Current Risk Assessment Tools Research Articles

Identifying novel biomarkers using proteomics to predict cancer-associated thrombosis.

Comprehensive protein analyses of plasma are made possible by high-throughput proteomic screens, which may help find new therapeutic targets and diagnostic biomarkers. Patients with cancer are frequently affected by venous thromboembolism (VTE). The limited predictive accuracy of current VTE risk assessment tools highlights the need for new, more targeted biomarkers. Although coagulation biomarkers for the diagnosis, prognosis, and treatment of VTE have been investigated, none of them have the necessary clinical validation or diagnostic accuracy. Proteomics holds the potential to uncover new biomarkers and thrombotic pathways that impact the risk of thrombosis. This review explores the fundamental methods used in proteomics and focuses on particular biomarkers found in VTE and cancer-associated thrombosis.

Symptomatic cancer diagnosis in general practice: a critical perspective of current guidelines and risk assessment tools.

Symptomatic cancer diagnosis in general practice: a critical perspective of current guidelines and risk assessment tools.

Cardiovascular Health and Disease in the Pakistani American Population.

This narrative review seeks to elucidate clinical and social factors influencing cardiovascular health, explore the challenges and potential solutions for enhancing cardiovascular health, and identify areas where further research is needed to better understand cardiovascular issues in native and American Pakistani populations. The prevalence of cardiometabolic disease is high not only in Pakistan but also among its global diaspora. This situation is further complicated by the inadequacy of current cardiovascular risk assessment tools, which often fall short of accurately gauging the risk among Pakistani individuals, underscoring the urgent need for more tailored and effective assessment methodologies. Moreover, social determinants play a crucial role in shaping cardiovascular health. The burden of cardiovascular disease and upstream risk factors is high among American Pakistani individuals. Future research is needed to better understand the heightened risk of cardiovascular disease among Pakistani individuals.

Prediction of Venous Thromboembolism in Diverse Populations Using Machine Learning and Structured Electronic Health Records.

Venous thromboembolism (VTE) is a major cause of morbidity and mortality worldwide. Current risk assessment tools, such as the Caprini and Padua scores and Wells criteria, have limitations in their applicability and accuracy. This study aimed to develop machine learning models using structured electronic health record data to predict diagnosis and 1-year risk of VTE. We trained and validated models on data from 159 001 participants in the Mount Sinai Data Warehouse. We then externally tested them on 401 723 participants in the UK Biobank and 123 039 participants in All of Us. All data sets contain populations of diverse ancestries and clinical histories. We used these data sets to develop small, medium, and large models with increasing features on a range of optimizing portability to maximizing performance. We make trained models publicly available in click-and-run format at https://doi.org/10.17632/tkwzysr4y6.6. In the holdout and external test sets, respectively, models achieved areas under the receiver operating characteristic curve of 0.80 to 0.83 and 0.72 to 0.82 for VTE diagnosis prediction and 0.76 to 0.78 and 0.64 to 0.69 for 1-year risk prediction, significantly outperforming the Padua score. Models also demonstrated robust performance across different VTE types and patient subsets, including ethnicity, age, and surgical and hospitalization status. Models identified both established and novel clinical features contributing to VTE risk, offering valuable insights into its underlying pathophysiology. Machine learning models using structured electronic health record data can significantly improve VTE diagnosis and 1-year risk prediction in diverse populations. Model probability scores exist on a continuum, affecting mortality risk in both healthy individuals and VTE cases. Integrating these models into electronic health record systems to generate real-time predictions may enhance VTE risk assessment, early detection, and preventative measures, ultimately reducing the morbidity and mortality associated with VTE.

Epigenetic age biomarkers and risk assessment in adult spinal deformity: a novel association of biological age with frailty and disability.

Surgery for spinal deformity has the potential to improve pain, disability, function, self-image, and mental health. These surgical procedures carry significant risk and require careful selection, optimization, and risk assessment. Epigenetic clocks are age estimation tools derived by measuring the methylation patterns of specific DNA regions. The study of biological age in the adult deformity population has the potential to shed insight onto the molecular basis of frailty and to improve current risk assessment tools. Adult patients who underwent deformity surgery were prospectively enrolled. Preoperative whole blood samples were used to assess epigenetic age and telomere length. DNA methylation patterns were quantified and processed to extract 4 principal component (PC)-based epigenetic age clocks (PC Horvath, PC Hannum, PC PhenoAge, and PC GrimAge) and the instantaneous pace of aging (DunedinPACE). Telomere length was assessed using both quantitative polymerase chain reaction (telomere to single gene [T/S] ratio) and a methylation-based telomere estimator (PC DNAmTL). Patient demographic and surgical data included age, BMI, American Society of Anesthesiologists Physical Status Classification System class, and scores on the Charlson Comorbidity Index, adult spinal deformity frailty index (ASD-FI), Edmonton Frail Scale (EFS), Oswestry Disability Index, and Scoliosis Research Society-22r questionnaire (SRS-22r). Medical or surgical complications within 90 days of surgery were collected. Spearman correlations and beta coefficients (β) from linear regression, adjusted for BMI and sex, were calculated. Eighty-three patients were enrolled with a mean age of 65 years, and 45 were women (54%). All patients underwent posterior fusion with a mean of 11 levels fused and 33 (40%) 3-column osteotomies were performed. Among the epigenetic clocks adjusted for BMI and sex, DunedinPACE showed a significant association with ASD-FI (β = 0.041, p = 0.002), EFS (β = 0.696, p = 0.026), and SRS-22r (β = 0.174, p = 0.013) scores. PC PhenoAge showed associations with ASD-FI (β = 0.029, p = 0.028) and SRS-22r (β = 0.159, p = 0.018) scores. PC GrimAge showed associations with ASD-FI (β = 0.029, p = 0.037) and SRS-22r (β = 0.161, p = 0.025) scores. Patients with postoperative complications were noted to have shorter telomere length (T/S 0.790 vs 0.858, p = 0.049), even when the analysis controlled for BMI and sex (OR = 1.71, 95% CI 1.07-2.87, p = 0.031). Epigenetic clocks showed significant associations with markers of frailty and disability, while patients with postoperative complications had shorter telomere length. These data suggest a potential role for aging biomarkers as components of surgical risk assessment. Integrating biological age into current risk calculators may improve their accuracy and provide valuable information for patients, surgeons, and payers.

Experts' Perspectives on the Current Practices and Gaps in the Management of Dyslipidemia: A Report from The United Arab Emirates

<P>Background: The prevalence of dyslipidemia in the United Arab Emirates (UAE) can be as high as 72.5%. Patients with dyslipidemia usually present with other metabolic risk factors, increasing their risk of acute cardiovascular diseases (CVD). Despite these alarming findings, local data regarding management practices in the UAE are lacking. The current report gathered the perspectives of experts from the UAE regarding the burden and current practices in managing dyslipidemia, as well as the current unmet needs and treatment gaps. <P> Methods: The present perspective is based on a consensus meeting that gathered the insights of six consultant cardiologists. All experts were affiliated with academic institutions representing the UAE's different geographical areas. <P> Findings and Conclusion: Data regarding the prevalence of dyslipidemia in the UAE and its associated ethnic differences are lacking. It is suggested that there needs to be more adherence to international recommendations for screening lipid profiles in primary care centers, which may lead to underestimating the burden of dyslipidemia. As CVD presents at a younger age in the UAE, screening for lipid profile at 35 is recommended, regardless of CVD risk. The panel emphasized that most dyslipidemia patients in the UAE are categorized as high or very high-risk groups for CVD; however, current risk assessment tools have not been validated in the UAE population and may not be tailored to the Middle Eastern and Asian populations. International guidelines have recommended the high-intensity statins-ezetimibe combination for high and very high-risk patients; the experts stated that the high-intensity statins-ezetimibe combination could be initiated in high-risk groups in the UAE. Several barriers to effectively controlling dyslipidemia in the UAE have been identified, including non-adherence, limited awareness about screening and management recommendations, and physician inertia toward achieving the recommended therapeutic goals.</P>

Violence risk assessment instruments in forensic psychiatric populations: a systematic review and meta-analysis

Although structured tools have been widely used to predict violence risk in specialist mental health settings, there is uncertainty about the extent and quality of evidence of their predictive performance. We aimed to systematically review the predictive performance of tools used to assess violence risk in forensic mental health, where they are routinely administered. In our systematic review and meta-analysis, we followed PRISMA guidelines and searched four databases (PsycINFO, Embase, Medline, and Global Health) from database inception to Nov 1, 2022, to identify studies examining the predictive performance of risk assessment tools in people discharged from forensic (secure) mental health hospitals. Systematic and narrative reviews were excluded from the review. Performance measures and descriptive statistics were extracted from published reports. A quality assessment was performed for each study using the Prediction Model Risk of Bias Assessment Tool. Meta-analysis was conducted on the performance of instruments that were independently externally validated with a sample size greater than 100. The study was registered with PROSPERO, CRD42022304716. We conducted a systematic review of 50 eligible publications, assessing the predictive performance of 36 tools, providing data for 10 460 participants (88% men, 12% women; median age [from 47 studies] was 35 years, IQR 33-38) from 12 different countries. Post-discharge interpersonal violence and crime was most often measured by new criminal offences or recidivism (47 [94%] of 50 studies); only three studies used informant or self-report data on physical aggression or violent behaviour. Overall, the predictive performance of risk assessment tools was mixed. Most studies reported one discrimination metric, the area under the receiver operating characteristic curve (AUC); other key performance measures such as calibration, sensitivity, and specificity were not presented. Most studies had a high risk of bias (49 [98%] of 50), partly due to poor analytical approaches. A meta-analysis was conducted for violent recidivism on 29 independent external validations from 19 studies with at least 100 patients. Pooled AUCs for predicting violent outcomes ranged from 0·72 (0·65-0·79; I2=0%) for H10, to 0·69 for the Historical Clinical Risk Management-20 version 2 (95% CI 0·65-0·72; I2=0%) and Violence Risk Appraisal Guide (0·63-0·75; I2=0%), to 0·64 for the Static-99 (0·53-0·73; I2=45%). Current violence risk assessment tools in forensic mental health have mixed evidence of predictive performance. Forensic mental health services should review their use of current risk assessment tools and consider implementing those with higher-quality evidence in support. Wellcome Trust.

Development and Validation of a Prediction Tool for Reoffending Risk in Domestic Violence.

Current risk assessment tools for domestic violence against family members were developed with small and selected samples, have low accuracy with few external validations, and do not report key performance measures. To develop new tools to assess risk of reoffending among individuals who have perpetrated domestic violence. This prognostic study investigated a national cohort of all individuals arrested for domestic violence between 1998 and 2013 in Sweden using information from multiple national registers, including National Crime Register, National Patient Register, Longitudinal Integrated Database for Health Insurance and Labour Market Studies Register, and Multi-Generation Register. Data were analyzed from August 2022 to June 2023. Arrest for domestic violence. Prediction models were developed for 3 reoffending outcomes after arrest for domestic violence: conviction of a new violent crime (including domestic violence), conviction of any new crime, and rearrest for domestic violence at 1 year, 3 years, and 5 years. The prediction models were created using sociodemographic factors, criminological factors, and mental health status-related factors, linking data from multiple population-based longitudinal registers. Cox proportional hazard multivariable regression was used to develop prediction models and validate them in external samples. Key performance measures, including discrimination at prespecified cutoffs and calibration statistics, were investigated. The cohort included 27 456 individuals (mean [SD] age, 39.4 [11.6] years; 24 804 men [90.3%]) arrested for domestic violence, of whom 4222 (15.4%) reoffended and were convicted for a new violent crime during a mean (SD) follow-up of 26.5 (27.0) months, 9010 (32.8%) reoffended and were convicted for a new crime (mean [SD] follow-up, 22.4 [25.1] months), and 2080 (7.6%) were rearrested for domestic violence (mean [SD] follow-up, 25.7 [30.6] months). Prediction models were developed with sociodemographic, criminological, and mental health factors and showed good measures of discrimination and calibration for violent reoffending and any reoffending. The area under the receiver operating characteristic curve (AUC) for risk of violent reoffending was 0.75 (95% CI, 0.74-0.76) at 1 year, 0.76 (95% CI, 0.75-0.77) at 3 years, and 0.76 (95% CI, 0.75-0.77) 5 years. The AUC for risk of any reoffending was 0.76 (95% CI, 0.75-0.77) at 1 year and at 3 years and 0.76 (95% CI, 0.75-0.76) at 5 years. The model for domestic violence reoffending showed modest discrimination (C index, 0.63; 95% CI, 0.61-0.65) and good calibration. The validation models showed discrimination and calibration performance similar to those of derivation models for all 3 reoffending outcomes. The prediction models have been translated into 3 simple online risk calculators that are freely available to use. This prognostic study developed scalable, evidence-based prediction tools that could support decision-making in criminal justice systems, particularly at the arrest stage when identifying those at higher risk of reoffending and screening out individuals at low risk of reoffending. Furthermore, these tools can enhance treatment allocation by enabling criminal justice services to focus on modifiable risk factors identified in the tools for individuals at high risk of reoffending.

Artificial intelligence to improve ischemia prediction in rubidium positron emission tomography

Abstract Funding Acknowledgements Type of funding sources: None. Background Patients are referred to coronary artery disease (CAD) testing based on their pre-test probability (PTP). The available, easy-to-use risk prediction tools use three variables only and their diagnostic accuracy is limited. Artificial intelligence-based tools incorporate multiple variables and could improve PTP assessment due to the resulting combinatorial power. The factors in the AI model are not seen as independent individual values, but are recognized as patterns derived by a combinatorial analysis of the individual metabolic profile of each patient. Purpose We aimed to compare the diagnostic accuracy of a novel, artificial intelligence-based tool with commonly used PTP tools to predict ischemia. Methods Consecutive patients (n = 2417) referred for Rubidium-Position Emission Tomography (PET) were evaluated. PTP was calculated using the ESC 2013/2019 and ACC 2012/2021 guidelines, and a memetic pattern-based algorithm (MPA) model incorporating symptoms, vitals, ECG and laboratory findings. Five risk categories (very low to very high risk) were defined (<5%, 5–15%, 15–50%, 50–85%, >85%). Ischemia was defined as summed difference score (SDS) ≥2 on PET. Receiver operator characteristics were calculated and compared using the DeLong method. Results Known CAD, ischemia and scar (detected on PET) were present in 46.3%, 37.1% and 24.6%, respectively. The MPA model was most accurate to assess PTP of ischemia (AUC: 0.758, ESC 2013: 0.661, ESC 2019: 0.673, ACC 2012: 0.585, ACC 2021: 0.667, p<0.0001 each) as depicted in figure 1. The MPA’s sensitivity and negative predictive value to rule-out ischemia were 99.1% and 96.4%, respectively. The model allocated patients more evenly across risk groups, reduced the proportion of patients in the 15–85% range by 29% (ACC 2012) to 51% (ESC 2013), and was the only risk tool to correctly estimate ischemia prevalence in the very low risk group (table 1). Conclusion The MPA model significantly improved PTP assessment of ischemia. The MPA model has originally been derived on the endpoint coronary stenosis >50% and validated in three clinical studies (1–3). This study now confirms that the MPA model applied to the endpoint ischemia outperforms the current risk assessment tools and is most accurate to assess PTP. It enables clinicians to safely exclude ischemia based on readily available variables without advanced testing. It could be used to improve risk stratification of patients and to significantly reduce unnecessary non-invasive functional tests.

Are Mental Disorders Associated with Recidivism in Men Convicted of Sexual Offenses?

In offending populations, prevalence rates of mental disorders are much higher than in the general population. Nevertheless, it is unclear whether mental disorders can improve the prediction of recidivism beyond actuarial risk assessment tools. The present prospective-longitudinal study was conducted between 2001 and 2021 and included 1066 men convicted of sexual offenses in Austria. All participants were evaluated with actuarial risk assessment tools for the prediction of sexual and violent recidivism and the Structured Clinical Interview for Axis I and Axis II disorders. Sexual and violent reconvictions were assessed. Exhibitionism and an exclusive pedophilia showed the strongest correlations with sexual recidivism in the total sample. In the child related offense subsample additionally a narcissistic personality disorder was correlated with sexual recidivism. The strongest correlation with violent recidivism was found for an antisocial and borderline personality disorder. None of the mental disorders could improve the prediction of recidivism beyond actuarial risk assessment tools. Common current actuarial risk assessment tools revealed good predictive accuracy in men convicted of sexual offenses. With few exceptions mental disorders were only weakly associated with recidivism, suggesting that there is no direct link between mental disorders and violent and sexual reoffending. Mental disorders should nevertheless be considered in treatment issues.

Urine CA125 and HE4 for the Detection of Ovarian Cancer in Symptomatic Women.

The symptoms of ovarian cancer are vague, and current risk assessment tools such as serum CA125 and transvaginal ultrasound scan fail to reliably detect the disease early. This study aimed to evaluate urine CA125 and HE4 as diagnostic biomarkers for ovarian cancer in symptomatic women. Paired urine and serum samples were collected from women undergoing treatment for ovarian cancer (cases) or investigations for gynaecological symptoms (controls). Biomarkers were measured using an automated chemiluminescent enzyme immunoassay analyser. Standard diagnostic accuracy metrics were calculated. In total, 114 women were included, of whom 17 (15%) were diagnosed with an epithelial ovarian malignancy. Levels of urine CA125 and HE4 were significantly elevated in women with ovarian cancer compared to controls [CA125: 8.5 U/mL (IQR: 2.4-19.5) vs. 2.3 U/mL (IQR: 1.0-6.4), p = 0.01. HE4: 12.0 nmol/L (IQR: 10.3-23.1) vs. 6.7 nmol/L (IQR: 3.4-13.6), p = 0.006]. Urine CA125 and HE4 detected ovarian cancer with an AUC of 0.69 (95% CI: 0.55-0.82) and 0.71 (95% CI: 0.69-0.82), respectively (p = 0.73). A combination of urine CA125 and HE4 at optimal thresholds had a sensitivity of 82.4% (95% CI: 56.6-96.2) and was comparable to the sensitivity of serum CA125 [88.2% (95% CI: 63.6-98.5)]. Larger studies are required to confirm our findings, standardise urine collection, and evaluate optimal biomarker thresholds. Urine CA125 and HE4 may be useful non-invasive diagnostic tools to triage women for formal ovarian cancer investigations.

MSR17 Modernizing Pressure Injury Risk Assessment in the ICU in the COVID Era: Ensemble Super-Learning and Explainable AI

Hospital-acquired pressure injuries (HAPrI) are areas of injury to the skin and/or underlying tissues. Risk stratification is essential for guiding prevention in the ICU, but current risk assessment tools require labor-intensive input. This motivates a tactical, parsimonious, and automatic risk profiling algorithm, that can be based on readily available clinical measures (e.g., COVID status, race, Medicare/Medicaid status). Additionally, International Pressure Injury Prevention guidelines call for the development of machine learning-based risk assessment algorithms that are clinician-interpretable and context-informed.

보호관찰 청소년 통합 재범위험성 평가도구 개발에 관한 연구

The current study attempts to investigate the predictive validities of the risk assessment tools that are currently being used in juvenile probation and parole in Korea. Since 2016, MOJ(Ministry of Justice) has used the JDRAI-S and JDRAI-D for a different purpose; JDRAI-S has been in use for the purpose of risk classification of juvenile delinquents while JDRAI-D has been utilized for case planning of juvenile probationers and parolees. With the two risk assessment tools being rolled out, concerns have been expressed over the predictive validity of the tools among the practitioners. More importantly, Probation officers have expressed the need to integrate the two different tools into a single, integrated risk assessment tool. Against this backdrop, the current study examines the predictive validities of JDRAI-S and JDRAI-D, and also investigates the possibility that a new tool could be constructed and demonstrate a good level of predictive validity. The results of the current study show that the two current risk assessment tools fail to demonstrate adequate levels of predictive validity. A series of ROC analyses showed that JDRAI-S has the value of AUC .64 and JDRAI-D has the value of AUC .61. Specially the predictive validity of JDRAI-D has posed a serious concern, because many items included in JDRAI-D have been shown not to be associated with recidivism. The second part of the study shows promising results. The newly constructed risk assessment tool, named as JDRAI-Ⅱ, demonstrates a very good level of predictive validity with the value of AUC .73. In addition, the sub factors of JDRAI-Ⅱ also show a good level of consistency and validity. At the end of the study, the authors provide some useful tips for future research and practices.

Abstract 11960: Right Ventricular Diameter in Risk Stratification of Sudden Cardiac Death

Introduction: Right ventricular (RV) abnormalities have been linked with the risk of sudden cardiac death (SCD) in general population. This study was designed to evaluate the additional value of right ventricular diameter (RVD) in predicting life-threatening ventricular tachycardia (VT)/ventricular fibrillation (VF) in patients with implantable cardioverter-defibrillators (ICD). Hypothesis: We hypothesized that higher value of RVD would be an easily accessible predictor in decision-making of ICD implantation. Methods and Results: We enrolled 954 chronic heart failure patients, with quantitative measurements of echocardiograms obtained before ICD implantation, and then divided them according to the median level of RVD. Fine-Gray subdistribution and Cox proportional hazard regression analyses were performed to estimate the predictive value of RVD in sustained VT/VF vs. non-arrhythmic mortality (defined as death without prior sustained VT/VF), respectively. Patients with higher median level of RVD had a significantly higher incidence of VT/VF (32.9% versus 26.7%; P=0.041) over the median follow-up of 2.83 years (interquartile range: 1.33-5.27 years). After adjustment for multiple clinically relevant variables, high RVD levels were significantly associated with an increased risk of VT/VF (per 1 SD increase, hazard ratio=1.16; 95% CI, 1.05-1.27; P=0.003), but not non-arrhythmic mortality (per 1 SD increase, hazard ratio=0.97; 95% CI, 0.82-1.14; P=0.680). When adding RVD to the Seattle Proportional Risk Model (SPRM) and the MADIT-ICD VT/VF risk score, the combined models yielded significant improvement in discrimination for VT/VF. Elevated RVD also contributed to residual risk beyond left ventricular parameters including left ventricular end-diastolic diameter (LVEDD) and left ventricular ejection fraction (LVEF). The restricted mean survival time measure indicated that life gained with ICD implantation in RVD higher group were 11.0, 28.4 and 29.1 days more than their lower counterparts in one, two, and three years’ time. Conclusions: RVD is associated with a significantly increased risk of sustained VT/VF in chronic heart failure patients, and provides greater prognostic power when combined with current risk assessment tools.

Machine Learning for Risk Prediction of Oesophago-Gastric Cancer in Primary Care: Comparison with Existing Risk-Assessment Tools.

Simple SummaryOesophago-gastric cancer is one of the commonest cancers worldwide, yet it can be particularly difficult to diagnose given that initial symptoms are often non-specific and routine screening is not available. Cancer risk-assessment tools, which calculate cancer risk based on symptoms and other risk factors present in the primary care record, can aid decisions on referrals for cancer investigations, facilitating earlier diagnosis. Diagnosing common cancers earlier could help improve survival rates. Using UK primary care electronic health record data, we compared five different machine learning techniques for probabilistic classification of cancer patients against a current widely used UK primary care cancer risk-assessment tool. The machine learning algorithms outperformed the current risk-assessment tool, with a higher overall accuracy and an ability to reasonably identify 11–25% more cancer patients. We conclude that machine-learning-based risk-assessment tools could help better identify suitable patients for further investigation and support earlier diagnosis.Oesophago-gastric cancer is difficult to diagnose in the early stages given its typical non-specific initial manifestation. We hypothesise that machine learning can improve upon the diagnostic performance of current primary care risk-assessment tools by using advanced analytical techniques to exploit the wealth of evidence available in the electronic health record. We used a primary care electronic health record dataset derived from the UK General Practice Research Database (7471 cases; 32,877 controls) and developed five probabilistic machine learning classifiers: Support Vector Machine, Random Forest, Logistic Regression, Naïve Bayes, and Extreme Gradient Boosted Decision Trees. Features included basic demographics, symptoms, and lab test results. The Logistic Regression, Support Vector Machine, and Extreme Gradient Boosted Decision Tree models achieved the highest performance in terms of accuracy and AUROC (0.89 accuracy, 0.87 AUROC), outperforming a current UK oesophago-gastric cancer risk-assessment tool (ogRAT). Machine learning also identified more cancer patients than the ogRAT: 11.0% more with little to no effect on false positives, or up to 25.0% more with a slight increase in false positives (for Logistic Regression, results threshold-dependent). Feature contribution estimates and individual prediction explanations indicated clinical relevance. We conclude that machine learning could improve primary care cancer risk-assessment tools, potentially helping clinicians to identify additional cancer cases earlier. This could, in turn, improve survival outcomes.

Creating a Standardized Post-Fall Debrief Tool: A Quality Improvement Project.

Performing post-fall debriefing improves patient outcomes through learning from defects and addresses adherence to fall prevention programs. While addressing an increase in fall rates, a quality improvement team discovered there was no standardized tool or process for completing post-fall debriefing. The team used the Plan-Do-Study-Act (PDSA) process to improve the post-fall debrief tool, with an analysis of pilot using the implementation science RE-AIM framework. Three units with a high focus on falls and an established debriefing culture participated in pilot to generate and standardize a post-fall debrief tool. Through 2 revisions with end user and champion feedback, the tool was refined to assess any contributing factors to the fall. Through use of the PDSA cycle, the team established content validity of the post-fall debrief tool. This tool is appropriate for inpatient adult and pediatric scale-up and complementary to current fall risk assessment tools.

The role of epicardial and pericoronary adipose tissue in cardiometabolic disease

Epicardial adipose tissue (EAT) is a metabolically active fat depot which modulates coronary arterial function. The measurement of EAT volume and attenuation from non-contrast cardiac computed tomography (CT) provides valuable information on cardiometabolic risk. Pericoronary adipose tissue (PCAT), a subset of EAT which directly surrounds the coronary vessels, may function as a sensor of coronary inflammation. PCAT attenuation on coronary CT angiography associates with coronary plaque morphology, stages of coronary artery disease, and risk of future cardiac events. Quantification of EAT and PCAT has incremental and complementary prognostic value over and above current risk assessment tools. This review covers the biological role of EAT in healthy and dysfunctional states, and CT-based assessment of EAT and PCAT towards the development of clinically meaningful imaging biomarkers.

The utility of risk assessment tools for acute pulmonary embolism in children.

The study consisted of patients under 18 years of age who were consulted to the Pediatric Pulmonology Clinic for the evaluation of PE. Patients were divided into two groups based on the confirmation of PE. The group with the presence of PE (n = 20) consisted of children who were diagnosed with PE. The group with the absence of PE (n = 28) consisted of children with clinically suspected PE but negative diagnostic imaging. Adult validated clinical decision PERC rule and Pediatric PE Model were retrospectively applied to the patients. In the study, PERC demonstrated a sensitivity of 60%and a specificity of 46% for the diagnosis of PE in children. When PE Model was evaluated for the children, it was found a 50% sensitivity and 75% specificity. Combining PE Model and PERC rule with D-dimer did not increase the specificity and sensitivity. Smoking was found to be relevant for PE in the childhood. Twenty-five percent of the patients had a genetic tendency for PE. All of the patients had an underlying disease as well. None of the current risk assessment tools (PE Model, PERC, D-dimer) were found to be accurate in predicting PE. Further larger population studies are still required to develop a better diagnostic approach.

Predicting crime during or after psychiatric care: Evaluating machine learning for risk assessment using the Danish patient registries

BackgroundStructural changes in psychiatric systems have altered treatment opportunities for patients in need of mental healthcare. These changes are possibly associated with an increase in post-discharge crime, reported in the increase of forensic psychiatric populations. As current risk-assessment tools are time-consuming to administer and offer limited accuracy, this study aims to develop a predictive model designed to identify psychiatric patients at risk of committing crime leading to a future forensic psychiatric treatment course. MethodWe utilized the longitudinal quality of the Danish patient registries, identifying the 45.720 adult patients who had contact with the psychiatric system in 2014, of which 474 committed crime leading to a forensic psychiatric treatment course after discharge. Four machine learning models (Logistic Regression, Random Forest, XGBoost and LightGBM) were applied over a range of sociodemographic, judicial, and psychiatric variables. ResultsThis study achieves a F1-macro score of 76%, with precision = 57% and recall = 47% reported by the LightGBM algorithm. Our model was therefore able to identify 47% of future forensic psychiatric patients, while making correct predictions in 57% of cases. ConclusionThe study demonstrates how a clinically useful initial risk-assessment can be achieved using machine learning on data from patient registries. The proposed approach offers the opportunity to flag potential future forensic psychiatric patients, while in contact with the general psychiatric system, hereby allowing early-intervention initiatives to be activated.

Sex-specific performance of the ASCVD pooled cohort equation risk calculator as a correlate of coronary artery calcium in Kampala, Uganda

Introduction: The prevalence of cardiovascular disease (CVD) is rising in Sub-Saharan Africa, but it is not known whether current risk assessment tools predict coronary atherosclerosis in the region. Furthermore, sex-specific performance and interaction with HIV serostatus has not been well studied. Methods: This cross-sectional study compared ASCVD risk scores and detectable coronary artery calcium (CAC>0) by sex in Kampala, Uganda (n = 200). The cohort was enriched for persons living with HIV, and all participants had at least one CVD risk factor. We fit log binomial regression models and constructed ROC curves to assess the correlation between ASCVD scores and CAC>0. Results: The mean age was 56. 62% were female and 50% of both men and women were living with HIV. The median 10-year ASCVD risk score was significantly higher in men (11.0%, IQR 7.6–19.4%) than in women (5.1%, IQR 3.2–8.7%), although the prevalence of CAC>0 was similar (8.1 vs 10.5%, p = 0.63). Each 10% increase in ASCVD risk was associated with increased risk of CAC>0 in men (PR 1.59, 95% CI 1.00–2.55, p = 0.05) but not women (PR 1.15, 95% CI 0.44–3.00, p = 0.77). ROC curves demonstrated an AUC of 0.57 for women vs 0.70 for men. Adjustment for HIV serostatus improved the predictive value of ASCVD in women only (AUC 0.78, p = 0.02). Conclusions: ASCVD risk score did not correlate with the presence of CAC in women. When HIV status was added to the ASCVD risk score, correlation with CAC was improved in women but not in men.