Time to move? Factors associated with burden of care among informal caregivers of cognitively impaired older people facing housing decisions: secondary analysis of a cluster randomized trial

Table 3 Two-level multiple regression analysis of factors associated with caregiver’s burden of care

Cluster-level (HSSCs random effect at intercept level: ICC = 13%)
Caregiver-level (Marginal R² = 26%)^a
Independent variables	β (95% CI)	p-value
Characteristics of the caregiver
Sex
Men	Reference	0.04
Women	4.15 (0.30, 7.99)
Caregiver’s housing preference for the cognitively impaired older person
Public nursing home	Reference	0.01
Stay at home	3.5 (−1.59, 8.59)
Caregiver’s home	9.18 (1.93, 16.44)
Private retirement residence	7.04 (1.76, 12.32)
Other (e.g. mixed private-public nursing homes)	11.22 (2.68, 19.77)
Time lapse (days) since the decision was made	−0.02 (−0.04, 0)	0.03
Decision regret^b
(< 5)	Reference	< 0.001
(≥5 à ≤25)	4.73 (0.76, 8.69)
(≥30)	10.7 (5.07, 16.33)
Decisional conflict^c	0.15 (0.03, 0.28)	0.02
Characteristics of the relationship between the caregiver and cognitively impaired older person
Relationship link between caregiver and the cognitively impaired older person
Other family member	Reference	< 0.001
Child	7.51 (2.56, 12.46)
Friend or other	6.78 (−5.09, 18.65)
Spouse	17.69 (11.83, 23.54)
Caregiver’s social support resources perception
The caregiver’s perception of the occurrence of a joint process in the decision-making (D-OPTION)^d	0.26 (0.17, 0.36)	< 0.001

HSSCs: health and social services centres; ICC: intra-class correlation coefficients; D-OPTION: Dyadic Observing Patient Involvement in Decision Making instrument
^aTotal variance explained by the model = 39%
^bDecisional conflict score < 5 = no regret, ≥5 to ≤25 = mild regret and ≥ 30 = moderate to strong regret [31]
^cFor one point increase score from 0 (no decisional conflict) to 100 (high decisional conflict)
^dfor one point increase, score from 0 = decision-making not joint to 100 = decision-making joint

ISSN: 1471-2318