Cohort Profile: The Irish Longitudinal Study on Ageing

Abstract

Ireland shares with other developed countries the prospect of rapid and sustained population ageing. The age distribution of the Irish population is undergoing a dramatic change at present and this trend is predicted to continue into the future.1 People are living longer, and older persons represent a larger proportion of the population. In Ireland, the proportion of the population aged ≥65 years has remained steady at ∼11% for the past 40 years. However, it is projected that this proportion will rise to 14% by 2021 and to 19% by 2031.2 The greatest increase will be in the oldest old, aged >80 years, which is expected to more than treble by 2036.2 This change in the demographic profile of the Irish population poses a major public health challenge. Unlike the situation in the USA,3,4 the UK5 and many other developed countries,6,7 there have been no large population-based cohort studies in Ireland to inform research on healthy ageing. Whereas a number of studies have provided population-based data on the health status of older people living in Ireland,8–10 many questions remain unanswered. The Irish Longitudinal Study on Ageing (TILDA) is a large prospective cohort study of ageing, which includes an assessment of the social, economic and health circumstances of community-resident older people living in Ireland. The study has been harmonized with leading international research so as to ensure adoption of best practice and comparability of results. The Irish government, The Atlantic Philanthropies and Irish Life plc have provided funding for the study. Ethical approval has been obtained from the Trinity College Dublin Research Ethics Committee. Recruitment for the first wave of TILDA began in October 2009 and fieldwork was completed in February 2011 when the target sample of more than 8000 participants had been achieved. The health assessment component (described below) is ongoing and will be completed by July 2011. The sampling frame is the Irish Geodirectory, a comprehensive and up-to-date listing and mapping of all residential addresses in the Republic of Ireland compiled by ‘an Post’ (the Irish Postal Service) and Ordnance Survey Ireland.11 An initial multi-stage probability sample of addresses was chosen by means of the RANSAM sampling procedure,12 developed by the Economic and Social Research Institute. Addresses were selected by means of a three stage process: (i) selection of first stage units (clusters which are subdivisions of District Electoral Divisions comprising between 500 and 1180 addresses) using proportionate stratification by socio-economic status (per cent in professional/managerial occupations), age structure (per cent of population aged ≥50 years) and geographical location. Selection of first stage units was based on probability proportionate to size, the size measure used being the estimated number of addresses containing a person aged ≥50 years in the cluster. (ii) Selection of a systematic random sample of fixed size (50 addresses) within each cluster. Each residential address in the country had an equal probability of selection. The selected addresses were randomly partitioned into two groups: an initial sample list of 25 600 addresses (40 addresses in each of the 640 clusters, Figure 1) for immediate issue and a reserve list of 6400 addresses (10 randomly selected from each of the 640 clusters). As the target sample size was achieved using the initial sample list, the reserve list was not utilized. A listing and location map of the selected residential addresses were provided to the social interviewers on Global Positioning System (GPS) devices which also had the coordinates of the selected addresses. Selected households with a unique address were sent a letter of invitation to participate in the study by post. Advance letters were hand delivered by the interviewers to those households with a non-unique address (∼40% of Irish addresses are non-unique and there is no national postcode or zipcode system). The letter of invitation was followed up after 1 week by a home visit from a member of the field staff to determine whether there was an eligible person living in the household. (iii) All household residents aged ≥50 years were eligible to participate in the study (primary respondent). The spouses/partners (of any age) of respondents were also invited to participate (secondary respondent). The sample was designed to give each household in the country an equal probability of selection and, since all members aged ≥50 years in each household were eligible, each person aged ≥50 years also had an equal probability of selection. The spouses aged <50 years were interviewed mainly to provide couple- or household-level data and will not in general be included in person-level analyses. Map of selected addresses—the location of the 640 clusters The study aims are broad and include: the description of the current economic, social and health circumstances of older people living in Ireland and the c