Background Our primary aim was to assess, among advanced heart failure patients, effectiveness of Advance care planning (ACP) in ensuring end of life (EOL) care consistent with patient wishes. Secondary aims were to assess its impact on patients’ decisional conflict, discussion of care preferences with surrogates, illness understanding, anxiety, depression and quality of life.

Methods We conducted a randomized controlled trial of ACP (based on Respecting Choices Model) versus usual care in Singapore. 282 patients hospitalized with heart failure and NYHA III and IV symptoms were randomized to ACP (93) or control (189) arm. They answered up to 6 follow-up surveys conducted every 4 months. Primary outcome was assessed in the deceased sample (89; 23 in ACP, 66 in control arm). Both intention-to-treat and per-protocol analyses were done.

Results 63% of ACP arm received intervention. Deceased patients in ACP arm were no more likely to have their wishes followed for EOL treatments (35% in ACP vs 44% in control; p=0.47) but were more likely to have their wishes followed for cardiopulmonary resuscitation (83% in ACP vs 62% in control, p=0.12) though the difference was not statistically significant. At first follow-up, ACP patients had lower decisional conflict (β=-10.8, p <0.01) and were more likely to discuss preferences with their surrogate decision maker (β=1.3, p=0.04). Both arms did not differ on other patient outcomes. Per-protocol analyses showed similar results.

Conclusion ACP has benefits in improving decision making. More needs to be done to improve implementation of ACP for patients with heart failure.