The Irish Longitudinal Study On Ageing (TILDA)

https://tilda.tcd.ie/

Overview

The Irish Longitudinal Study on Ageing (TILDA) is a major inter-institutional initiative led by Trinity College Dublin (TCD) which aims to produce a massive improvement in the quantity and quality of data, research and information relating to older people and ageing in Ireland. It involves detailed interviews on a two yearly basis with a cohort of people aged 50 and over, charting their health, social and economic circumstances over a 10-year period. The study aims to determine: (1) the health, social and economic status and needs of older people, (2) the biological and environmental components of “successful ageing”, (3) the contributions that older people are making to society, and (4) how each of these key components interact and inform public health and social policy ensuring that Ireland meets the needs and choices of its citizens.

Data Collection

Wave 1 of TILDA was completed between October 2009 and February 2011 and collected data from a cohort of 8,504 respondents residing in the Republic of Ireland aged 50 years and over. Households were selected in geographic clusters from a list of all residential addresses in Ireland. Each selected household was visited by an interviewer and residents aged 50 or older, as well as their spouses or partners, were invited to participate. The household response rate was 62.0% (Whelan & Savva, 2013). Each respondent provided written informed consent. Those with cognitive impairments that prevented meaningful consent being given, were not included in the study. Respondents were interviewed in their homes by trained professional interviewers on many aspects of health, lifestyle, social interactions, and financial circumstances. They were also asked to answer a self-completion questionnaire (SCQ) and 85% returned the SCQ by post at wave 1. Each respondent was then invited to travel to one of two health centers, in Dublin and Cork, for a comprehensive health assessment or received a home health assessment; 72% of participants underwent a health assessment at wave 1. Venous blood (plasma and Buffy coats) from 69% of wave 1 participants was stored at -80^oC at Trinity Biobank for future analytics. The sampling procedure, the home interview, and the health assessments have all been described in detail (Kearney et al., 2011a; Kearney et al., 2011b).

Wave 2 of TILDA was completed between February 2012 and March 2013 and included CAPI data from 7,375 self-respondents, a response rate of 88% of eligible and cognitively capable respondents, with 85% of these respondents returning an SCQ. This wave did not include a comprehensive health assessment. However, 97% of CAPI respondents performed the Timed Up and Go (TUG) and hand grip strength (baseline dynamometer reading on dominant hand) tests during the home interview.

Wave 3 of the study was completed between June 2014 and December 2015 and involved 6,566 self-respondents, a response rate of 85%, with 85% of these returning an SCQ. All participants at wave 3 were invited to undergo a health assessment, which was either a full assessment at a center in Dublin, or a partial assessment in their own homes; 81% of participants underwent a comprehensive health assessment. The various clinical measurements taken at the health center were repeated from wave 1 with the inclusion of some new additional function and performance tests. Venous blood (whole blood, plasma and Buffy coat) from 75% of wave 3 participants was stored at -80^oC at Trinity Biobank for future analytics. PAXgene RNA blood samples were also collected from 874 participants and stored at -80^oC for future analyses at wave 3.

Wave 4 of TILDA was completed between January and December 2016 and included 5,856 self-respondents, a response rate of 84%, with 86% of these returning an SCQ. This wave did not include a comprehensive health assessment. However, 91% of CAPI respondents performed the Timed Up and Go (TUG) and 95% performed the hand grip strength (baseline dynamometer reading on dominant hand) test during the home interview.

Wave 5 of TILDA was completed between March and December 2018 and included 5,103 self-respondents, a response rate of 84%, with 86% of these returning an SCQ. This wave did not include a comprehensive health assessment. However 89% of CAPI respondents performed the Timed Up and Go (TUG) and 91% performed the hand grip strength (baseline dynamometer reading on dominant hand) test during the home interview.

Wave 6 (2020) will be undertaken in 2020. This wave will incorporate the CAPI, SCQ and a comprehensive health assessment. A replenishment of the TILDA cohort will also be undertaken with a target recruitment sample of 2,000 respondents aged 50-59 years.

Biomarkers

TILDA biomarker data collection was undertaken during the comprehensive health assessments at Wave 1 and 3. The Timed Up and Go (TUG) and hand grip strength (baseline dynamometer reading on dominant hand) were assessed during the home interviews at waves 2, 4 and 5.

The biomarkers collected to date are outlined in Tables1 and 2. TILDA has been designed such that many biomarkers map onto similar international studies of older adults such as the Health and Retirement Study (HRS), the Survey of Health, Ageing and Retirement in Europe (SHARE) and the English Longitudinal Study of Ageing (ELSA).

Publications

A description of the TILDA cohort profile is available in the following publications:

Donoghue, O. A., McGarrigle, C. A., Foley, M., Fagan, A., Meaney, J., & Kenny, R. A. (2018). Cohort Profile Update: The Irish Longitudinal Study on Ageing (TILDA). International journal of epidemiology, 47(5), 1398-1398l.

Kearney, P. M., Cronin, H., O’regan, C., Kamiya, Y., Savva, G. M., Whelan, B., & Kenny, R. (2011). Cohort profile: the Irish longitudinal study on ageing. International journal of epidemiology, 40 (4), 877-884.

Data Access

Table 1. Tests included in the TILDA health assessment at each health wave
Neuropsychological function	Gait and mobility
Mini-Mental State Examination (MMSE)^1,a	Timed up-and-go (TUG)^1,3,d
Montreal Cognitive Assessment (MOCA)^1,3	Repeated chair stands³
Sustained Attention to Response Task (SART)^1,3	GAITRite assessment
Picture memory test¹	– normal pace^1,3,c
National Adult Reading Test (NART)³	– dual task (manual)^1,c
Visual reasoning¹	– dual task (cognitive)^1,3,c
Choice reaction time^1,3	– maximum pace^3,c
Colour trails test^1,3
Centre for Epidemiological Studies Depression (CES-D) 8-item scale^1,3,b	Blood Biomarkers^1,3
State trait anxiety inventory³	Lipid profile^1,3
	HbA1c^1,3
Cardiovascular function	C-reactive protein¹
Blood pressure (Omron)^1,3	Creatinine¹
Pulse wave velocity^1,3,c	Cystatin C¹
Heart rate variability^1,3,c	Folate¹
Phasic blood pressure^1,3,c	Vitamin B12¹
Near-infrared spectroscopy (NIRS)^3,c	Vitamin D¹
	Lutein¹
Sensory function	Zeaxanthin¹
Visual acuity^1,3,c	Interluekin-18^1e
Contrast sensitivity^1,3,c	Complement Factor 3^1e
Macular pigment optical density^1,c
Retinal photograph^1,3,c	Hair sample³
Sound-induced flash illusion (SHAMS)^3,c	Cortisol³
	Cortisone³
Physical Activity	Testosterone³
Physical activity question^3,c	Progesterone³
Accelerometry^3,g	Dehydroepiandrosterone³

Other	Genetics/Genomics
Oral health^3,c,g	Genomic DNA extracted¹
MRI^3,c,e	7 Age-related Macular Degeneration candidate SNPs genotyped¹
	12 Blood pressure SNPs genotyped¹
Anthropometrics
Grip strength^1,3,d	Epigenetics
Height^1,3	DNA methylation ^1,e
Weight^1,3
Waist and hip circumference^1,3	Transcriptomics
	PAXgene RNA Blood Tubes^3,e
Bone health
Heel bone ultrasound^1,3,c

^1,3 Indicate the wave in which data were collected (i.e. Wave 1, Wave 3).
^a MMSE was included in Waves 2, 3 and 4 CAPI.
^b CESD-20 was included in Waves 1 and 2 CAPI; CESD-8 was included in Waves 3 and 4 CAPI.
^cIncluded in the health centre assessment only.
^dTUG and grip strength were included in Waves 2, 4 and 5 CAPI.
^e Included for a sub-study only, due to resource availability (i.e. limited accelerometers, availability of dentist, limited MRI funding).

Table 2. New health assessment tests introduced in Wave 3
Test or equipment used	Measures
National Adult Reading Test (NART) 50 words	Pre-morbid intelligence
State trait anxiety inventory (STAI) 20-item scale	Transient worry and heightened emotionality
Near-infrared spectroscopy (NIRS) (PortaMon, Artinis Medical Systems, The Netherlands) –non-invasive sensor on prefrontal cortex (approximate EEG site FP1)	Trends in regional tissue oxygen saturations providing an indirect measure of cerebral perfusion
Sound-induced flash illusion (SHAMS test)	Multisensory integration i.e. the ability of the nervous system to integrate sensory information obtained from auditory and visual sources
Repeated chair stands 5 times sit to stand timed with stopwatch	Lower limb strength, coordination and endurance
Gait assessment (maximum pace) GAITRite^®walkway, sensored area = 4.88 m	Maximum walking speed
Self-reported physical activity question used in an algorithm along with age, sex, BMI, resting heart rate	Cardiorespiratory fitness
Hair samples minimum length 3 cm, weight 10 mg	Cortisol secretion and other stress hormones
Accelerometers (GENEActiv Original Accelerometers, Activinsights Ltd, UK) worn on non-dominant wrist where possible; 24 , 7 days^a	Amount, intensity and duration of activity and rest
Oral health assessment conducted by dentist^a	Oral health indicators e.g. number of teeth, gum and tooth health etc.
Magnetic resonance imaging (MRI)^a 32-channel head detector coil on a 3 T MRI scanner (Achieva TX, Philips, The Netherlands)	Anatomical and functional brain images

^aMRI, accelerometry and the oral health assessment were provided to a random subsample of participants due to resource availability (i.e. limited funding, limited accelerometers, availability of dentist).