Randomized Evaluation of Default Access to Palliative Services (REDAPS)

Inpatient palliative care (PC) consultation can improve the quality, cost, and outcomes of care for patients with life-limiting illnesses. However, there is limited experimental evidence available to quantify the effects of inpatient PC consultation in generalizable, real-world settings.

This large, real-world trial assessed the effects of ordering PC consultation by default (with an opt-out) for hospitalized patients over 65 years of age with one of three advanced chronic serious illnesses: chronic obstructive pulmonary disease (COPD), dementia, and end-stage renal disease (ESRD). The trial was conducted between May 2016 and November 2018 at 11 hospitals across 8 states within one health system. Hospitals were randomly assigned to change from usual care (clinicians can opt-in to a PC consult) to the intervention (clinicians can opt-out of an automatically ordered PC consult). Clinicians were alerted in the electronic health record (EHR) to the default PC consult order and provided up to 24 hours to opt-out, and patients or family members could also decline these services.

We conducted a concurrent qualitative study among 28 Hospitalist physicians at 7 participating hospitals to understand their views on consulting PC services for hospitalized patients with advanced dementia for whom default PC consult opt-out rates were lowest.

We enrolled 34,239 patient-encounters overall, 24,065 of which had a hospital length of stay of at least 72 hours and were included in primary analyses.

This study found that a default order tripled the rate of PC consultations and led to earlier consultation by 1.2 days compared to usual care, without requiring increased PC staffing. The intervention did not reduce hospital length of stay overall. However, among the subset of patients who only received a PC consult because of the default order intervention, there was a nearly 10% reduction in length of stay. The intervention also led to increased rates of hospice use and do-not-attempt-resuscitation orders, importantly without an increased risk of dying in the hospital.

REDAPS was the largest study ever conducted in the field of PC, and produced the first large-scale, experimental evidence of the effectiveness of inpatient PC as delivered within the course of routine real-world care. In contrast to prior PC trials, these results reflect a broad and generalizable population from 11 community hospitals that resemble the settings where most Americans receive their care.

Our embedded qualitative study among Hospitalists who received the EHR alert for a default consult order revealed that their decision-making for patients with dementia was influenced by factors at multiple levels: patient, family caregivers, clinician, and hospital. Specifically, they were more likely to agree with inpatient PC consultation for patients with very advanced disease, receiving aggressive care, or those with family-clinical team conflicts around goals of care and prognostic understanding. Hospitalists’ limited time, and for some, a lack of confidence in PC skills were strong drivers of consultation. Additional barriers to PC referral included a hospital culture that conflated palliative care with hospice and end-of-life care or had PC teams perceived to be too busy.

Ascension

National Institute on Aging