Nurse-involved advance care planning for adults receiving maintenance dialysis: a systematic review of decision-quality and psychosocial outcomes

Adults receiving maintenance dialysis face high mortality and may receive end-of-life care that does not reflect their preferences. We evaluated whether nurse-involved advance care planning (ACP) - delivered by nurses alone, within a multidisciplinary team, or by nurses or other permitted professionals - improves decision-quality and psychosocial outcomes. We searched nine databases and three trial registries from inception to 26 March 2026. Maintenance-dialysis trials formed the primary synthesis; pre-dialysis trials were examined separately as sensitivity evidence. Risk of bias was assessed with the revised Cochrane tool; because the two extractable agreement estimates were different estimands, outcomes were synthesized without meta-analysis. Nine randomized trials were included: seven in maintenance dialysis and two in pre-dialysis stage 5 chronic kidney disease. Nurses delivered all core encounters in seven of nine trials; one used a multidisciplinary team; the cluster trial permitted nurses or social workers to deliver it without reporting the nurse share. The 2024 cluster trial reported higher patient-surrogate agreement about life-sustaining-treatment preferences in hypothetical scenarios at 2 weeks (adjusted odds ratio 1.61, 95% confidence interval 1.12–2.31); in the 2015 efficacy trial the primary time-adjusted estimate across 2–12 months was 1.4 (0.9–2.1), and the 2-month effect of 1.89 (1.1–3.3) attenuated by 12 months; the estimands were not pooled. Decisional-conflict findings were mixed: clear benefit in one cluster trial, model- and time-dependent findings in one efficacy trial, and no clear difference in two pilots or one short-term trial’s total score. One feasibility trial reported ACP documentation descriptively and found lower quality-of-life scores after immediate than after deferred ACP (22 completers); no trial measured goal-concordant care received; most other outcomes came from one or two trials. Low-certainty evidence suggests that nurse-involved ACP may improve preparation for future end-of-life decisions; the most consistent signal, short-term patient-surrogate agreement about hypothetical preferences, is distinct from ACP documentation and goal-concordant care. The evidence base is small, mainly hemodialysis-based, concentrated in one intervention family, and includes many secondary comparisons without multiplicity adjustment; findings should be interpreted cautiously. International Prospective Register of Systematic Reviews (PROSPERO), CRD420261404839; retrospectively registered on 10 June 2026.