Assessing restructuration of perinatal care in Ile de

Assessing restructuration of perinatal care in Ile de France 1998 -2009 C. Crenn Hebert, APHP, Colombes, Perinatal Network Hauts de Seine North, PERINAT-ARHIF, France ORAH 2009

Create health information system at regional level Dr C. Crenn Hebert, Dr C. Menguy, E. Lebreton, G. Echardour, Dr A. Serfaty APHP, J. Zeitlin APHP, Perinatal Network Hauts de Seine North, France Hal Montreuil Seine Saint-Denis, France « PERINAT-ARHIF » , Paris, France ARHIF, Paris, France Hal Trousseau, Perinatal Network Eastern Paris, France INSERM, UMR S 953, UPMC Univ Paris 06, France

Content • Background: Ile de France region and policies • Method for creating information system: – Health certificate – Hospital discharge summary • Results: – Perinatal structures – Perinatal indicators – Focus on Hauts de Seine district • Discussion • Conclusion

Ile de France : > 1/5 French births

Background: perinatal policies Perinatal care: a priority by Regional health authority ØObjective: Ø 80% very preterm births in level III perinatal centre (PC III) Ø Level I: maternity unit without neonatal care Ø Level IIA: on-site neonatal care Ø Level IIB: on-site neonatal intensive care unit Ø Level III: on-site neonatal resuscitation Ø Specialised PCIII: foetal medicine or neonatal surgery *L'action collective périnatalité en Ile de France, 1996 -2000, A. Serfaty, E. Papiernik, ed ENSP 2005

Regional Health Organization Schemes: ØAssess and elaborate successive schemes (1994, 1999, 2006) ØIn partnership with the regional health authority and perinatal health clinicians ØThe need for data …. Ø Regional birth commission Survey in 1998

1998: 8 Perinatal Centres level III + 2 specialised PCIII (foetal medicine or neonatal surgery) 135 maternity units- unequal distribution of PCIII over the 8 districts

1998: Regional birth commission Survey: only 60% of very preterm births in PC III 57% 82% 70% 59% 64% 40% 52% Regionalisation of very preterm birth sites in the Paris region in 1998 ; Annie Serfaty et al. Santé publique 2003, vol 15, no 4, pp. 491 -502

Creating information system: method • Criteria: ØUse of existing databases ØRoutine data ØData are available quickly ØData are validated • Involving perinatal health professionals: >Analysis to be shared between regional health authority and data producer

1 - Health certificates • First one at birth (live birth): indicators of birth certificate (BC) Ø parental demographics, Ø antenatal period, Ø delivery, Ø newborn characteristics (gestational age, birth weight…), Ø first week events (followed by 2 nd certificate at 9 months, 3 rd certificate at 24 months) • Sent to maternal and child protection services in district of parents residence • Common regional database from all districts: in progress from 2002

Hospital Maternity unit, Neonatal unit Birth certificates MCPS of residence district (anonymous files) Health ministry HEALTH CERTIFICATES: Each livebirth (BC) MATERNAL AND CHILD PROTECTION SERVICES: MCPS District of residence Validation Modification “Perinat-arhif” statistician Epidemiological research unit Regional health authority Data management Common regional data base Statistics Analysis validated by regional medical group of MCPS Drs

2 - Hospital Discharge Data System : (French PMSI) • Hospital Discharge Summary : perinatal indicators set in IDF region, for every hospital, depending of place of residence Ø pregnant or mother stay Ø newborn (live or stillborn > 2009) or neonate stay Ø until 2008, no gestational age is notified • But without any link Mother-Baby! “PERINAT-ARHIF” system

Hospital Maternity unit, Neonatal unit Hospital Discharge Data System : HDDS Unit Discharge Summary: UDS HDDS UDS MID Hospital Discharge Summary LINKING FILE ( anonymous files) e-PMSI Health Ministery Quality Control Audits Validation Modification Data management Perinatal data base Statistics Analysis validated by Perinatal health networks Medical committees Medical Information Department: MID Hospital Discharge Summary + Linking File: mother and baby + direct validation process+ audits = “PERINAT-ARHIF” system “Perinat-arhif” team Regional health authority

www. perinat-arhif. org

Download application programs 2006, 2007 Patient flows But also: - Regional summary tables - Hospital summary tables … 85 perinatal indicators compatible: . AUDIPOG (www. audipog. net). EUROPERISTAT (www. europeristat. com)

Completeness of the data Comparison with civil registration « INSEE » • Birth Certificates (BC) 2007 93. 3% • Hospital Discharge Summary (HDS) 2007 95. 4% Great heterogeneity between each district But giving feed-back to professionals seems to improve the reliability

Completeness BC/district/ 2006 and 2007 / source INSEE civil registration

Completeness HDS 2006 AND 2007 / INSEE

Results Perinatal structures Perinatal indicators Focus on Hauts-de-Seine district

Restructuring perinatal care in IDF 1998 Ø 170 000 births Ø by IDF residents : source civil registration INSEE 2007 Ø 179 264 births by IDF residents : (INSEE) Ø 135 maternity units Ø 105 maternity units Ø 8 PC III Ø 14 PC III Ø 15 in 2008 Ø 16 in 2009

2007 – 14 PC III (incl spec PCIII)

Deliveries and perinatal centre type (source HDD) PC I : decrease PC II : stable PC III: increase

Newborns with birth weight <1500 g (source HDD)

2007 80% Newborns <1500 g born in CP III (source HDD) 81% 76% 83% 81% 83% 87% 74% 77%

2007 78% Newborns <32 wk born in PC III (source BC) 80% 73% 81% 83% 78% 72% 75% 0% to 2, 8% gestational age missing

Focus on Hauts-de-Seine (Hd. S) district

Differences in populations between north and south of district (2007) source HDD Residents in North • <1500 g : 1, 04%* * p= 0. 03 Residents in South • <1500 g : 0, 77%*

Impact of not having a PC III? Differences in health care Hauts de Seine North/ Hauts de Seine South (2007) (source HDD) Residents in North Residents in South Ø < 1500 g born in PCIII: 91, 3%* 74, 4%* Ø Multiple delivery in PCIII: 23, 5%** * p =0. 003 **p<0. 0001 Ø Multiple delivery in PCIII 62, 7%**

2008 15 th PC III in Hauts de Seine North

2008 Newborns <1500 g born in PC III (source HDD) 84% 72% 77% 85% 86% 90% 81% 77%

Results of opening PC III in North 2008 Hauts de Seine North/ Hauts de Seine South (source HDD) Residents in North Residents in South 92 • < 1500 g born in PCIII: 83, 6% (ns) • < 1500 g born in PCIII: 87, 5% (ns) • Multiple delivery in PCIII: 45, 3%** • Multiple delivery in PCIII: 59, 7%** **p=0. 001

Results of opening PC III in North 2008 Hauts de Seine North/ Hauts de Seine South (source HDD) Residents in North 92 Residents in South 92 • < 1500 g born in PC I or IIA: 7, 0% • < 1500 g born in PC I or IIA: 8, 0% • Multiple delivery in PC I or IIA: 39, 9%** • Multiple delivery in PCI or IIA: 26, 8%** **p=0. 002

Discussion • Change in perinatal structures may be monitored at different levels • Routine data information important for planning of maternity care • Evolution in perinatal health indicators may also reflect: – Changes in health of the population – Changes in medical practices among health professionals

Discussion • Relevance of the initial indicator is questionable (rate of in-born very preterm birth) • In Hauts de Seine district, previous PC level IIB is upgraded to PC level III • Hospital discharge data provided only birth weight , but similar data are obtained from birth certificates with gestational age in 2007 Thoroughness and accuracy of the data: hard to achieve…

Conclusion • Combining analysis from 2 routine information systems helps to provide validated data for a regional vision of restructuration • Analysis at perinatal network level and smaller territorial scales is important to access population needs and the adequacy of health care provision.

Thank you for your attention Any Questions?

1994 -2009: Government policies in France Ø Change in organization of perinatal care – – Level I: maternity unit without neonatal care Level IIA: on-site neonatal care Level IIB: on-site neonatal intensive care unit Level III: on-site neonatal resuscitation Ø Change in medical practice • Toward regionalisation: place of birth according to risk level • In utero-transport to avoid negative effect of post-natal transport

2007 – 14 PC III Paris 30820 27743 18624 24937 20331 19899 18978 17932 Living Births district volumes differ, PC III capacities also

Réseaux de périnatalité d'Ile de France Réseau Périnatal 92 Nord CPN III CPN II B CPN II A CPN I Établissement AP-HP Établissement Public de Santé Établissement PSPH Établissement Privé à but non lucratif Établissement Privé à but lucratif © 2008 ARHIF-APHP-CRAMIF tous droits réservés

Réseaux de périnatalité d'Ile de France Possibilité de collaboration avec Réseau Périnatal 92 Nord Réseau Périnatal 92 Sud CPN III CPN II B CPN II A CPN I Établissement AP-HP Établissement Public de Santé Établissement PSPH Établissement Privé à but non lucratif Établissement Privé à but lucratif © 2008 ARHIF-APHP-CRAMIF tous droits réservés