Patient and Caregiver Outcomes of Health System, Community-Based, and Usual Dementia Care: A Prespecified Analysis of the Dementia Care Study (D-CARE) Randomized Clinical Trial.

Importance: Despite the large numbers of persons living with dementia, the best approach to providing dementia care is unknown. Objective: To compare the effectiveness of health system dementia care (HSDC), community-based dementia care (CBDC), and usual care (UC) on person living with dementia and caregiver outcomes. Design, Setting, and Participants: The Dementia Care Study (D-CARE) was a pragmatic, 18-month, 3-armed, assessor-blinded randomized clinical trial conducted from June 2019 through August 2023 at 4 clinical trial sites in North Carolina, Texas, and Pennsylvania. Person living with dementia-caregiver dyads were included. Data were analyzed from December 2024 to June 2025. Interventions: HSDC comanagement by nurse practitioners or physician assistants or CBDC provided telephonically by a social worker, nurse, or licensed therapist for 18 months. Main Outcomes and Measures: Prespecified outcomes included person living with dementia cognition, functional status, and quality of life; caregiver ratings of quality of care (10 items; range, 0-10) and satisfaction with care (11 items; range, 11-55), including how helpful the dementia care was, access to services, and support; positive aspects of caregiving; and overall caregiver burden as well as a measure of whether either the person living with dementia or caregiver benefitted. Results: A total of 2176 person living with dementia-caregiver dyads were enrolled; 1271 persons living with dementia (58.4%) and 1650 caregivers (75.8%) were female, and their mean (SD) age were 80.6 (8.5) years and 65.2 (12.3) years, respectively. There were no treatment differences between groups in person living with dementia functional status, cognition, or quality of life or in overall caregiver burden or positive aspects of caregiving. Caregiver satisfaction with care was higher with both interventions compared with UC (HSDC: least-squares mean difference, 2.6 points; 98.3% CI, 1.0-4.2; P < .001; CBDC: least-squares mean difference, 3.3 points; 98.3% CI, 1.7-4.9; P < .001). The difference in caregiver satisfaction with care between the interventions and UC was apparent by 3 months and persisted through the study. Caregiver-rated quality of care was higher in the CBDC group compared with UC (least-squares mean ratio, 1.1; 98.3% CI, 1.0-1.3; P = .046). Conclusions and Relevance: In this randomized clinical trial, HSDC and CBDC did not differ from UC on most person living with dementia and caregiver measures. However, caregivers reported higher satisfaction with both interventions compared with UC. These findings can help further refine comprehensive dementia care programs. Trial Registration: ClinicalTrials.gov Identifier: NCT03786471.