Background At the beginning on the COVID-19 pandemic, advance care planning (ACP) was widely encouraged and endorsed for adults with serious illness to ensure their treatment and care preferences would be honoured, including location of death, often considered a surrogate quality indicator for end-of-life care. Coordinate My Care (CMC) represents the UK’s largest Electronic Palliative Care Coordination System that comprises an ACP component.

We aimed to examine the impact of ACP on place of death for people who died during the COVID-19 pandemic with a CMC record.

Methods Retrospective cohort analysis of CMC records for people aged over 18 who died between 20/03/20 and 05/03/21 with recorded place of death. Socio-demographic, clinical and ACP-related factors associated with achieving preferred place of death (PPD) were examined using logistic regression.

Results 11,913 records were included. 76.9% patients died in their preferred place location of death (57.7% Home, 31.4% Care Home, 7.5% Hospice, 3.3% Hospital, 0.1% Other). An increased likelihood of dying in PPD was associated with a ‘Not for resuscitation’ (DNACPR) status (OR=1.51, 95% CI 1.17 to 1.93), a Ceiling of Treatment for Symptomatic Treatment (when compared to Full active treatment, OR=3.52, 95% CI 2.77 to 4.50), documented family discussions regarding resuscitation recommendations (OR=1.51, 95% CI 1.33 to 1.72) and 2+ non-urgent care record views in the 30 days before death (OR=1.27, 95% CI 1.13 to 1.43). People from materially deprived areas had a decreased likelihood of dying in their PPD (OR= 0.65, 95% CI 0.54 to 0.79).

Conclusions Modifiable elements of ACP significantly influence place of death, even when controlling for socio-economic and demographic determinants. In times of crisis, effective ACP is central to delivering high quality end-of-life care; ACP related factors must be considered in ongoing research on end-of-life outcomes.