Development and 10-year validation of QRISK4-Lifetime for cardiovascular risk prediction in 13 million adults
Year: 2026
Type: preprint
Study design: Observational Study
Abstract: Objective To develop QRISK4-Lifetime to estimate cardiovascular disease (CVD) risk to age 95, compare its 10-year performance with QRISK4 and examine differences in risk classification between the two time horizons. Methods We used the linked electronic health datasets and cohort framework of the original QRISK4 study. We fitted sex-specific cause-specific Cox models for incident CVD and death from other causes, using attained age at cohort entry. Predicted risks combined both cause-specific hazards. We assessed discrimination and calibration at 10 years overall, by age and ethnicity. We compared characteristics of patients with a 10-year QRISK4-Lifetime risk ≥ 10% with an equal-sized group selected using lifetime risk to age 95 Results There were 202,424 CVD events in the English derivation cohort of 9,976,306 people and 65,253 events in the validation cohort of 3,246,602 people. Overall C-indices were 0.860 (95% CI 0.858 to 0.862) in women and 0.845 (0.843 to 0.847) in men for both QRISK4-Lifetime and QRISK4. Observed risks were close to predicted risks. Each risk classification approach selected 310,648 patients (9.6%); 262,417 (8.1% of the cohort) met only the lifetime threshold. Of these, 72.8% were younger than 40 years and 79.5% were men. Conclusions QRISK4-Lifetime had similar performance to QRISK4 when evaluated at 10 years. Lifetime assessment identified a substantially younger population with increased lifetime risks but below the specified 10-year threshold of 10%, supporting its potential complementary role in prevention discussions. The validity of lifetime estimates and the clinical consequences of using both time horizons require further evaluation.
Source: medRxiv
Authors: Julia Hippisley-Cox, Carol Coupland
Institution: Queen Mary University of London
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: green
Funder: Queen Mary University of London
OpenAlex ID: W7225648999