Improving a Full Range of Advance Care Planning Behavior Change and Action Domains: The PREPARE Randomized Trial

Abstract

Context: Advance care planning (ACP) engagement includes a wide range of behaviors and actions related to discussions and documentation, yet few ACP intervention studies measure the full process.

Objectives: The objective of the study was to compare the effects of an easy-to-read advance directive (AD) versus an ACP web site plus the AD (PREPARE + AD) on Behavior Change Processes and Actions, including discussions and documentation.

Methods: Secondary data were from a completed ACP trial. Participants were primary care patients, ≥60 years old, with two comorbidities. We used the validated ACP Engagement Survey to examine six-month change in subscales measuring Behavior Change Processes (knowledge, contemplation, self-efficacy, readiness) and Actions (decision makers, quality of life, flexibility for decision makers, asking clinicians questions), specifically related to discussions and documentation. We used adjusted mixed-effects linear models to compare mean change and engagement over time.

Results: Compared to the AD-only, PREPARE + AD resulted in greater increases in all Behavior Change Processes subscales and Actions related to decision makers, quality of life, and flexibility (all P-values ≤0.005). Both interventions significantly increased the proportion of participants who engaged in ACP discussions (PREPARE + AD, 99.5%; AD-only, 93.3%) and documentation (PREPARE + AD, 99.5%; AD-only, 90.4%), with greater increases for PREPARE + AD (all P-values <0.001).

Conclusion: Both PREPARE plus an easy-to-read AD and an AD-only markedly increased ACP engagement in a full range of ACP behaviors, including discussions and documentation, and engagement was nearly 100% with PREPARE + AD. Future ACP studies should examine a full range of ACP behaviors beyond ADs and the impact of PREPARE and easy-to-read AD implementation on health care systems.

Trial registration: ClinicalTrials.gov NCT01550731.

Keywords: Advance care planning; advance directive; behavior change; communication; decision aid.

Published by Elsevier Inc.

Figures

Figure 1. Impact of PREPARE plus Advance Directive (AD) compared to an AD-only on Advance Care Planning Behavior Processes

Process subscale scores for PREPARE+AD (n=205, solid line) versus an AD-only (n=209, dashed line) over 6-months and showing percent change in scores. A) Knowledge Subscale, PREPARE+AD effect size 0.60; B) Contemplation Subscale, PREPARE+AD effect size 0.70; C) Self-Efficacy Subscale, PREPARE+AD effect size 0.44; D) Readiness Subscale, PREPARE+AD effect size 0.44. All scores are on a Likert scale 0–5. P values reflect significance for group x time interactions, using repeated measures, mixed-effects linear regression models, adjusted for race, health literacy, baseline ACP documentation, and clustering by physician. AD indicates advance directives; PREPARE indicates a patient-centered ACP website. The percent change for each study group is calculated as the difference between the 6-month value minus the baseline value divided by the baseline value*100.

Figure 2. Impact of PREPARE plus Advance Directive (AD) compared to an AD-only on Advance Care Planning Actions

Action subscale scores for PREPARE+AD (n=205, solid line) versus an AD-only (n=209, dashed line) over 6-months and showing percent change in scores. A) Decision Maker Subscale (range 0–5), PREPARE+AD effect size 0.56; B) Quality of Life Subscale (range 0–10), PREPARE+AD effect size 0.53. Quality of Life includes acceptable health states and care desired at the end of life. C) Flexibility in Decision Making Subscale (range 0–5), PREPARE+AD effect size 0.87; D) Ask Clinicians Questions Subscale (range 0–5), PREPARE+AD effect size −0.33. P values reflect significance for group x time interactions, using repeated measures, mixed-effects linear regression models, adjusted for race, health literacy, baseline ACP documentation, and clustering by physician. AD indicates advance directives; PREPARE indicates a patient-centered ACP website. The percent change for each study group is calculated as the difference between the 6-month value minus the baseline value divided by the baseline value*100.