Dodds, Richard M; Bunn, Jonathan G; Hillman, Susan J; Granic, Antoneta; Murray, James; Witham, Miles D; Robinson, Sian M; Cooper, Rachel; Sayer, Avan A; (2023) Simple approaches to characterising multiple long-term conditions (multimorbidity) and rates of emergency hospital admission: Findings from 495,465 UK Biobank participants. Journal of internal medicine, 293 (1). pp. 100-109. ISSN 0954-6820 DOI: https://doi.org/10.1111/joim.13567
Permanent Identifier
Use this Digital Object Identifier when citing or linking to this resource.
Abstract
BACKGROUND: Numerous approaches are used to characterise multiple long-term conditions (MLTC), including counts and indices. Few studies have compared approaches within the same dataset. We aimed to characterise MLTC using simple approaches, and compare their prevalence estimates of MLTC and associations with emergency hospital admission in the UK Biobank. METHODS: We used baseline data from 495,465 participants (age 38-73 years) to characterise MLTC using four approaches: Charlson index (CI), Byles index (BI), count of 43 conditions (CC) and count of body systems affected (BC). We defined MLTC as more than two conditions using CI, BI and CC, and more than two body systems using BC. We categorised scores (incorporating weightings for the indices) from each approach as 0, 1, 2 and 3+. We used linked hospital episode statistics and performed survival analyses to test associations with an endpoint of emergency hospital admission or death over 5 years. RESULTS: The prevalence of MLTC was 44% (BC), 33% (CC), 6% (BI) and 2% (CI). Higher scores using all approaches were associated with greater outcome rates independent of sex and age group. For example, using CC, compared with score 0, score 2 had 1.95 (95% CI: 1.91, 1.99) and a score of 3+ had 3.12 (95% CI: 3.06, 3.18) times greater outcome rates. The discriminant value of all approaches was modest (C-statistics 0.60-0.63). CONCLUSIONS: The counts classified a greater proportion as having MLTC than the indices, highlighting that prevalence estimates of MLTC vary depending on the approach. All approaches had strong statistical associations with emergency hospital admission but a modest ability to identify individuals at risk.
Item Type | Article |
---|---|
Faculty and Department | Faculty of Epidemiology and Population Health > Dept of Medical Statistics |
PubMed ID | 36131375 |
Elements ID | 213983 |
Official URL | http://dx.doi.org/10.1111/joim.13567 |
Download
Filename: Dodds-etal-2022-Simple-approaches-to-characterising-multiple-long-term-conditions.pdf
Licence: Creative Commons: Attribution 4.0
Download