Impact of structured advance care planning program on patients’ wish items and healthcare utilization

Abstract

Background: Structured advance care planning (ACP) program is an important service in the end-of-life care for patients with advanced medical illness. We pioneered a structured and coordinated ACP program for patients with advanced malignancies and end-stage organ failure in Hong Kong. This study investigated the impact of a structured ACP program on the concordance rate for patients’ final wishes, patient/family satisfaction, and the number of acute admissions (AA) and length of stay (LOS) in hospital.

Methods: Patients with advanced malignancy or end-stage organ failure who were able to complete ACP forms during the current admission to medical units were recruited. Patients who could not complete ACP forms or <18 years of age were excluded. The ACP program comprised the following components: (I) baseline education (workshop/role play) in ACP sessions for linked nurses of different medical units; (II) structured ACP discussions with recruited patients and their proxies during admission, after any change in clinical status, and also at monthly intervals; (III) formal structured review of patients’ goals at regular team meetings; (IV) “flagging” of advance directive (AD) in hospital computer system and (V) feedback to linked nurse on the congruence of care. Mentally competent patients who did not receive ACP and matched for disease and demographics were selected as controls in a 1:2.5 ratio.

Results: Two hundred forty-three patients were included for analysis between August 2016 and July 2017, of which 69 patients joined the ACP program and 174 of them did not. Two hundred and one patients (83%) had advanced cancer. All had done do-not-attempt-cardiopulmonary-resuscitation (DNACPR) order in the ACP group. The concordance rates for patients’ wishes on quality of life, end-of-life and funeral arrangements were 95%, 100% and 100% respectively. Over 70% of patients and their families (N=10) showed satisfaction with the program. The ACP group also had lower mean AA and shorter LOS (0.78±0.23 vs. 1.2±0.8 episode/patient, 4.6±1.7 vs. 7.5±2.5 days, P=0.037 and P=0.023 respectively) in the last 3 months of life compared with the non-ACP group.

Conclusions: This ACP program achieved high concordance rate for patients’ wish items and reduced healthcare utilization.