrichospitals in prenatal care. In: EQffectiveness andsatisfaction in antenatal care, Enkin M, Chalmers I (eds).London: SIMP/Heinemann, 1982: 254-65.

27 Robinson S, Golden J, Bradley S. The role of the midwifein the provision of antenatal care. In: Effectiveness andsatisfaction in antenatal care. Enkin M, Chalmers I (eds).London: SIMP/Heinemann, 1982: 234-46.

28 Hall M, Macintyre S, Porter M. Antenatal care assessed.Aberdeen: Aberdeen University Press, 1985.

Accepted for publication July 1987

328



j.com/

J Epidem

iol Com

munity H


ber 1987. Dow

nloaded from


contribution of to antenatal care in france: impact level ...during the following trimester or...

Documents