Mortality outcomes after removal from end-of-life registers: A prospective cohort-controlled study

BackgroundEarlier identification and registration of people in the last year of life improves care quality and outcomes in general practice. However, there is little evidence on patients who subsequently no longer require end-of-life registration, nor on the safety or outcomes of de-registration following clinical review. AimTo determine the prevalence, safety, and prognostic validity of GP-led removal from the end-of-life register (EOL_R) using a systematic digital review process. Design and SettingObservational cohort study in eight practices in Wolverhampton, UK, using a whole-population integrated primary and secondary care dataset. MethodAll adults on the EOL_R were systematically reviewed using a digital end-of-life pathway (PRADA) incorporating robotic process analysis of recognised end-of-life care markers. GPs recorded a binary decision to retain or remove patients from the register. Mortality outcomes were compared with those retained on the EOL_R, a tightly propensity-matched cohort not on the register, and the residual general population over 15 months. ResultsOf 422 registered patients, 33 (7.8%) were removed following GP assessment. One-year survival in the removed group was statistically indistinguishable from the propensity-matched control cohort, and survival was significantly higher versus those retained on the EOL_R (60.4%, p<0.001). Removal demonstrated a negative predictive value for mortality of 90.9%. ConclusionGP-led removal from the end-of-life register can be undertaken safely and identifies a distinct group with substantially better prognosis. Digital systems that support systematic review, documentation, and follow-up should be incorporated into routine practice and reflected in national guidance and the Quality and Outcomes Framework. Statement boxesO_ST_ABSWhat is knownC_ST_ABSEarlier identification and registration of people in the last year of life improves care coordination and outcomes. However, little is known about patients who subsequently no longer require end-of-life registration, or whether removal following clinical review is associated with adverse outcomes. What this study addsIn a whole-population primary care cohort, General Practitioners removed 33 patients (7.8% of those registered) from the end-of-life register following structured clinical review. One-year survival in this group was equivalent to a tightly matched cohort not on the register, and survival was substantially higher than among patients retained on the register. A simple robotic based review process can prompt systematic reassessment, capture GP clinical judgement, and enable prospective monitoring following removal. ImplicationsEnd-of-life registration should be treated as a dynamic process requiring ongoing clinical review. Digital systems can support safe removal from registers by documenting decisions and embedding follow-up for patients whose prognosis remains uncertain. Evidence-based guidance and governance processes for a review process are needed to ensure people are not retained on registers unnecessarily. How this fits inEnd-of-life registers are intended to support proactive care for people in their last year of life, yet there is little evidence about patients who later stabilise and may no longer require registration. In eight UK practices, a systematic GP review supported by a digital end-of-life pathway identified a small but clinically important group suitable for removal, without adverse mortality outcomes. Beyond improving register accuracy, structured review creates opportunities for meaningful dialogue with patients, shared reassessment of care goals, and appropriate de-escalation of end-of-life labelling. Embedding routine, structured review with documented decisions and follow-up may also reduce unnecessary clinical workload and strengthen governance in primary care. Novelty StatementSystematic GP review of end-of-life registers, as stipulated in GMC guidance, is rarely evaluated. We propose a digitally driven, systematic and dynamic clinical-governance-led approach to register review.