Formal health care costs among older people in Ireland: methods and estimates using The Irish Longitudinal Study on Ageing (TILDA) [version 1; peer review: 3 approved, 1 approved with reservations]

Background: Reliable data on health care costs in Ireland are essential to support planning and evaluation of services. New unit costs and high-quality utilisation data offer the opportunity to estimate individual-level costs for research and policy.

Methods: Our main dataset was The Irish Longitudinal Study on Ageing (TILDA). We used participant interviews with those aged 55+ years in Wave 5 (2018) and all end-of-life interviews (EOLI) to February 2020. We weighted observations by age, sex and last year of life at the population level. We estimated total formal health care costs by combining reported usage in TILDA with unit costs (non-acute care) and public payer reimbursement data (acute hospital admissions, medications). All costs were adjusted for inflation to 2022, the year of analysis. We examined distribution of estimates across the population, and the composition of costs across categories of care, using descriptive statistics. We identified factors associated with total costs using generalised linear models.

Results: There were 5,105 Wave 5 observations, equivalent at the population level to 1,207,660 people aged 55+ years and not in the last year of life, and 763 EOLI observations, equivalent to 28,466 people aged 55+ years in the last year of life. Mean formal health care costs in the weighted sample were EUR 8,053; EUR 6,624 not in the last year of life and EUR 68,654 in the last year of life. Overall, 90% of health care costs were accounted for by 20% of users. Multiple functional limitations and proximity to death were the largest predictors of costs. Other factors that were associated with outcome included educational attainment, entitlements to subsidised care and serious chronic diseases.

Conclusions: Understanding the patterns of costs, and the factors associated with very high costs for some individuals, can inform efforts to improve patient experiences and optimise resource allocation.