Introduction: Inter-hospital transfer (IHT) patients have higher in-hospital mortality, higher healthcare costs, and worse outcomes compared to non-transferred patients. Goals of care (GoC) discussions prior to transfer are necessary in patients at high risk for decline to ensure that the intended outcome of transfer is goal concordant. However, the frequency of these discussions is not well understood. This study was intended to assess the prevalence of GoC discussions in IHT patients with early mortality, defined as death within 72 hours of transfer, and prevalence of primary diagnoses associated with in-hospital mortality. Methods: This was a retrospective study of IHT patients aged 18 and older who died within 72 hours of transfer to Wake Forest Baptist Medical Center between October 1, 2016-October 2018. Documentation of GoC discussions within the electronic health record (EHR) prior to transfer was the primary outcome. We also assessed charts for primary diagnosis associated with in-hospital mortality, code status changes prior to death, in-hospital healthcare interventions, and frequency of palliative care consults. Results: We included in this study a total of 298 patients, of whom only 10.1% had documented GoC discussion prior to transfer. Sepsis (29.9%), respiratory failure (28.2%), and cardiac arrest (27.5%) were the top three diagnoses associated with in-hospital mortality, and 73.2% of the patients transitioned to comfort measures prior to death. After transfer, 18.1% of patients had invasive procedures performed with 9.7% undergoing major surgery. Palliative care consultation occurred in only 4.4%. Conclusion: The majority (89.9%) of IHT patients with early mortality did not have GoC discussion documented within EHR prior to transfer, although most transitioned to comfort measures prior to their deaths, highlighting that additional work is needed in this area. [West J Emerg Med. 2020;21(4)935-942.]

e24003 Background: Immune checkpoint inhibitors (ICI) can lead to durable responses in some patients with advanced non-small cell lung cancer (NSCLC) with less toxicity than chemotherapy. However, ICI can cause idiosyncratic adverse effects and radiographic responses can be difficult to interpret, making prognostication difficult and potentially impeding a timely transition to hospice. Our purpose was to explore end of life (EOL) outcomes in NSCLC patients treated with ICI. Methods: Retrospective data were collected on all patients with NSCLC started on ICI at single center (2014-2018) and who died before last known follow-up. EOL outcomes included hospitalizations, ICU admissions, timing/location of hospice referral and death. Charts were reviewed to identify barriers to hospice referrals or enrollment. Outcomes were compared to published data of similar cohort (2008-2010) from same institution treated with chemotherapy. Results: Out of 143 patients who died, 83 (58%) had internal hospice referral by cancer center; 15 (11%) were referred by external provider. Hospitalization frequency was associated with higher likelihood of internal hospice referral (p 0.04). Internal hospice referral was not associated with differences in age, sex, race, ethnicity, smoking history, cancer subtype, treatment response/toxicity, or overall survival. Internal hospice referral was associated with decreased rates of death on the hospital floor (p < 0.001) and the intensive care unit (ICU, p < 0.001). When compared to chemotherapy cohort, there was similar rate of hospice referral (68% vs. 74%, p 0.33) but higher rates of starting new systemic therapy within 30 days of death (17% vs. 6%, p 0.001) and last dose within 14 days of death (13% vs. 5%, p 0.005). Other EOL outcomes (hospitalization frequency, death in ICU) were similar, although ICI cohort trended towards a lower rate of death on the floor than chemotherapy cohort (16% vs. 25%, p 0.06). Barriers to hospice referral were not well documented by providers but lack of family support for home EOL care was a common reason patients declined hospice. Conclusions: ICI was associated with a higher rate of systemic treatment at EOL as compared to a historical chemotherapy cohort, although the rates of hospice referral and other outcomes were similar.

Background Immune checkpoint inhibitors (ICIs) can lead to durable responses in patients with lung cancer but may delay transitions to hospice at the end of life (EOL). We aimed to test the association of continuity of care with EOL outcomes in the ICI era. Methods We collected retrospective data on all patients with lung cancer who started ICI treatment at a single comprehensive cancer center in the United States (1/1/14–5/1/18) and subsequently died. We defined a hospice referral as having continuity of care if placed by a provider from the patient’s multidisciplinary cancer team (e.g., a medical oncologist, palliative care specialist, intensivist, and hospitalist). Results In this cohort of 143 patients, 58% had a team-based hospice referral which was associated with a lower risk of death in the hospital. The most common reason patients declined hospice at EOL was an unwillingness to discontinue cancer-directed therapy. As compared to a similar historical cohort of patients treated with chemotherapy alone (2008–2010), there was a similar rate of hospice referral (68% vs 74%) but higher rates of new systemic therapy initiated within 30 days of death (17% vs 6%, p .001) and last dose within 14 days of death (13% vs 5%, p .005). Conclusions Future studies should test the continuity of care at EOL as a new quality metric for advanced NSCLC.