Determining Palliative Care Penetration Rates in the Acute Care Setting

Abstract

ContextIntermountain Healthcare, in collaboration with Cerner Corporation, developed a hospital-based electronic palliative care algorithm. ObjectivesThis study aims to improve identification of patients who would benefit from palliative care services, and calculate palliative care penetration rates. MethodsThis study used a mixed-methods nonrandomized retrospective study design. Three 30-day iterations of clinical data were analyzed for patients identified by the electronic algorithm. During the second and third 30-day iterations, palliative care clinicians conducted chart reviews on a weekly basis for identified patients and determined whether the patients were appropriate for a palliative care consult. Positive predictive values (PPVs) were calculated. Based on the PPV, palliative care consult penetration rates were also calculated. ResultsDuring the first iteration, the algorithm triggered 2995 times on 1384 unique patient encounters (69.3% of the total inpatient population). In the second iteration, the algorithm triggered 851 times on 477 unique patient encounters (26.4% of the total inpatient population). Eight hundred twenty-one chart reviews were completed on 420 unique patient encounters. The PPV was 68.3%. Based on the PPV, the projected palliative care penetration rate was 17.6%. During the third iteration, the algorithm triggered 1229 times on 539 unique patient encounters (33.3% of the total inpatient population). Nine hundred sixty-seven chart reviews were completed on 505 unique patient encounters. The PPV was 80.1%. Based on the PPV, the projected palliative care penetration rate was 26.4%. ConclusionThis study successfully optimized an electronic palliative care identification algorithm with a PPV of 80.1% and indicates appropriate palliative care penetration rates may be as high as 26.4% of the total inpatient population.