A recent study published in the journal, BMJ Supportive & Palliative Care, reports discrepancy between death date recorded by general practitioners (GPs) and that recorded in national data in almost a quarter of cases. Chances of discrepancy in death date were more likely in individuals who died young, those with an external death cause and in practices with historically large discrepancies.
Authors linked GP reported death date from the Clinical Practice Research Datalink with national data from the Office for National Statistics (ONS) for 118,571 patients (aged between 18 and 105 years at death) who died between September 2010 and September 2015.
Death dates recorded in GPs record preceded ONS dates in 2.9% of cases, and followed in 20.3% of cases, whereas both dates perfectly matched in 76.8% of cases. In 1.5% cases, GP-recorded death date was ≥4 weeks later than ONS dates. Chances of GP-recorded death date >4 weeks, later than ONS date, were higher in those who died at 18-29 vs 80-89 years of age (aOR, 1.69; 95% CI, 1.40-2.04) and those with an external cause of death (aOR, 2.28; 95% CI, 2.10-2.48). General practices with greatest vs lowest discrepancies had higher odds of substantially recording discordant dates (OR, 189; 95% CI, 112-340).
Authors suggest that although GP date is adequate for majority of research and policy needs, ONS date should be favoured where available. However, cases where a more accurate date of death is required, GP date of death should be treated with caution. Authors comment: “Further research into the cause of such wide interpractice variation is needed.”
