Table 3. Multinomial logistic regression analysis result for association of frailty subtypes with factors^@.

Frailty subtypes	Variables	Walds	P value	OR	95% CI
Multi-frail	Intercept	30.57	<0.001
	Age
	80~	24.07	<0.001	6.43	3.06~13.51
	60~*	.	.	1	.
	Educational level
	Illiteracy	4.25	0.04	2.59	1.05~6.41
	Junior high school and above	.	.	1	.
	Marital status
	Widowed / Divorced / Unmarried	16.87	<0.001	2.88	1.74~4.76
	Married*	.	.	1	.
	Relationship with family
	Harmonious	11.41	0.001	0.21	0.08~0.52
	Not harmonious*	.	.	1	.
	Exercise frequency
	Every day	11.30	0.001	0.39	0.23~0.68
	Never exercised*	.	.	1	.
	Physical activity level
	Low level	52.02	<0.001	35.30	13.40~92.98
	Medium level	10.26	0.001	4.94	1.86~13.12
	High level*	.	.	1	.
Cognitive and functionally frail	Intercept	35.34	<0.001
	Age
	80~	127.11	<0.001	11.72	7.64~17.98
	70~	21.14	<0.001	2.28	1.60~3.23
	60~*	.	.	1	.
	Sex
	Male	8.63	0.003	0.59	0.42~0.84
	Female*	.	.	1	.
	Educational level
	Illiteracy	6.47	0.01	1.86	1.15~3.01
	Junior high school and above	.	.	1	.
	Marital status
	Widowed / Divorced / Unmarried	12.21	<0.001	1.67	1.25~2.22
	Married*	.	.	1	.
	Housing satisfaction
	Dissatisfied	4.80	0.03	2.93	1.12~7.66
	Satisfied*	.	.	1	.
	Whether pensions are the main source of income
	Yes	13.20	<0.001	1.58	1.24~2.03
	No*	.	.	1	.
	Exercise frequency
	Every day	20.18	<0.001	0.51	0.38~0.68
	Never exercised*	.	.	1	.
	Physical activity level
	Low level	39.69	<0.001	3.35	2.30~4.88
	Medium level	12.55	<0.001	1.73	1.28~2.35
	High level*	.	.	1	.
Psychologically frail	Intercept	34.12	<0.001
	Age
	80~	6.34	0.01	1.64	1.12~2.41
	60~*	.	.	1	.
	Sex
	Male	9.76	0.002	0.64	0.48~0.84
	Female*	.	.	1	.
	Educational level
	Illiteracy	4.13	0.04	1.48	1.01~2.15
	Junior high school and above*	.	.	1	.
	Marital status
	Widowed / Divorced / Unmarried	18.84	<0.001	1.73	1.35~2.21
	Married*	.	.	1	.
	The number of times children visit per month
	0~2	27.07	<0.001	2.99	1.98~4.51
	3~4	13.45	<0.001	1.99	1.38~2.87
	5~*	.	.	1	.
	Housing satisfaction
	Dissatisfied	19.90	<0.001	4.91	2.44~9.88
	General	20.70	<0.001	1.76	1.38~2.25
	Satisfied*	.	.	1	.
	Whether pensions are the main source of income
	Yes	8.89	0.003	1.36	1.11~1.67
	No*	.	.	1	.
	Physical activity level
	Medium level	14.25	<0.001	1.53	1.23~1.91
	High level*	.	.	1	.
Physiologically frail	Intercept	53.88	<0.001
	Age
	80~	13.51	<0.001	1.74	1.30~2.34
	70~	7.76	0.01	1.28	1.08~1.52
	60~*	.	.	1	.
	Educational level
	Illiteracy	10.72	0.001	1.61	1.21~2.14
	Primary school	9.58	0.002	1.38	1.13~1.69
	Junior high school and above*	.	.	1	.
	Housing satisfaction
	Dissatisfied	7.46	0.01	2.56	1.30~5.03
	General	28.03	<0.001	1.65	1.37~1.99
	Satisfied*	.	.	1	.
	Physical activity level
	Medium level	87.52	<0.001	2.18	1.85~2.56
	High level*	.	.	1	.

^@: Multinomial logistic regression was used to evaluate the associations between the 12 baseline characteristics and the frailty subtypes. “Relatively healthy” was used as a reference class for the four frailty subtypes.

*Reference category of variables.

Abbreviations: OR, odds ratios; CI, confidence interval.