Objectives
To determine the occurrence and clinicodemographic associations of hospital-based specialist palliative care (SPC) referral before and during the COVID-19 pandemic, timing of completed SPC consultation and comparative prevalence of ‘no cardiopulmonary resuscitation (CPR)’ orders, and end-of-life medication use, according to SPC involvement.
Design
Cross-sectional secondary analysis of a retrospective cohort study with a pre-pandemic (November 2019 to February 2020) group (Pre-COVID, n=170) and two intra-pandemic (March to August 2020) groups, one without (COVID–ve, n=170) and one with COVID-19 infection (COVID+ve, n=85). In the cohort study, Pre-COVID and COVID–ve group decedents were matched 2:1 on age, sex and care service (internal medicine/intensive care unit (ICU)) at death with COVID+ve decedents. In our current secondary analysis, clinicodemographic variables associated with SPC referral were examined in multivariable logistic regression, reporting adjusted ORs (aORs) and 95% CIs.
Setting
One quaternary and two tertiary acute care hospitals.
Participants
Decedent cohort with a terminal hospital admission (N=425).
Main outcome measures
SPC referral (yes/no) and timing of completed SPC consultation before death. Additional outcomes included ‘no CPR’ status and end-of-life medication prescription and dosing.
Results
SPC referral occurred in 70 (41.2%), 71 (41.8%) and 26 (30.6%) of the Pre-COVID, COVID–ve and COVID+ve groups, respectively (p=0.18). The aORs for SPC referral were lower for deaths in ICU (0.07; 95% CI 0.03 to 0.16) and admissions from nursing homes/long-term care (0.45; 95% CI 0.23 to 0.9), and higher for active cancer (2.5; 95% CI 1.39 to 4.39). Recipients of SPC consultation, compared with non-recipients, more frequently had a ‘no CPR’ order, had it placed earlier and were more frequently prescribed palliative end-of-life medications.
Conclusions
Hospital SPC consultation rates early in the COVID-19 pandemic were largely maintained at pre-pandemic levels. Having active cancer was positively associated with SPC referral, whereas both ICU death and having a nursing home/long-term care pre-admission source were negatively associated with referral.