Digital phenotyping and additive multimorbidity scoring to operationalise very-high-risk secondary prevention in primary care
Z Mathieson, R Simmonds-Rosten, E Shuker, W Yang, R VarmaAbstract
Background
Patients with established atherosclerotic cardiovascular disease (ASCVD) frequently remain at high residual risk due to suboptimal lipid control and clustering of cardiometabolic comorbidities. Scalable, implementation-ready approaches to systematically identify and prioritise these individuals in UK primary care remain elusive(1). We evaluated a digitally enabled phenotyping strategy, combined with a pragmatic additive cardiovascular risk management (CVRM) care-gap score, to quantify modifiable residual burden.
Methods
Within a primary care population of 10,306 adults, a standardised electronic search (CVDPL2F) identified patients with established ASCVD (CHD, PCI/stent, PAD, stroke/TIA), excluding familial hypercholesterolaemia and those using inclisiran, with LDL-C >1.8 mmol/L or non-HDL >2.5 mmol/L (2). Structured EHR data were extracted cross-sectionally. A CVRM care-gap score was defined as: baseline 2 (ASCVD + LDL above target) plus 1 point each for diabetes (HbA1c >53 mmol/mol), hypertension (SBP >140 mmHg), CKD (eGFR <60), and obesity (BMI >30 kg/m²). Risk modifiers (smoker, COPD, heart failure, AF, fatty liver) were flagged but not scored. Primary analysis quantified multimorbidity burden and treatment gaps.
Results
Of 10,306 patients, n=173 (1.7%) met criteria. Mean age 71.1±12.2 years; 57.2% male. Mean LDL-C 2.7mmol/L; 44.5% >2.6 mmol/L. SBP >140 mmHg in 7.5%; diabetes in 24.9% (HbA1c >53: 11.6%); CKD in 26.0%; BMI >30 in 24.9%. Mean CVRM score 3.2±0.7; 33.5% scored 4 or more. Two or more risk modifiers were present in 34.7% of cases.
Conclusion
Digital phenotyping identified a concentrated, very-high-risk ASCVD cohort with substantial multimorbidity and modifiable care gaps. The CVRM score provides a scalable prioritisation framework for structured community-based secondary prevention. A prospective longitudinal evaluation is underway to quantify intervention intensity (pharmacological and lifestyle) and the associated change in mean CVRM score, providing an objective measure of individualised care-gap modification within primary care.Figure 1