Frequently Asked Questions and indicator definitions

Frequently asked questions

Why is there missing information for a result in an NHS Hospital?
Either there were too few births to produce a reliable statistic (<30 births in denominator, or <5 events); or the data for that measure was poorly recorded (<70% complete); or the result was implausibly high or low, indicating a data entry / data quality issue; or all results for the Hospital were excluded because there were <10 births recorded in any one month.

If an NHS Hospital has higher or lower results for a maternity outcome than expected, does it mean their maternity care is better or worse than other hospitals?
Not necessarily. The characteristics of service users such as their age and pre-existing conditions (the service user 'case mix'), and their care preferences, can also affect the results. Rapid Quarterly Reporting doesn't currently adjust for these factors, nor is it possible to completely account for all of them. Additionally, while high percentages of outcomes like 3rd/4th degree tears are bad, a very low percentage may indicate underdiagnosis or underreporting which is also a concern. For other outcomes, like elective caesarean section, there is no 'ideal' percentage and a higher or lower result doesn't indicate better or worse performance.

How should I use Rapid Quarterly Reporting results when considering using maternity services in an NHS Hospital?
In general, lower percentages of negative outcomes may indicate a hospital is performing well, and higher percentages may indicate a problem. However, as other factors like the characteristics of service users can also affect the results, there is some uncertainty. To further investigate the care provided by an NHS Hospital, you can read about safety and patient experience there on the Care Quality Commission website. You may also wish to prepare specific questions about the care you will receive, including how complications are handled and your treatment options, to discuss at your antenatal appointment.

How is it decided that a result is higher or lower than expected?
A result is categorised as higher or lower than expected if there is statistical evidence it is different to the national average. This happens when the difference is large enough that there is only a 5% (or less) chance of getting a result so extreme. Whether evidence is found also depends on how many births are analysed: with more births, smaller differences from the national average can be reliably detected as statistically significant at the 5% level. Another way of saying this is that we’re better able to compare outcomes and practices from bigger hospitals to the national average, because we have more data for the comparison.

Can I use Rapid Quarterly Reporting data in quality improvement projects, research, or evaluations? ?
Yes, the results can be cited as 'Rapid Quarterly Reporting, National Maternity and Perinatal Audit, Royal College of Obstetricians and Gynaecologists'.

Why are data not available in ‘real time’?
Some time is needed to clean, check, and analyse the data before it is published, and updates only happen quarterly (once every 3 months).

Where can I find information on other maternity outcomes?
The NMPA Clinical Audit has information on a wider variety of measures at NHS Hospital level than reported here (see Table 1 below), though the data is less recent. The Clinical Audit also includes results for Scotland and Wales as well as England. Other sources of information on maternity and neonatal outcomes include the Care Quality Commission website, the NHS Maternity Dashboard and the National Perinatal Epidemiology Unit.

Does the NMPA Quarterly Reporting include information for Wales and Scotland?
It does not currently, as the datasets required are not as immediately available to the NMPA as Hospital Episodes Statistics. The NMPA Clinical Audit does however have results for these countries, and data from them may be added to Quarterly Reporting in the future.

How is the Quarterly Reporting data different from data in the NMPA national Clinical Audit?
NMPA Rapid Quarterly Reporting uses a database of routinely collected NHS hospital admission records called Hospital Episodes Statistics (HES). By contrast, the NMPA Clinical Audit uses both HES data and data from a specialised maternity database called the Maternity Services Data Set (MSDS). Both datasets are valid for analysing maternity outcomes. However, HES data can be processed more quickly, whilst the MSDS captures a wider range of outcomes, so these data sources are used by the NMPA for different purposes. The below table compares the measures reported in the most recent NMPA Clinical Audit Report and the Rapid Quarterly Reporting:

Table 1. Comparison of measures included in the most recent Clinical Audit report, and the Rapid Quarterly Reporting.


Measure
Clinical Audit Report 2021
Rapid Quarterly Reporting

Stillbirth


Baby born small-for-gestational age


Preterm birth


3rd/4th degree tear


Elective caesarean section


Unassisted vaginal birth (without forceps or vacuum)


Assisted vaginal birth (with forceps or vacuum)


Emergency caesarean section


Induction of labour


Quality of data on gestational age, fetus outcome (birth status), and birthweight


Birth without intervention(spontaneous labour onset with no caesarean section, forceps or vacuum, episiotomy, or anaesthesia)


Vaginal birth after caesarean birth


Smoking cessation


Episiotomy


Postpartum haemorrhage of 1500ml or more


Unplanned maternal readmission


Skin-to-skin contact within 1 hour of birth


Breast milk at first feed and discharge


5 minute Apgar score of less than 7

 

Indicator definitions

NMPA Rapid Quarterly Reporting uses Hospital Episodes Statistics (HES) inpatient admissions data. HES collects a detailed record of each episode of admitted patient care delivered by the NHS in England. Diagnoses for each patient are recorded using the International Classification of Diseases, 10th edition (ICD-10). Procedures performed during an admission episode are coded using the Office of Population, Censuses and Surveys Classification of Surgical Operations and Procedures, 4th revision (OPCS-4). Each maternity episode also contains specific information regarding the labour and delivery (e.g. parity, mode of delivery, gestational age, birthweight) in supplementary data fields known as the HES ‘maternity tail’. Together, all these data are used to calculate the indicators in NMPA Rapid Quarterly Reporting. The technical details for each are presented below, where the denominator is the count of patients who could have had the indicator, and the numerator is the count of patients who actually did.

Proportion of still births 
Measure Description Technical specification
Denominator All singleton births Singleton births are defined as birth records without an ICD code for a multiple birth (Z37.2–7) AND without strong evidence of a multiple birth in the maternity tail (>1 birthweight [birweit], birth order [birord] AND >1 in the number of babies [numbaby] field). 
Numerator Number of stillbirths Stillbirths are defined by an ICD-10 code (Z37.1) OR, in providers with over 95% data completeness, by birth status field (birstat_1=2,3,4). 
Proportion of small-for-gestational-age (SGA) births 
Measure Description Technical specification
Denominator All singleton births with gestational age and birthweight and sex recorded Singleton births with gestat_1, sexbaby_1, birweit_1 not missing
Numerator Number of babies born small for gestational age (<10th centile) SGA is defined as less than the 10th birthweight centile using the WHO-UK 1990 charts. Birthweight centiles are calculated using birthweight (birweit_1), sex (sexbaby_1) and gestational age (gestat_1) fields in HES maternity tail
Proportion of preterm births 
Measure Description Technical specification
Denominator All singleton births with gestational age recorded Singleton births with gestat_1 not missing
Numerator Number of preterm births Preterm birth is defined using gestational age field in the HES maternity tail (gestat_1<37)
Proportion of vaginal births with a third- or fourth-degree tear 
Measure Description Technical specification
Denominator All singleton vaginal births  A vaginal birth is defined as any record in which a vaginal mode of birth is recorded in the procedure fields (OPCS-4 codes: R19–R24). 
Numerator Number of women with a third- or fourth-degree perineal tear  A tear is defined by the presence of an ICD-10 code for a third- or fourth-degree tear (O70.2; O70.3) or an OPCS-4 procedure code for repair of a third- or fourth-degree tear (R32.2; R32.5) 
Proportion of women who had an elective caesarean birth 
Measure Description Technical specification
Denominator All singleton births  Singleton births are defined as birth records without an ICD code for a multiple birth (Z37.2–7) AND without strong evidence of a multiple birth in the maternity tail (>1 birthweight [birweit], birth order [birord] AND >1 in the number of babies [numbaby] field).
Numerator Number of elective caesarean sections (ELCS)  ELCS is defined using OPCS-4 code R17
Proportion of women who had an unassisted vaginal birth
Measure Description Technical specification
Denominator All singleton births excluding elective caesarean sections Singleton births excluding ELCS defined by OPCS-4 code R17
Numerator Number of unassisted vaginal births Unassisted birth is defined using OPCS-4 codes R23 or R24, and also includes breech deliveries (R19 and R20)
Proportion of women who had an assisted vaginal birth
Measure Description Technical specification
Denominator All singleton births excluding elective caesarean sections Singleton births excluding ELCS defined by OPCS-4 code R17
Numerator Number of assisted vaginal births Assisted birth is defined using OPCS-4 codes R21 (forceps delivery) or R22 (vacuum delivery)
Proportion of women who had an emergency caesarean birth 
Measure Description Technical specification
Denominator All singleton births excluding elective caesarean sections Singleton births excluding ELCS defined by OPCS-4 code R17
Numerator Number of emergency caesarean sections (EMCS)  EMCS is defined using OPCS-4 codes R18/R25.1
Proportion of women with induced labour 
Measure Description Technical specification
Denominator All singleton births with delivery onset recorded excluding elective caesarean sections Singleton births with delonset not missing, excluding ELCS defined by OPCS-4 code R17
Numerator Number of induced labours Induction is defined using the delivery onset field (delonset=3, 4, 5) from the HES maternity tail. Failed induction (ICD-10 code O61) is also included in the numerator.
Proportion of women with recorded gestational age, fetus outcome, and birthweight 
Measure Description Technical specification
Denominator All singleton births  Singleton births are defined as birth records without an ICD code for a multiple birth (Z37.2–7) AND without strong evidence of a multiple birth in the maternity tail (>1 birthweight [birweit], birth order [birord] AND >1 in the number of babies [numbaby] field).
Numerator Number of women with recorded gestational age, fetus outcome, and birthweight Requires valid (non-missing and also not taking values indicating unknown or not recorded data e.g. 9, 99, 9999) recorded values in the gestat_1, birstat_1­, and birweit_1 fields