scholarly journals How High-Risk Comorbidities Co-Occur in Readmitted Patients With Hip Fracture: Big Data Visual Analytical Approach

10.2196/13567 ◽  
2020 ◽  
Vol 8 (10) ◽  
pp. e13567
Author(s):  
Suresh K Bhavnani ◽  
Bryant Dang ◽  
Rebekah Penton ◽  
Shyam Visweswaran ◽  
Kevin E Bassler ◽  
...  
Keyword(s):  
Hip Fracture ◽  
High Risk ◽  
Prediction Models ◽  
Significant Risk ◽  
Training Data ◽  
Integrated Analysis ◽  
Bipartite Network ◽  
Data Set ◽  
Targeted Interventions ◽  
Patient Subgroups

Background When older adult patients with hip fracture (HFx) have unplanned hospital readmissions within 30 days of discharge, it doubles their 1-year mortality, resulting in substantial personal and financial burdens. Although such unplanned readmissions are predominantly caused by reasons not related to HFx surgery, few studies have focused on how pre-existing high-risk comorbidities co-occur within and across subgroups of patients with HFx. Objective This study aims to use a combination of supervised and unsupervised visual analytical methods to (1) obtain an integrated understanding of comorbidity risk, comorbidity co-occurrence, and patient subgroups, and (2) enable a team of clinical and methodological stakeholders to infer the processes that precipitate unplanned hospital readmission, with the goal of designing targeted interventions. Methods We extracted a training data set consisting of 16,886 patients (8443 readmitted patients with HFx and 8443 matched controls) and a replication data set consisting of 16,222 patients (8111 readmitted patients with HFx and 8111 matched controls) from the 2010 and 2009 Medicare database, respectively. The analyses consisted of a supervised combinatorial analysis to identify and replicate combinations of comorbidities that conferred significant risk for readmission, an unsupervised bipartite network analysis to identify and replicate how high-risk comorbidity combinations co-occur across readmitted patients with HFx, and an integrated visualization and analysis of comorbidity risk, comorbidity co-occurrence, and patient subgroups to enable clinician stakeholders to infer the processes that precipitate readmission in patient subgroups and to propose targeted interventions. Results The analyses helped to identify (1) 11 comorbidity combinations that conferred significantly higher risk (ranging from P<.001 to P=.01) for a 30-day readmission, (2) 7 biclusters of patients and comorbidities with a significant bicluster modularity (P<.001; Medicare=0.440; random mean 0.383 [0.002]), indicating strong heterogeneity in the comorbidity profiles of readmitted patients, and (3) inter- and intracluster risk associations, which enabled clinician stakeholders to infer the processes involved in the exacerbation of specific combinations of comorbidities leading to readmission in patient subgroups. Conclusions The integrated analysis of risk, co-occurrence, and patient subgroups enabled the inference of processes that precipitate readmission, leading to a comorbidity exacerbation risk model for readmission after HFx. These results have direct implications for (1) the management of comorbidities targeted at high-risk subgroups of patients with the goal of pre-emptively reducing their risk of readmission and (2) the development of more accurate risk prediction models that incorporate information about patient subgroups.

scholarly journals How High-Risk Comorbidities Co-Occur in Readmitted Patients With Hip Fracture: Big Data Visual Analytical Approach (Preprint)

2019 ◽  
Author(s):  
Suresh K Bhavnani ◽  
Bryant Dang ◽  
Rebekah Penton ◽  
Shyam Visweswaran ◽  
Kevin E Bassler ◽  
...  
Keyword(s):  
Hip Fracture ◽  
High Risk ◽  
Prediction Models ◽  
Significant Risk ◽  
Training Data ◽  
Integrated Analysis ◽  
Bipartite Network ◽  
Data Set ◽  
Targeted Interventions ◽  
Patient Subgroups

BACKGROUND When older adult patients with hip fracture (HFx) have unplanned hospital readmissions within 30 days of discharge, it doubles their 1-year mortality, resulting in substantial personal and financial burdens. Although such unplanned readmissions are predominantly caused by reasons not related to HFx surgery, few studies have focused on how pre-existing high-risk comorbidities co-occur within and across subgroups of patients with HFx. OBJECTIVE This study aims to use a combination of supervised and unsupervised visual analytical methods to (1) obtain an integrated understanding of comorbidity risk, comorbidity co-occurrence, and patient subgroups, and (2) enable a team of clinical and methodological stakeholders to infer the processes that precipitate unplanned hospital readmission, with the goal of designing targeted interventions. METHODS We extracted a training data set consisting of 16,886 patients (8443 readmitted patients with HFx and 8443 matched controls) and a replication data set consisting of 16,222 patients (8111 readmitted patients with HFx and 8111 matched controls) from the 2010 and 2009 Medicare database, respectively. The analyses consisted of a supervised combinatorial analysis to identify and replicate combinations of comorbidities that conferred significant risk for readmission, an unsupervised bipartite network analysis to identify and replicate how high-risk comorbidity combinations co-occur across readmitted patients with HFx, and an integrated visualization and analysis of comorbidity risk, comorbidity co-occurrence, and patient subgroups to enable clinician stakeholders to infer the processes that precipitate readmission in patient subgroups and to propose targeted interventions. RESULTS The analyses helped to identify (1) 11 comorbidity combinations that conferred significantly higher risk (ranging from <i>P</i>&lt;.001 to <i>P</i>=.01) for a 30-day readmission, (2) 7 biclusters of patients and comorbidities with a significant bicluster modularity (<i>P</i>&lt;.001; Medicare=0.440; random mean 0.383 [0.002]), indicating strong heterogeneity in the comorbidity profiles of readmitted patients, and (3) inter- and intracluster risk associations, which enabled clinician stakeholders to infer the processes involved in the exacerbation of specific combinations of comorbidities leading to readmission in patient subgroups. CONCLUSIONS The integrated analysis of risk, co-occurrence, and patient subgroups enabled the inference of processes that precipitate readmission, leading to a comorbidity exacerbation risk model for readmission after HFx. These results have direct implications for (1) the management of comorbidities targeted at high-risk subgroups of patients with the goal of pre-emptively reducing their risk of readmission and (2) the development of more accurate risk prediction models that incorporate information about patient subgroups.

scholarly journals Development of a Gene-Based Prediction Model for Recurrence of Colorectal Cancer Using an Ensemble Learning Algorithm

2021 ◽  
Vol 11 ◽  
Author(s):  
Han-Ching Chan ◽  
Amrita Chattopadhyay ◽  
Eric Y. Chuang ◽  
Tzu-Pin Lu
Keyword(s):  
Colorectal Cancer ◽  
High Risk ◽  
Adjuvant Chemotherapy ◽  
Prediction Models ◽  
Stage I ◽  
Training Data ◽  
Differentially Expressed ◽  
Support Vector ◽  
Data Set ◽  
Risk Of Recurrence

It is difficult to determine which patients with stage I and II colorectal cancer are at high risk of recurrence, qualifying them to undergo adjuvant chemotherapy. In this study, we aimed to determine a gene signature using gene expression data that could successfully identify high risk of recurrence among stage I and II colorectal cancer patients. First, a synthetic minority oversampling technique was used to address the problem of imbalanced data due to rare recurrence events. We then applied a sequential workflow of three methods (significance analysis of microarrays, logistic regression, and recursive feature elimination) to identify genes differentially expressed between patients with and without recurrence. To stabilize the prediction algorithm, we repeated the above processes on 10 subsets by bagging the training data set and then used support vector machine methods to construct the prediction models. The final predictions were determined by majority voting. The 10 models, using 51 differentially expressed genes, successfully predicted a high risk of recurrence within 3 years in the training data set, with a sensitivity of 91.18%. For the validation data sets, the sensitivity of the prediction with samples from two other countries was 80.00% and 91.67%. These prediction models can potentially function as a tool to decide if adjuvant chemotherapy should be administered after surgery for patients with stage I and II colorectal cancer.

Prognostic Model for Stratification of Radioresistant Nasopharynx Carcinoma to Curative Salvage Radiotherapy

2018 ◽  
Vol 36 (9) ◽  
pp. 891-899 ◽  
Author(s):  
You Quan Li ◽  
Yun Ming Tian ◽  
Sze Huey Tan ◽  
Ming Zhu Liu ◽  
Grace Kusumawidjaja ◽  
...  
Keyword(s):  
High Risk ◽  
Prognostic Index ◽  
Intensity Modulated Radiotherapy ◽  
Training Data ◽  
Cancer Center ◽  
Weight Discrimination ◽  
Data Set ◽  
Training Cohort ◽  
Risk Patients ◽  
Grade 3

Purpose To investigate for a prognostic index (PI) to personalize recommendations for salvage intensity-modulated radiotherapy (IMRT) in patients with locally recurrent nasopharyngeal carcinoma (lrNPC). Methods Patients with lrNPC from two academic institutions (Sun Yat-Sen University Cancer Center [SYSUCC-A; n = 251 (training cohort)] and National Cancer Centre Singapore [NCCS; n = 114] and SYSUCC-B [n = 193 (validation cohorts)]) underwent salvage treatment with IMRT from 2001 to 2015. Primary and secondary clinical end points were overall survival (OS) and grade 5 toxicity-free rate (G5-TFR), respectively. Covariate inclusion to the PIs was qualified by a multivariable two-sided P < .05. Discrimination and calibration of the PIs were assessed. Results The primary PI comprised covariates that were adversely associated with OS in the training cohort (gross tumor volumerecurrence hazard ratio [HR], 1.01/mL increase [ P < .001], agerecurrence HR, 1.02/year increase [ P = .008]; repeat IMRT equivalent dose in 2-Gy fractions [EQD2] ≥ 68 Gy HR, 1.42 [ P = .03]; prior radiotherapy-induced grade ≥ 3 toxicities HR, 1.90 [ P = .001]; recurrent tumor [rT]-category 3 to 4 HR, 1.96 [ P = .005]), in ascending order of weight. Discrimination of the PI for OS was comparable between training and both validation cohorts (Harrell’s C = 0.71 [SYSUCC-A], 0.72 [NCCS], and 0.69 [SYSUCC-B]); discretization by using a fixed PI score cutoff of 252 determined from the training data set yielded low- and high-risk subgroups with disparate OS in the validation cohorts (NCCS HR, 3.09 [95% CI, 1.95 to 4.89]; SYSUCC-B HR, 3.80 [95% CI, 2.55 to 5.66]). Our five-factor PI predicted OS and G5-TFR (predicted v observed 36-month OS and G5-TFR, 22% v 15% and 38% v 44% for high-risk NCCS and 26% v 31% and 45% v 46% for high-risk SYSUCC-B). Conclusion We present a validated PI for robust clinical stratification of radioresistant NPC. Low-risk patients represent ideal candidates for curative repeat IMRT, whereas novel clinical trials are needed in the unfavorable high-risk subgroup.

scholarly journals POSSUM and P-POSSUM Scoring in Hip Fracture Mortalities

2020 ◽  
Vol 11 ◽  
pp. 215145932093167
Author(s):  
William L. Johns ◽  
Benjamin Strong ◽  
Stephen Kates ◽  
Nirav K. Patel
Keyword(s):  
Hip Fracture ◽  
High Risk ◽  
Mortality Rate ◽  
Clinical Decision Making ◽  
Tertiary Care ◽  
Significant Risk ◽  
Screening Tools ◽  
Morbidity And Mortality ◽  
Mortality And Morbidity ◽  
Operative Severity Score

Introduction: Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity (POSSUM) and Portsmouth POSSUM (P-POSSUM) are general surgical tools used to efficiently assess mortality and morbidity risk. Data suggest that these tools can be used in hip fracture patients to predict morbidity and mortality; however, it is unclear what score indicates a significant risk on a case-by-case basis. We examined the POSSUM and P-POSSUM scores in a group of hip fracture mortalities in order to assess their accuracy in identification of similar high-risk patients. Materials and Methods: Retrospective analysis of all consecutive mortalities in hip fracture patients at a single tertiary care center over 2 years was performed. Patient medical records were examined for baseline demographics, fracture characteristics, surgical interventions, and cause of death. Twelve physiological and 6 operative variables were used to retrospectively calculate POSSUM and P-POSSUM scores at the time of injury. Results: Forty-seven hip fracture mortalities were reviewed. Median patient age was 88 years (range: 56-99). Overall, 68.1% (32) underwent surgical intervention. Mean predicted POSSUM morbidity and mortality rates were 73.9% (28%-99%) and 31.1% (5%-83%), respectively. The mean predicted P-POSSUM mortality rate was 26.4% (1%-91%) and 53.2% (25) had a P-POSSUM predicted mortality of >20%. Subgroup analysis demonstrated poor agreement between predicted mortality and observed mortality rate for POSSUM in operative (χ2 = 127.5, P < .00001) and nonoperative cohorts (χ2 = 14.6, P < .00001), in addition to P-POSSUM operative (χ2 = 101.9, P < .00001) and nonoperative (χ2 = 11.9, P < .00001) scoring. Discussion/Conclusion: Hip fracture patients are at significant risk of both morbidity and mortality. A reliable, replicable, and accurate tool to represent the expected risk of such complications could help facilitate clinical decision-making to determine the optimal level of care. Screening tools such as POSSUM and P-POSSUM have limitations in accurately identifying high-risk hip fracture patients.

scholarly journals Plasma Protein Profiles Differ between Women Diagnosed with Cervical Intraepithelial Neoplasia (CIN) 1 and 3

2006 ◽  
Vol 2 ◽  
pp. 117693510600200 ◽  
Author(s):  
Chandrika J. Piyathilake ◽  
Denise K. Oelschlager ◽  
Sreelatha Meleth ◽  
Edward E. Partridge ◽  
William E. Grizzle
Keyword(s):  
Cervical Cancer ◽  
High Risk ◽  
Cervical Intraepithelial Neoplasia ◽  
Plasma Protein ◽  
Intraepithelial Neoplasia ◽  
Cervical Cancer Prevention ◽  
Training Data ◽  
Treatment Programs ◽  
Protein Profiles ◽  
Data Set

Early detection of precancerous cells in the cervix and their clinical management is the main purpose of cervical cancer prevention and treatment programs. Cytological findings or testing for high risk (HR)-human papillomavirus (HPV) are inadequately sensitive for use in triage of women at high risk for cervical cancer. The current study is an exploratory study to identify candidate surface-enhanced laser desorption/ionization (SELDI) time of flight (TOF) mass spectrometry (MS) protein profiles in plasma that may distinguish cervical intraepithelial neoplasia (CIN 3) from CIN 1 among women infected with HR-HPV. We evaluated the SELDI-TOF-MS plasma protein profiles of HR-HPV positive 32 women with CIN 3 (cases) and 28 women with CIN1 (controls). Case-control status was kept blinded and triplicates of each sample and quality control plasma samples were randomized and after robotic sample preparations were run on WCX2 chips. After alignment of mass/charge (m-z values), an iterative method was used to develop a classifier on a training data set that had 28 cases and 22 controls. The classifier developed was used to classify the subjects in a test data set that has six cases and six controls. The classifier separated the cases from controls in the test set with 100% sensitivity and 100% specificity suggesting the possibility of using plasma SELDI protein profiles to identify women who are likely to have CIN 3 lesions.

scholarly journals Melanoma risk prediction models

2014 ◽  
Vol 71 (8) ◽  
pp. 757-766
Author(s):  
Jelena Nikolic ◽  
Tatjana Loncar-Turukalo ◽  
Srdjan Sladojevic ◽  
Marija Marinkovic ◽  
Zlata Janjic
Keyword(s):  
Risk Factors ◽  
At Risk ◽  
High Risk ◽  
Prediction Models ◽  
Significant Risk ◽  
Prognostic Models ◽  
Hair Color ◽  
Melanoma Risk ◽  
Data Mining Algorithms ◽  
Red Hair

Background/Aim. The lack of effective therapy for advanced stages of melanoma emphasizes the importance of preventive measures and screenings of population at risk. Identifying individuals at high risk should allow targeted screenings and follow-up involving those who would benefit most. The aim of this study was to identify most significant factors for melanoma prediction in our population and to create prognostic models for identification and differentiation of individuals at risk. Methods. This case-control study included 697 participants (341 patients and 356 controls) that underwent extensive interview and skin examination in order to check risk factors for melanoma. Pairwise univariate statistical comparison was used for the coarse selection of the most significant risk factors. These factors were fed into logistic regression (LR) and alternating decision trees (ADT) prognostic models that were assessed for their usefulness in identification of patients at risk to develop melanoma. Validation of the LR model was done by Hosmer and Lemeshow test, whereas the ADT was validated by 10-fold cross-validation. The achieved sensitivity, specificity, accuracy and AUC for both models were calculated. The melanoma risk score (MRS) based on the outcome of the LR model was presented. Results. The LR model showed that the following risk factors were associated with melanoma: sunbeds (OR = 4.018; 95% CI 1.724- 9.366 for those that sometimes used sunbeds), solar damage of the skin (OR = 8.274; 95% CI 2.661-25.730 for those with severe solar damage), hair color (OR = 3.222; 95% CI 1.984-5.231 for light brown/blond hair), the number of common naevi (over 100 naevi had OR = 3.57; 95% CI 1.427-8.931), the number of dysplastic naevi (from 1 to 10 dysplastic naevi OR was 2.672; 95% CI 1.572-4.540; for more than 10 naevi OR was 6.487; 95%; CI 1.993-21.119), Fitzpatricks phototype and the presence of congenital naevi. Red hair, phototype I and large congenital naevi were only present in melanoma patients and thus were strongly associated with melanoma. The percentage of correctly classified subjects in the LR model was 74.9%, sensitivity 71%, specificity 78.7% and AUC 0.805. For the ADT percentage of correctly classified instances was 71.9%, sensitivity 71.9%, specificity 79.4% and AUC 0.808. Conclusion. Application of different models for risk assessment and prediction of melanoma should provide efficient and standardized tool in the hands of clinicians. The presented models offer effective discrimination of individuals at high risk, transparent decision making and real-time implementation suitable for clinical practice. A continuous melanoma database growth would provide for further adjustments and enhancements in model accuracy as well as offering a possibility for successful application of more advanced data mining algorithms.

scholarly journals Predictive Modeling of 30-Day Emergency Hospital Transport of German Patients Using a Personal Emergency Response: Retrospective Study and Comparison with the United States (Preprint)

2020 ◽  
Author(s):  
Jorn op den Buijs ◽  
Marten Pijl ◽  
Andreas Landgraf
Keyword(s):  
United States ◽  
Health Care ◽  
High Risk ◽  
Predictive Model ◽  
The United States ◽  
Data Set ◽  
Targeted Interventions ◽  
Emergency Hospital ◽  
The Us ◽  
Hospital Transport

BACKGROUND Predictive analytics based on data from remote monitoring of elderly via a personal emergency response system (PERS) in the United States can identify subscribers at high risk for emergency hospital transport. These risk predictions can subsequently be used to proactively target interventions and prevent avoidable, costly health care use. It is, however, unknown if PERS-based risk prediction with targeted interventions could also be applied in the German health care setting. OBJECTIVE The objectives were to develop and validate a predictive model of 30-day emergency hospital transport based on data from a German PERS provider and compare the model with our previously published predictive model developed on data from a US PERS provider. METHODS Retrospective data of 5805 subscribers to a German PERS service were used to develop and validate an extreme gradient boosting predictive model of 30-day hospital transport, including predictors derived from subscriber demographics, self-reported medical conditions, and a 2-year history of case data. Models were trained on 80% (4644/5805) of the data, and performance was evaluated on an independent test set of 20% (1161/5805). Results were compared with our previously published prediction model developed on a data set of PERS users in the United States. RESULTS German PERS subscribers were on average aged 83.6 years, with 64.0% (743/1161) females, with 65.4% (759/1161) reported 3 or more chronic conditions. A total of 1.4% (350/24,847) of subscribers had one or more emergency transports in 30 days in the test set, which was significantly lower compared with the US data set (2455/109,966, 2.2%). Performance of the predictive model of emergency hospital transport, as evaluated by area under the receiver operator characteristic curve (AUC), was 0.749 (95% CI 0.721-0.777), which was similar to the US prediction model (AUC=0.778 [95% CI 0.769-0.788]). The top 1% (12/1161) of predicted high-risk patients were 10.7 times more likely to experience an emergency hospital transport in 30 days than the overall German PERS population. This lift was comparable to a model lift of 11.9 obtained by the US predictive model. CONCLUSIONS Despite differences in emergency care use, PERS-based collected subscriber data can be used to predict use outcomes in different international settings. These predictive analytic tools can be used by health care organizations to extend population health management into the home by identifying and delivering timelier targeted interventions to high-risk patients. This could lead to overall improved patient experience, higher quality of care, and more efficient resource use.

Three Biomarker Panel at Day 7 and 14 Can Predict Development of Grade II-IV Acute Graft-Versus-Host Disease

Blood ◽  
2010 ◽  
Vol 116 (21) ◽  
pp. 675-675 ◽  
Author(s):  
Sophie Paczesny ◽  
Thomas Braun ◽  
Mark Vander Lugt ◽  
Andrew Harris ◽  
Bryan Fiema ◽  
...  
Keyword(s):  
High Risk ◽  
Acute Gvhd ◽  
Training Data ◽  
Data Set ◽  
Graft Versus Host ◽  
Specificity And Sensitivity ◽  
Predicted Probability ◽  
Grade Ii ◽  
Full Intensity ◽  
Acute Graft

Abstract Abstract 675 We have previously identified five biomarker proteins that have diagnostic and prognostic value for acute graft-versus-host disease (GVHD) (Blood 113:273-278, Sci Transl Med 2:50-57). In order to determine whether biomarkers can predict GVHD before the appearance of clinical symptoms, we evaluated the three most informative biomarkers of the five (IL2-Receptor-α, TNFR1, elafin) in patient samples prospectively collected between 2000 and 2010 from 513 unrelated (URD) hematopoietic cell transplant (HCT) patients. We focused on URD HCT recipients because they are most likely to develop acute GVHD and could potentially benefit from a predictive laboratory assay and subsequent preemptive intervention. We measured biomarker plasma levels by sequential enzyme-linked immunosorbent assay on samples obtained prior to conditioning (pre-HCT), and at day +7 and day +14 after HCT. In designing the analytical approach, we took into consideration that median time to GVHD onset is delayed after reduced intensity conditioning, that there may be limited opportunity to preemptively intervene in patients on the verge of developing GVHD, and that there have been changes in GVHD prophylaxis agents over the past decade. Therefore, we randomly divided the patients into training (N=342) and validation (N=171) data sets that were balanced for (i) full intensity conditioning, (ii) onset of grade II-IV GVHD earlier than day +21 and (iii) HCT performed after 2005. In order to create a prediction model for acute GVHD, we used the biomarker levels in the training data set to simulate biomarker values for a hypothetical 50,000 patients using the following assumptions: (1) the incidence of GVHD by day 100 is 55%, (2) the median day of GVHD onset after full intensity URD HCT is day 21, with 10% of patients developing GVHD prior to day +7, and 3% developing GVHD after day +56. These assumptions were based on historical URD HCT data at our center. We used logistic regression to compute a predicted probability, p7, of developing grade II-IV GVHD for each of the 50,000 patients based upon the biomarker levels pre-HCT and at day +7. Any patient with p7≥0.64 was categorized as high risk for development of GVHD. For all low risk patients (p7<0.64), who had not developed GVHD between day +7 and +14, we measured the biomarkers at day +14 and created another logistic regression model to compute an additional predicted probability, p14, of development of grade II-IV GVHD after day +14. Any patient in the initial low risk group with p14≥0.41 was now categorized as high risk. For the training data set, the combination of these two tests correctly predicted the absence of grade II-IV GVHD by day 56 in 77% (95% CI: 71%-83%) of patients without GVHD (i.e. specificity), and correctly predicted the development of grade II-IV GVHD by day 56 in 50% (95% CI: 42%-58%) of patients who had not already developed GVHD by the time of the test sample (i.e. sensitivity). Applying these two diagnostic tests to the validation data set gave a specificity of 75% (95% CI: 67%-83%) and a sensitivity of 57% (95% CI: 44%-69%), similar to the values found in the training data set. Table 1 shows the specificity and sensitivity of the two combined tests for different values of p7 and p14. The table is ordered by descending combined specificity. The bolded numbers, which we propose could be used for designing a GVHD preemptive clinical trial, correspond to the values of combined specificity and sensitivity, which lead to the greatest percentage of patients correctly predicted for GVHD occurrence among those listed. In conclusion, measurement of a three-biomarker panel pre-HCT, at day +7, and day +14 predicts grade II-IV GVHD with a specificity of 75% and sensitivity of 57%, with the median time between a high risk test result and onset of grade II-IV GVHD by day 56 equal to 14 days, with a range of (2-41) days. If these results are confirmed in multicenter samples, we could propose to preemptively treat acute grade II-IV GVHD in the unrelated HCT population. Disclosures: No relevant conflicts of interest to declare.

Identification of Patients at High Cardiovascular Risk: a Critical Appraisal of Applicability of Statistical Risk Prediction Models

2001 ◽  
Vol 40 (01) ◽  
pp. 6-11 ◽  
Author(s):  
I. Colombet ◽  
P. Degoulet ◽  
G. Chatellier ◽  
H. Dréau
Keyword(s):  
Cardiovascular Risk ◽  
High Risk ◽  
Prediction Models ◽  
Evaluation Study ◽  
World Health ◽  
Clinical Value ◽  
Data Set ◽  
High Risk Patients ◽  
Risk Patients ◽  
Mild Hypertensive

AbstractAssessment of cardiovascular risk is widely proposed as a basis for taking management decisions about patients presenting with hypertension or hypercholesterolemia. Our aim was to critically assess the use of risk equations derived from epidemiological studies for the purpose of identifying high-risk patients.Risk equations were retrieved from the MEDLINE database and then applied to a data set of 118 patients. This data set was an evaluation study of the clinical value of the World Health Organization 1993 hypertension guidelines for the decision to treat mild hypertensive patients. We calculated agreement: 1) between equations and 2) between equations and the decision to treat taken by the physician.Most models were not applicable to our population, mainly because the original population had a narrow age range or comprised only males. Between-model agreement was better for the lower and upper risk quintiles than for the three other risk quintiles (0.58, 0.33, 0.34, 0.45, 0.70, from the lower to the upper risk quintile). When using an arbitrary threshold for defining high-risk patients (i.e. >2% per year), we observed a huge variation of the proportion of patients classified at high risk (from 0 to 17%). There was a poor agreement between risk models and the decision to treat taken by the physician. These results suggest that risk-based guidelines should be validated before their diffusion.

scholarly journals Methylation differences reveal heterogeneity in preterm pathophysiology: results from bipartite network analyses

2018 ◽  
Vol 46 (5) ◽  
pp. 509-521 ◽  
Author(s):  
Suresh K. Bhavnani ◽  
Bryant Dang ◽  
Varun Kilaru ◽  
Maria Caro ◽  
Shyam Visweswaran ◽  
...  
Keyword(s):  
Dna Methylation ◽  
Network Analysis ◽  
Gestational Age ◽  
Bipartite Network ◽  
Bipartite Networks ◽  
Gestational Length ◽  
Specific Patient ◽  
Targeted Interventions ◽  
Network Analyses ◽  
Patient Subgroups

AbstractBackground:Recent studies have shown that epigenetic differences can increase the risk of spontaneous preterm birth (PTB). However, little is known about heterogeneity underlying such epigenetic differences, which could lead to hypotheses for biological pathways in specific patient subgroups, and corresponding targeted interventions critical for precision medicine. Using bipartite network analysis of fetal DNA methylation data we demonstrate a novel method for classification of PTB.Methods:The data consisted of DNA methylation across the genome (HumanMethylation450 BeadChip) in cord blood from 50 African-American subjects consisting of 22 cases of early spontaneous PTB (24–34 weeks of gestation) and 28 controls (>39 weeks of gestation). These data were analyzed using a combination of (1) a supervised method to select the top 10 significant methylation sites, (2) unsupervised “subject-variable” bipartite networks to visualize and quantitatively analyze how those 10 methylation sites co-occurred across all the subjects, and across only the cases with the goal of analyzing subgroups and their underlying pathways, and (3) a simple linear regression to test whether there was an association between the total methylation in the cases, and gestational age.Results:The bipartite network analysis of all subjects and significant methylation sites revealed statistically significant clustering consisting of an inverse symmetrical relationship in the methylation profiles between a case-enriched subgroup and a control-enriched subgroup: the former was predominantly hypermethylated across seven methylation sites, and hypomethylated across three methylation sites, whereas the latter was predominantly hypomethylated across the above seven methylation sites and hypermethylated across the three methylation sites. Furthermore, the analysis of only cases revealed one subgroup that was predominantly hypomethylated across seven methylation sites, and another subgroup that was hypomethylated across all methylation sites suggesting the presence of heterogeneity in PTB pathophysiology. Finally, the analysis found a strong inverse linear relationship between total methylation and gestational age suggesting that methylation differences could be used as predictive markers for gestational length.Conclusions:The results demonstrate that unsupervised bipartite networks helped to identify a complex but comprehensible data-driven hypotheses related to patient subgroups and inferences about their underlying pathways, and therefore were an effective complement to supervised approaches currently used.

Sign in / Sign up

Export Citation Format

Share Document