Eleanor Morgan, Paul Cleary, Tom Wingfield
INTRODUCTION: People diagnosed with tuberculosis (TB) after death (postmortem) experience the ultimate diagnostic delay. We aimed to identify social and health factors associated with postmortem TB diagnosis in England. METHODS: We conducted a national retrospective cohort study using routinely collected surveillance data from the UK Health Security Agency's National TB Surveillance System (NTBS) between 1 January 2010 and 31 December 2022. Of 72 058 people notified with TB, 19 without a recorded diagnostic outcome were excluded, leaving 72 039 for analysis. The primary outcome was postmortem versus antemortem TB diagnosis. Associations were assessed using univariate and multivariable logistic regression. RESULTS: Of 72 039 participants, 574 were diagnosed postmortem, averaging 0.84 cases per week. With the exception of age 0-4 years (adjusted odds ratio, aOR, 5.36 (95%CI 1.62-17.67), the likelihood of postmortem diagnosis increased markedly with age, from aOR 2.96 (95% CI 1.14 to 7.64) at age 35-39 years to 85.67 (95% CI 32.34 to 226.97) at age ≥90 years. Male sex (aOR 1.62; 95% CI 1.31 to 2.00), UK birth (aOR 1.53; 95% CI 1.10 to 2.14), alcohol misuse (aOR 3.08; 95% CI 1.30 to 7.30) and drug misuse (aOR 2.48; 95% CI 1.21 to 5.12) were associated with increased odds of postmortem diagnosis. Compared with London, all other National Health Service regions had increased odds of postmortem diagnosis. BCG vaccination (aOR 0.49; 95% CI 0.29 to 0.84) and pulmonary TB (aOR 0.60; 95% CI 0.48 to 0.75) were associated with lower odds. DISCUSSION: Postmortem TB diagnosis in England is frequent and associated with being in early childhood (0-4 years), older age, male sex, UK birth, substance misuse and notification outside London. Recognising postmortem TB diagnosis as a 'never event' and developing targeted interventions to reduce diagnostic delays are essential to achieve England's TB elimination goals.