Comparative performance of six multimorbidity indices to discriminate intrinsic capacity impairments among older adults

1709
Chen Yang1, Xi Cao1, Yihan Mo2, June Zhang1, Xiuhua Wang3
1 School of Nursing, Sun Yat-sen University, Guangzhou, China; 2 Cicely Saunders Institute of Palliative Care, Policy & Rehabilitation, King’s College London, London, UK; 3 School of Nursing, Central South University, Changsha, China
Measurement of health status

Abstract

Introduction: Optimal intrinsic capacity (IC) is crucial for preserving the functional abilities of older adults. The presence of multimorbidity is closely associated with IC impairments. Various multimorbidity indices have been developed for diverse health outcomes. This study aimed to compare the performance of six commonly used multimorbidity indices to discriminate IC impairments among community-dwelling older adults.

Method: We used data from a multidimensional geriatric assessment program including 627 community-dwelling older adults in five cities of Hunan, China. Six multimorbidity indices were extracted from the data, including disease counts, Functional Comorbidity Index (FCI), the Deyo Charlson comorbidity index, two indices (total score and comorbidity index) derived from the Cumulative Illness Rating Scale-Geriatric (CIRS-G), and medication counts. The IC was measured with five individual domains, i.e., locomotion, vitality, sensory, cognition, and psychological capacity. Individuals were regarded as having IC impairments if they had impairments in one or more domains. Associations between multimorbidity indices and IC impairments were examined using logistic regression analyses. The discriminative ability of multimorbidity indices for IC impairments was compared using the c-statistics.

Results: A total of 374 (59.6%) participants had IC impairments. All multimorbidity indices were significantly associated with IC impairments after adjusting for confounding factors. All indices showed acceptable discriminative power (c-statistic ranged from 0.711 to 0.759) for IC impairments. The comorbidity index derived from CIRS-G resulted in the highest c-statistic, followed by the total score of CIRS-G and FCI.

Conclusions: Our study results suggest that multimorbidity indices differed in their ability to discriminate IC impairments. The comorbidity index derived from CIRS-G performed better than other multimorbidity indices included in this study. The comorbidity index has the potential as a simple proxy measure of indicating the need for interventions to optimise IC for older adults in community settings.