NMPA methods
Using the NMPA results
Information governance and data access
About the NMPA switch to the Maternity Services Data Set (MSDS) - England only
About the NMPA
What is the National Maternity and Perinatal Audit?
The National Maternity and Perinatal Audit (NMPA) is a large scale audit of the NHS maternity services across England, Scotland and Wales. It began in July 2016 and was initially commissioned for three years. Funding has been extended until July 2021.
Using high quality data, the audit aims to evaluate a range of care processes and outcomes in order to identify good practice and areas for improvement in the care of women and babies. The audit consists of three elements:
- an organisational survey of maternity and neonatal care provision, and services available to women
- an annual clinical audit of a number of key measures to identify unexpected variation between maternity services
- a programme of periodic ‘sprint’ audits on specific topics
What is the purpose of the audit?
The aim of the audit is to provide relevant and comprehensive information about maternity and neonatal services provided by the NHS in England, Scotland and Wales. This information will allow clinicians, NHS managers, commissioners and women to compare and evaluate services, and can be used to inform care quality improvements.
We have developed an interactive online facility whereby maternity service providers and commissioners can benchmark the care provided by their service against other, similar services, against regional and national averages, or against local or national standards.
Together, these outputs from the audit will allow healthcare professionals, NHS managers, commissioners and policy makers to examine the extent to which current practice meets guidelines and standards and to identify areas for improvement.
Who is responsible for the the NMPA and how is the audit funded?
Commissioned by the Healthcare Quality Improvement Partnership (HQIP), the audit is led by the Royal College of Obstetricians and Gynaecologists in partnership with the Royal College of Midwives, the Royal College of Paediatrics and Child Health and the London School of Hygiene and Tropical Medicine. The NMPA is commissioned as part of HQIP’s National Clinical Audit and Patient Outcomes Programme (NCAPOP) on behalf of NHS England, the Welsh Government and the Health Department of the Scottish Government.
Is participation mandatory?
Trusts are required to participate and report on all National Clinical Audit and Patient Outcomes Programme projects as part of their NHS Standard contract. All NHS maternity units in England, Scotland and Wales are therefore expected to participate in the NMPA. This wide participation will give a rich overview of the services and will allow meaningful comparisons to be made.
Are women using the maternity services involved and can they access the audit results?
A number of organisations representing women using maternity services have been involved in the design and implementation of the audit and will continue to be consulted on the running of the audit and its priorities. Women who have used the maternity services are also consulted via our Women and Families Involvement Group.
Women using the maternity services and the general public can access all NMPA reports and results via the NMPA website (www.maternityaudit.org.uk).
You can download our information leaflets for women whose data are used in the NMPA here.
Where does the NMPA fit in with other HQIP-funded projects, such as the National Neonatal Audit Programme (NNAP) and MBRRACE-UK?
The NMPA works closely with both NNAP and MBRRACE, who are represented on our Clinical Reference Group.
NNAP uses the Neonatal Research Database, derived by the Neonatal Data Analysis Unit at Imperial College from the BadgerNet system which is used to record care of babies admitted to almost all neonatal units in the UK. From this it establishes key measures of neonatal care in the UK. We are linking the NMPA dataset with the same neonatal dataset to find out more about what factors in pregnancy or during birth might cause babies to need neonatal care.
MBRRACE uses a different methodology; instead of using routinely collected data, clinicians submit detailed information directly about cases meeting the MBRRACE eligibility criteria. For example, if a mother dies in the UK, the clinicians looking after her make a phone call to the MBRRACE team, and subsequently follow a case reporting process. This is an excellent way of looking at relatively rare events in detail. The NMPA looks instead at routinely collected data, which is collected about every birth, and is thus able to report on more common events, such as caesarean section or induction of labour, for which a case reporting process would be impractical.
NMPA Methods
How are the data collected?
-
Clinical audit
It will not be necessary for the maternity services to collect any additional clinical data specifically for the clinical audit as we make use of data which are already collected routinely as part of women’s and babies’ care.
- For England, data for 2017/18 births onwards will now come from the Maternity Services Data Set (MSDS), with the MSDS data supplied directly to the NMPA by NHS Digital (for 2015/16 and 2016/17, English trusts provided a data extract from their local maternity record system directly to the NMPA).
- For Wales, the NHS Wales Informatics Service (NWIS) supplies the NMPA with data from the Maternity Indicators data set (MIds), linked to the Patient Episode Database for Wales (PEDW).
- For Scotland, the Information Services Division (ISD Scotland) provides data from the Scottish Birth Record (SBR) and Scottish Morbidity Record (SMR02) datasets.
-
Sprint audits
Bespoke data collection is not required for the topic-specific sprint audits, which make extensive use of data linkage to combine existing data sources.
-
Organisational Survey
As part of the audit, an organisational survey of NHS Trusts and Health Boards was conducted early in 2017 and 2019. The organisational surveys collect information on service delivery and the organisation of maternity care and contribute to a better understanding of the care provided to pregnant women.
How were the measures for the report selected?
The selection of measures was guided by a panel of clinical and academic experts, including obstetricians, midwives, statisticians and health service researchers, as well as the NMPA Women and Families Involvement Group and organisations representing maternity and neonatal service users.
Where can I get more detail about the measures?
Information about the exact definitions of the NMPA measures, and about the data sources, fields and codes used, can be found in the NMPA Measures Technical Specification.
What is case mix adjustment?
Case mix adjustment (or risk adjustment) is a statistical process to take into account the characteristics of the women a maternity service cares for, such as age, ethnic background, BMI and socio-economic status. This makes the results more comparable between different services. Despite this, differences can still remain if some conditions are not well recorded, due to other data quality issues, or due to differences in the care provided.
All NMPA measures are adjusted for case mix except smoking cessation, skin to skin contact and babies given breast milk.
The NMPA Measures Technical Specification provides details on the case mix factors used.
Why is there missing information?
Not every trust was able to provide all the necessary information for every measure, and a small number of NHS trusts in England were unable to provide any data. The majority of trusts and boards failed the NMPA data quality checks for at least one measure, and data quality remains a national issue. Further work and investment is required to increase data quality and completeness, both at the point where data are recorded, and in the flow from local to central datasets.
Why was the 2017 clinical report revised and what has changed?
Following publication of the first clinical audit report in November 2017, it came to light that some sites were affected by a data quality issue which impacted on results for modes of birth, VBAC, labour induction and obstetric haemorrhage. This prompted further analysis and the publication of a revised report in March 2018, which can be found here.
To address the data quality issue, the revision included redefinition of the affected measures to include all fetal presentations, as opposed to cephalic only. As a consequence, some previously included sites have been excluded on data quality grounds, while others could now be included. It also means that results for most sites included in the audit changed to some degree, as did national rates. All trusts and boards participating in the audit were given the opportunity to check their results. Further information can be found here.
Does the NMPA cover all of the UK?
Currently, the NMPA includes England, Scotland and Wales (Great Britain). Northern Ireland are not participating in the NMPA at present but the audit may expand its coverage in the future.
What further resources are available for participating maternity services?
For further technical documentation relating to the NMPA, please see our Resources for maternity service providers. If there are any other resources that would be helpful to make available on this website, please contact us.
Using the NMPA results
How can I view the clinical results?
The clinical results can be viewed in several ways:
- View by site or trust/board, with spine charts
- View by measure (at site, trust/board, region/LMS or country level), with funnel or scatter plots, or tables
- Interactive results tables covering all maternity services and all measures, which can be filtered to view those of interest (at site or trust/board level)
The full reports and summaries on the Reports page provide aggregated data at country and national (Great Britain) level.
How can I view the organisational results?
The organisational results can be viewed in several ways:
- Overviews
- a) Summary of information by site or trust
- b) Maps of maternity units and neonatal units across England, Scotland and Wales
- c) Services and facilities available by country
- Detailed comparisons with charts or tables (at site, trust/board, region/LMS, neonatal network or country level)
The full reports and summaries on the Reports page provide aggregated data at country and national (Great Britain) level.
How do I compare between services?
For the clinical results, you can select display of a single location or of all locations in table format; all locations are always displayed on the funnel and scatter plots. When you hover over a point, it tells you which location this is, and the result for this location. Underneath the results table or plot, there is also some contextual information about the selected location.
To compare a limited number of specific locations, use the Interactive results tables.
For the organisational results you can select either all, or up to four locations for display in tables and charts. Maps show all sites, and have pop-ups with site information.
Why can’t I find separate information about labour wards versus alongside midwifery units on the same site?
Where obstetric units and midwifery units are co-located, data are presented at site level because it is currently not possible to reliably determine exact location of birth for all births on sites with a co-located obstetric unit and alongside midwifery unit.
How should I interpret these results?
In order to gain a full understanding of a service, you should look at the whole pattern of their results and explore the relationships between the indicators, rather than focusing on individual results that may stand out as being high or low. Many of the indicators are inter-related (for example, a site with a higher caesarean section rate may have a lower instrumental birth rate) and as such it is important to consider all results together, rather than in isolation. The full reports discuss the variation seen in each group of measures.
How does a funnel plot work and what is normal variation?
Watch a short video introduction to funnel plots.
A funnel plot is a graphical method for comparing the
performance of organisations which takes the size of each organisation into
account. This is important because the amount by which the result of an
individual service may vary from the national mean is influenced by random
fluctuations related to the number of births within the service. Some variation in results is expected in healthcare, even when differences between the services in terms of the characteristics of the women in their care are taken into account as much as possible (case mix adjustment). Funnel plots provide a way to represent this variation and indicate where this is within or outside of expected limits.
The straight central horizontal line is the overall mean. The values within the inner curved dotted lines are within the expected range, given this mean and the sample size of each site or trust/board. 1 in 20 values are expected to fall outside of the inner dotted lines by chance alone. 1 in 500 values are expected to fall outside of the outer lines by chance alone. Larger samples allow greater precision about the expected range.
If the result of a maternity service falls outside of the expected range, this does not necessarily mean that this is due to care quality; it could be (wholly or partly) due to data quality, remaining differences in case mix or organisational issues. However, it is a trigger to look into all the possible reasons for variation outside of the expected range.
The results for smoking cessation, skin to skin contact and babies given breast milk are displayed on scatter plots as data quality is not sufficient to allow comparison of performance between organisations with funnel plots.
What is a potential outlier?
A trust or board is classed as a potential outlier if it has a higher than expected result for one of the three NMPA outlier indicators: blood loss of 1500ml or more, Apgar score of less than 7 at 5 minutes, and third and fourth degree tears. Having a high rate does not necessarily mean that trust or board is providing ‘sub-standard’ care; better detection of tears or blood loss may be a contributing factor.
What is a spine chart?
Spine charts are useful to view results for measures in the context of other relevant measures at a selected site or trust/board, rather than in isolation.
Spine charts show the rates of all measures for the site or trust/board you have selected, and how these compare to the average of all sites or trusts/boards included in each measure. They also give an idea of the overall distributions of rates, and if the rates of the selected site or trust/board fall outside of the expected range. They are essentially slices of the funnel plot for each measure at the point of the site or trust/board you have selected.
Can women use these results to choose which hospital to give birth in?
The NMPA results can be used by women using the maternity services to find out more about rates of events surrounding childbirth, and about the availability of services and facilities in local maternity units. However, decisions about where to give birth should always be made in discussion with a midwife or obstetrician.
Information governance and data access
Who uses details about me and my care for the NMPA?
The NMPA team members who use your information are employees at the Royal College of Obstetricians and Gynaecologists (this organisation is the Data Processor for the audit). Information that could identify you is only accessible to a small number of team members. The information is then anonymised before it is used by the rest of the team. NMPA team members are fully trained so they know how to keep your information secure. The contact details of the NMPA team are given on the Contact Uspage.
Who controls how the NMPA uses details about me and my care?
The Healthcare Quality Improvement Partnership, who commission the NMPA, decide how the NMPA uses the information collected. The contact details of the Healthcare Quality Improvement Partnership can be found on the GDPR Statement page of their website. This includes details of their Data Protection Officer.
Can I access the details the NMPA holds about me and my care?
If your care details are included in the NMPA (i.e. if you gave birth after 1 April 2014) and you would like to request access to the information we hold about you please contact the Healthcare Quality Improvement Partnership, our commissioners, who decide what information we can share. For a summary of what information we collect about women and babies, what we do with the information and why, please see the NMPA information leaflets for women.
What is the legal basis for the NMPA holding my data?
Under the new EU rules on confidentiality (the General Data Protection Regulations) we have a lawful basis to hold information on women and their babies. The purpose of the audit is to provide information to NHS maternity care providers to help them improve the quality of care women and their babies receive during pregnancy and childbirth. These uses are classified as justifiable purposes for using personal information (or a legitimate interest, GDPR Article 6(1)(e) and 9 (2) (i)).
English and Welsh law allows the NMPA to handle data on pregnancies and births to audit NHS services without the informed consent of each woman and baby covered by the audit (Section 251 of the National Health Service Act 2006 – Control of Patient Information). For a copy of our Section 251 approval letter, see our Resources page.
How does the NMPA keep information secure?
The NMPA team make sure that information we hold on women and their babies is kept confidential and secure. We have comprehensive procedures and assurances in place, for example:
• Everyone in the NMPA team is fully trained in keeping information confidential and secure. We take the training every year.
• We have information security and information governance policies in place, which are regularly reviewed.
• We have done an official assessment of how well we are doing at keeping information secure (a Data Protection Impact Assessment, or DPIA).
• When we use birth and death registration information and information about your hospital appointments or stays, we are covered by Section 42(4) of the Statistics and Registration Service Act (2007) as amended by section 287 of the Health and Social Care Act (2012), which says that we can use this information to help NHS providers provide and improve their services.
• We store all NMPA data in a secure location, and all data stays inside the secure NHS network (the N3 network).
• We will securely destroy all information we hold 5 years after our contract for the audit ends. The reason for keeping the information after the contract is so that we can respond to enquiries about publications, for example from hospitals who want to investigate their results further.
What are my rights if the NMPA holds my information?
Under the new data protection laws (the General Data Protection Regulations, or GDPR) you have the right to:
• Request a copy of the information we hold about you – you have the right to receive this free of charge within 1 month,
• Request that we correct any inaccuracies in the information we hold about you,
• Request that we destroy information that we hold about you.
• Complain about the way we are handling your information. If you think there is a problem please contact us in the first instance, and if you are unhappy with our response you can contact the Information Commissioner’s Office.
If you would like to request any of these, please contact the Healthcare Quality Improvement Partnership, who are responsible for what happens to the information we hold (they are the data controller).
You also have the right to request that no one uses your birth registration information or information about your hospital appointments or stays. To request this please contact NHS Digital (for births or hospital appointments/stays in England) or The NHS Wales Informatics Service (for information (for births or hospital appointments/stays in Wales). They are responsible for what happens to information on births and hospital appointments and stays.
How can I request access to NMPA data for secondary use?
If you would like to request access to data from the NMPA that isn’t available on our website, for purposes of quality improvement, including research, service evaluation and audit, please visit the Healthcare Quality Improvement Partnership Data Access Request Group (HQIP DARG) website for more information. The Healthcare Quality Improvement Partnership are our commissioners, who decide what information we can share. For certain information that we hold you will need extra permissions from the organisation who provided us with the data. We will let you know if this is the case.
About the NMPA switch to the Maternity Services Data Set (MSDS) - England only
When will the NMPA switch to using the MSDS?
From the next round of the audit onwards, i.e. starting with MSDS data covering births between 1 April 2017 and 31 March 2018.
Are there any changes for Scotland or Wales?
No; the NMPA already uses central datasets for births in Scotland and Wales.
Why is the NMPA switching to the MSDS?
From the start of the NMPA, it has always been the plan to use central datasets like the MSDS in order to minimise the burden of data submission on trusts and boards. The switch will enable trusts to focus on optimising MSDS data quality in the longer term, which will support a range of national reporting activities to benefit maternity care. It will bring England in line with Scotland and Wales, for which the NMPA already uses central maternity datasets.
Does the trust need to submit any data to the NMPA in future?
It will not be necessary to submit a clinical data extract to the NMPA as this information will now come from the MSDS, with the MSDS extract supplied directly to the NMPA by NHS Digital. However, we will conduct an organisational survey during winter 2018/19, for which participation is required.
What happens if you find problems with data quality in our MSDS data?
As before, the NMPA team will be undertaking careful data quality assessments. Our approach to these assessments is informed by our previous work in this area and draws on both methodological and clinical expertise. As with our previous publications, where we discover data quality problems for a particular variable within a trust, the trust will be excluded from audit measures that require this variable and results will not be published. The same principle applies to site level results. We appreciate that some trusts may be concerned that insufficient MSDS data completeness for 2017/18 could result in exclusion from some or all NMPA measures, but HQIP will not impose penalties on trusts.
Can we resubmit corrected data either to NHS Digital to pass on, or directly to the NMPA?
No; while NHS Digital’s two month MSDS submission window is open, it is possible to make as many submissions as needed for the relevant month. Once it is closed, no further submissions are possible and the data cannot be refreshed later. Because the data will have been ‘locked’ following submission to NHS Digital, and because the NMPA works to a strict timescale, it will not be possible for trusts to resubmit data either to NHS Digital or to the NMPA. While all MSDS data for the year 2017/18 will already have been submitted and their submission window will have been closed, careful scrutiny prior to future MSDS submissions and action on data quality feedback will help ensure your trust’s data can be fully used.
What feedback is there on our data quality?
Trusts can access detailed reports on their MSDS data quality prior to, and at the point of submission, via NHS Digital’s submission portal. Individual feedback is also provided to the trust following each MSDS submission via the data quality scorecard emailed to trusts. The NMPA site level results of the data quality assessments for the last NMPA reporting period (based on data submitted directly from your trust’s electronic maternity information system) can be found on the NMPA website; these will also be a useful source of data quality information.
Will any NMPA measures change as a result of the switch?
From the start of the NMPA, it has always been the plan to use central datasets like the MSDS to minimise the burden of data submission on trusts. To this end, the previous NMPA data extract specification followed the MSDS format closely. This means that changes to the construction of the measures are unlikely. However, data availability may impact on the number of measures the NMPA will be able to report and individual trust data quality and completeness may impact on the number of trusts or sites which can be included in each measure.
What will happen when the MSDS v2.0 is introduced in spring 2019?
We have been working closely with NHS Digital to ensure that the data items needed to construct the NMPA measures are present in the MSDS v2.0. As with the current version of the MSDS, careful mapping of data items is needed to ensure the correct flow of information. This may be most effectively undertaken by clinicians/IT midwives and data specialists in collaboration, and in discussion with your electronic maternity system supplier.
Will the MSDS extract contain personal identifiable data?
MSDS data will be linked to Hospital Episode Statistics (HES), and the Office for National Statistics (ONS) birth and mortality registers for English trusts by the Data Linkage Service at NHS Digital. The MSDS extract and any linked data will be provided in pseudonymised form to the NMPA by NHS Digital and will not contain any identifiers. Pending approvals, the MSDS extract will contain delivery date and time.
How do we know that data will be kept secure by the NMPA?
The RCOG holds all record level NMPA data on a secure server within the NHS N3 network. The server is operated by the RCOG, and is leased to the RCOG by RedCentric. RedCentric have provided NHS network services for many years and are contracted under an agreement that guarantees secure data processing. All organisations involved in data processing have an up-to-date NHS IG Toolkit assessment. Further details on data security are available on request.
Do we need to inform all women who used our maternity service during the data collection period of the NMPA about how their information will be used?
There is no need to individually contact the women who have used your maternity service, as this was deemed to be impracticable by the Health Research Authority’s Confidentiality Advisory Group. We were instead asked to make ‘reasonable efforts’ to inform women of the processing of their data (including details of how to opt out), by providing a leaflet for trusts to display on their website. Individual leaflets can also be printed out and displayed in appropriate locations around the trust. We regret we are unable to provide printed copies of our information leaflet for women to trusts. We will soon be updating our leaflet to reflect the switch to the MSDS.
Has our Caldicott Guardian approved this?
We informed all Caldicott Guardians about the NMPA in writing at the start of the audit. No approval was required from Caldicott Guardians, since this is a National Clinical Audit and Patient Outcomes Programme audit, commissioned on behalf of NHS England.
Is there a data sharing agreement between our trust and the NMPA?
The NMPA will have a data sharing agreement with NHS Digital in order to access the MSDS and link it to other datasets. Therefore, there is no need for a Data Sharing Agreement directly between the NMPA and NHS trusts.
Will women who booked with us and then gave birth elsewhere be included in our results?
No. Your results will only be based on the MSDS data for births that took place under the care of your trust, i.e. on one of your maternity unit sites or at home under the care of your midwives. Until the introduction of the MSDS v2.0, it will not be possible to distinguish between births supervised by NHS midwives and those supervised by independent midwives contracted by the NHS, as may happen in some trusts. If the birth has been submitted to the MSDS it will be treated as an NHS birth for the purpose of the NMPA.