Association between frailty and incident risk of disability in community-dwelling elder people: evidence from a meta-analysis

• Published in: Public health (London) Vol. 175; pp. 90 - 100
• Main Authors: Liu, H.X., Ding, G., Yu, W.J., Liu, T.F., Yan, A.Y., Chen, H.Y., Zhang, A.H.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ltd 01.10.2019; Elsevier Science Ltd
• Subjects: Aged; Analysis; Bias; Candidates; Cohort analysis; Cohort studies; Communities; Comorbidity; Confidence intervals; Disability; Disability risk; Disabled Persons - statistics & numerical data; Education; Frail; Frail Elderly - statistics & numerical data; Frailty; Frailty - epidemiology; Geriatrics; Heterogeneity; Humans; Incidence; Independent Living - statistics & numerical data; Meta-analysis; Older people; Prevention programs; Risk analysis; Risk Factors; Statistical analysis; Subgroups; Test procedures; Frailty; Disability risk; Cohort studies; Meta-analysis
• ISSN: 0033-3506; 1476-5616; 1476-5616
• DOI: 10.1016/j.puhe.2019.06.010

Frailty is considered to be one of the risk factors of disability. However, the results of original reported studies are not consistent with respect to the frailty and incidence of disability, and previously published meta-analyses have also shown inconsistent results. This meta-analysis was conducted to investigate the relationship between the different stages of frailty and the incidence of disability by examining updated overall trends in community-dwelling elders. Cohort studies in English or Chinese based on associations between frailty and incident disability risks that were published from 2000 until the current date were researched using PubMed, Embase, Web of Science, and CENTRAL databases. The Q test and I2 statistic were used to examine between-study heterogeneity. Random-effect models were adopted to synthesize the results based on the study heterogeneity. Subgroup analyses were also conducted to explore the possible sources of between-study heterogeneity based on the characteristics of participants. Eighteen cohort studies with 88,906 participants were included in our meta-analyses. Compared with the non-frailty category, the combined relative risks (RRs) (95% confidence interval [CI]) of the disability were 1.66 (1.49–1.85) and 2.53 (2.01–3.14) for the category of prefrailty and frailty, respectively. Results suggested that the incident risk of disability at follow-up times <5 (RR = 3.19, 95% CI = 2.25–4.53) was significantly higher than for follow-up times ≥5 in the frailty category (RR = 2.00, 95% CI = 1.55–2.56). The risk in a sample size of ≥1000 (RR = 2.78, 95% CI = 2.04–3.14) was significantly higher than that when the sample size was <1000 (RR = 1.91, 95% CI = 1.53–2.37) in the frailty group. Compared with a value adjusted for comorbidity, the unadjusted comorbidity was significantly higher in the prefrailty category (1.90 vs. 1.52). Compared with a value adjusted for education, the unadjusted education was significantly higher in the prefrailty category (1.81 vs. 1.46). No publication bias was observed. The overall meta-analysis confirms that frailty has significantly increased the incident risk of disability. Frail, elderly people are at the highest risk of future disability and may be adequate candidates for taking part in prevention and intervention programs. •Prospective design, large number of subjects living in the community, ability to examine diverse medical conditions and factors previously reported as being associated with disability, complete follow-up.•Our study excluded the poorer methodological quality studies that were potentially one of the causes of heterogeneity, and included four high-quality cohort literatures published after 2017.•We conducted synthesized pooled risk estimate analysis about frailty and disability.