The ALSPAC fetal and neonatal resource: detailed data abstracted from the clinical records of the new-born [version 1; peer review: awaiting peer review]

In a previous Data Note, we outlined the data obtained from clinical obstetric records concerning many details of the pregnancies resulting in the births of the children in the Avon Longitudinal Study of Parents and Children (ALSPAC). Here we describe the data that have been abstracted from medical records concerning the fetus and neonate. Full details concerning the selection biases regarding the data abstracted are outlined in the previous Data Note. The records that have been abstracted, and described in this Data Note, concern the health of the fetus (measured in relation to the results of fetal monitoring, presentation at various stages of pregnancy, and the method of delivery) as well as the status of the newborn immediately post-delivery. Details of signs, symptoms and treatments of this population of new-born babies, as recorded in the clinical records, are described for the time during which they were in hospital or under the care of a designated midwife. These data add depth to the information collected from elsewhere concerning this period of the child’s life: from the questionnaires completed at the time by the mother; and clinical details from neonatal intensive or special care units which will be detailed in a further Data Note.


Introduction
The UK's large influential National Perinatal Mortality Survey of 1958 identified fetal asphyxia as responsible for almost half of the 35 perinatal deaths per 1000 births occurring at that time (Butler & Alberman, 1969;Butler & Bonham, 1963). By the 1990s this rate had fallen dramatically to <1 per 1000 births (Mori et al., 2008). This improvement was assumed to be largely as the result of advances in the monitoring of the fetus during late pregnancy as well as more efficient methods of resuscitation. However, the consequences of these interventions (for example, inducing or augmenting labour; delivery by caesarean section [CS]; vaginal delivery using forceps or vacuum techniques; and vigorous methods of resuscitation) may have had long-term effects on the developing child. Similar questions concerning long-term effects can be asked of other problems experienced by the newborn, including the degree of jaundice, the duration of phototherapy or the exposure to antibiotics or other medications.
One of the original aims of the Avon Longitudinal Study of Parents and Children (ALSPAC) was to determine possible effects of early exposures on later health and development in childhood, adolescence and throughout adulthood (Golding et al., 2001). The prevalence of many outcomes has changed over time, the extent of which has not been fully explained. These include increases in chronic childhood disorders such as diabetes (e.g., Patterson et al., 2012), autism spectrum disorder (e.g., Rosenberg et al., 2009) and obesity (Wang & Lim, 2012) as well as in maternal prenatal depression (e.g., Pearson et al., 2018).
Very few longitudinal studies prior to ALSPAC had collected detailed in-depth information to allow epidemiological analyses to test whether details of fetal or neonatal exposures may be associated with these and other outcomes. Therefore, we have described here the information covering the health of the fetus that has been abstracted from obstetric records and the treatment and health of the neonate that has been abstracted from paediatric records. The data described in this Data Note differ from that in a published Data Note which describes the mother during pregnancy and the puerperium (Birmingham et al., 2021a; subsequently referred to as the Obstetric Data Note). Here we describe the data based on whether there could be different answers for different members of a multiple pregnancy. If so, the data is described here (and is therefore primarily child based). Examples concern the presentation of the fetus, fetal monitoring, and method of delivery.

Methods
As already indicated, ALSPAC was originally designed to determine the ways in which aspects of the environment (possibly interacting with the individual's genes) influenced child health and development (Fraser et al., 2013). The Study enrolled pregnant women who had an expected date of delivery between 1 st April 1991 and 31 st December 1992. The women had to be resident in that part of the old administrative county of Avon in south-west England comprising three District Health Authorities (Southmead, Frenchay, and Bristol and Weston) (Boyd et al., 2013). Data have been collected using a variety of methods including questionnaires completed during pregnancy by the pregnant woman and her partner, abstraction of details from medical records, accurate measuring of the offspring and collection and analysis of biological samples including cord blood.
All women resident in the area at the time they were pregnant were eligible, provided that their expected date of delivery lay between 1 st April 1991 and 31 st December 1992. 14,541 pregnant women resident in the area were recruited into ALSPAC. From these pregnancies, there were a total of 14,676 fetuses and 14,062 live births. Of these children, 13,988 were still alive at one year of age. Mothers were considered enrolled if they had returned at least one questionnaire or attended a "Children in Focus" clinic by 19 th July 1999.
Most of the deliveries took place in either Southmead Hospital (53%), Bristol Maternity Hospital (now known as St Michael's Hospital) (38%) or Weston General Hospital (4%). A few deliveries took place at the mother's home (2%) in a hospital out of the area due to the mother unexpectedly going into labour while, for example, on holiday or very occasionally when in transit to the hospital (2%).
In regard to the obstetric and neonatal records, a shortage of funding resulted in only slightly under two-thirds of the original Study sample having had data abstracted from medical records to date. The data abstraction form, abstraction instructions and checking instructions include data relating to the fetus, the newborn immediately after delivery and signs, symptoms and treatments during the first weeks of life (see Extended data). The choice of which records were abstracted are described in more detail in the accompanying Obstetric Data Note (Birmingham et al., 2021a). That paper includes details of the likely biases incurred in analysing the data, and the possible methods of analysis. The sample includes almost all the caesarean sections, the instrumental vaginal deliveries, the pre-term deliveries, the multiple births and the fetal and neonatal deaths. There has always been the intention to complete this data extraction to include the whole of the ALSPAC enrolled population but that awaits further funding. It should be noted that extra data are available on the infants admitted to intensive or special care within the neonatal period; the detailed records are available for analysis and will be documented in a further Data Note.  (Birmingham, 2018). Ethical approval was less formal at that time, with the LRECs only recently established.

Ethics approval
The data collected are governed by strict ethical criteria (see Birmingham, 2018) to ensure that no personal identifying information is revealed. Nevertheless, within these ethical strictures the Study encourages access to the data by bona fide scientists. Please note that the Study website contains details of all the data that is available through a fully searchable data dictionary and variable search tool, and a detailed proposal form for access to specified data.

The variable numbering system
The variable numbers for most of this data set start with either the letters 'DEL_P' or 'DEL_B' followed by a number. The distinction relates to whether the variable refers to the fetus up until the birth (DEL_P), or the baby after delivery (DEL_B). For simplicity these will be known as the P (pregnancy) and B (baby) numbers throughout this paper. In addition, the question number is quoted -i.e. the actual question asked on the data abstraction form (Extended data (Birmingham et al., 2021b)).

Data available
Due to funding restrictions, to date only 8369 pregnancies have had detailed data abstraction using the proforma shown in the data abstraction form (Extended data (Birmingham et al., 2021b)). Details of the case selection and the possible biases generated are detailed in the Obstetric Data Note (Birmingham et al., 2021a).
The fetus in distress prior to delivery Monitoring the fetal heart rate. In all, there were five different types of monitoring used on the pregnant women during labour (Table 1a), the most common being the use of a cardiotocography (CTG) monitor, continuously or intermittently. Continuous CTG monitoring does not allow the mother to move freely or change position hence the common use of intermittent CTG monitoring. A small number of records (n=38) described a different type or method of monitoring. However, for about a third of the pregnancies (n=2539) the method of monitoring was not recorded. It is likely that most of these mothers would have had intermittent CTG as the hospital protocols clearly state that normal or 'low risk' pregnancies should have intermittent external CTG monitoring performed on admission to the Delivery Suite and for approximately 20 minutes every 2-3 hours thereafter.

Abnormalities of the fetal heart rate.
Of the fetal heart rates monitored, a total of 11 different fetal heart rate patterns or abnormalities were recorded in each of the first and second stages of labour (Table 1b, Table 1c, Table 1d). There were a small number of additional fetuses for which heart rate abnormalities were recorded but it was not clear from the records as to whether they had occurred in the first or second stage of labour (data not shown); variables have been derived to indicate whether the abnormal heart rate had occurred during labour, by combining the first and second stage occurrences with those where the timing was unknown (Table 1d).

Other signs of fetal compromise.
Apart from the heart rate abnormalities, other signs of fetal distress were recorded including whether intrauterine growth restriction (IUGR) was suspected, whether meconium was seen in the liquor (and whether this was old or new), and whether an abnormal fetal blood pH had been recorded. For those with a low pH, the level is available together with the time from that level to delivery of the baby (Table 1e).
The fetus during delivery Position of the fetus. The presentation of the fetus had been recorded during pregnancy, at the start of labour and at delivery (Table 2a). During pregnancy, the presentation was recorded on several occasions often as the result of an ultrasound      examination. Consequently, there are separate variables concerning whether the baby was in a potentially problematic position prior to the onset of labour (i.e., breech, transverse lie, or an unstable lie). The actual presentation at the start of labour is denoted by one variable (vertex, breech or 'other' -the latter being described in text). The fifth variable indicates the actual position at the time of delivery (whether vaginal or by CS).

Method of delivery.
The method of delivery is described by one variable (DEL_P1210), distinguishing between assisted breech and breech extraction; assisted vaginal birth using forceps or vacuum extraction; CS, and spontaneous delivery. Further variables classify the types of forceps used and whether the CS was elective or emergency (Table 2b).
The cord and placenta. Although the number of babies being born with their umbilical cord around their neck was relatively common (1933 births; 23%), the more dangerous cord prolapse only occurred on 24 occasions. The delivery of the placenta (or placentae in the case of most multiple births) often incurred problems of retained placenta, with consequent manual removal (Table 2c). The actual length of the third stage of labour ranged from immediately to 275 minutes (>4 hours) as shown in Table 15 of the Obstetric Data Note (Birmingham et al., 2021a).
Most of the placentae were retained by ALSPAC if the birth took place in either of the two Bristol-based major maternity hospitals, provided the mother did not object. This involved placing the placenta immediately post-delivery into formalin in a container supplied especially for the purpose. There was rarely any standardisation at the time as to whether the membranes were retained, but a length of umbilical cord was cut and frozen at -20°C separately. Weight of the placenta was not standardised in any way and was available for only 3208 of the births in this sample. Subsequent examinations of some of the placentae stored in formalin have produced measurements using a standard procedure (see Holroyd et al., 2016).

The baby at birth
Outcomes of pregnancy including details of stillbirths and other deaths and measurements of the newborn will be included in a further Data Note (in preparation). See also Treatments and procedures below. Other baseline characteristics of the pregnancy and child are available in the ALSPAC resource with the variable names shown in Box 1.

Alive at 28 days Kz011a
Alive at 1 year Kz011b

Birthweight Kz030
Condition at birth. Three different indicators of asphyxia at birth were used, including whether the baby: cried immediately; time taken before first breath (<1 minute, 1-3 minutes or >3 minutes); or was resuscitated. In addition, non-binary measurements were recorded of the time taken to establish regular respirations, and the Apgar scores at 1 and 5 minutes.
Although it was relatively unusual for hospitals to record whether the baby cried immediately (26%), the other indicators were recorded in at least 85% of births (Table 3a). Table 3b and Table 3c. Of the 8181individuals with detailed information 67.5% (n=5521) had no resuscitation or other treatments or investigations at, or shortly after, delivery.

Treatments at birth. Methods of resuscitation and other treatments/investigations are shown in
The neonatal period Place and care of neonate. Table 4a indicates that approximately 10% of newborns were transferred immediately after delivery to a Special Care Baby Unit (SCBU) or to the transitional care ward in St Michael's Hospital (Ward 76). This is a ward where babies can be cared for alongside their mothers who also remain in hospital. The newborns may have needed extra observations or help to feed particularly if they had been born prematurely or weighed less than 2.5 kg at birth. The duration of these admissions (if more than 24 hours) was also recorded. Those admitted to SCBU, or Neonatal Intensive Care have had detailed data collected covering their admission which will be described in a further Data Note. Nearly all the babies (98%) were discharged to their mother, with approximately 5% being re-admitted before their six-week postnatal check.
The paediatric examination of the newborns included an assessment of gestation, the mean number of weeks in this cohort is estimated to be 39.5. Abnormalities of the babies' hips were noted in just under 5% (Table 4b).
Signs and symptoms. Signs and symptoms from minor to serious were noted in the babies' first 14 days and are documented in Table 4c. The most common being jaundice (56%), 'unsettled' (38%) and pyrexia (28%). Other conditions that were noted were: apnoeic attacks, cyanotic attacks, high pitched or abnormal cry and sticky or moist eyes. The number of Although the placentae were often weighed in the delivery room, there was no conformity as to whether or not the membranes or cord were included. We therefore recommend that the weight of the placenta should not be used unless measured under controlled conditions by the ALSPAC path team (see kz033); b described in text, data not currently available.; c in vaginal deliveries.   babies recorded as suffering from convulsions, umbilical infection or 'twitching' were in single figures only. Only 293 (4%) of the babies had no problems at all recorded. Table 4d, 69% of neonates were breast fed including those that were both breast and bottle fed with 11% recorded as having difficulties with feeding.

Feeding and nutrition. As indicated in
Vitamin K. Table 4e shows that Vitamin K was administered to 67% of the neonates although it is likely to have been considerably more as 2463 (32%) had no indication in the medical records that it had been administered. It is known that in one hospital, staff would document on the babies' name cards that this vitamin had been given. These cards, which were usually attached to the mothers' bed or babies' cots, were     frequently taken home by the mother, or discarded, but not filed in the medical records. Of those that were documented in the records, 68% were given the vitamin orally and 30% intramuscularly.
Protocols from the two main hospitals state that all babies should be given the vitamin: 1mg orally for full-term normal deliveries and 0.5 mg intramuscularly for others. The hospitals differed on where the vitamin should be administered (i.e., delivery suite or ward) (see Box 2 Management Guidelines and Midwifery Operational Policies). These protocols can be found in the ALSPAC Archive in the University of Bristol Library (Special Collections Archive Box 784).

KONAKION (PHYTOMENADIONE/VITAMIN K)
All infants should receive Vitamin K orally or IM after birth with maternal consent. The dose (

KONAKION (PHYTOMENADIONE/VITAMIN K)
All babies should be given Vitamin K soon after delivery. This must be recorded in the appropriate column of page 2 of the baby notes with date, time and route of administration.
All normal term infants to be given: The following babies should be given:  The article linking IM administration of Vitamin K with childhood cancer (Golding et al., 1992) was not published until August 1992 and would not have influenced the route of Vitamin K administration for this cohort as policy change was slow due to considerable controversy. However, oral administration had long been used in some Bristol units.

Treatments and procedures.
The neonates that had some form of treatment or investigation were 78% as can be seen in Table 4f. Although most of the types of treatment are described in text, 14% were nursed in an incubator and information on numbers using cot shields or light meters or receiving phototherapy is documented in this table also. 59% of the neonates are known to have had no drugs given while in the hospital. The mother documented the drugs given to the neonate at home in the paper questionnaire sent to her four weeks after the baby was born. These questionnaires were not sent to mothers if it was known that the baby had died or was seriously ill. Details of treatments to babies admitted to SCBU or Neonatal Intensive Care will be outlined elsewhere (in preparation).

Strengths and limitations
There are four major strengths of these data. Firstly, each item was abstracted from the paper medical record with a strict protocol and meticulous checking; secondly, the data collected were documented at the time so that there was no element of retrospective recall; thirdly, these data can be augmented by information from the mothers' self-completion questionnaires; and fourthly, the data provide an important baseline from which to assess the long-term benefits and possible hazards of the various facets of care.
There is one major limitation of the data -many aspects of fetal exposures, neonatal conditions and treatments are missing for over 5000 ALSPAC newborns. Admittedly, by the selection criteria used on the 8369, the majority of the more complex cases have already been abstracted, but for valid epidemiological analysis the population of all the others are also needed. It is hoped that efforts can be made in the future to fill this important gap.

Data availability
Underlying data ALSPAC data access is through a system of managed open access. The steps below highlight how to apply for access to the data included in this Data Note and all other ALSPAC data: 1. Please read the ALSPAC access policy which describes the process of accessing the data and samples in detail, and outlines the costs associated with doing so.
2. You may also find it useful to browse our fully searchable research proposal database which lists all research projects that have been approved since April 2011.
3. Please submit your research proposal for consideration by the ALSPAC Executive Committee. You will receive a response within 10 working days to advise you whether your proposal has been approved.
If you have any questions about accessing data, please email alspac-data@bristol.ac.uk.
The Study website also contains details of all the data that is available through a fully searchable data dictionary.

Extended data
Although the abstraction form, instructions and checking instructions are labelled "ALSPAC Mother during pregnancy and the pueperium", these documents also cover data abstraction relating to the fetus, the new-born immediately after delivery and signs, symptoms and treatments during the first weeks of life.  (Birmingham et al., 2021d).
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

Consent
Consent to abstract data from medical records was obtained on an 'opt out' basis which was acceptable to the LRECs at that time. The Study Mothers were informed in the initial information booklet that their medical records would be accessed.