Survival and predictors of survival of children born with congenital heart disease-BINOCAR

What organisation instigated the work and why?

This project was instigated by researchers at Newcastle University to address the paucity of information on long-term survival of individuals born with congenital heart disease (CHD). This information can be used for counselling parents when a diagnosis of CHD is made either prenatally or postnatally. Additionally the information can be used to calculate estimates of the number of individuals living with CHD, which is important for health care commissioning.

How are other organisations involved? Why?

The data being requested is to be linked to death registrations is collected by congenital anomaly registries. Data will be sent from these registries to NHS Digital to be linked with death registrations. The data will then be sent to Public Health England Northern Office (which houses the congenital anomaly register covering the North of England), where it will be analysed by an employee of Newcastle University who has an honorary contract with PHE.

Which organisations will have access to the record level data supplied by NHS Digital?

Only three nominated individuals, who are either employed by PHE or employed by Newcastle University with an honorary contract with PHE, will have access to the data when it is stored at the PHE Northern Office.

How was this work instigated?

This project was originally funded by British Heart Foundation as part of a PhD. However, due to the extreme delays in obtaining the data the PhD has been finished (Oct 2015) and the project is now being funded by a Newcastle University Faculty Fellowship.

Aim

The aim of this study is to produce robust survival estimates, up to 25 years, for children born with congenital heart disease (CHD) between 1985-2012 in England and Wales. A further objective is to identify predictors of long-term survival of children born with CHD. This application is to link together data on cases of CHD notified to six congenital anomaly registers to death registrations, using several patient identifiable variable. Using the death registrations, NHS Digital will add the following variables to the data sets: survival status (alive/ deceased), year of death, number of days survived. Public Health England (PHE) also request that NHS Digital add IMD score at delivery to the dataset so that they can examine socioeconomic inequalities in CHD survival.

What does the work aim to achieve

With advances in surgical and medical techniques, survival of individuals with CHD has increased dramatically over the last 20 to 30 years. However, there is a paucity of information on CHD survival. This information is useful for counselling parents when a diagnosis of CHD is made prenatally or postnatally. This information can also be used to estimate the number of individuals living with CHD in the UK, which can aid health service commissioning.

Background

PHE originally applied to complete this data linkage in 2013. However, they have not yet received any data. The objectives remain the same as in their original agreement.

The applicant requires NHS Digital to link in death registrations using the following patient identifiable variables: Name, Address, sex, maternal age. The applicant also requires NHS Digital to provide Index of Multiple Deprivation scores and ranks for each case (based on postcode).

Data from the 6 different Congenital anomaly registers (CARs) centres – Northern Congenital abnormality survey, East Midlands and South Yorkshire CAR, CAR and information service for Wales, CAR for Oxfordshire, Berkshire and Buckinghamshire, South West CAR, Wessex antenatal detected anomalies register- has been securely transferred to NHS Digital.

The data from each register contained information on all children born with a CHD in their region. Each case had data on the following variables: ID no, infant forename and surname, postcode at delivery, mothers age at delivery, sex, number of foetuses, date of birth, gestational age at delivery (weeks), birth weight, ethnicity, register, antenatal diagnosis (ICD codes and text), postnatal diagnosis (ICD codes and text).

Babies have been matched and NHS Digital has calculated the number of days the baby survived and recoded postcode into IMD. NHS Digital has removed infant forename, surname and postcode and recoded date of birth/death into year of birth/death only. The data is therefore pseudonymised. The pseudonymised data set will be securely sent to the Regional Maternity Survey Office, Public Health England (now known as the NCARDRS PHE Northern Office) where it will be stored on a password protected computer and analysed by the applicant. Of the three users who will have access to the data one is employed by PHE, while two are employed by Newcastle University but hold honorary contracts at PHE.

All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract i.e.: employees, agents and contractors of the Data Recipient who may have access to that data).

PHE Northern Office stores the data on a server which can only be accessed at the PHE Northern Office. This data is stored in a separate location to the participant identifiers. The two datasets will not be re-linked and the data will remain pseudonymised. Data will only be accessed by individuals within the PHE Northern Office who have authorisation from CAG to access the data for the purpose described, all of whom are substantive employees or hold honorary contracts at PHE.

PHE will use Kaplan-Meier curves to estimate survival at age: one, five, ten and twenty of children born with any CHD. Similarly they will use Kaplan-Meier curves to estimate survival of CHD subtypes separately, as these vary in severity. Additionally, the NCARDRS North will use the demographic data collected by the BINOCAR (including maternal age, gestational age at delivery, birth weight, sex, index of multiple deprivation) to investigate predictors of survival. This will involve the use of Cox regression, and also use modelling to predict survival in the future.

In order to inform parents and clinicians, the applicant intends to present the outputs to and liaise with the British Congenital Cardiac Association, which is affiliated with the British Cardiovascular Society.

The results will be presented at Epidemiology and perinatal medicine conferences (such as Society for Pediatric and Perinatal Epidemiologic Research and Society for Social Medicine). They will also be published in peer reviewed medical journals such as Circulation. A summary of the results will be put on the BINOCAR website http://www.binocar.org/. The results will be analysed within the first year after the data is received. The data will be kept for three years. All outputs stated here will be aggregated where there are less than five cases in order to ensure anonymity.

All outputs are aggregated with small number suppression in line with the HES Analysis Guide.

Children with CHD require highly specialised healthcare, and it is important that health services are in place to address this. Robust estimates of prevalence and the long-term survival of CHD are required to accurately plan for the provision of health care and range of services required by these children in the future. However, very few studies have reported the long-term survival of children born with CHD.

This research will be of benefit not only to support planning and administration of the provision made for health and social care for pregnancies and infants affected by congenital anomalies, but the information on long-term survival is also important for health professionals who need these statistics to counsel parents when their child is diagnosed with CHD. This information is also important for parents, particularly when a diagnosis of CHD is made prenatally.

PHE have only received the final data set in 2018 and so are still in the process of analysing it and writing up with results. Therefore there are not yet any benefits.