Summary
A systemic review can be written on different thresholds for different age in CVD prediction models. The existing models for CVD risk prediction use varying cut-off points for waist circumference, BMI, blood pressure and other risk factors for predicting CVD events. These cut-off points are largely derived from studies in Western populations and may not accurately stratify risk in other ethnic groups. The commonly used cut-off points for waist circumference and BMI to define obesity may need to be lowered for some ethnic groups. Different waist circumference cut-off points may be needed for men and women. There is considerable heterogeneity in the cut-off points used to define high risk in different studies. The cut-off points that optimise sensitivity and specificity may differ from those that maximise AUC. The choice of cut-off point depends on the purpose and consequences of risk stratification. Higher cut-off points can be used when false positives have less serious consequences. Lower cut-off points are needed when it is important not to miss high risk individuals. Recent studies have developed new cut-off points for waist circumference and BMI in some ethnic groups. The cut-off points for blood pressure also vary between guidelines. Lower cut-off points for high normal blood pressure and hypertension tend to be used in recent guidelines. Different cut-off points may be needed for older and younger adults. Some studies suggest lower blood pressure cut-off points should be used to define high risk in young and middle aged adults. Close monitoring or treatment may be needed for younger adults with blood pressure previously considered normal or high normal. Most existing CVD risk prediction models have not considered age-specific thresholds. The development of more comprehensive models that incorporate age and ethnicity-specific cut-off points could improve risk stratification.
A laboratory-based risk score was used to estimate cardiovascular disease risk in different countries, taking into account factors such as age, sex, smoking, blood pressure, diabetes, and cholesterol levels. Different thresholds were applied to determine high risk levels based on country income.
Published By:
P Ueda, M Woodward, Y Lu, K Hajifathalian… - The lancet Diabetes & …, 2017 - thelancet.com
Cited By:
0
Adding extra variables to CKD risk models has not been well studied, and the impact of using these models on patient care and outcomes is unknown. Incorporating easily assessable risk factors into CKD prediction models can be useful, particularly in the context of automatic reporting of estimated GFR.
Published By:
JB Echouffo-Tcheugui, AP Kengne - PLoS medicine, 2012 - journals.plos.org
Cited By:
0
Minimum sample size for studies was set at 20 participants to reduce small study effects. The analysis included 42 cross sectional studies, 19 randomized controlled trials, one cohort study, and one case-control study.
Published By:
C Friedemann, C Heneghan, K Mahtani, M Thompson… - Bmj, 2012 - bmj.com
Cited By:
0
Clinical prediction models help estimate a patient's risk of disease based on their characteristics. However, determining the appropriate risk threshold for intervention can be challenging. Presenting results for multiple risk thresholds can improve outcomes.
Published By:
, … 'Evaluating diagnostic tests and prediction models' of … - BMC medicine, 2019 - Springer
Cited By:
0
Global Burden of Disease Study 2016 conducted a systematic analysis on 84 risks or clusters of risks from 1990 to 2016.
Published By:
S Kaptoge, L Pennells, D De Bacquer… - The Lancet global …, 2019 - thelancet.com
Cited By:
0
The population attributable fraction for cardiovascular events due to high blood pressure is 23.8%. Different blood pressure ranges have varying number needed to treat to prevent one cardiovascular event.
Published By:
D Luo, Y Cheng, H Zhang, M Ba, P Chen, H Li, K Chen… - bmj, 2020 - bmj.com
Cited By:
0
Risk factors for developing colon, rectal or colorectal cancer were identified using a combination of two or more risk factors to determine individuals at higher risk. Details on the development and performance of the risk model were outlined.
Published By:
JA Usher-Smith, FM Walter, JD Emery, AK Win… - Cancer prevention …, 2016 - AACR
Cited By:
0
Heterogeneity in BMI, WC, and WHR was found in studies based on AUC, sensitivity, and specificity. Optimal cut-off points varied widely between 49.0 and 99.7.
Published By:
M Darbandi - Preventing chronic disease, 2020 - cdc.gov
Cited By:
0
None of the traditional hypertension risk prediction models considered genetic factors, but recent models are starting to incorporate them. AUC ranges from 0.64 to 0.97, and C-statistic ranges from 60% to 90%.
Published By:
D Sun, J Liu, L Xiao, Y Liu, Z Wang, C Li, Y Jin, Q Zhao… - PloS one, 2017 - journals.plos.org
Cited By:
0
The study analyzed three risk scores and found that QRISK2 may not accurately predict individual cardiovascular disease risk in high-risk populations. Factors such as calendar year, ethnicity, diabetes mellitus, and the number of medical records influenced the deviations between QRISK2 and individual predicted risks.
Published By:
TP van Staa, M Gulliford, ESW Ng, B Goldacre… - PloS one, 2014 - journals.plos.org
Cited By:
0