Evaluation of an initiative to improve advance care planning for a home-based primary care service

Background Advance care planning (ACP) is a process that enables individuals to describe, in advance, the kind of health care they would want in the future. There is evidence that ACP reduces hospital-based interventions, especially at the end of life. ACP for frail older adults is especially important as this population is more likely to use hospital services but less likely to benefit from resource intensive care. Our study goal was to evaluate whether an approach to ACP developed for frail older adults, known as the Palliative and Therapeutic Harmonization or PATH, demonstrated an improvement in ACP. Methods The PATH approach was adapted to a primary care service for homebound older adults in Vancouver, Canada. This retrospective chart review collected surrogate measures related to ACP from 200 randomly selected patients enrolled in the service at baseline (prior to June 22, 2017), and 114 consecutive patients admitted to the program after implementation of the PATH ACP initiative (October 1, 2017 to May 1, 2018). We compared the following surrogate markers of ACP before and after implementation of the PATH model, chart documentation of: frailty stage, substitute decision-maker, resuscitation decision, and hospitalization decision. A composite ACP documentation score that ascribed one point for each of the above four measures (range 0 to 4) was also compared. For those with documented resuscitation and hospitalization decisions, the study examined patient/ substitute decision-maker expressed preferences for do-not-resuscitate and do-not-hospitalize, before and after implementation. Results We found the following changes in ACP-related documentation before and after implementation: frailty stage (27.0% versus 74.6%, p < .0001); substitute decision-maker (63.5% versus 71.9%, p = 0.128); resuscitation decision documented (79.5% versus 67.5%, p = 0.018); and hospitalization decision documented (61.5% versus 100.0%, p < .0001); mean (standard deviation) composite ACP documentation score (2.32 (1.16) versus 3.14 (1.11), p < .0001). The adjusted odds ratios (95% confidence intervals) for an expressed preference of do-not-resuscitate and do-not-hospitalize after implementation were 0.87 (0.35, 2.15) and 3.14 (1.78, 5.55), respectively. Conclusions Results suggest partial success in implementing the PATH approach to ACP in home-based primary care. Key contextual enablers and barriers are important considerations for successful implementation.