Table 5 A subgroup analysis of 55 participants with prefrailty at baseline using a multivariable linear regression model to estimate the difference between 2014 and 2019 in three component summary scores for each group stratified by the 5-year change in frailty status.

Multivariable regression models	Trends in frailty scores from 2014 to 2019 among participants with prefrailty
	Improved (n = 14)			Maintained (n = 36)			Worsened (n = 5)
	β	95% CI	P-value	β	95% CI	P-value	β	95% CI	P-value
Physical Component Summary score
Model 1: Unadjusted	17.3	3.9 to 30.6	0.012	10.7	−1.4 to 22.9	0.084	Reference	Reference	n/a
Model 2: Age- and sex-adjusted	17.4	4.0 to 31.0	0.012	11.1	−1.2 to 23.4	0.076	Reference	Reference	n/a
Model 3: Fully-adjusted^a	20.2	4.2 to 36.2	0.015	11.8	−1.9 to 25.5	0.090	Reference	Reference	n/a
Mental Component Summary score
Model 1: Unadjusted	−0.1	−11.8 to 11.6	0.98	−3.8	−14.5 to 6.9	0.48	Reference	Reference	n/a
Model 2: Age- and sex-adjusted	0.3	−11.5 to 12.1	0.96	−3.5	−14.3 to 7.3	0.52	Reference	Reference	n/a
Model 3: Fully-adjusted^a	0.6	−12.7 to 13.8	0.93	−4.4	−15.7 to 6.9	0.44	Reference	Reference	n/a
Role/Social Component Summary score
Model 1: Unadjusted	22.0	9.3 to 34.6	0.001	17.0	5.4 to 28.5	0.005	Reference	Reference	n/a
Model 2: Age- and sex-adjusted	22.2	9.2 to 35.1	0.001	17.1	5.3 to 28.9	0.005	Reference	Reference	n/a
Model 3: Fully-adjusted^a	20.8	5.5 to 36.2	0.009	16.7	3.6 to 29.9	0.009	Reference	Reference	n/a

^aThe fully adjusted model for multivariable regression analysis includes age, sex, body mass index, hypertension, diabetes mellitus, chronic kidney disease, cardiovascular disease, and cancer as covariates.
CI, confidence interval; n/a, not applicable.

Quick links

Search