PubMed · 42537597
Beyond multidimensionality: a systematic review of recurrent frailty archetypes in community-dwelling older adults.
Abstract
BACKGROUND: Frailty is a clinically heterogeneous geriatric syndrome commonly summarised using physical or multidomain severity scores. Whether person-centred analyses identify recurring within-frailty configurations has not been systematically examined in community-dwelling older adults. METHODS: We searched PubMed, Embase, MEDLINE, and CINAHL (January 2000-November 2025) for cross-sectional studies using latent class, latent profile, or analogous clustering methods to derive frailty subgroups. Quality was assessed using the AHRQ checklist and a purpose-built appraisal of person-centred model reporting. Study-derived classes were mapped in duplicate to a structured archetype framework developed through comparison of class-defining features across studies. RESULTS: Fourteen reports representing 12 independent datasets from eight countries were included. Six configurations were identified: minimally impaired reference, mobility-physical, nutritional-metabolic, cognitive-predominant, combined cognitive-physical, and psychosocial/mood-predominant. Convergence was measurement-dependent. The reference and mobility-physical configurations recurred across physical-only and multidomain indicator sets, while the combined cognitive-physical configuration appeared across several multidomain frameworks but required cognition to be measured. The remaining configurations emerged only when their defining domains were included. Evidence of prognostic value beyond aggregate frailty severity came from one deficit-index study. Collapsing shared-provenance reports and excluding the boundary-eligible study did not alter recurrence; excluding the Croatian dataset left five configurations recurrent, with the cognitive-predominant configuration supported by one independent dataset. CONCLUSIONS: Person-centred analyses identify recurring within-frailty configurations, but their apparent stability is partly measurement-dependent. A five-configuration core persisted after exclusion of the Croatian dataset, whereas the cognitive-predominant configuration remained weakly replicated. Harmonised indicators and rigorous external validation are needed before clinical application.
Explore related subjects
Keep this discovery
Yu Liang Lim, Hua Wang Tan, Daniel Hengyi Tang, Tran Hien Anh Anna, Chin Yee Cheong, Lian Leng Low, Gerald Choon Huat Koh, Qin Xiang Ng. 2026-07-23. Beyond multidimensionality: a systematic review of recurrent frailty archetypes in community-dwelling older adults.. https://doi.org/10.1016/j.archger.2026.106365
Cite the original work for its findings. Save a collection to share your selection of sources.
Discover connections
Connections use source metadata and explicit phrase matches, not verified experimental comparisons.