Pragmatic Cluster Randomized Trial of Advance Care Planning for Nursing Home Residents with Dementia

Key Message : APPROACHES was an embedded pragmatic randomized clinical trial that tested whether an intervention that trained nursing home staff to implement advance care planning reduced hospital transfers versus usual care for residents with dementia. The trial found no difference in the rate of hospital transfers in intervention versus control facilities.

Abstract:

Background: Advance care planning (ACP) aims to support decision-making by families for nursing home (NH) residents with dementia, but it is inconsistently delivered in this setting. The embedded pragmatic randomized clinical trial ‘Aligning Patient Preferences – a Role Offering Alzheimer’s patients, Caregivers, and Healthcare providers Education and Support (APPROACHES)’ was designed to address gaps in NH staff training and systematize processes for conducting and documenting ACP conversations.

Objectives: To evaluate whether residents in NHs randomized to implement an intervention that trained staff to facilitate and implement ACP had reduced hospital transfer rates (primary outcome) compared to those in NHs randomized to usual care. Secondary outcomes included hospice enrollment, death in hospital, and burdensome treatments near end-of-life.

Methods: 128 NHs across 8 states were randomized to either the intervention or control arm (N=64 NHs/arm). Staff in intervention facilities were trained to facilitate ACP conversations and implement them systematically among residents with dementia. Outcomes were assessed using Medicare claims data.

Results: Characteristics of residents with dementia in the intervention (n = 7,474) and control facilities (n = 7,358) were well-balanced. 1,877 residents (25%) in intervention facilities received ACP. Hospital transfer rates in the intervention NHs were 0.99 (95% confidence interval (CI), 0.88-1.12) and 1.05 (95% CI, 0.93-1.19) in the control NHs. Transfer rates did not significantly differ between the two arms (Incident rate ratio = 0.94, 95% CI, 0.84-1.06). Secondary outcomes of hospice use, death in hospital, ventilator use, feeding tubes, and receiving burdensome treatment at the end of life also did not significantly differ between arms.

Conclusion: In this pragmatic trial evaluating an ACP intervention among NH residents with dementia, there were no differences in the hospital transfer rate between residents in intervention and control group facilities. Inconsistent implementation of the intervention across nursing homes is a potential factor impacting results.

References: N/A