Differential Impact of a Multicomponent Goals-of-Care Program in Patients with Hematologic and Solid Malignancies

Compassionate Communication and Advance Care Planning to improve End-of-life Care in Treatment of Haematological Disease 'ACT': Study Protocol for a Cluster-randomized trial

INTRODUCTION

To support the implementation of advance care planning and serious illness conversations in haematology, a previously developed conversation intervention titled 'Advance Consultations Concerning your Life and Treatment' (ACT) was found feasible. This study aims to investigate the effect of ACT on the quality of end-of-life care in patients with haematological malignancy and their informal caregivers.

METHODS AND ANALYSIS

The study is a nationwide 2-arm cluster randomised trial randomising 40 physician-nurse clusters across seven haematological departments in Denmark to provide standard care or ACT intervention. A total of 400 patients with haematological malignancies and their informal caregivers will be included. The ACT intervention includes an ACT conversation that centres on discussing the patient's prognosis, worries, hopes and preferences for future treatment. The intervention is supported by clinician training and supervision, preparatory materials for patients and informal caregivers, and system changes including dedicated ACT-conversation timeslots and templates for documentation in medical records.This study includes two primary outcomes: (1) the proportion of patients receiving chemotherapy within the last 30 days of death and (2) patients' and informal caregivers' symptoms of anxiety (General Anxiety Disorder-7) at 3 6, 9, 12 and 18 months follow-up. Mixed effects models accounting for clusters will be used.

ETHICS AND DISSEMINATION

The Declaration of Helsinki and the European GDPR regulations as practised in Denmark are followed through all aspects of the study. Findings will be made available to the participants, patient organisations, funding bodies, healthcare professionals and researchers at national and international conferences and through publication in peer-reviewed international journals.

REGISTRATION DETAILS

The study is registered at ClinicalTrials.gov (NCT05444348). The Regional Ethics Committee of the Capital Region of Denmark (record no: 21067634) has decided that approval is not necessary as per Danish legislation. Study approval has been obtained from The Capital Region of Denmark Data Protection Agency (record no: P-2022-93).

TRIAL REGISTRATION NUMBER

NCT05444348.

Cost of Hospitalization Associated with Inpatient Goals-of-Care Program Implementation at a Comprehensive Cancer Center: A Propensity Score Analysis

The impact of goals-of-care programs on acute hospitalization costs is unclear. We compared the hospitalization cost in an 8-month period before implementation of a multimodal interdisciplinary goals-of-care program (1 May 2019 to 31 December 2019) to an 8-month period after program implementation (1 May 2020 to 31 December 2020). Propensity score weighting was used to adjust for differences in potential covariates. The primary outcome was total direct cost during the hospital stay for each index hospitalization. This analysis included 6977 patients in 2019 and 5964 patients in 2020. The total direct cost decreased by 3% in 2020 but was not statistically significant (ratio 0.97, 95% CI 0.92, 1.03). Under individual categories, there was a significant decrease in medical oncology (ratio 0.58, 95% CI 0.50, 0.68) and pharmacy costs (ratio 0.86, 95% CI 0.79, 0.96), and an increase in room and board (ratio 1.06, 95% CI 1.01, 1.10). In subgroup analysis, ICU patients had a significant reduction in total direct cost after program implementation (ratio 0.83, 95% CI 0.72, 0.94). After accounting for the length of ICU admission, we found that the total direct cost per hospital day was no longer different between 2019 and 2020 (ratio 0.986, 95% CI 0.92, 1.05), suggesting that shorter ICU admissions likely explained much of the observed cost savings. This study provides real-world data on how "in-the-moment" GOC conversations may contribute to reduced hospitalization costs among ICU patients.

The impact of a multidisciplinary goals-of-care program on unplanned readmission rates at a comprehensive cancer center

PURPOSE

This study examined the 30-day unplanned readmission rate in the medical oncology population before and after the implementation of an institution-wide multicomponent interdisciplinary goals of care (myGOC) program.

METHODS

This retrospective study compared the 30-day unplanned readmission rates in consecutive medical patients during the pre-implementation period (May 1, 2019, to December 31, 2019) and the post-implementation period (May 1, 2020, to December 31, 2020). Secondary outcomes included 7-day unplanned readmission rates, inpatient do-not-resuscitate (DNR) orders, and palliative care consults. We randomly selected a hospitalization encounter for each unique patient during each study period for statistical analysis. A multivariate analysis model was used to examine the association between 30-day unplanned readmission rates and implementation of the myGOC program.

RESULTS

There were 7028 and 5982 unique medical patients during the pre- and post-implementation period, respectively. The overall 30-day unplanned readmission rate decreased from 24.0 to 21.3% after implementation of the myGOC program. After adjusting for covariates, the myGOC program implementation remained significantly associated with a reduction in 30-day unplanned readmission rates (OR [95% CI] 0.85 [0.77, 0.95], p = 0.003). Other factors significantly associated with a decreased likelihood of a 30-day unplanned readmission were an inpatient DNR order, advanced care planning documentation, and an emergent admission type. We also observed a significant decrease in 7-day unplanned readmission rates (OR [95% CI] 0.75 [0.64, 0.89]) after implementation of the myGOC program.

CONCLUSION

The 30-day and 7-day unplanned readmission rates decreased in our hospital after implementation of a system-wide multicomponent GOC intervention.

Impact of an Institution-Wide Goals of Care Program on the Timing of Referrals to Outpatient Palliative Care

CONTEXT

Palliative care has received increased interest since the COVID-19 pandemic due to its role in guiding goals of care (GOC) discussions.

OBJECTIVES

We assessed the change in the timing of outpatient palliative care referrals before and after implementing an institution-wide multicomponent interdisciplinary GOC (myGOC) program.

METHODS

We reviewed 200 random supportive care center (SCC) consult visits each from June to November 2019 (before myGOC) and June to November 2020 (after myGOC). Data regarding Edmonton Symptom Assessment Scale (ESAS) scores, time from hospital registration to SCC visit, SCC visit until death/last follow-up, and advance care planning (ACP) notes were collected. Kaplan-Meier curves were used to evaluate overall survival (OS).

RESULTS

The median OS from the SCC consult visit was 15.2 months (95% CI:11.7-19.7) before and 14.0 months (95% CI:10.8-17.9) after the myGOC program (P = 0.646). There were no significant differences in the median time between the SCC consult visit to death/last follow-up (11.95 vs. 12.0 months after myGOC; P = 0.841) and the first visits to our cancer center and SCC (6.1 vs. 5.29 months after myGOC; P = 0.689). Patients seen after myGOC had significantly lower ESAS symptom scores, better performance status (2 [1-2] vs. 2 [1-3]; P = 0.018], and more ACP notes composed by medical oncology teams (25.5% vs. 4.5%; P < 0.001).

CONCLUSION

There were no significant differences in OS among patients seen in the SCC before and after myGOC, likely related to a ceiling effect. More oncologists had ACP discussions with patients, and patients had lower symptom scores on ESAS after myGOC, likely indicating that more patients were referred for GOC discussions and ACP rather than for symptom distress.