1890P - End-of-Life (EOL) systemic anticancer treatment (SACT) and health services use before and during the COVID-19 pandemic

Background

EOL chemotherapy is considered poor quality of care, in part due to its association with increased EOL health services use, but newer SACT agents are less well characterized in this regard. We examined rates of EOL SACT type and their association with EOL health services use over time, both before and during the COVID-19 pandemic.

Methods

64,271 adult patients received SACT within 5 years of death from 03/2015-03/2021 (pre-pandemic: 16/03/2015-15/03/2020; pandemic period: 16/03/2020-15/03/2021) in Ontario, Canada. We measured EOL SACT as the proportion of patients who received the last dose of SACT within 30 days of death, and classified SACT into chemotherapy alone (CT) and any immunotherapy/targeted therapy (IT/TT). We measured health services use at EOL as the proportion of patients with >1 ED visit, >1 hospitalization, or ≥1 ICU admission within 30 days of death. Segmented linear regression was used to estimate monthly changes in trends of outcomes before the pandemic, and immediate (first month), and sustained (first year) effects of the pandemic.

Results

16,962 (26.4%) patients received EOL SACT (CT: 10,935; IT/TT: 6,027). Rate of CT remained stable during the study period; rate of IT/TT increased pre-pandemic (0.09%/month, P<0.001), with an immediate decrease during the pandemic (-1.36%, P=0.03), but no sustained change (P=0.99 compared to pre-pandemic). Patients receiving EOL SACT (vs no SACT) were more likely to use any health service (42.2% vs 24.5%), regardless of SACT type (CT: 42.8%; IT/TT: 41.1%). Among those receiving CT, rate of ≥1 ICU admission at EOL increased by 0.06%/month (P=0.03) pre-pandemic; during the pandemic there was no immediate change (P=0.54) and no sustained change (P=0.93). Among those receiving IT/TT, rate of >1 hospitalization at EOL increased by 0.03%/month (P=0.03) pre-pandemic; during the pandemic there was an immediate decrease (-4.67%, P=0.005) but no sustained change (P=0.09).

Conclusions

In this cohort study, rates of CT at EOL remained unchanged, while rates of IT/TT increased over time. Quality guidelines should take into account that health service use for IT/TT was similar to that of CT at EOL, although patterns of use differed both before and during the COVID-19 pandemic.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Canadian Institutes of Health Research (CIHR), Canada Graduate Scholarship - Doctoral (CGS-D), Peterborough KM Hunter Charitable Foundation Award in Cancer Research.

Disclosure

J. Lau: Financial Interests, Institutional, Funding, Dr Lau receives funding from the Health Canada Health Care Policy and Strategies Program outside the submitted work: Health Canada Health Care Policy and Strategies Program. M.K. Krzyzanowska: Financial Interests, Institutional, Principal Investigator, Dr Krzyzanowska is a principal investigator for trials with Eli Lilly, Exelixis, and Novartis and receives advisory board fees from Ipsen outside the submitted work: Eli Lilly, Exelixis, Novartis; Financial Interests, Institutional, Advisory Board, Dr Krzyzanowska is a principal investigator for trials with Eli Lilly, Exelixis, and Novartis and receives advisory board fees from Ipsen outside the submitted work: Ipsen. All other authors have declared no conflicts of interest.