New QR4 algorithm outperforms previous models in predicting cardiovascular disease risk

In a recent study published in Nature Medicine, researchers introduced a new cardiovascular risk prediction algorithm.

Study: Development and validation of a new algorithm for improved cardiovascular risk prediction. Image Credit: Basicdog/Shutterstock.comStudy: Development and validation of a new algorithm for improved cardiovascular risk prediction. Image Credit: Basicdog/Shutterstock.com

Background

Cardiovascular disease (CVD) remains the leading cause of death worldwide. International guidelines recommend using risk prediction tools targeting at-risk populations for interventions.

European, the United States (US), and the United Kingdom (UK) guidelines recommend systematic coronary risk evaluation 2 (SCORE2), atherosclerotic CVD (ASCVD) score, and QRISK3, respectively.

Notably, recent studies have highlighted conditions linked to high CVD risk, such as cancer, down syndrome, and learning disability, among others, that these tools do not capture.

Current tools will underestimate risk if these conditions are independently associated with higher CVD risk. As such, people diagnosed with these conditions may not have the opportunity for beneficial interventions. Moreover, if overestimated, people may receive unnecessary interventions.

The study and findings

In the present study, researchers developed and validated a new CVD risk prediction tool, QR4. They used the data from the Clinical Practice Research Datalink (CPRD) GOLD and QResearch databases. Derivation and validation cohorts were generated using QResearch practices in England.

In addition, a second validation cohort was formed using the CPRD GOLD practices from Wales, Northern Ireland, and Scotland. Individuals aged 18­–84 between 2010 and 2021 were included.

Subjects with preexisting CVD, those missing deprivation data, and those taking statins were excluded. Participants were followed up until the diagnosis of CVD, death, or the end of the study.

The primary outcome was incident CVD, i.e., non-fatal or fatal myocardial infarction, transient ischemic stroke, ischemic/hemorrhagic stroke, or ischemic heart disease.

Secondary outcomes included coronary heart disease-related death, non-fatal myocardial infarction, and non-fatal or fatal stroke.

Tertiary outcomes were similar to secondary outcomes but additionally included cardiac arrhythmias, hypertension, and fatal congestive cardiac failure. The performance of QR4, ASCVD, and SCORE2 was compared using three outcome definitions.

Established risk factors from SCORE2, ASCVD, and QRISK3, as well as new candidate variables from the literature, were included as predictor variables. Cause-specific Cox models estimated the 10-year CVD risk, accounting for non-CVD mortality as a competing risk for males and females. Besides, three additional models (A–C) were developed.

Model A covered QRISK3 parameters without accounting for competing risks, and model B was similar to the main model, but the follow-up ended before the coronavirus disease 2019 (COVID-19) pandemic. In contrast, time since cancer diagnosis was included as a predictor variable in model C.

A decision curve analysis evaluated the net benefit of QR4 relative to model A and QRISK3, accounting for competing risks.

Findings

The QResearch derivation and validation cohorts comprised more than 9.97 and 3.24 million individuals, respectively, while the CPRD validation cohort comprised 3.54 million subjects.

The cohorts were generally similar, except that the QResearch cohorts had more complete data on body mass index (BMI), cholesterol, smoking, and ethnicity than the CPRD cohort. Within the derivation cohort, there were 202,424 cases of incident CVD.

In 2020, CVD rates were lower at 4.03 per 1,000 person-years but returned to pre-COVID-19 levels (4.31) in 2021. Non-CVD mortality rates increased between 2019 (3.45 per 1000 person-years) and 2020 (3.84) and remained elevated in 2021.

The team identified seven novel CVD predictors for females and males – lung, blood, brain, and oral cancers, learning disability, Down syndrome, and chronic obstructive pulmonary disease (COPD).

Further, there were two additional predictors for females – postnatal depression and preeclampsia. CVD risk in both sexes was not associated with asthma, hypothyroidism, hyperthyroidism, and antiphospholipid antibody syndrome, among others.

In females, CVD risk was not associated with endometriosis, in vitro fertilization, miscarriage, gestational diabetes, placental abruption, and polycystic ovarian syndrome.

The adjusted hazard ratios for several predictors, except for lung cancer, were higher at younger ages in females. The adjusted hazard ratios for blood and brain cancers declined with age in males. Estimates from the three additional models were similar to the main model.

The decision curve analysis suggested a slightly greater net benefit with QR4 than models A and QRISK3. QR4 was also more accurate than SCORE2 and ASCVD risk scores.

Conclusions

The researchers developed and validated QR4, a new CVD risk score incorporating nine novel predictors.

It can predict the 10-year CVD risk in both males and females. Its performance was more accurate than other CVD risk scores. Besides, QR4 accounts for competing risks (non-CVD death), reducing the overprediction of risks.

Overall, these findings may lead to significant improvements in health outcomes.

Journal reference:
Tarun Sai Lomte

Written by

Tarun Sai Lomte

Tarun is a writer based in Hyderabad, India. He has a Master’s degree in Biotechnology from the University of Hyderabad and is enthusiastic about scientific research. He enjoys reading research papers and literature reviews and is passionate about writing.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Sai Lomte, Tarun. (2024, April 23). New QR4 algorithm outperforms previous models in predicting cardiovascular disease risk. News-Medical. Retrieved on October 31, 2024 from https://www.news-medical.net/news/20240423/New-QR4-algorithm-outperforms-previous-models-in-predicting-cardiovascular-disease-risk.aspx.

  • MLA

    Sai Lomte, Tarun. "New QR4 algorithm outperforms previous models in predicting cardiovascular disease risk". News-Medical. 31 October 2024. <https://www.news-medical.net/news/20240423/New-QR4-algorithm-outperforms-previous-models-in-predicting-cardiovascular-disease-risk.aspx>.

  • Chicago

    Sai Lomte, Tarun. "New QR4 algorithm outperforms previous models in predicting cardiovascular disease risk". News-Medical. https://www.news-medical.net/news/20240423/New-QR4-algorithm-outperforms-previous-models-in-predicting-cardiovascular-disease-risk.aspx. (accessed October 31, 2024).

  • Harvard

    Sai Lomte, Tarun. 2024. New QR4 algorithm outperforms previous models in predicting cardiovascular disease risk. News-Medical, viewed 31 October 2024, https://www.news-medical.net/news/20240423/New-QR4-algorithm-outperforms-previous-models-in-predicting-cardiovascular-disease-risk.aspx.

Comments

The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of News Medical.
Post a new comment
Post

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.

You might also like...
COVID-19 vaccination lowers risk of developing severe cardiovascular conditions