Investigating the interplay between frailty and healthcare utilization in older adults

Frailty, a syndrome characterized by physiological decline and below average resilience to stressors, often results in clinical complexity and poor quality of life. In this thesis, we examine how frailty has evolved over time in the older population, its association with acute clinical events, and how it relates to avoidable hospitalizations and late-life healthcare use. We used data from the longitudinal population-based Swedish National study on Aging and Care in Kungsholmen (SNAC-K), involving 3363 adults aged 60 and above.

In Study I, frailty was measured with a data-driven 40 deficit frailty index (FI) among 2941 participants over the first six SNAC-K waves (median 9.6-year follow-up). Temporal trends in frailty state transitions were investigated, and life expectancies from age 60 in frailty states across five birth years (1900, 1910, 1920, 1930, 1940) were predicted. We found that those born more recently were less likely to transition from mild as well as moderate and severe frailty to death. Moreover, the number and proportion of years lived with frailty after age 60 expanded and occurred later in life in more recent years.

Including 3146 SNAC-K participants, Study II explored the relationship between frailty trajectories, measured with a data-driven 40 deficit FI over five waves of SNAC-K, and acute clinical events, specifically MIs (myocardial infarctions), LRTIs (lower respiratory tract infections), and falls, resulting in hospitalization from five years before to twelve years after SNAC-K baseline (median follow-up 11 years). We found higher levels of frailty from age 75 to 95 with each additional acute event. The frailty trajectories differed most by the number of falls, with greater differences in the trajectories by number of MIs and LRTIs once we imputed frailty at death.

Study III examined the association between physical frailty status and incident avoidable hospitalizations (median 7.6-year follow-up), defined as hospitalizations for specific chronic or acute conditions that could have been prevented through adequate outpatient care. Among the included 3168 SNAC-K participants, we found the relative and absolute risk of avoidable hospitalization to be higher among those who were pre-frail and frail. Moreover, we found this association to be greater among older adults aged <78 and those with relatively fewer chronic diseases, but the association diminished among those receiving formal social care.

Study IV included 1467 participants that died in the first 15 years of SNAC-K. Latent class trajectories of walking speed, as a proxy of frailty, were generated based on the first five waves of SNAC-K (median 11-year follow-up) and the associations of the different classes with healthcare utilization (total and avoidable hospitalizations, outpatient specialist care visits, days in hospital) in the last year of life were evaluated. We identified three classes: 1) 'faster speed and gradual decline', 2) rapidly declining', and 3) 'slower speed and gradual decline.' Inpatient hospitalizations and length of stay were less frequent in the slower speed and gradual decline class, avoidable hospitalization rates were higher in the rapidly declining class, and outpatient specialist care was less common in both, compared to the faster speed and gradual decline class. We found evidence that avoidable hospitalization rates were mitigated by informal and formal social care.

With trends of living longer with frailty, preventing its progression at earlier stages is essential. Prevention of acute clinical events and frailty should be promoted, and provision of informal and formal social care could reduce excessive healthcare utilization among frail adults. Further investigations should explore whether and how the care of individuals with frailty can be optimized.

Keywords: frailty, health trends, myocardial infarction, lower respiratory tract infections, falls, hospitalization, avoidable hospitalization, healthcare utilization, walking speed, end of life

List of scientific papers

I. Tazzeo C, Rizzuto D, Calderón-Larrañaga A, Dekhtyar S, Zucchelli A, Xia X, Fratiglioni L, Vetrano DL. Living longer but frailer? Temporal trends in life expectancy and frailty in older swedish adults. J Gerontol A Biol Sci Med Sci. 2024;79(1):glad212. https://doi.org/10.1093/gerona/glad212

II. Tazzeo C, Gregorio C, Rizzuto D, Fratiglioni L, Maggi S, Welmer A-K, Zucchelli A, Calderón-Larrañaga A, Vetrano DL. Acute clinical events and trajectories of frailty after age 60: A population-based cohort study. [Manuscript]

III. Tazzeo C, Rizzuto D, Calderón-Larrañaga A, Gentili S, Lennartsson C, Xia X, Fratiglioni L, Vetrano DL. Avoidable hospitalizations in frail older adults: The role of sociodemographic, clinical, and care-related factors. J Am Med Dir Assoc. 2024;25(11):105225. https://doi.org/10.1016/j.jamda.2024.105225

IV. Tazzeo C, Gregorio C, Rizzuto D, Fratiglioni L, Gentili S, Welmer A-K, Calderón-Larrañaga A, Vetrano DL. Trajectories of walking speed after age 60 and healthcare utilization in the last year of life. [Manuscript].