Presentations from this strand that have been made available for download can be accessed from the hyperlink in the title line of the abstract. Copyright in all presentations remains with the author(s), to whom permissions requests should be made. All presentations represent the views of the author(s).
Population health across space and time: the geographical harmonisation of the ONS Longitudinal Study for England and Wales
Paul Norman, University of Leeds; Mylene Riva, Durham University
There is a need in health research to identify whether inequalities are increasing or improving. Both cross-sectional and cohort studies contribute to our knowledge with the ONS Longitudinal Study (LS) for England and Wales a major resource. However, any research into geographical change over time is hampered by boundary change or when data are not available for the geography relevant to an analysis. We use population-weighted centroids to estimate an LS member's location at previous time points and link this to 2001 Census Output Areas, not so that analyses can be carried out at this scale but for records to be linked to larger geographies or area classifications. In terms of reliability, we find that accuracy improves with increasing size of geographical units and when area typologies are used. In example analyses we find that in a time-series of cross-sections, mortality improves across all area types, but not to the same extent. A longitudinal analysis indicates that changes in the area types in which people were living leads to steeper health gradients than if people had stayed living in the same type of area. Differences though are small suggesting that there is little mobility between area types. We recommend that longitudinal and cohort studies retain members' postcodes so that ongoing linkages can be made when boundary changes occur and for relevance to application relevant geographies. Our method can be used to enhance past records and maximise previous investment in the collection of data.
Email: p.d.norman@leeds.ac.uk
Estimating migration flows in Northern Ireland by health characteristics measured in the Census
James J. Brown, University of Southampton; John W. McDonald, Institute of Education
This presentation reports on work to estimate migration flows using record linkage between the sample of Census data in the Northern Ireland Longitudinal Study (NILS) and movements measured by changes of addresses based on the health card registration data. The link to the NILS data opens the possibility of estimating the annual cross-flows between areas and by characteristics not measured on the health card registration data, such as self-reported health status on the previous Census using statistical models that calibrate different sources of data to be consistent.
Email: j.j.brown@soton.ac.uk
Micro-level modelling to identify the separate effects of migrant status and other personal characteristics on people's job-status change
Tony Champion, University of Newcastle; Mike Coombes, University of Newcastle; Ian Gordon, London School of Economics
This paper investigates the relationship between social mobility and geographical mobility in England and Wales using the linked census records of the ONS Longitudinal Study. It builds on the insights originally achieved by Fielding some two decades ago, which among other things showed that in 1971-1981 there were particularly high levels of upward social mobility among those moving to the 'escalator region' of south-east England. Novel features of our analysis include updating the analysis to 1991-2001, adopting an urban rather than regional perspective, covering a wider set of cities than just London and using a single-scale measure of job status based on imputed hourly income as opposed to a set of inter-class transitions. Micro-level modelling is used to gauge the separate effect on career progression of being an inter-city migrant after allowing for people's other attributes and characteristics, notably age and gender. Inter-city migrants are also distinguished by reference to the type of city region that they moved between, notably in terms of the hierarchical level and regional location of the city.
Email: tony.champion@newcastle.ac.uk
What can the ONS Longitudinal Study tell us about the mortality of the 'Golden Cohert'?
Shayla Goldring, Nigel Henretty, Julie Mills, Kate Johnson, Office for National Statistics
It is well documented that the generations born around 1930 (between 1925 and 1934) are consistently exhibiting higher rates of mortality improvement than the generations either side of them. There is currently no evidence that these differentials are declining. Similar cohort effects seen in other countries suggest that these differentials may well persist into the oldest ages. In current ONS National Population Projections, it is assumed that these cohorts will continue to experience higher rates of improvement. However, it is not yet precisely clear why this is so. Understanding what is causing the significant decrease in mortality for this generation is particularly important for forecasting the population at older ages. This paper details preliminary research carried out using the ONS Longitudinal Study to try to better understand why the members of the generation born around the early 1930s in England and Wales have been enjoying higher rates of mortality improvement throughout their adult life.
Email: shayla.goldring@ons.gsi.gov.uk
Using the Northern Ireland Longitudinal Study to analyse socio-economic and demographic determinants of antibiotic prescribing patterns: lessons for antibiotic stewardship
Fiona Johnston, Northern Ireland Statistics & Research Agency; Michael Rosato, Queen's University, Belfast
The Northern Ireland Longitudinal Study (NILS) is a large scale, representative data-linkage study consisting of approximately 500,000 people (28% of the population). The NILS follows major life events using information from sources including the 2001 Census, vital events data (e.g. births, deaths) from the General Register Office and demographic data derived from health card registration data. The NILS is an innovative research resource which also allows for the opportunity to link to distinct Health and Social Care administrative datasets. For this analysis, the NILS database was linked to the Enhanced Prescribing Database, an electronic record of drugs prescribed in Northern Ireland. This linkage required specific ethical approval and only an anonymised dataset was made available to researchers. The aim of this project is to examine for individual and area level characteristics influencing variation in antibiotic usage to inform the management of antibiotic prescribing and consumption in Northern Ireland. A multinomial logistic regression modelling framework was used to examine the relative risks of receiving an antibiotic prescription in relation to both (a) individual level demographic and socio-economic characteristics; and (b) area level attributes such as urban/rural classifications and deprivation measures. Preliminary results show clear emerging patterns in relation to the prescribing of antibiotics: women; those not married; those living in deprived circumstances; and those from a Catholic community background all show significantly increased risks of being prescribed more antibiotic drugs when compared with their associated reference groups.
Email: fiona.johnston@dfpni.gov.uk
Deriving trends in life expectancy by National Statistics Socio-economic class (NS-SEC) using the ONS Longitudinal Study
Brian Johnson, Office for National Statistics
There is a need for continuous measures of socio-economic health inequality which can be tracked over time. For some time, ONS has published a series "Trends in life expectancy by social class using the ONS Longitudinal Study" which was updatedperiodically. Registrar General's Social Class (RGSC) based on occupation was used owing to the need for continuity over a relatively long period. In 2001, the National Statistics Socio-economic Classification (NS-SEC) replaced RGSC in official statistics. It was not possible to produce a medium-term series of life expectancy by NS-SEC, since 2001 was the first year where occupation was coded by NS-SEC at the census and at death. In order to produce a series of trends over more than 20 years based on NS-SEC, it was necessary to link NS-SEC to occupations in the 1980's and 1990's and then to measure subsequent mortality rates for different NS-SEC classes using the ONS Longitudinal Study. For the 1980s occupations, there was no recognised method for assigning NS-SEC. This presentation describes how such an approximation was derived, the challenges involved and the methods used to minimise error. Conceptual and measurement issues are discussed and the latest results are shown including trends in life expectancy by socio-economic class and other supplementary findings.
Email: brian.johnson@ons.gsi.gov.uk
Survivorship 2001-2008 among residents of communal establishments in 2001 in England and Wales: Results from the Office for National Statistics Longitudinal Study
Emily Grundy, London School of Hygiene & Tropical Medicine
The size of the population in long-term residential and nursing home care at any one point in time depends on rates of admission and length of stay. Information on both is therefore needed for planning purposes. Additionally, although an individual's length of stay in residential or nursing home care will depend on his or her particular health status and circumstances, people making long term financial plans for themselves or their families may also wish to know about average durations of stay in institutional care. However, there are few sources of recent nationally representative data on admissions to and exits from residential and nursing homes and other long-term care facilities which allow durations of stay to be calculated. In this paper data from the Office for National Statistics Longitudinal Study (ONS LS) are used to partially fill this gap. I present information on the survival of older people who in the 2001 Census were recorded as residents of residential care homes, nursing homes, or other types of communal establishment. I then examine differentials in the survival of this population by characteristics including broad type of establishment (residential, nursing, or other); gender and marital status in 2001. Finally I use information on place of death to assess the assumption that residents in communal establishments of various types in 2001 remained in institutional care throughout the follow-up period (from the 2001 Census to the end of 2008).
Email: emily.grundy@lshtm.ac.uk