In an Australian first, Lumos data have been used to estimate cardiovascular risk in Australian patients.
Cardiovascular disease (CVD) is the leading cause oaf morbidity and mortality in Australia. Risk equations are pivotal in guiding clinical decision-making, targeting preventive measures and optimising resource allocation for individuals at high risk of CVD. The current national risk equations (AusCVDRisk) are derived from New Zealand data and have not been validated using Australian data, limiting confidence in their application to local populations.
The Lumos program provides a secure, state-wide data linkage between general practice, hospital and mortality records, offering population-level coverage and longitudinal continuity of care data. This infrastructure allows for large-scale, representative analyses of real-world healthcare data, enabling the development and validation of risk models that reflect actual clinical practice patterns and patient diversity across New South Wales.
Validated sex-specific 5-year cardiovascular disease risk tool.
Based on Australian data across the complete patient journey.
Demonstrates improved classification of CVD risk.
Kuo, Barbieri et al.1 present a study using Lumos data to develop and validate sex-specific 5-year CVD risk equations.
The study used patient data extracted for 5.6 million unique patients who had attended 680 general practices2. At the time, this comprised approximately 25% of NSW general practices.
Individuals aged 30 – 74 years on 1 January 2017 with no prior CVD history and at least one record for a clinical measurement or pathology test were included in the analysis.
Sex-specific Cox proportional hazards models were used to estimate 5-year risk of a fatal or non-fatal CVD event. Predictors included demographics, smoking, chronic conditions, clinical variables and medications.
This study developed and validated sex-specific 5-year CVD risk equations that performed consistently across demographic and geographic subgroups.
The new model demonstrated improved classification of CVD risk compared to the current AusCVDRisk equation. This means that more patients with a true CVD risk are correctly recognised as having higher risk, while patients with a low CVD risk are less likely to be flagged for unnecessary treatment or intervention.
Inform and strengthen cardiovascular prevention strategies
Embed as an automated application in general practice to assist in identifying and managing patients at risk of CVD.
Health policymakers, primary-care networks, and digital-health developers seeking to strengthen cardiovascular prevention strategies through evidence-based, automated risk assessment embedded in Australian general practice.
Options are being explored to operationalise the new models within general practice software for automated, locally validated CVD risk estimation. For the first time, this would provide general practice with 5-year CVD risk estimates based on comprehensive Australian data.
Lumos is the largest collaboration between NSW general practices, NSW Health and NSW Primary Health Networks (PHNs) to date.
Lumos combines deidentified data from general practices, hospitals and other health services. For more information on the steps between data collection and release of data for analytics, see Lumos Data Timelines Explained.
Lumos does not currently include Aboriginal Medical Services and the Aboriginality of patients cannot be identified in the data. Lumos is working with Aboriginal leaders and communities to implement Indigenous Data Sovereignty and Governance to address this gap.
Information contained in this publication is based on knowledge and understanding at the time of writing and is subject to change.