Intervention to improve care at life's end in inpatient settings: the BEACON trial

Abstract

Background: Widespread implementation of palliative care treatment plans could reduce suffering in the last days of life by adopting best practices of traditionally home-based hospice care in inpatient settings.

Objective: To evaluate the effectiveness of a multi-modal intervention strategy to improve processes of end-of-life care in inpatient settings.

Design: Implementation trial with an intervention staggered across hospitals using a multiple-baseline, stepped wedge design.

Participants: Six Veterans Affairs Medical Centers (VAMCs).

Intervention: Staff training was targeted to all hospital providers and focused on identifying actively dying patients and implementing best practices from home-based hospice care, supported with an electronic order set and paper-based educational tools.

Main measures: Several processes of care were identified as quality endpoints for end-of-life care (last 7 days) and abstracted from electronic medical records of veterans who died before or after intervention (n = 6,066). Primary endpoints were proportion with an order for opioid pain medication at time of death, do-not-resuscitate order, location of death, nasogastric tube, intravenous line infusing, and physical restraints. Secondary endpoints were administration of opioids, order/administration of antipsychotics, benzodiazepines, and scopolamine (for death rattle); sublingual administration; advance directives; palliative care consultations; and pastoral care services. Generalized estimating equations were conducted adjusting for longitudinal trends.

Key results: Significant intervention effects were observed for orders for opioid pain medication (OR: 1.39), antipsychotic medications (OR: 1.98), benzodiazepines (OR: 1.39), death rattle medications (OR: 2.77), sublingual administration (OR: 4.12), nasogastric tubes (OR: 0.71), and advance directives (OR: 1.47). Intervention effects were not significant for location of death, do-not-resuscitate orders, intravenous lines, or restraints.

Conclusions: This broadly targeted intervention strategy led to modest but statistically significant changes in several processes of care, indicating its potential for widespread dissemination to improve end-of-life care for thousands of patients who die each year in inpatient settings.

Trial registration: ClinicalTrials.gov NCT00234286.

Figures

Figure 1

Odds ratios and 95 % confidence intervals for process of care endpoints adjusting for longitudinal trends (ICU = intensive care unit)