Cardiovascular risk increases dramatically with age, leading to nearly universal risk-based statin eligibility in the elderly population. To limit overtreatment, elderly individuals at truly low risk ...need to be identified.
Discovering “negative” risk markers able to identify elderly individuals at low short-term risk for coronary heart disease and cardiovascular disease.
In 5,805 BioImage participants (mean age 69 years; median follow-up 2.7 years), the authors evaluated 13 candidate markers: coronary artery calcium (CAC) = 0, CAC ≤10, no carotid plaque, no family history, normal ankle-brachial index, test result <25th percentile (carotid intima-media thickness, apolipoprotein B, galectin-3, high-sensitivity C-reactive protein, lipoprotein(a), N-terminal pro–B-type natriuretic peptide, and transferrin), and apolipoprotein A1 >75th percentile. Negative risk marker performance was compared using patient-specific diagnostic likelihood ratio (DLR) and binary net reclassification index (NRI).
CAC = 0 and CAC ≤10 were the strongest negative risk markers with mean DLRs of 0.20 and 0.20 for coronary heart disease (i.e., ≈80% lower risk than expected from traditional risk factor assessment) and 0.41 and 0.48 for cardiovascular disease, respectively, followed by galectin-3 <25th percentile (DLR 0.44 and 0.43, respectively) and absence of carotid plaque (DLR 0.39 and 0.65, respectively). Results obtained by other candidate markers were less impressive. Accurate downward risk reclassification across the Class I statin-eligibility threshold defined by the American College of Cardiology/American Heart Association was largest for CAC = 0 (NRI 0.23) and CAC ≤10 (NRI 0.28), followed by galectin-3 <25th percentile (NRI 0.14) and absence of carotid plaque (NRI 0.08).
Elderly individuals with CAC = 0, CAC ≤10, low galectin-3, or no carotid plaque had remarkable low cardiovascular risk, calling into question the appropriateness of a treat-all approach in the elderly population.
Display omitted
Timely identification of patients at a high risk of clinical deterioration is key to prioritizing care, allocating resources effectively, and preventing adverse outcomes. Vital signs-based, ...aggregate-weighted early warning systems are commonly used to predict the risk of outcomes related to cardiorespiratory instability and sepsis, which are strong predictors of poor outcomes and mortality. Machine learning models, which can incorporate trends and capture relationships among parameters that aggregate-weighted models cannot, have recently been showing promising results.
This study aimed to identify, summarize, and evaluate the available research, current state of utility, and challenges with machine learning-based early warning systems using vital signs to predict the risk of physiological deterioration in acutely ill patients, across acute and ambulatory care settings.
PubMed, CINAHL, Cochrane Library, Web of Science, Embase, and Google Scholar were searched for peer-reviewed, original studies with keywords related to "vital signs," "clinical deterioration," and "machine learning." Included studies used patient vital signs along with demographics and described a machine learning model for predicting an outcome in acute and ambulatory care settings. Data were extracted following PRISMA, TRIPOD, and Cochrane Collaboration guidelines.
We identified 24 peer-reviewed studies from 417 articles for inclusion; 23 studies were retrospective, while 1 was prospective in nature. Care settings included general wards, intensive care units, emergency departments, step-down units, medical assessment units, postanesthetic wards, and home care. Machine learning models including logistic regression, tree-based methods, kernel-based methods, and neural networks were most commonly used to predict the risk of deterioration. The area under the curve for models ranged from 0.57 to 0.97.
In studies that compared performance, reported results suggest that machine learning-based early warning systems can achieve greater accuracy than aggregate-weighted early warning systems but several areas for further research were identified. While these models have the potential to provide clinical decision support, there is a need for standardized outcome measures to allow for rigorous evaluation of performance across models. Further research needs to address the interpretability of model outputs by clinicians, clinical efficacy of these systems through prospective study design, and their potential impact in different clinical settings.
The risk of thermal runaway in lithium-ion battery (LIB) attracts significant attention from domains of society, industry, and academia. However, the thermal runaway prediction in the framework of ...system safety requires further efforts. In this paper, we propose a methodology for dynamic risk prediction by integrating fault tree (FT), dynamic Bayesian network (DBN) and support vector regression (SVR). FT graphically describes the logic of mechanism of thermal runaway. DBN allows considering multiple states and uncertain inference for providing quantitative results of the risk evolution. SVR is subsequently utilized for predicting the risk from the DBN estimation. The proposed methodology can be applied for risk early warning of LIB thermal runaway.
Over the last dozen years, the topic of small and medium enterprise (SME) default prediction has developed into a relevant research domain that has grown for important reasons exponentially across ...multiple disciplines, including finance, management, accounting, and statistics. Motivated by the enormous toll on SMEs caused by the 2007–2009 global financial crisis as well as the recent COVID-19 crisis and the consequent need to develop new SME default predictors, this paper provides a systematic literature review, based on a statistical, bibliometric analysis, of over 100 peer-reviewed articles published on SME default prediction modelling over a 34-year period, 1986 to 2019. We identified, analysed and reviewed five streams of research and suggest a set of future research avenues to help scholars and practitioners address the new challenges and emerging issues in a changing economic environment. The research agenda proposes some new innovative approaches to capture and exploit new data sources using modern analytical techniques, like artificial intelligence, machine learning, and macro-data inputs, with the aim of providing enhanced predictive results.
Abstract The focus of this paper is on what role dynamic risk factors should play in the development of explanations of crime. Following a discussion of the nature of explanation we propose that in ...their current form dynamic risk factors should not be regarded as causes of crime because they cannot be coherently conceptualized as causal mechanisms. We then examine the issue of how best to ascertain whether risk factors are causes and a number of methodological guidelines are suggested to assist in this evaluation process. Finally, we conclude that dynamic risk factors are valuable predictors of recidivism and that, additionally, suitably reconstructed they can serve an important methodological function in identifying the causes of crime and reoffending.
•Early mortality after radical RT/CRT in HNSCC patients is often not cancer-related.•An early non-cancer risk prediction model was developed.•After internal validation, the AUC was 0.74 and ...calibration was good.•Pneumonia was estimated to be the cause of death in 32% of early non-cancer deaths.•The provided risk estimates can help select patients for supportive interventions.
In patients with head and neck squamous cell carcinoma (HNSCC), curative-intent radiotherapy (RT) and chemoradiotherapy (CRT) are associated with substantial acute morbidity and 5–10% of patients die within 180 days of treatment initiation. Most of these early deaths occur without HNSCC recurrence or progression and may therefore be preventable to some extent. We developed a prediction tool to estimate the risk of non-HNSCC mortality occurring within the first 180 days followingRT/CRT initiation.
Patients with HNSCC treated with RT/CRT, including postoperative RT/CRT, at Rigshospitalet or Herlev Hospitals between 2010–2017 were identified in the Danish Head and Neck Cancer Group (DAHANCA) database. Predictor variables included age, stage, performance status, tumor subsite including p16 status, comorbidity, postoperative status, smoking and pre-treatment albumin levels. The 180-day non-HNSCCmortality risk was estimated by combining cause-specific Cox regression models.
We included 2209 patients. The 180-day non-HNSCCmortality rate was 4.4% and almostone third (31.6%) of non-HNSCCdeathswere caused by pneumonia.After internal model validation, the area under the receiver operating curve was 0.74 (95% CI: 0.66–0.81) and calibration was good for risk predictions ranging from 0% to 20%.
We developed a prediction tool to estimate the 180-day non-HNSCC mortality risk. This tool can be used to select high-risk patients for supportive interventions aiming to improve survival rates, and is availablefor interactive use at https://emriskpred.shinyapps.io/EMNED_App/.
The low density lipoprotein cholesterol to Apolipoprotein B (LDL-C/ApoB) ratio is a validated proxy for low density lipoprotein (LDL) particle size that can be easily calculated from a standard ...lipid/apolipoprotein profile. Whether it is predictive of cardiovascular events in patients with established atherosclerosis is not known and is addressed in the present investigation.
We determined the LDL-C/ApoB ratio in a cohort of 1687 subjects with established atherosclerosis. Prospectively, major cardiovascular events (MACE) including cardiovascular death, non-fatal myocardial infarction and non-fatal stroke were recorded over a period of 9.9 ± 4.6 years. The study covers >16,000 patient-years.
At baseline, the LDL-C/ApoB ratio was 1.36 ± 0.28 in our cohort. During follow up, a total of 558 first MACE were recorded. The LDL-C/ApoB ratio predicted MACE in univariate Cox proportional hazard analysis (HR 0.90 0.82–0.98; p = 0.014); this finding was confirmed after adjustment for age, gender, intensity of statin treatment, hypertension, history of smoking, type 2 diabetes, body mass index and ApoB (HR 0.87 0.78–0.97; p = 0.013).
The LDL-C/ApoB ratio is independently predictive of MACE in subjects with established atherosclerosis.
Display omitted
•LDL-C and ApoB are recommended targets for therapy in the latest ESC/EAS guidelines on dyslipidemia.•The atherogenicity of LDL-C particles depends on their size, but measuring LDL particle size in clinical routine is not feasible.•The LDL-C/ApoB ratio is an easily obtainable proxy of LDL size in clinical practice.•We show that in patients with established cardiovascular disease this ratio independently predicts MACE.•With new lipid-lowering drugs, e.g. triglyceride lowering agents, the LDL-C/ApoB ratio may be a valuable risk indicator.
This study aimed to evaluate employing the German Registry of Acute Aortic Dissection Type A (GERAADA) score to predict 30-day mortality in an aortic centre in the USA.
Between January 2010 and June ...2021, 689 consecutive patients underwent surgery for acute type A dissection at a single institution. Excluded were patients with missing clinical data (N = 4). The GERAADA risk score was retrospectively calculated via a web-based application. Model discrimination power was calculated with c-statistics from logistic regression and reported as the area under the receiver operating characteristic curve with 95% confidence intervals. The calibration was measured by calculating the observed versus estimated mortality ratio. The Brier score was used for the overall model evaluation.
Included were 685 patients mean age 60.6 years (SD: 13.5), 64.8% male who underwent surgery for acute type A aortic dissection. The 30-day mortality rate was 12.0%. The GERAADA score demonstrated very good discrimination power with an area under the receiver operating characteristic curve of 0.762 (95% confidence interval 0.703-0.821). The entire cohort's observed versus estimated mortality ratio was 0.543 (0.439-0.648), indicating an overestimation of the model-calculated risk. The Brier score was 0.010, thus revealing the model's acceptable overall performance.
The GERAADA score is a practical and easily accessible tool for reliably estimating the 30-day mortality risk of patients undergoing surgery for acute type A aortic dissection. This model may naturally overestimate risk in patients undergoing surgery in experienced aortic centres.
Introduction
Hepatic, pancreatic, and complex biliary (HPB) surgery can be associated with major morbidity and significant mortality. For the past 5 years, the American College of Surgeons–National ...Surgical Quality Improvement Program (ACS–NSQIP) has gathered robust data on patients undergoing HPB surgery. We sought to use the ACS–NSQIP data to determine which preoperative variables were predictive of adverse outcomes in patients undergoing HPB surgery.
Methods
Data collected from ACS–NSQIP on patients undergoing hepatic, pancreatic, or complex biliary surgery between 2005 and 2009 were analyzed (
n
= 13,558). Diagnoses and surgical procedures were categorized into 10 and eight groups, respectively. Seventeen preoperative clinical variables were assessed for prediction of 30-day postoperative morbidity and mortality. Multivariate logistic regression was utilized to develop a risk model.
Results
Of the 13,558 patients who underwent an HPB procedure, 7,321 (54%) had pancreatic, 4,881 (36%) hepatic, and 1,356 (10%) biliary surgery. Overall, 70.3% of patients had a cancer diagnosis. Post-operative complications occurred in 3,850 patients for an overall morbidity of 28.4%. Serious complications occurred in 2,522 (18.6%) patients; 366 patients died for an overall peri-operative mortality of 2.7%. Peri-operative outcome was associated with diagnosis and type of procedure. Hepatic trisectionectomy (5.8%) and total pancreatectomy (5.4%) had the highest 30-day mortality. Of the preoperative variables examined, age >74, dyspnea with moderate exertion, steroid use, prior cardiac procedure, ascites, and pre-operative sepsis were associated with morbidity and mortality (all
P
< 0.05).
Conclusions
While overall morbidity and mortality for HPB surgery are low, peri-operative outcomes are heterogeneous and depend on diagnosis, procedure type, and key clinical factors. By combining these factors, an ACS–NSQIP “HPB Risk Calculator” may be developed in the future to help better risk-stratify patients being considered for complex HPB surgery.
Aim: There is currently no consensus on the best risk assessment technique for predicting complications after hip surgery in the elderly, which is hindering the accuracy of surgical risk assessment. ...The goal of this study was to build a risk assessment model and evaluate its predictive value using the modified frailty index (5-mFI) and the prognostic nutritional index (PNI).
Methods: A retrospective investigation was undertaken on 150 patients (aged ≥60 years) who had hip fracture surgery. Using univariate and multivariate logistic regression models, the relationship between combined 5-mFI and PNI and the evaluation of postoperative unfavorable outcomes such as infection and unscheduled intensive care unit (ICU) admission was investigated. Finally, utilizing receiver operating characteristic (ROC) curve analysis, the model's predictive value for adverse outcomes following hip fracture surgery in elderly patients was assessed.
Results: Univariate and multivariate logistic analyses revealed that preoperative PNI, 5-mFI, ASA, and gender acted as independent predictors of adverse outcomes after hip fracture surgery in the elderly. According to the ROC curve analysis, the predictive model demonstrated a high predictive value for total postoperative complications (AUC: 0.788; 95%CI: 0.715-0.860; p<0.01), infectious complications (AUC: 0.798; 95% CI: 0.727-0.868; P<0.001), and unplanned ICU admission (AUC: 0.783; 95% CI: 0.705-0.861; P<0.001).
Conclusions: The multivariable evaluation model, which included 5-mFI and PNI, showed a high predictive value and can hence be applied to predict the adverse outcomes in elderly patients undergoing hip fracture surgery.