NHS Digital Data Release Register - reformatted

National Centre For Social Research

Opt outs honoured: N

Basis: Informed Patient consent to permit the receipt, processing and release of data by the HSCIC

Format: Identifiable Sensitive

How often: Ongoing

When: unknown — 11/2016

HSCIC Id: DARS-NIC-148332-V6CY3-v0.0

Data: MRIS - Cohort Event Notification Report

Data: MRIS - Scottish NHS / Registration

Data: MRIS - Cause of Death Report

Data: MRIS - Personal Demographics Service

Data: Hospital Episode Statistics Admitted Patient Care

Data: MRIS - Flagging Current Status Report

Output: The primary objective of ELSA is to collect longitudinal data on health, disability, economics, and social participation and networks, from a broad-based sample of the English population aged 50 and older. This includes a unique coverage of biomedical, genetic, performance and psychosocial measures. Participants are approached biennially for the main interview. All waves included an interviewer-administered questionnaire and self-completion form; in waves 2, 4 and 6 there was also a nurse visit for biomedical measures.

ELSA is not only a study of health but encompasses many facets of ageing. The focus of the study is to provide data necessary for an exploration of the unfolding dynamic relationships between health and functioning, social networks and economic position, as people plan for, move into and progress beyond retirement. It therefore has extensive components measuring financial status and social participation as well as health and cognition.

Some of the questions that ELSA can address are: the nature and timing of retirement and post retirement labour market activity; the determinants of economic well-being at older ages; cognitive functioning and its impact on decision making among older people; disability and the compression of morbidity; the evolution of economic, social and health inequalities in an ageing population; social participation and social productivity at older ages; the impact of good quality health care on future health and well-being.

The main output from this data request would be the creation of a rich and extremely useful ELSA database to be used as a resource for research analysis. Future outputs would include those described as collaborators (who are applying to HSCIC separately) and the planned sub licencing agreements to bona fide 3rd party researchers.

Activities: The requested data from the HSCIC would be linked to the survey data from the English Longitudinal Survey of Ageing (ELSA) . The linking process consists of matching the variables provided by HSCIC to the survey data via unique identifiers, termed ‘serial numbers’, for each participant who has consented to linkage.

NatCen creates a new set of serial numbers for the linked data, and maintains a look up file which allows this to be linked to the survey data.

The ELSA dataset and HSE dataset are not linked together and are linked to HSCIC separately and via separate agreements.

Simple data flow:

• NatCen provide NHS number, DOB, sex, postcode and ELSA member number to HSCIC
• HSCIC provides NHS number, latest demographic data, Exits/re-entries to NHS, ONS mortality data and cancer registration data and pseudo HES data
• No onward sharing is being requested in this application.

Objective: NatCen collects permission from respondents for data linkage to Hospital Episode Statistics (HES), Personal Demographics Service (PDS), Cancer Registrations and ONS Mortality data on the English Longitudinal Study of Ageing (ELSA). They also request the same data for the Health Survey of England (HSE), but an application to link data to the HSE has been submitted separately.

Being able to link the survey data to the administrative data is an important feature of the survey, because it allows association between the health and lifestyle information and measurements collected in the survey at a particular time with mortality, cancer and HES data over a longer time period.

This application is to keep data already supplied by the HSCIC and to request further data in order to maintain and further develop the ELSA databases. A request to be able to share the linked data with 3rd parties via sub licensing agreements will made via an amendment application to HSCIC via DAAG in the near future. There would be no standard outputs, but all would be for the purposes of research. Requests might relate to whether people have died, age at death, cause of death, frequency of hospital episodes in total or for particular diagnoses, etc – see below. The linked data required, and the purpose for which it is required, would be set out in the Data Release Form submitted to NatCen.

NatCen collects the survey data and acts as the data controller for providing access to the survey data (and in the future any associated linked data) to researchers who will use the data for specific research projects.

The purpose of requesting linked data is to provide important information relating to the health of respondents. This includes details which would be too burdensome to ask them to provide in an interview. It also allows NatCen to understand the health of ELSA respondents who have not been able to continue with the study owing to ill-health. Also to create a rich dataset by linking HES, PDS, Cancer Registrations and ONS Mortality to the survey data for planned onward sharing with researchers.

Linkage with HSCIC data is needed for many analyses to investigate the relationship between health conditions and behavioural and social characteristics reported in the survey and subsequent mortality. For instance, the linked data has been used to explore relationships between survey data on social class and raised blood pressure and mortality rates from coronary heart disease; the analyses also took into account demographic characteristics and other risk factors, such as smoking, family history of heart disease and cholesterol levels.

ELSA is a continuation of HSE, not only being based on the HSE sample of older people, but also in having considerable overlap in the type of health information collected such that, for some analyses, the HSE interview is being treated as the baseline (wave 0). The two datasets, however, are never linked.

There are 5 forms of usage that the researchers wish to address.
1. The maintenance and development of the ELSA database to be of the most benefit for health and social care
2. Administrative use when issuing the sample to interviewers, to avoid upsetting bereaved individuals by being prepared in advance
3. For calculation of response rates - NatCen need to know who has become ineligible because of death; the type of attrition is also important in understanding the study sample and the interpretations they can draw
4. For future 3rd party users more generally to have access to mortality information. Many longitudinal analyses will be completely misleading unless they can take account of censorship through death. For this the minimum needed is the month and year of death. This will be applied for separately
5. The interview data are deposited with the UK Data Service, a condition of ELSA funding. Linked data will be applied for separately

Benefits: The primary objective of ELSA is to collect longitudinal data on health, disability, economics, and social participation and networks, from a broad-based sample of the English population aged 50 and older. This includes a unique coverage of biomedical, genetic, performance and psychosocial measures.

ELSA is not only a study of health but it produces a great number of analyses and outputs relevant to or aimed at improving the provision of health or adult social care, or the promotion of health. ELSA plays an important role in providing high quality data from a multidisciplinary perspective that integrates information about the economic, social, psychological, community and health experience of older people in England.

The ELSA research team publish a comprehensive report analysing each new wave of data that is collected. These are available to download from the study website - http://www.elsa-project.ac.uk/ - which itself hosts a wide range of information about the study and its findings.

Of particular note is ELSA’s publication list, maintained on the website: http://www.elsa-project.ac.uk/publications . This provides a comprehensive listing of 250 outputs which have included findings based on analysis of ELSA. These include journal articles, working papers, book chapters and conference papers/presentations.

Some examples relevant to the provision of health or adult social care or the promotion of health are included below:

• Sexual health and well-being among older men and women in England: findings from the English Longitudinal Study of Ageing, David Lee , James Nazroo , Daryl O'Connor , Margaret Blake and Neil Pendleton , Archives of Sexual Behavior , Epub ahead of print] , January 2015

• Taking up physical activity in later life and healthy ageing: the English longitudinal study of ageing., Mark Hamer , Kim Lavoie and Simon Bacon , British Journal Of Sports Medicine , Vol: 48 (3), pp:239-43 , February 2014

• Limited health literacy is a barrier to colorectal cancer screening in England: Evidence from the English Longitudinal Study of Ageing., Lindsay C. Kobayashi , Jane Wardle and Christian von Wagner , Preventative medicine , November 2013 Journal Articles

• The SES health gradient on both sides of the Atlantic, ELSA Working Paper, James Banks , Michael Marmot , Zoë Oldfield and James Smith , January 2007

• Association between low functional health literacy and mortality in older adults: longitudinal cohort study, Sophie Bostock and Andrew Steptoe , British Medical Journal , doi:10.1136/bmj.e.1602 , March 2012 Journal Articles

The ELSA survey dataset is also a key output. Similar to HSE, ELSA survey data can be accessed by bonafide researchers who register to use with the UK data archive and download data sets to do their own analysis. Selected summary tables are also made available via the study website in Excel format - http://www.elsa-project.ac.uk/data_elsa. There is considerable use of the data in these ways.
Like HSE, ELSA findings are used widely by those involved in the development of health and social care policy. As a result, questionnaire and related study content (such as the particular objective measurements used) are adjusted at each wave in response to policy concerns and to emerging social and health issues. For example, wave 7 included questions on hearing, oral health, extended questions on cognitive function and new questions on expectations and perceptions of the costs of social care.
Selected examples of the benefits to improving the provision of health or adult social care, or the promotion of health which ELSA data can or has provided are included below:
• Consequences of improved survival with serious illness and the rise in chronic disease. The longitudinal nature of ELSA makes it possible to monitor the experience of people as they acquire chronic illnesses, and evaluate the consequences of ill-health from economic, social and well-being perspectives. New measures of cognitive function included at waves 7 and 8 will allow us to estimate the prevalence of mild cognitive impairment and dementia throughout the country, as well as investigate their determinants and consequences for individuals and their families. We can use the detailed data on health, disability and functioning to monitor progress towards achievement of extended healthy life years. This information will also be relevant for government to inform the debate on key issues include the importance of improving care provision for the elderly, managing dementia more effectively through better treatment and research, and deciding how to pay for social care.
• Social care. ELSA provides data permitting an understanding of the impact of changes in the range of social services available to older people on their well-being, health and social integration. Developing this understanding requires the collection of robust data that not only covers all relevant features of the care received, but also covers in detail the characteristics of those receiving, and not receiving, care, and that measures short, medium and long-term outcomes. The multidisciplinary and longitudinal nature of ELSA means that it has great potential in this regard. Data from the survey’s questionnaire module on social care can be used to address questions such as: older people’s receipt of and payment for care; the pattern of take-up of Direct Payments and Personal Budgets by older people; the provision of informal care; and the relationships between receipt of formal care, informal care and care needs among older people.
• Public health and transport policy. ELSA data was used to explore the link between the introduction of a bus pass allowing free local travel during off-peak hours for those aged 60 or over and public health. The findings showed that the policy was associated with increased use of public transport and also that older people who used public transport were less likely to be obese, as were those eligible for free local bus travel. The results were noted by members of the Parliamentary Select Committees for Health and Transport, as well as policymakers at the Department for Transport, for whom they are valuable in future planning of concessionary bus travel policy.
• Organisational changes in the NHS. The health data collected in ELSA, particularly when coupled with linked administrative data on healthcare, will provide important information about the use of services by the elderly before and after the transition from PCT to GP commissioning of care, the quality of health care, the interface between primary and secondary case, and the identification of trends in health that will impact on future demand.
• Social isolation and loneliness. ELSA has provided valuable information about the relationship of social isolation and loneliness to well-being, health and cognition. The survey is an important resource for monitoring these experiences as people age.
• Subjective well-being and public policy. ELSA includes measures of the different elements of subjective well-being since its inception, and these have been supplemented by the core questions from the ONS experimental module. It provides unique information about the trajectories of well-being among older people in England, and relationships with economic position, social factors, health, and cognition.

The creation of these linked databases will provides further benefits in the near future when NATCEN requests that they be allowed to offer psuedonymised data via sub licence to 3rd party researchers. Each researcher would bring different benefits to health and social care, but by way of example the following are wishing to use linked ELSA /HSCIC (they are currently applying for the linked data separately with the HSCIC)

Imperial - have been funded by DH to do a 2 year project on the potential health and economic benefits of the free bus pass, and the ELSA-linked data is the central part of this. The linked data will allow the researcher to examine whether there are any differences in hospitalisation/mortality associated with the bus pass, and this information will be the main driver of the economic modelling (and a lot of the policy discussion focuses on costs to benefits). DH are expecting the report in about a year’s time.

University of East Anglia - The ELSA analysis is part of a larger programme of research which aims to develop an intervention to improve access to primary care for older people and test it within a feasibility study. The ELSA linked with HES analysis is vital because it will map risk factors along the patient pathway from recognising problems, accessing primary care and use of secondary care. The analysis will also be used to test concepts which come out of the interviews and literature review. Results from the ELSA linked with HES analysis will be triangulated with the other components to develop a new intervention to improve access to primary care for deprived older people in rural areas which will in turn should improve community healthcare

Source: NHS Digital.