Cross-sectional association between medical expenses and intellectual activity in community-dwelling older adults

Table 3 Prevalence ratios for impaired IA (IA score ≤ 3) associated with medical expenses

	Prevalence of impaired IA	Crude	Model 1^a	Model 2^b	Model 3^c	Model 4^d	Model 5^e
		(n = 12,458)	(n = 12,305)	(n = 11,679)	(n = 11,498)	(n = 11,228)	(n = 11,318)
		PR (95% CI)	PR (95% CI)	PR (95% CI)	PR (95% CI)	PR (95% CI)	PR (95% CI)
Medical expenses in the past year
Low	25.5%	1.00	1.00	1.00	1.00	1.00	1.00
Medium	25.1%	0.98 (0.91–1.06)	0.98 (0.90–1.05)	0.93 (0.86–1.00)	0.94 (0.87–1.01)	0.94 (0.87–1.02)	0.93 (0.86–1.01)
High	32.1%	1.26 (1.17–1.35)	1.16 (1.08–1.25)	0.97 (0.90–1.05)	0.99 (0.91–1.07)	0.96 (0.89–1.04)	0.98 (0.90–1.06)
None	32.5%	1.27 (1.16–1.40)	1.23 (1.11–1.35)	1.19 (1.08–1.31)	1.08 (0.98–1.20)	1.12 (1.01–1.24)	1.14 (1.03–1.26)
Health behaviors
Health checks: no (ref: yes)					1.07 (1.003–1.14)
Smoking: current (ref: never/ex-smoker)					1.28 (1.20–1.38)
Frequency of exercise (h/w): < 1 (ref: ≥ 1)					1.46 (1.38–1.56)
Dietary variety: low (ref: high)					1.27 (1.19–1.35)
Oral health
Use of extra cleaning devices: no (ref: yes)						1.16 (1.09–1.24)
Bedtime brushing: not daily (ref: daily)						1.17 (1.10–1.26)
Difficulty with chewing hard foods: yes (ref: no)						1.10 (1.02–1.17)
Use of dentures: yes (ref: no)						1.02 (0.96–1.09)
Social capital
Social participation: no (ref: yes)							1.44 (1.35–1.53)
Social support: no (ref: yes)							1.19 (1.10–1.29)
Social networks: no (ref: yes)							1.28 (1.18–1.38)

PRs and 95% CIs were calculated using Poisson regression analyses
CI confidence interval, IA intellectual activity, ref. reference, PR prevalence ratio
^aModel 1 is adjusted for age, gender, health insurance, accessibility to public facilities in their residential area, family size, and body mass index
^bModel 2 is adjusted for the covariates in Model 1 plus physical and mental functioning (self-rated health, health-related QOL (PCS and MCS), sleep disturbance, depression, and cognitive functioning)
^cModel 3 is adjusted for the variables in Model 2 plus health behaviors (health checks, smoking, frequency of exercise, and dietary variety)
^dModel 4 is adjusted for the variables in Model 2 plus oral health (use of extra cleaning devices, frequency of bedtime brushing, difficulty with chewing hard foods, and use of dentures)
^eModel 5 is adjusted for the variables in Model 2 plus social capital (social participation, social support, and social network)

ISSN: 1347-4715