Advance Care Planning Coaching in CKD Clinics: A Pragmatic Randomized Clinical Trial

Document Type

Journal Article

Publication Date



American Journal of Kidney Diseases




Advance care planning (ACP); advance directives; chronic kidney disease (CKD); end-of-life care; goals of care; health care decision making; motivational interviewing; palliative care; POLST; predialysis CKD; randomized controlled trial (RCT); treatment wishes


Rationale & Objective: Although guidelines recommend more and earlier advance care planning (ACP) for patients with chronic kidney disease (CKD), scant evidence exists to guide incorporation of ACP into clinical practice for patients with stages of CKD prior to kidney failure. Involving nephrology team members in addition to primary care providers in this important patient-centered process may increase its accessibility. Our study examined the effect of coaching implemented in CKD clinics on patient engagement with ACP. Study Design: Multicenter, pragmatic randomized controlled trial. Setting & Participants: Three CKD clinics in different states participated: 273 patients consented to participate, 254 were included in analysis. Eligible patients were 55 years or older, had stage 3-5 CKD, and were English speaking. Intervention: Nurses or social workers with experience in nephrology or palliative care delivered individualized in-person ACP sessions. The enhanced control group was given Make Your Wishes About You (MY WAY) education materials and was verbally encouraged to bring their completed advance directives to the clinic. Outcome: Primary outcome measures were scores on a 45-point ACP engagement scale at 14 weeks and a documented advance directive or portable medical order at 16 weeks after enrollment. Results: Among 254 participants analyzed, 46.5% were 65-74 years of age, and 54% had CKD stage 3. The coached patients scored 1.9 points higher at 14 weeks on the ACP engagement scale (β = 1.87 [95% CI, 0.13-3.64]) adjusted for baseline score and site. Overall, 32.8% of intervention patients (41 of 125) had an advance directive compared with 17.8% (23 of 129) of patients in the control group. In a site-adjusted multivariable model, coached patients were 79% more likely to have a documented advance directive or portable medical order (adjusted risk ratio, 1.79 [95% CI, 1.18-2.72]), with the impact principally evident at only 1 study site. Limitations: Small number of study sites and possible unrepresentativeness of the broader CKD population by study participants. Conclusions: Individualized coaching may be effective in enhancing ACP, but its impact may be influenced by the health care environment where it is delivered. Funding: The Patrick and Catherine Weldon Donaghue Medical Research Foundation, via the Greater Value Portfolio. Trial Registration: Registered at with study number NCT03506087.