Housing, family and environmental risk factors for hospital admissions in children
Patient Notification/Fair Processing Notice
Who are we?
We are researchers at University College London (UCL). UCL is the data controller for this project, together with City, University of London.
Why are we doing this project?
Emergency hospital admission rates in children aged less than 15 years have increased by 30% since 2000. These admissions are highly stressful for children and their families, as well as costly for the National Health Service (NHS). In this project, we will look at whether certain household, parental and environmental factors are associated with the probability of a child being admitted to hospital. This will tell us areas where investments in public health, or efforts to improve healthcare access are most likely to lead to improvements in child health, which would reduce the need for emergency hospital admissions.
We will focus on the following two areas:
1) We will examine how building characteristics (such as damp and mould), household overcrowding, and air pollution exposure during pregnancy are associated with the probability of hospital admissions for respiratory infections in babies 2) We will determine whether children whose parents were born abroad, or are less proficient in English, have similar access to primary and community health services as children whose parents were born in the UK.
We will use statistical techniques to show how these factors are associated with the probability of hospital admissions in children. Our methods will take into account parents’ socio-economic circumstances, and whether children have underlying health problems.
The data used in this project
We will use a database containing birth and death registration and hospital records (Hospital Episode Statistics) for 6.7 million children. The children were all born in England between 2005 and 2014. The data were linked by researchers from City, University of London for a previous study, see our Fair Processing Notice. They are held on a secure server at the Office for National Statistics (ONS). The data held include the baby’s sex, birth weight and gestational age, hospital diagnoses and operations for babies and mothers, and parents’ country of birth.
The ONS will link this existing database to 2011 Census data from mothers and their partners (if any), which means information about housing, mother’s and her partner’s education and knowledge of English/English proficiency (self-assessed from the Census data) proficiency will be available for this research project. The existing database will also be linked to data on outdoor air pollution (from the Department for Environment, Food and Rural Affairs: https://uk-air.defra.gov.uk/data/), and building standards data (from Energy Performance Certificates: https://epc.opendatacommunities.org/).
All data are kept on highly secure servers at the ONS and can only be accessed by a small number of accredited individuals.
The current birth database does not contain any identifying information, such as names, addresses or NHS numbers.
Patient Notification/Fair Processing Notice
The identifying information (such as NHS numbers) are kept in a separate database, away from the health data, and cannot be accessed by researchers, only a small number of ONS staff. The linkage to Census data will be carried out by ONS staff using secure methods which are described in detail on the ONS website.
The air pollution and building characteristics data, which do not include any information about individuals, will be linked to the existing birth database using the mothers’ full postcodes. This will be done by ONS staff who do not have access to any clinical data. Linkage will be carried out by extracting a list of residential postcodes and study numbers (one for each mother) from the identifier database. The postcode list will be linked to buildings and air pollution data before postcodes are removed. The study numbers will be used to link the buildings and air pollution data to the hospital and birth registration data.
At no point will identifying information about you or your child, such as NHS numbers, full postcodes or dates of birth, be kept together with health, Census, air pollution or buildings data. Researchers will not be able to see this identifying information, only de-personalised (referred to as pseudonymised) data about health, family and living conditions.
All staff working on this project have a legal duty of confidentiality to protect personal information about mothers and children. All staff working with the data have had special training in keeping data confidential and secure.
Where to find the results
Our results will be published in open access journals so that they can be read online and downloaded for free. We will announce these publications and provide lay summaries of our results on our website: https://www.ucl.ac.uk/child-health/research/population-policy-and-practice/child-health-informatics-group/ . All publications will also be released to the press via the UCL Media Engagement Office.
Legal and ethical permissions
The legal basis for processing personal data for this purpose data at UCL falls under Article 6(1)(e) of the General Data Protection Regulations (GDPR), i.e. the “public task” of healthcare research. It also falls under Article 9(2)(j), “processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes”.
Personally identifying information (including names, addresses, dates of birth and NHS numbers) will be processed by the Office for National Statistics in order to link the data. This is conducted under section 251 of the National Health Services Act (2006) and its Regulations, the Health Service (Control of Patient Information) Regulations (2002), with approval from the Secretary of State for Health and Social Care via the Confidentiality Advisory Group (CAG reference: 18/CAG/0159).
Ethical approval for this research was granted by the Health Research Authority's London – Queen’s Square Research Ethics Committee (reference: 18/LO/1514).
Data storage and retention
Different parts of the data may be subject to different retention periods as agreed with the different data providers (ONS and NHS Digital). These retention periods are still being negotiated but are expected to be until at least 2022. Retention of data is necessary to ensure that analyses can be checked and to adhere to guidelines on reproducibility of research.
If you would like to know more about the project and how we using your data, please contact us at firstname.lastname@example.org. You may also contact the principal investigator, Dr Pia Hardelid, by phone or post, via the UCL Great Ormond Street Institute of Child Health. Contact details are available here: https://www.ucl.ac.uk/child-health/
You are entitled to object to your data, or your child’s data (if you are their legal guardian), being used for this research. If you do not want us to use your or your child’s data, please get in touch using the contact details above. However, note that the UCL research team will not be able to identify you and cannot remove your records from the study directly at your request.
You have the right to tell NHS Digital if you do not want the information you provide to them to be used beyond the purpose of providing healthcare. This is known as an "opt-out". Please visit NHS Digital’s website for further details: https://www.nhs.uk/your-nhs-data-matters/manage-your-choice/. Your choice will not affect the health care you receive.
University College London is one of the Data Controllers for this study. If you have any questions about how UCL uses personal data, you can contact the Data Protection Officer: email@example.com.
You also have the right to complain directly to the Information Commissioner’s Office, which is an independent regulatory authority set up to uphold information rights.