Effect of Goals-of-Care Program for Inpatients at a Comprehensive Cancer Center During the COVID-19 Pandemic

Get Permission

David Hui, MD, MSc

David Hui, MD, MSc

As reported in the Journal of Clinical Oncology by David Hui, MD, MSc, and colleagues, an interdisciplinary goals-of-care program instituted at The University of Texas MD Anderson Cancer Center in response to anticipated COVID-19–related increases in need for intensive care unit (ICU) use resulted in reduced ICU mortality, length of stay in the ICU, and in-hospital mortality among inpatients with cancer.

Study Details 

The program (myCOG), implemented in March 2020, consisted of six main components:

  • Risk stratification to systematically identify patients at high risk of mortality
  • Immediate goals-of-care discussions with oncologists
  • Real-time monitoring with daily feedback
  • Goals-of-care education and assistance with goals-of-care conversations by palliative care staff, including deployment of goals-of-care rapid response team if needed
  • Longitudinal monitoring of goals-of-care program metrics
  • Institutional leadership support.

The analysis compared outcomes among consecutive adult patients with cancer admitted to medical units at the center during an 8-month pre-implementation period of May 1, 2019, to December 31, 2019, and an 8-month postimplementation period of May 1, 2020, to December 31, 2020. The primary outcome measure was ICU mortality. Outcomes were assessed using propensity score weighting to adjust for differences in potential covariates, including age, sex, cancer diagnosis, race/ethnicity, and Sequential Organ Failure Assessment score.

Key Findings

The analysis included 56,977 patients admitted during the pre-implementation period and 55,964 admitted during the postimplementation period. There were 727 ICU admissions in the pre-implementation period and 638 in the postimplementation period. 

After program implementation, there were significant reductions in ICU mortality (21.9% vs 28.2%; change = –6.3%, 95% confidence interval [CI] = –9.6% to –3.1%, P = .0001), length of ICU stay (median = 3 vs 4 days, mean change = –1.4 days, 95% CI = –2.0 to –0.7 days, P < .0001) and in-hospital mortality (6.1% vs 7%, mean change = –0.9%, 95% CI = –1.5% to –0.3%, P = .004).

After implementation, the proportion of patients with an in-hospital do-not-resuscitate order increased significantly, from 14.7% to 19.6% (odds ratio [OR] = 1.4, 95% CI = 1.3–1.5, P < .0001), and do-not-resuscitate orders were established earlier during stay (mean difference = –3.0 days, 95% CI = –3.9 to –2.1 days, P < .0001).

Significant increases were also observed in proportions of patients with documentation of goals-of-care discussions before (33% vs 6%, OR = 7.2, 95% CI = 6.5–8.1, P < .0001) and during  (49% vs 11%, OR = 7.6, 95% CI = 7.0–8.4, P <.0001) index admission.

The investigators concluded, “This study showed improvement in hospital outcomes and care plan documentation after implementation of a system-wide, multicomponent goals-of-care intervention.”

Dr. Hui, of the Department of Palliative, Rehabilitation & Integrative Medicine, The University of Texas MD Anderson Cancer Center, is the corresponding author for the Journal of Clinical Oncology article.

Disclosure: The study was supported by grants from the National Institutes of Health and National Cancer Institute. For full disclosures of the study authors, visit

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.