Article Text
PDF
Abstract
Objective To determine whether birth outside a level-3 centre (outborn) is associated with a difference in the combined outcome of mortality or moderate-to-severe neurological impairment at 5.5 years of age compared with birth in a level-3 centre (inborn) when antenatal steroids and gestational age (GA) are accounted for.
Design Individual matched study nested within a prospective cohort. Each outborn infant was matched using GA and antenatal steroids with a maximum of four inborns. Conditional logistic regression was used to calculate ORs before being adjusted using maternal and birth characteristics. Analyses were carried out after multiple imputation for missing data.
Setting EPIPAGE-2 French national prospective cohort including births up to 34 weeks GA inclusive.
Patients Outborn and inborn control infants selected between 24 and 31 weeks GA were followed in the neonatal period and to 2 and 5.5 years. 3335 infants were eligible of whom all 498 outborns and 1235 inborn infants were included—equivalent to 2.5 inborns for each outborn.
Main outcome measure Survival without moderate-to-severe neurodevelopmental impairment at 5.5 years.
Results Chorioamnionitis, pre-eclampsia, caesarian birth and small-for-dates were more frequent among inborns, and spontaneous labour and antepartum haemorrhage among outborns. There was no difference in the main outcome measure at 5.5 years of age (adjusted OR 1.09, 95% CI 0.82 to 1.44); sensitivity analyses suggested improved outcomes at lower GAs for inborns.
Conclusion In this GA and steroid matched cohort, there was no difference in survival without moderate-to-severe neurodevelopmental impairment to 5.5 years of age between inborn and outborn very preterm children. This suggests steroids might be important in determining outcomes.
- Mortality
- Neonatology
- Epidemiology
Data availability statement
Data are available on reasonable request. The EPIPAGE-2 data are accessible to all research teams, public or private, French or foreign, subject to authorisation by the cohort Data Access Committee. The 2016 law for modernisation of the French Public Health System provides a legal framework for access to and reuse of already collected cohort data by complying with ‘Reference Methodology MR-004’. Only non-nominative data defined as having a low risk of reidentification are freely accessible. All data access requests must be submitted to the EPIPAGE 2 Data Access Committee; this committee evaluates the research projects based on the following criteria: (1) methodological strengths and weaknesses (feasibility, choice of methods to achieve the objectives); (2) absence of overlap with other ongoing projects - in caseof overlap, discussions with the different teams are organised; (3) relevance of the requested data for the project and respect for confidentiality. The study protocol, the data access charter and the data access procedure can be found on the EPIPAGE-2 website (https://epipage2.inserm.fr/index.php/fr/cote-recherche/235-acces-aux-donnees-et-questionnaires). Questionnaires and data catalogues are available at (https://pandora-epipage2.inserm.fr/public/).
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Excess mortality and/or neurological complications have been demonstrated in preterm infants born outside neonatal intensive care units compared with inborns in many countries.
Pre-eclampsia, premature rupture of membranes and small for gestational age are frequent in inborns whereas spontaneous birth and placental abruption are frequent in outborns.
Administration of antenatal steroids is lower in outborn than inborn pregnancies, and is independently associated with worse outcomes.
WHAT THIS STUDY ADDS
There were no differences in survival to 5.5 years without moderate/severe neurodevelopmental impairment between inborns and outborns when gestational age and antenatal steroids were accounted for.
There were improved outcomes at 5.5 years with inborn delivery for births at 24–27 weeks of gestation, but not below 24 weeks.
There were no differences in outcomes at 2 years of age or at hospital discharge–other than for necrotising enterocolitis, which was more frequent among outborns.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Improved protocols should be implemented to enhance antenatal transfer of women delivering at extremely preterm gestations.
Administration of antenatal steroids to women presenting to non-level 3 units who risk premature delivery should be enhanced.
Further research about the impact of place of birth on outcomes to school age for children born very preterm (below 32 weeks of gestation) is required.
Introduction
Excess mortality and increased neurological complications have been demonstrated in preterm infants born in non-tertiary (level 1 and level 2) hospitals without neonatal intensive care facilities in the USA,1–3 Finland,4 Spain,5 Australia,6 7 Canada8 and England.9 10 Possible reasons that are not mutually exclusive are that outborn infants were more unwell prior to delivery, that perinatal care was suboptimal for those outborn or that postnatal transfer is deleterious to the infants’ well-being. Previous studies have investigated the longer-term neurological consequences of outborn very preterm infants up to 2–3 years of age.7 11–13
Non-level-3 hospitals are often able to transfer mothers with ‘stable’ pathologies (where birth is not imminent) such as pre-eclampsia,8 spontaneous rupture of membranes without labour2 6 and small for gestation age.2 6 8 ‘Unstable’ pathologies such as spontaneous labour and antepartum haemorrhage are more frequent in outborn births. This creates distortion in caseload between transferring and receiving units which introduces selection bias when trying to understand differences in outcomes according to place of birth.2 6 8 9 In addition, the use of antenatal steroids is more prevalent in inborn births2 7 9 10 14–17 and represents an important source of confounding between the outcomes of outborn and inborn infants; steroids being independently associated with reduced mortality and neuromorbidity.18
We used the EPIPAGE-219–21 national, population-based prospective study to examine the hypothesis that, when the influences of gestational age (GA) and antenatal steroids are controlled for, the combined outcome of survival without moderate or severe neurodevelopmental impairment at 5.5 years is not different between outborn and inborn infants.
Methods
Study population
EPIPAGE-2 prospectively collected data on all births occurring in France in 2011 under 35 weeks GA.19–21 Infants born in mainland France between 24 and 31 weeks of GA were eligible for inclusion in this study. Infants born at home, who had lethal congenital abnormalities or for whom antenatal steroids data were missing were excluded.
Exposure
In France, perinatal centres are categorised into level-3 units, which provide neonatal intensive care for infants requiring ongoing invasive ventilation,and non-level-3 units, which may provide some short-term invasive ventilation. National guidelines in 2011 recommended that births at less than 32 weeks of gestation or less than 1500 g occur in level-3 units,22 building on the regionalised system of care first mandated in 1998.23
Outcomes
The primary outcome was survival without moderate or severe neurodevelopmental impairment at 5.5 years of age. Evaluations were performed by trained paediatricians and developmental child psychologists, and included cerebral palsy (CP), sensory (vision and hearing) and cognitive abilities. CP was classified using the Global Motor Function Classification System (GMFCS), with level 1 being mild, levels 2–3 moderate and levels 4–5 severe impairment.24 Moderate or severe visual impairment was defined as bilateral binocular visual acuity 3.2/10, and moderate or severe hearing impairment as unilateral or bi-lateral hearing loss >40 dB not corrected or only partially corrected by hearing aids. Cognitive deficiency was measured using the full scale intelligence quotient (FSIQ) from the French version of the Wechsler Preschool and Primary Scale of Intelligence fourth Edition,25 with moderate-to-severe impairment defined as an FSIQ 2 SD below the mean of a term-born reference group.21 We present results for the primary outcome and survival, and for moderate-to-severe neurodevelopmental impairment among survivors.
We also investigated earlier outcomes. At discharge, we examined survival without severe neonatal morbidities (intraventricular haemorrhage grades III–IV, or cystic periventricular leukomalacia; necrotising enterocolitis, Bell stages 2–3; severe bronchopulmonary dysplasia, requiring ≥30% oxygen and/or ventilatory support at 36 weeks’ GA; retinopathy of prematurity stage 3 or higher) as well as individual morbidities.20 At age 2, we examined survival without moderate or severe sensorimotor impairment defined as CP (GMFCS levels 2–5) and/or unilateral or bilateral deafness and/or blindness, as reported by attending paediatricians. We additionally report the number of children with a parentally reported Ages and Stages Questionnaire (ASQ) score below the reference threshold in at least one of the five domains (communication abilities, gross motor skills, fine motor skills, problem solving abilities and personal-social skills).26
Statistical analysis
We compared potential confounding factors between outborn and inborn groups of infants, then performed conditional logistic regression to explore outcome differences using a matched cohort design to eliminate differences in GA and antenatal steroid administration between the two groups. A greedy matching algorithm without replacement was used, with exact matching on week of GA and antenatal steroid administration in the 2 weeks prior to birth (three categories: full course, defined as two doses more than 24 hours before birth, partial course, or no steroids); neither variable had missing data. As there were missing data among outcomes and covariates, all analyses were performed after multiple imputation; hence unique identifiers of matched infants were noted for subsequent use with the imputed data. The imputation model included the exposure, outcomes and matching variables, as well as variables potentially predicting non-response or the outcome. Categorical variables were imputed using logistic or multinomial regression and continuous variables using predictive mean matching. Full details are shown in online supplemental table 1. Using the entire population of infants eligible for study inclusion; we generated 50 independent imputed datasets with 30 iterations each. Data for infants previously selected by the matching process were then retained from each imputed data set for use in analyses. Data were also weighted to account for imbalances due to varying numbers of matched inborn children; outborns were assigned a weight of one and each control subject had a weight proportional to the number of cases in its matched set divided by the number of controls in the set. Conditional logistic regression was performed with adjustment for maternal age (years), parents’ socioeconomic status (using the highest level of either parent, or the mother only if she was single), nulliparous, use of infertility treatment, multiple pregnancy, tocolysis, antenatal administration of magnesium sulfate, premature rupture of membranes (PROM), spontaneous onset of labour, chorioamnionitis, pre-eclampsia, antepartum haemorrhage, caesarean section, cephalic presentation, small-for-GA and sex (all variables binary unless stated).
Supplemental material
Analyses were performed in SAS using ‘proc psmatch’ for matching and ‘proc MI’ for imputation. Results were pooled according to Rubin’s rule.27 All statistical tests were two sided and the probability of a type 1 error (α) was set at <0.05.
Sensitivity analyses
We assessed the primary outcome in restricted populations of 24–27 weeks and 28–31 weeks of gestation to counter for potential effects from infants who were not transferred and from differences in attitude to extreme preterm birth; we also provide estimates by week of GA. Because reasons for delivery may vary between fetuses in multiple pregnancies, we repeated these analyses using the complete GA range (24–31 weeks) but restricted to singleton births only. Finally, to account for the influence of obstetric decision-making, we examined the population of 24–31 week GA fetuses alive at maternal admission to hospital. We also repeated the principal analyses using only subjects with complete data.
Patient and public involvement
There was no patient involvement in this study. However, EPIPAGE-2 maintains contact with the cohort through letters, newsletters and its website, and follow-up is ongoing. National parents’ associations assist with the dissemination of results.
Results
Of 3816 live born infants enrolled between 24 and 31 weeks of gestation in EPIPAGE-2, 3355 were eligible for inclusion in the study (figure 1); no important differences were seen with the 102 children missing information about antenatal steroid administration (online supplemental table 2). A total of 498 outborn infants were matched with 1235 of the 2857 inborn infants (2.48 inborns per outborn). A comparison of the characteristics of the matched and unmatched inborn infants is contained in online supplemental table 3.
Flow chart of children eligible for inclusion.
Baseline maternal and newborn characteristics are presented in table 1. Administration of a complete course of steroids was lower for outborns (24.7%) than inborns (72.3%) in the whole population. After matching, important differences remained for maternal age, mother’s country of birth, use of infertility treatment, maternal pregnancy complications and obstetric treatments, and in fetal size at birth.
Perinatal characteristics of the initial cohort and matched outborn infants with inborn controls
Primary outcome
There was no difference in the primary outcome of survival without moderate or severe neurodevelopmental impairment at 5.5 years of age using imputed data (adjusted OR 1.09, 95% CI 0.82 to 1.44, p=0.56), as shown in table 2.
Outcomes at age 5.5 years in a matched study of inborn and outborn children from the EPIPAGE-2 cohort according to gestational age at birth (analysis using data after multiple imputation)
Secondary outcomes
Table 2 also shows results for the secondary outcomes. Prior to adjustment, there was a suggestion of improved survival at 5.5 years for inborns, with an unadjusted OR 1.36 (95% CI 0.99 to 1.88). This association was mitigated after accounting for potential confounders. At 2 years corrected age, survival without moderate or severe sensorimotor impairment was higher in the inborns in unadjusted analysis (OR 1.45, 95% CI 1.09 to 1.91), but this was no longer apparent after adjustment (adjusted OR 1.27, 95% CI 0.93 to 1.72). Among survivors at 2 years, there were no differences in CP (adjusted OR 0.66, 95% CI 0.21 to 1.34) or for the ASQ (adjusted OR 1.05, 95% CI 0.78 to 1.42). Prior to hospital discharge, there were no statistically significant differences in survival without severe neonatal morbidity or for individual pathologies except necrotising enterocolitis which was less frequent among inborns (3.6% compared with 5.8%, adjusted OR 0.56, 95% CI 0.32 to 0.98).
Sensitivity analyses
No statistically significant effects were seen in sensitivity analyses among those alive at the beginning of labour or when the population was restricted to singletons only. Among those born at 24–27 weeks’ GA, there was evidence of higher survival without moderate-to-severe impairment at 5.5 years of age for inborns compared with outborns (adjusted OR 1.71, 95% CI 1.05 to 2.81) but not for those born at 28–31 weeks’ gestation (table 3). Between 26 and 31 weeks gestation, there was a trend towards increasing ORs with decreasing GA although all of the 95% CI included 1.0 except for at 26 weeks where the 95% CI in adjusted analysis was 1.01 to 8.24 (p=0.048). Below 26 weeks, there was no evidence of a difference between inborns and outborns (online supplemental table 4).
Sensitivity analyses for survival without moderate-to-severe impairment at age 5.5 years in a matched study of inborn and outborn children from the EPIPAGE-2 cohort according to gestational age at birth (analysis using data after multiple imputation)
Results among cases with complete data (table 4) were broadly the same as the main analysis other than for CP, which showed a markedly lower odds at ages 2 (adjusted OR 0.22, 95% CI 0.10 to 0.47) and at age 5.5 (adjusted OR 0.41, 95% CI 0.19 to 0.87) years and for hearing impairment at 2 years of age (adjusted OR 0.17, 95% CI 0.07 to 0.44).
Outcomes at age 5.5 years in a matched study of inborn and outborn children from the EPIPAGE-2 cohort according to gestational age at birth (analysis using only subjects with complete data)
Discussion
Principal results
When GA and antenatal steroids are accounted for, in a large national cohort with neurological follow-up to 5.5 years of age, no differences were identified in survival without moderate or severe neurological impairment for very preterm outborn infants compared with inborns, nor were any differences seen in individual components. This was also the case at 2 years of age. With the exception of necrotising enterocolitis, no differences were seen in neonatal morbidities.
Strengths and limitations
The major strength of the EPIPAGE 2 study is that it is a large national cohort with detailed, prospective, long-term follow-up to preschool age carried out by trained investigators. Other recent publications2 6 9 10 28 have looked exclusively at mortality and neonatal neurological complications, which do not necessarily extrapolate to later functional ability. The follow-up rate—around 70% at 5.5 years—means our study compares favourably with other studies of outborns, for example, 62% of outborns and 69% of inborns at 1 year of age in Western Australia7; 57% of outborns and 77% of inborns were seen at 2–3 years in a cohort from New South Wales11; and under 50% were seen at 3 years of age in a Japanese study.29 However, the loss to follow-up is potentially a limitation: not all children completed all of the tests, meaning that there were up to 40% missing data for some components of the outcome; this was addressed using multiple imputation, which has been shown to be robust even when the missing-at-random assumption is violated.30
Another strength is the inclusion of all live births, plus the sensitivity analysis accounting for the entire population of at risk fetuses alive at maternal admission to hospital. Results from studies not including these data are difficult to interpret.2 10 12 13 For instance, one recent study which did not include pretransfer deaths used proportional hazards regression after propensity score matching and reported increased in-hospital mortality for inborns.2 Our analyses showed no differences between the inborn and outborn groups regardless of which baseline population was used. We were, however, limited by an inability to consider factors like the Apgar score or cord gas results that reflect status at birth and which may be less than ideal in non-level 3 units due to suboptimal obstetric care. We also could not evaluate the potential impact of the presence of paediatric transport teams in non-level 3 units prior to delivery as these data were not collected. Furthermore, despite the importance of studying long-term outcomes, management of infants in the EPIPAGE-2 cohort may no longer be representative of current care standards. For instance, there is an increasing emphasis on the neonatal ‘Golden Hour’,31 which may not have been uniformly implemented at the time in non-level-3 units. We unfortunately did not have the data to examine this, nor do we have more recent data with which to evaluate evolution of practices over time. However, long-term follow-up can only be based on historic practices; hence, the importance of trying to understand which factors might affect later outcomes.
Interpretation
The decision to match for antenatal steroids was based on their association with reductions in neonatal death and intraventricular haemorrhage.18 Previous studies of outborn preterm infants have noted similar patterns of increased neonatal deaths and neurological complications and the disproportional use of steroids in inborn versus outborn births.3 5–10 15 17 Our two-step methodology, matching then adjustment, has the advantage of initially neutralising confounding by matching steroids at the expense of tending to harmonise the characteristics of inborns and outborns. However, despite the matching process, the inborn group retained a similar predominance of ‘stable’ pathologies such as pre-eclampsia,8 intrauterine growth retardation6,8 10 preterm PROM6 10 and chorioamnionitis10 as seen in other studies. Similarly, the outborn deliveries featured more ‘unstable’ pathologies such as placental abruption6 and delivery was more frequent following spontaneous labour.6
The disadvantage of matching is the undesirable, and foreseeable, effect of potentially introducing a selection bias because of the assumption that there is no random error attributable to sampling variability, which is unlikely to be true.32 Differences exist between the matched and unmatched inborns for several antenatal characteristics—particularly, the use of antenatal steroids which was higher for inborns who were not matched. There were also higher levels of tocolysis, small for GA and PROM in unmatched inborns, but more pre-eclampsia among those who were matched.
The second step of our method therefore was to adjust the point estimate using maternal and pregnancy characteristics to reduce the potential selection bias induced by matching. Both before and after adjustment, we found no difference in outcome between the groups. This suggests that, after accounting for other factors, differences in antenatal steroid administration between inborn and outborn births play an important role in influencing long-term outcomes, and that public health policies should promote greater administration of antenatal steroids for women presenting with threatened preterm delivery.
Our sensitivity analyses looked at excluding both the lowest and highest GA infants. The high level of mortality observed among outborn infants of the lowest GAs has previously been reported.7 The 24-31 weeks' GA population had an estimated OR for the principal outcome not far from one. When we looked at only the 24–27 week infants, the adjusted estimate showed improved outcomes for inborns. Analyses performed by week of GA suggested a trend favouring inborn infants at lower GAs, although not at the GAs of 24 and 25 weeks for which attitudes in France in 2011 were known to be more mixed.33 34 These results are consistent with other studies that suggest benefits of antenatal transfer are greater with decreasing GA.1 7 35 Moreover, while over 80% of children were inborn and a regionalised system of care has existed in France since the 1990s,23 there remain a subset of outborns that could potentially be avoided.34
At 2 years of age, there were no differences in outcomes between our inborn and outborn groups, similar to two smaller Australian studies that looked at infants of 23–28 weeks’ GA11 and 23–25 weeks’ GA.7 Similarly, there were no differences in neonatal outcomes other than for necrotising enterocolitis which was of borderline statistical significance (p=0.043) and may be a chance finding due to multiple testing; this warrants further investigation.
Generalisability
There is no consensus regarding which GAs to include in studies of outborns, with other studies using different GA ranges,2 6 8–11 14 16 17 28 and some studies instead using a birth weight cut-off.3–5 36 We included infants of 24–31 weeks due to the recommendation that births of less than 32 weeks should occur in a Level-3 unit22 and because of the known poor outcomes of births at <24 weeks in France.19 Many European and other high-income countries have similar guidelines and approaches to those of France in 2011.37 Our results should, therefore, be relevant for such countries too.
Conclusions
Overall, there was no difference in longer-term survival to 5.5 years of age without moderate or severe neurological impairment between inborn and outborn very preterm children when GA and maternal antenatal steroids were accounted for. This suggests that antenatal steroids may be an important determinant of differences in outcome found between inborns and outborns in other studies. For infants born at the lowest GAs, however, an improved outcome was associated with birth in level 3 hospitals; this should motivate the implementation of improved protocols to enhance antenatal transfer of women delivering at these gestations.
Data availability statement
Data are available on reasonable request. The EPIPAGE-2 data are accessible to all research teams, public or private, French or foreign, subject to authorisation by the cohort Data Access Committee. The 2016 law for modernisation of the French Public Health System provides a legal framework for access to and reuse of already collected cohort data by complying with ‘Reference Methodology MR-004’. Only non-nominative data defined as having a low risk of reidentification are freely accessible. All data access requests must be submitted to the EPIPAGE 2 Data Access Committee; this committee evaluates the research projects based on the following criteria: (1) methodological strengths and weaknesses (feasibility, choice of methods to achieve the objectives); (2) absence of overlap with other ongoing projects - in caseof overlap, discussions with the different teams are organised; (3) relevance of the requested data for the project and respect for confidentiality. The study protocol, the data access charter and the data access procedure can be found on the EPIPAGE-2 website (https://epipage2.inserm.fr/index.php/fr/cote-recherche/235-acces-aux-donnees-et-questionnaires). Questionnaires and data catalogues are available at (https://pandora-epipage2.inserm.fr/public/).
Ethics statements
Patient consent for publication
Ethics approval
As required by French law and regulations, this study was approved by the national data protection authority (Commission Nationale de l’Informatique et des Libertés, CNIL no 911009) and by the appropriate ethics committees, that is, the advisory committee on the treatment of personal health data for research purposes (CCTIRS: Comité Consultatif sur le Traitement de l’Information en matière de Recherche, approval granted 18 November 2010; reference number 10.626) and the committee for the protection of people participating in biomedical research (CPP: Comité de Protection des Personnes, approval granted 18 March 2011, reference CPP SC-2873). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors would like to express their appreciation to the children and families who participated in EPIPAGE-2 without whom this study would have been impossible. The authors also wish to thank everyone involved in the EPIPAGE-2 Study Groups: EPIPAGE-2 Study group – 2 year follow-up: Alsace: D Astruc, P Kuhn, B Langer, J Matis, C Ramousset; Aquitaine: X Hernandorena, P Chabanier, L Joly-Pedespan, MJ Costedoat, A Leguen; Auvergne: B Lecomte, D Lemery, F Vendittelli; Basse-Normandie: G Beucher, M Dreyfus, B Guillois, Y Toure; Bourgogne: A Burguet, S Couvreur, JB Gouyon, P Sagot, N Colas; Bretagne: J Sizun, A Beuchée, P Pladys, F Rouget, RP Dupuy, D Soupre, F Charlot, S Roudaut; Centre: A Favreau, E Saliba, L Reboul; Champagne-Ardenne: N Bednarek, P Morville, V Verrière; Franche-Comté: G Thiriez, C Balamou; Haute-Normandie: L Marpeau, S Marret, C Barbier; Ile-de-France: G Kayem, X Durrmeyer, M Granier, M Ayoubi, O Baud, B Carbonne, L Foix L’Hélias, F Goffinet, PH Jarreau, D Mitanchez, P Boileau, C Duffaut, E Lorthe, L Cornu, R Moras; Languedoc-Roussillon: P Boulot, G Cambonie, H Daudé, A Badessi, N Tsaoussis; Limousin: A Bédu, F Mons, C Bahans; Lorraine: MH Binet, J Fresson, JM Hascoët, A Milton, O Morel, R Vieux, L Hilpert; Midi-Pyrénées: C Alberge, C Arnaud, C Vayssière, M Baron; Nord-Pas-de-Calais: ML Charkaluk, V Pierrat, D Subtil, P Truffert, S Akowanou; PACA et Corse: C D’Ercole, C Gire, U Simeoni, A Bongain, M Deschamps; Pays de Loire: B Branger, JC Rozé, N Winer, V Rouger, C Dupont; Picardie: J Gondry, G Krim, B Baby; Rhône-Alpes: M Debeir, O Claris, JC Picaud, S Rubio-Gurung, C Cans, A Ego, T Debillon, H Patural, A Rannaud; Guadeloupe: E Janky, A Poulichet, JM Rosenthal, E Coliné; Guyane: A Favre, N Joly; Martinique: S Châlons, J Pignol, PL Laurence; La Réunion : PY Robillard, S Samperiz, D Ramful. Inserm UMR 1153: PY Ancel, V Benhammou, B Blondel, M Bonet, A Brinis, A Coquelin, M Durox, M Kaminski, K Khemache, B Khoshnood, C Lebeaux, L Marchand-Martin, J Rousseau, MJ Saurel-Cubizolles, D Tran, J Zeitlin. EPIPAGE-2 Study group – 5 year follow-up : Alsace: D Astruc, P Kuhn, B Langer, J Matis, C Ramousset; Aquitaine: X Hernandorena, P Chabanier, L Joly-Pedespan, M Rebola, MJ Costedoat, A Leguen, C Martin; Auvergne: B Lecomte, D Lemery, F Vendittelli, E Rochette; Basse-Normandie: G Beucher, M Dreyfus, B Guillois, V. Datin-Dorrière, Y Toure, D Rots; Bourgogne: A Burguet, S Couvreur, JB Gouyon, P Sagot, N Colas, A Franzin; Bretagne: J Sizun, A Beuchée, P Pladys, F Rouget, RP Dupuy, D Soupre, F Charlot, S Roudaut; Centre: A Favreau, E Saliba, L Reboul, E Aoustin; Champagne-Ardenne: N Bednarek, P Morville, V Verrière; Franche-Comté: G Thiriez, C Balamou, C Ratajczak; Haute-Normandie: L Marpeau, S Marret, C Barbier, N Mestre; Ile-de-France: G Kayem, X Durrmeyer, M Granier, M Ayoubi, O Baud, B Carbonne, L Foix L’Hélias, F Goffinet, PH Jarreau, D Mitanchez, P Boileau, C Duffaut, E Lorthe, L Cornu, R Moras, D Salomon, S Medjahed, K Ahmed; Languedoc-Roussillon: P Boulot, G Cambonie, H Daudé, A Badessi, N Tsaoussis, M Poujol; Limousin: A Bédu, F Mons, C Bahans; Lorraine: MH Binet, J Fresson, JM Hascoët, A Milton, O Morel, R Vieux, L Hilpert; Midi-Pyrénées: C Alberge, C Arnaud, C Vayssière, M Baron; Nord-Pas-de-Calais: ML Charkaluk, V Pierrat, D Subtil, P Truffert, S Akowanou, D Roche, M Thibaut; PACA et Corse: C D’Ercole, C Gire, U Simeoni, A Bongain, M Deschamps, M Zahed; Pays de Loire: B Branger, JC Rozé, N Winer, V Rouger, C Dupont, H Martin; Picardie: J Gondry, G Krim, B Baby, I Popov; Rhône-Alpes: M Debeir, O Claris, JC Picaud, S Rubio-Gurung, C Cans, A Ego, T Debillon, H Patural, A Rannaud; Guadeloupe: E Janky, A Poulichet, JM Rosenthal, E Coliné, C Cabrera; Guyane: A Favre, N Joly, Stouvenel A; Martinique: S Châlons, J Pignol, PL Laurence, V Lochelongue; La Réunion : PY Robillard, S Samperiz, D Ramful. Inserm UMR 1153: PY Ancel, H Asadullah, V Benhammou, B Blondel, A Brinis, ML Charkaluk, A Coquelin, V Delormel, M Durox, M Fériaud, L Foix-L’Hélias, F Goffinet, M Kaminski, G Kayem, K Khemache, B Khoshnood, C Lebeaux, E Lorthe, L Marchand-Martin, AS Morgan, L Onestas, V Pierrat, J Rousseau, MJ Saurel-Cubizolles, D Tran, D Sylla, L Vasante-Annamale, J Zeitlin.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors P-YA and FG obtained funding for EPIPAGE-2 and initiated the cohort and follow-up. This research project was conceived by PJ and ASM, and further developed with input from CD, JF and TD. Initial analyses were performed by LM-M in discussion with ASM, PJ, TD and complementary analyses were conducted by LM-M in conjunction with ASM. The first draft of the paper was written by PJ and ASM, with critical review and input from all other authors. All authors critically revised the manuscript and agreed to the published version of the manuscript. ASM and PJ supervised all aspects of the work; ASM is the guarantor.
Funding The project has been funded with support from: (1) The French Institute of Public Health Research/Institute of Public Health and its partners: the French Health Ministry, the National Institute of Health and Medical Research (INSERM), the National Institute of Cancer, and the National Solidarity Fund for Autonomy (CNSA). (2) The National Research Agency through the French EQUIPEX program of investments in the future (reference ANR-11-EQPX-0038 and ANR-19-COHO-001). (3) The PREMUP Foundation. (4) Fondation de France (Reference 11779). (5) Fondation pour la Recherche Médicale (SPF20160936356—ASM). (6) Programme Hospitalier de Recherche Clinique Epinutri (DGOS13-040). (7) Ministère de l'Enseignement Supérieur, De La Recherche et de L'Innovation (G13129KK). (8) Apicil Foundation (R20065KK).
Competing interests Neil Marlow declares consultancy fees from Novartis and InfanDx GMBH during the past 5 years. There are no other competing interests.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.