Implications of time of day and day of the week for the outcome of birth

One of our team had been involved in research from the 1970s onwards which showed that the birth of babies was spread unevenly through the days of the week. Our previous project, known as 'Birth Timing', which analysed over seven million births in England and Wales from 2005-2014, was the first national study of both the day and time of births and this new proposal will build on this work.

Our research team linked three large sets of routinely collected data to combine mother and baby records from three sources. These were birth registration, birth notification when NHS numbers are allocated to babies, and records about maternity care during labour and birth. A database was created to enable us to analyse the data.

Our results showed a regular weekly cycle with numbers of births increasing each day from Mondays to Fridays, lower numbers of births on Saturdays and lowest weekly numbers of births on Sundays. Fewest births occurred at Christmas. Births also varied seasonally, with a peak in late September. When labour involved no medical assistance, numbers of births were highest from midnight to 7am, peaking from 4am to 6am. They were lowest during the day and the hours up to midnight. Planned caesarean births were usually on weekday mornings. Births after induced labour were more likely to occur around midnight on Tuesdays to Saturdays, whether or not any medical assistance was needed, with a peak before public holidays.

As the variation in the timing of births by place and type of birth has implications for midwifery and medical staffing and for the organisation of maternity care more generally, further research is needed to use the new database to investigate the implications. We therefore propose research to investigate whether death rates at birth of children and their mothers and varies according to the time of day and day of the week. In addition, we hope to analyse variations in the rates of serious complications. We plan to move beyond the simple subdivision into time inside and outside 'normal working hours' and analyse outcomes at different times of day and night. Asthere are signs that some births are brought forward so that they occur before weekends and public holidays, we plan to analyse the implications of these selection processes.

It is also known that rates of planned caesareans and induced labours vary widely between maternity units and the extent of these variations is not explained by differences in the groups of women giving birth. The suggestion is that these differences reflect differences in obstetricians' and midwives' policies. We therefore plan to see how times of birth vary between units and whether they appear to be associated with these differences in policies.

Who will benefit from this research?

There are two categories of potential beneficiaries from this research. The first is professionals, both clinicians working in the maternity services and commissioners who need to plan and pay for maternity provision. It is noteworthy that this area of research was first suggested to us by the Royal College of Obstetricians and the Royal College of Midwives and both organisations provided letters of support when we applied for funds for our previous project 'Births and their outcomes by time, day and year: a retrospective birth cohort data linkage study', to undertake the data linkage and analysis.

This research is equally relevant to women using maternity services. In the consultations we had with service users during our previous project, 'Births and their outcomes by time, day and year: a retrospective birth cohort data linkage study' we were told that press reports both of published research and also of speculations not based on evidence have led to anxiety among women going into labour at night and at weekends.

How will they benefit from this research?

The overall patterns of births, their possible associations with mortality and the extent to which these patterns are likely to vary between maternity units are likely to be important information for people responsible for managing midwifery staffing and availability of obstetric care at appropriate levels when it is needed. Although it will not, of course, enable quantification of staffing levels, it should provide additional insights to inform decision making.

For service users, the research should provide more robust information. If it points to specific factors, it should enable service user representatives in making their case to the services about specific changes which could lead to improvements in the availability of high quality maternity care irrespective of the time of day or night.