NHS Digital Data Release Register - reformatted

London Ambulance Service NHS Trust projects

14 data files in total were disseminated unsafely (information about files used safely is missing for TRE/"system access" projects).


MPDS Maternity Study — DARS-NIC-320147-M9K6K

Type of data: information not disclosed for TRE projects

Opt outs honoured: Yes - patient objections upheld, Anonymised - ICO Code Compliant, Yes (Section 251 NHS Act 2006)

Legal basis: Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii), Health and Social Care Act 2012 – s261(1) and s261(2)(b)(ii), Health and Social Care Act 2012 – s261(2)(b)(ii)

Purposes: No (NHS Trust)

Sensitive: Non Sensitive, and Sensitive, and Non-Sensitive

When:DSA runs 2020-02-27 — 2023-02-26 2020.07 — 2020.07.

Access method: One-Off

Data-controller type: LONDON AMBULANCE SERVICE NHS TRUST

Sublicensing allowed: No

Datasets:

  1. Hospital Episode Statistics Admitted Patient Care
  2. Hospital Episode Statistics Accident and Emergency
  3. Emergency Care Data Set (ECDS)
  4. Hospital Episode Statistics Accident and Emergency (HES A and E)
  5. Hospital Episode Statistics Admitted Patient Care (HES APC)

Objectives:

Maternity emergencies can occur at any stage during pregnancy and emergency ambulance services are often the first point of call for pregnant women seeking emergency help. Time critical questions and life-saving medical advice is given as early as the 999 call by emergency medical dispatchers (EMD) who are trained to follow a scripted protocol. In the London Ambulance Service (LAS) NHS Trust, 999 calls are triaged using the Advanced Medical Priority Dispatch System (MPDS). This triage software assists in determining the speed and type of response a woman will receive from the ambulance service.

MPDS consists of 36 Protocols; each protocol asks specific questions that prioritise callers in order of urgency. Protocol 24 is used triage callers with pregnancy, childbirth and miscarriage related issues. The questions and advice given through Protocol 24 is based on expert consensus and user experience. There is currently no published research evaluating telephone triage for maternity calls. The London Ambulance Service NHS Trust is currently running the MPDS Maternity Study which aims to:

• To determine the accuracy of MPDS Protocol 24 in identifying maternity emergencies.
• To identify predictors of maternity emergencies at the time of 999 call that can alert EMDs to potentially life threatening conditions developing.
• To identify which of the current key questions on MPDS predict maternity emergencies.
• To understand the experiences of Emergency Medical Dispatchers in using Protocol 24 that would not be gained from analysing clinical data.
• To explore if MPDS Protocol 24 triage has any impact on how ambulance clinicians manage women using this service.

For women who called 999 from April 2019 to July 2019 and were triaged by the LAS NHS Trust, using the MPDS. The LAS NHS Trust aims to retrospectively review the 999 call records and clinical patient records to determine whether the woman was considered by the paramedics to be suffering a maternity emergency and required the ambulance response that was allocated.

The aim of processing is also to understand the demographics of women who experience maternity emergencies In the UK, Black women are five times more likely and Asian women are twice as likely to die in pregnancy than other groups and the number of new-born stillbirths and deaths in the first week is higher in socioeconomically deprived women. Identifying women who may be at higher risk will help strengthen emergency medical services ability to predict maternity emergencies during telephone triage.

The study is funded by the International Academy of Emergency Dispatch (IAED), a non-profit standard-setting organisation that maintains the MPDS triage protocols and promotes safe and effective emergency dispatch services worldwide. The study’s Principle Investigator is the LAS NHS Trust’s Head of Research who is a PhD graduate, Honorary Research Fellow and Professor. She has successfully participated in a number of landmark studies including PARAMEDIC-2 and the RIGHT-2 trials which were recently published in the New England Journal of Medicine and the Lancet respectively and is the Chair of the National Ambulance Research Steering Group (NARSG). The remainder of the research team includes the LAS NHS Trust’s Clinical Audit Manager who has specialist knowledge of the ambulance service and research methodology, a Research Paramedic, an independent statistician from the London School of Hygiene and Tropical Medicine (LSHTM), an advisor with expertise in emergency medical dispatch, a consultant midwife, a research administrator and a lay person as the women and public representative.

The research project has received approval from the Surrey Borders Research Ethics Committee (18/LO/2027) and Confidentiality Advisory Group (18/CAG/0202) to process patient data under section 251 of the National Health Service Act 2006.

Legal basis
The LAS NHS Trust uses Article 6(1)(e) - public task, as the legal basis for the processing of identifiable data in this study (processing is necessary for the performance of a task in the public interest or in the exercise of official authority vested in the controller).

The LAS NHS trust is a public authority. The ICO states: “Public authorities in the National Health Service (NHS) range from trusts to individual practitioners who provide services under contract to the NHS”.
https://ico.org.uk/media/fororganisations/documents/1152/public_authorities_under_the_foia.pdf.

The UK policy framework for health and social care research (v3.3 07/11/17) refers to healthcare providers (including Foundation Trusts) ‘having regard to this policy framework according to their legal duty under Section 111(7) of the Care Act 2014 and contributing to the fulfilment of their commissioners’ legal duties to promote research under the Health and Social Care Act 2012.’

Article 9(2)(j) (processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject).

In order to fully understand whether or not women experiencing a maternity emergency, the LAS NHS Trust need to refer to the final patient outcome from hospital records as these will provide the true diagnosis. To achieve this, the LAS NHS Trust will request the women’s hospital data from NHS Digital HES database. Once the hospital outcomes data are obtained, the LAS NHS Trust will provide a pseudonymised spreadsheet of grouped maternity emergencies and non-emergencies cases to the statistician at LSHTM to determine the accuracy of MPDS maternity triage using sensitivity and specificity analysis.

The outcomes of interest in the study (diagnosis, investigations, treatments, quality of life measures, mortality, demographics) for approx. 3,000 mothers (and their babies if born whilst under ambulance service care) are only available through NHS Digital and would be addressable using the single HES data request.

The research team anticipates women outcome data will be collected from both of the following cohorts of HES datasets:

1. Accident and Emergency (AE) and Emergency Care (EC) - Ambulance crews convey women who are under 20 weeks pregnant to the emergency department.

2. Admitted Women Care (APC) – when appropriate ambulance crews may convey a woman directly to the labour or maternity ward if the woman is over 20 weeks pregnant.

The LAS NHS Trust are requesting a one-off release of record level data for each women during the data collection period (1st April – 31st July 2019) only from NHS Digital.

The LAS NHS Trust is the data controller and determine how personal data will be processed by the data processors. The London Ambulance Service NHS Trust and the LSHTM are the joint data processors.

The LAS NHS Trust as a joint processor is responsible for receiving the pseudonymised HES data from NHS Digital, matching the correct 999 call record with women hospital data, and sending securely to The LSHTM for analysis. The LSHTM is a joint data processor and is responsible for analysing pseudonymised data on behalf of the data controller.

Expected Benefits:

The results of the MPDS Maternity Study will be relevant to the 850,000 estimated pregnant women each year in the UK and many more around the world who may need treatment from emergency medical services. The results are also relevant to emergency medical services who receive thousands of emergency maternity calls each year and are responsible for delivering the right response, at the right time, to women who need help first.

The results of the MPDS Maternity Study are expected to identify the accuracy of MPDS triage for maternity calls, the incidence of prehospital maternity emergencies, and women demographics that can be used to predict with callers at higher risk of maternity emergencies.

Determining the sensitivity will help understand how accurately MPDS correctly identifies women experiencing maternity emergencies. A high sensitivity is clearly important because MPDS is used to identify serious but treatable maternity emergencies. Understanding the sensitivity of MPDS can lead to steps to maintain or improve its accuracy through the IAED Proposals for Change process. Determining the specificity will help understand how accurately MPDS correctly identifies women who are not experiencing maternity emergencies. An understanding of the specificity is useful for the emergency medical services who use MPDS. It can provide the evidence to support the decision on how to appropriately allocate limited resources to each call, without compromising women safety and reduce the burden of over-dispatching to a call where possible.

The incidence of pregnant women with maternity emergencies attended by emergency medical services is not known, yet the cost to the NHS for litigation cover against maternity claims totalled £482 million in 2012-13 according to the National Institute for Health and Care Excellence (NICE). Litigation in maternity care across the NHS is rising, accounting for the highest value and the second highest number of claims in 2012-13 and similar may be said of other health care systems around the world. The results of the study are expected to improve emergency medical services (EMS) ability to identify maternity emergencies so there should be less chance of adverse outcomes for women. By also identifying those who do not need an emergency ambulance, EMS are able to reach other women with more serious, potentially life-threatening, conditions more quickly and are able to save the NHS the cost of dispatching an ambulance unnecessarily.

Previous work related to child and maternal confidential enquiries in the UK suggest disadvantaged and vulnerable women and their babies are at higher risk of poor and unequal health outcomes. In the UK for example, pregnant women from black and ethnic minority groups are four times more likely to die in childbirth than other groups and the number of stillbirths and deaths in the first week is higher in socioeconomically deprived women. These inequalities in maternal health are a result not only of biological and individual risk factors but also of other factors like wealth, ethnic background, religion, geographic location, education and so on. Identifying and managing higher risk factors will help strengthen emergency medical services capacity to effectively triage and allocate resources to manage maternity emergencies. The results of the MPDS Maternity Study will identify which factors are associated with maternity emergencies and can be used to improve the predictive ability of telephone triage to identify maternity emergencies.

Beyond the MPDS Maternity Study, the data linkage methods using NHS Digital would inform study design and produce benefits to future research that links ambulance and hospital records. Learning from the data linkage methods developed in the MPDS Maternity Study will be shared by methodological publications that do not include the data in open-access, peer reviewed journals for the benefit of future studies.

Outputs:

This area of emergency dispatch has not been studied before and the results, expected mid-2020, will hopefully provide an evidence base to amend the questions used in the telephone triage of pregnant women who call emergency medical services around the world. MPDS is used by 3,000 agencies and translated in 21 languages, and the results have the potential to change the number, order and wording of the triage questions, as well as the medical instructions given to the caller by the EMD before an ambulance arrives.

The research team aim to disseminate the results widely, including presentation at relevant conferences and publication in an open access, high impact medical journal. The research team has a track record of publishing in high quality peer reviewed journals and aims to publish in journals in the fields of prehospital medicine and emergency dispatch; subject-specific journals will be used where appropriate. Previously this has included The Lancet, New England Journal of Medicine and The British Medical Journal.

A full report with aggregated outputs with small number suppressed in line with the HES Analysis Guide will also be shared with The IAED who maintain the standards of MPDS and are funding the study. The IAED will not receive raw data and data will not be used for sales and marketing purposes.

A non-technical summary of the main study findings will be published on the LAS’s internal and external websites, as well as the UK National Ambulance Research Steering Group’s website. The main findings will be presented at key conferences, such as International Academy of Emergency Dispatch Navigator Conference and the annual Emergency Medical Services Conference, with any outputs aggregated with small numbers suppressed in line with the HES Analysis Guide. The research team will continue to involve the women and public representative to help develop a dissemination strategy for a lay audience.

This study will be registered with the LAS Clinical Audit & Research Unit and the National Institute for Health Research (NIHR) on behalf of NHS England.

Processing:

The following women identifiers will be extracted from each LAS NHS Trust 999 call record and sent to NHS Digital for linking to the hospital records as permitted by the CAG outcome (18/CAG/0202):

- NHS Number
- Date of Birth
- Postcode
- Gender

It is necessary to process these identifiers as they will produce the highest possible rate of linkage by NHS Digital and ensure that ambulance records are linked to hospital records for the same women, as accurately as possible. The study is interested in the hospital outcomes for each woman following the specific call to the ambulance service that led to their transport to hospital. The LAS NHS Trust are requesting the data items for each woman during the data collection period (1st April – 31st July 2019) only from NHS Digital.

The research team will securely transfer the cohort of women identifiers for the entire sample to NHS Digital via the NHS Portal. The NHS Digital Portal is accessed via a secure authentication method to named users only. Once the linking is complete, the data will be returned to the research team pseudonymised, using only study ID and the requested data items. Identifiers sent to NHS Digital for initial linkage will be kept separately and not shared with LSHTM.

The excel spreadsheet of data items will be pseudonymised and shared securely with the independent statistician at the LSHTM for analysis. The excel spreadsheet of data items will be encrypted and sent securely to the independent statistician at LSHTM using Egress Switch v4.25 – a secure email encryption service for sharing sensitive files. The National Cyber Security Centre has accredited Egress Switch for the protection of Official Information which would include NHS Digital Data. Egress Switch v4.0 (and above) has been certified as satisfying the requirements of National Cyber Security Centre’s Commercial Product Assurance (CPA) Foundation Grade. That CPA certification is for the Egress Switch’s e-mail encryption functionality and permits the LAS NHS Trust to send encrypted e-mails using Microsoft Outlook. Egress Switch also allows the LAS NHS Trust to control who has access to the encrypted e-mail, even after it has been sent. The LAS NHS Trust will restrict access to the named independent statistician at the LSHTM. Egress Switch logs all access requests for keys, allowing the LAS NHS Trust to monitor when the encrypted e-mail was accessed by the named independent statistician at the LSHTM.

The statistician at the LSHTM will return the pseudonymised results of the analysis via email and a manuscript summarising the results of these analyses will be produced and submitted for publication in a peer-reviewed journal.

The LAS NHS Trust on receipt of the hospital data for the period (1st April - 31st July 2019) will use the study ID and time of the 999 call to ensure the correct 999 call record is matched to the correct hospital data for each women on a pseudonymised spreadsheet. The LAS NHS Trust will provide the pseudonymised spreadsheet to the LSHTM statistician to determine the accuracy of MPDS maternity triage using sensitivity and specificity analysis.

The LSHTM on receipt of the pseudonymised spreadsheet will estimate the sensitivity, specificity, positive/negative predictive values, test accuracy, and positive/negative likelihood ratios of the dispatchers’ initial diagnoses (emergency vs non-emergency) compared with the clinicians’ final diagnoses (emergency vs non-emergency). The pseudonymised data will also be used to describe the association between each of the telephone triage questions and the clinicians’ final diagnoses (emergency vs non-emergency).

Alternative linking methods that use fewer or no identifiers are not appropriate or adequate because not only would they lead to inaccurate linkage, compromising data accuracy, but would also reduce the number of linked records. The sample size of 3000 women has been calculated by the statistician at the LSHTM as the minimum number of women the LAS NHS Trust requires to show the sensitivity and specificity of the MPDS protocol. Failing to achieve a high level of matching accuracy will mean that the LAS NHS Trust are unable to achieve the aims of the project.

The research team do not need to receive any women identifiers back from NHS Digital, as no attempts to re-identify these women will be made.

On receipt of the linked data from NHS Digital, the research team at the LAS NHS Trust will store data on the LAS NHS Trust encrypted secure area network and restrict access to only those individuals who are working on the MPDS Maternity Study (i.e. the Principal Investigator who receives the data, and the Lead Investigator and Research Paramedic who enter the data). The LAS NHS Trust trust-wide system of role-based access means data folders are only accessible to those that require access to it. The research team access data via a secure server and team are granted access only after they successfully complete Information Governance training mandated by the LAS NHS Trust on an annual basis. The team must work on the data on LAS NHS Trust encrypted servers and any output will be saved to the same servers. They are instructed that no copies of the data are to be taken and placed in any other location.

Once the spreadsheet has been downloaded by the independent statistician, it will be stored on a secure, backed-up server at the LSHTM. The spreadsheet will be accessed by the independent statistician using their dedicated personal computer (i.e. not shared by others) that is encrypted, password protected and located within a locked office at the LSHTM. The independent statistician will be the only individual granted access to the relevant area of the secure server and the data will not be accessible remotely. The independent statistician at the LSHTM has undertaken ICH Good Clinical Practice (GCP), Information Security Awareness, and General Data Protection Regulation (GDPR) training.

A manuscript summarising the results of these analyses will be produced and submitted for publication in a peer-reviewed journal. Twelve months after publication, the pseudonymised data will be securely deleted from the secure server at the LSHTM and the personal computer used to store the data will be securely wiped and re-imaged. A data destruction certificate will be issued. The LAS NHS Trust as data controller will securely delete pseudonymised data from the secure server within 36 months of the data sharing agreement with NHS Digital.

IAED maintain a single unified MPDS protocol, with no variants, that introduces changes to the standards of each protocol in an orderly manner, which is then provided to all users of MPDS across the world. Changes or improvements are introduced via the submission of Proposals for Change (PfC). It is anticipated that the IAED will use the results of this study as an evidence base to generate several PfC for its maternity protocol, which will improve the triage questions as well as the medical instructions given to the caller before an ambulance arrives.

The IAED will not determine the way the data will be processed as set out in the funding contract where the responsibility of data processing lies solely with the LAS NHS Trust. The IAED will only receive a full report with aggregated outputs in line with the HES Analysis from the LAS NHS Trust. Data will not be used for sales and marketing purposes, but to improve the accuracy of the maternity triage protocol.

IEAD will not have influence on the outcomes nor suppress any of the findings of the research.

https://www.emergencydispatch.org/proposal-for-change-becomes-a-protocol