Table 5.

Associations with cost of care in community-dwelling persons with dementia: societal perspective based on the GLM

	B	Wald	p	EXP (B)	95% Wald Confidence Interval for Exp (B)
					Lower	Upper
Amount of caregivers	−0.037	0.846	0.358	0.964	0.892	1.042
Gender of the caregiver
[gender = male]	−0.137	1.807	0.179	0.872	0.714	1.065
[gender = female]	Refa			1
Gender of the person with dementia
[gender = male]	−0.015	0.018	0.892	0.985	0.791	1.227
[gender = female]	Refa			1
Educational level caregiver in 3 categories
[Educational level caregiver = low]	−0.193	4.695	0.030b*	0.824	0.692	0.982
[Educational level caregiver = moderate]	−0.329	15.065	0.000*	0.720	0.610	0.850
[Educational level caregiver = high]a	Refa			1
Professional situation in 3 categories
[Professional situation = active]	0.429	11.921	0.001*	1.535	1.204	1.958
[Professional situation = nonactive]	0.311	7.270	0.007*	1.365	1.089	1.712
[Professional situation = retired]a	Refa			1
Relationship between caregiver and person with dementia
[Relationship = partner/spouse]	−0.090	0.294	0.588	0.914	0.659	1.267
[Relationship = daughter/son]	−0.335	5.031	0.025b*	0.715	0.533	0.959
[Relationship = other family member/friend]a	Refa			1
Person with dementia level of dependency - KATZ scale	0.069	126.625	0.000*	1.072	1.059	1.085
Age of the person with dementia	−0.002	0.197	0.657	0.998	0.988	1.007
Frequency of behavioral problems	0.060	0.628	0.428	1.062	0.915	1.233
Impact of behavioral problems on caregiver	0.019	0.137	0.712	1.019	0.921	1.127
Caregiver intention to institutionalize	0.023	1.077	0.229	1.023	0.98	1.068
Caregiver quality of life - derived from EQ5D	−0.515	8.980	0.003*	0.597	0.426	0.837

^areference category, *significant different, ^bnot significant in the linear regression with bootstrapping