NHS Maternity Care for Women with Multiple Births and Their Babies A study on feasibility of assessing care using data from births between 1 April 2015 and 31 March 2017 in England, Wales and Scotland.

Introduction This report focuses on the maternity care for women with multiple births during the period 1 April 2015 to 31 March 2017 and their babies. The purpose of this report is to describe the feasibility of assessing maternity care for women with multiple births and their babies, using routinely collected data. Methods This study examines the feasibility of using existing data sources and linkages within NMPA to report the characteristics and outcomes of twin pregnancy and birth and to assess the care of women with multiple birth. National guidelines from the National Institute for Health and Care Excellence (NICE) and the Royal College of Obstetricians and Gynaecologists (RCOG) and recommendations from Twins Trust and the Multiple Births Foundation were consulted in the development of audit measures. Maternal outcomes were reported per pregnancy. Perinatal outcomes were reported per pregnancy or per baby, as appropriate. The characteristics and outcomes of higher order births were assessed and reported separately from those of twins. Given that many of the national guidelines referred to local service configuration, an evaluation of the availability of specific services for women with multiple birth was conducted, by linking the results of the NMPA Organisational Survey 2017 with the location of birth of the women with multiple pregnancy.2 Key findings We have demonstrated that an audit of maternity and neonatal care for women and babies affected by multiple birth is feasible using NMPA methodology and data sources, but such an audit will be limited by data availability and quality issues. We identified 41608 babies born from multiple pregnancies in 20 655 women from England, Scotland and Wales. When compared with the number of multiple births reported in data from the Office for National Statistics, this represented an estimated case ascertainment of 89.5%, compared with case ascertainments of 92% in 2015/16 and 97% in 2016/17 for singleton births. Case ascertainment is affected by inaccuracies in the recorded number of infants born to each woman and by unavailability of data on the number of fetuses identified in the first trimester of pregnancy. Only two of 174 clinical guideline statements can be directly assessed using NMPA methods. These relate to recommendations that mothers should be supported to breastfeed and that neonatal networks should aim to reduce term neonatal admissions. The most common reason that recommendations or clinical guideline statements cannot be assessed isthe absence of information on chorionicity and amnionicity in the data. This information is not routinely collected in maternity datasets. Challenges were also identified in classifying caesarean section into categories according to whether the procedure was planned or the procedure was urgent or an emergency, particularly in the context of risk of spontaneous preterm labour in multiple pregnancies in women with planned caesarean birth. It is also not possible to assess provision of specialist services for twin babies with fetal complications (e.g. twin-to-twin transfusion syndrome) because these diagnoses and related therapeutic procedures are poorly recorded in the data. Assessing maternity care for women with multiple births: feasibility study viii Case mix adjustment using standard NMPA methods is more complex for women with twin births, compared with those with singleton births. For each pregnancy, a choice has to be made which of the two birthweights need to be included in the case mix adjustment. There is also a small number of women affected by less common comorbidities (e.g. hypertension) and antenatal complications (e.g. placenta praevia), usually included in the established NMPA adjustment method. A study of variation in measures of maternity care between NHS trusts or boards, or between hospital sites, is only possible for measures where the outcome is common (e.g. prelabour caesarean birth). When relevant features of care or outcomes are rare, maternity and neonatal care can only be assessed at regional or national level. For the evaluation of maternity and neonatal care that is specific to those babies admitted to a neonatal unit, successful linkage of NMPA maternity data with the NNRD was slightly lower for twin births before 32+0 weeks of gestation than the existing linkage of singleton neonates. For example, the linkage rate at 30+0 to 31+6 weeks of gestation was 87.7% for liveborn twins compared with 94.9% for all liveborn babies. This was particularly noted at gestations less than 28+0 weeks. The most likely explanation for this lower linkage rate is less complete and maybe less accurate data entry, including possible errors or omissions with neonatal NHS numbers. It is possible to assess the availability of specialist services at the level of NHS trust or board, or hospital site, for women giving birth following multiple pregnancy. However, this can currently only be studied according to the place of birth, as information on where antenatal care was received is not available. It should be noted that a similar problem exists for singleton births. Recommendations R1 Maternity service providers should consider the local reasons for inaccuracies in the recording of ‘number of infants’ at birth and work to correct these by the end of the 2020/21 reporting year. This might require auditing local data, mandating the ‘number of infants’ data item and checking data download reports for national datasets to ensure that ‘birth order’ has not been mislabelled as ‘number of infants’. R2 Maternity service providers and national organisations responsible for collating and managing maternity datasets should request/record data on the number of fetuses in the first trimester of pregnancy, in addition to number at birth, for women with multiple pregnancy, and should plan to be compliant with this for the next version of the national data specification. R3 Maternity service providers and national organisations responsible for collating and managing maternity datasets should make chorionicity and amnionicity a compulsory data item in maternity information systems and national datasets for women with multiple pregnancy. This should be implemented in the next version of the national data specification. R4 Maternity service providers who offer specialist fetal procedures, such as intrauterine fetal laser therapy, should work with their coding departments to ensure that the fetal complications and procedures are properly coded into HES, SMR and PEDW by the end of the 2020/21 reporting year. R5 Maternity service providers and national organisations responsible for collating and managing maternity datasets should work to include a compulsory field on planned mode of birth, to enable distinction between those women who have an urgent caesarean birth following labour onset for new clinical reasons and those who have planned caesarean birth. This should be implemented in the next version of the national data specification. R6 Maternity service providers should put local systems in place by the end of the 2020/21 reporting year to ensure that the NHS number for every newborn baby is stored in the maternity information system and linked to the mother’s number. Particular care must be taken to ensure that the baby’s NHS number is not linked to the baby record of the other twin.