Advance Care Planning Among Older Adults With Advanced Non-Dialysis-Dependent CKD and Their Care Partners: Perceptions Versus Reality?

Rationale & Objective: Older patients with advanced chronic kidney disease (CKD) use intensive care at the end of life and die in a hospital more frequently than patients with cancer or heart disease. Advance care planning (ACP) can help align treatment with patient preferences and improve patient-centered care, yet ACP quality and experiences among older patients with CKD and their care partners remain incompletely understood, particularly among the non-dialysis-dependent population. Study Design: In-person interviewer-administered surveys of patients 70 years and older with non-dialysis-dependent CKD stage 4 or 5 and their self-identified care partners. Setting & Participants: 42 participants (31 patients, 11 care partners) at 2 clinical sites in greater Boston. Outcomes: Completion of advance directives and self-reported perceptions, preferences, and experiences of ACP. Analytical Approach: Descriptive analysis of patient and care partner surveys. McNemar test analysis to compare patient and care partner responses. Results: Most patients had written advance directives (64%) and surrogate decision makers (81%). Although patients reported positive perceptions and high trust in their clinicians' judgment, few (16%) had actually discussed preferences for life-sustaining treatment with their nephrologists. Few ACP discussions included components reflective of high-quality ACP: 16% of patients had been asked about their values concerning end-of-life care and 7% had discussed issues of decision-making capacity and consent to care should their health decline. When presented with 2 hypothetical scenarios (stroke/heart attack or dementia), nearly all patients and care partners reported a preference for comfort care over delaying death. Care partners were more likely than patients to report that they had experienced discussion components reflective of high-quality ACP with the clinical team. Limitations: Single metropolitan area; most patients did not identify a care partner; nonresponse bias and small sample size. Conclusions: Patients often believed that their clinicians understood their end-of-life wishes despite not having engaged in ACP conversations that would make those wishes known. Improving clinical ACP communication may result in end-of-life treatment that better aligns with patient goals.