Post‐Mortem Community Surveillance of COVID‐19: Implementation and Evaluation of a Pilot System in the Funeral Sector in England, UK, January 2021 to February 2022

TL;DR

This study tested a new method to detect missed COVID-19 deaths by collecting samples from funeral homes in the UK, finding some cases not identified before death.

Contribution

The study introduces and evaluates a novel post-mortem surveillance system using funeral directors to detect undiagnosed community deaths from SARS-CoV-2.

Findings

Eight post-mortem SARS-CoV-2 cases were detected, three of which were not identified during life.

The program was acceptable to funeral directors and families, but participation was limited by workload and legal constraints.

Data completeness and timeliness were good, but consent rates varied significantly over time.

Abstract

Early in the COVID‐19 pandemic, due to limited testing, a potential gap in capturing SARS‐CoV‐2‐positive community deaths was identified. Post‐mortem testing for respiratory viruses had never been implemented in the United Kingdom.

Through implementing and evaluating a pilot, we aimed to establish feasibility and acceptability of post‐mortem SARS‐CoV‐2 surveillance using funeral directors (FDs) to capture ‘missed’ COVID‐19 community deaths.

Between January 2021 and February 2022, four FDs took upper respiratory tract samples from eligible people who died outside hospital. We tested for SARS‐CoV‐2 and other respiratory viruses using reverse transcription‐polymerase chain reaction and matched results to the national COVID‐19 mortality dataset. We evaluated the pilot for acceptability, data completeness and timeliness, and simplicity, using semi‐structured interviews, a questionnaire, and data audit.

Two thousand eight hundred sixty‐five deaths were handled by FDs: 998 were assessed for eligibility, 342 were eligible 81 were tested. Eight were SARS‐CoV‐2‐positive, of which three were not identified by ante‐mortem clinical testing. The programme was acceptable in principle to FDs and families, but FDs' participation was limited by the burden of legal requirements and existing workloads. Families' willingness to consent fluctuated (monthly consent rate 4–83%, overall 30%); fewer consented when overall cases were low. Completeness and timeliness of data was good. FDs judged the programme simple.

The pilot established feasibility and demonstrated, even with small numbers, the ability to detect ‘missed’ deaths. There were significant obstacles to implementation. Alternative settings for taking specimens are being explored instead to address this gap in national surveillance.

Figures