https://journal.chestnet.org/article/S0012-3692(12)60129-9/pdf

Antithrombotic therapy for VTE disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines.

Regression models are frequently used to develop diagnostic, prognostic, and health resource utilization models in clinical, health services, outcomes, pharmacoeconomic, and epidemiologic research, and in a multitude of non-health-related areas. Regression models are also used to adjust for patient heterogeneity in randomized clinical trials, to obtain tests that are more powerful and valid than unadjusted treatment comparisons.

/pdf/regression-modeling-strategies-1jyn0f4r32.pdf

Regression Modeling Strategies

https://www.healthcare.uiowa.edu/familymedicine/fpinfo/Docs/Chest%20Rx%20VTE%20Feb%202016.pdf

Antithrombotic Therapy for VTE Disease: CHEST Guideline and Expert Panel Report

The performance of prediction models can be assessed using a variety of methods and metrics. Traditional measures for binary and survival outcomes include the Brier score to indicate overall model performance, the concordance (or c) statistic for discriminative ability (or area under the receiver operating characteristic [ROC] curve), and goodness-of-fit statistics for calibration.Several new measures have recently been proposed that can be seen as refinements of discrimination measures, including variants of the c statistic for survival, reclassification tables, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Moreover, decision-analytic measures have been proposed, including decision curves to plot the net benefit achieved by making decisions based on model predictions.We aimed to define the role of these relatively novel approaches in the evaluation of the performance of prediction models. For illustration, we present a case study of predicting the presence of residual tumor versus benign tissue in patients with testicular cancer (n = 544 for model development, n = 273 for external validation).We suggest that reporting discrimination and calibration will always be important for a prediction model. Decision-analytic measures should be reported if the predictive model is to be used for clinical decisions. Other measures of performance may be warranted in specific applications, such as reclassification metrics to gain insight into the value of adding a novel predictor to an established model.

/pdf/assessing-the-performance-of-prediction-models-a-framework-2soq09lsy5.pdf

Assessing the performance of prediction models: a framework for traditional and novel measures.

The TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis) Statement includes a 22-item checklist, which aims to improve the reporting of studies developing, validating, or updating a prediction model, whether for diagnostic or prognostic purposes. The TRIPOD Statement aims to improve the transparency of the reporting of a prediction model study regardless of the study methods used. This explanation and elaboration document describes the rationale; clarifies the meaning of each item; and discusses why transparent reporting is important, with a view to assessing risk of bias and clinical usefulness of the prediction model. Each checklist item of the TRIPOD Statement is explained in detail and accompanied by published examples of good reporting. The document also provides a valuable reference of issues to consider when designing, conducting, and analyzing prediction model studies. To aid the editorial process and help peer reviewers and, ultimately, readers and systematic reviewers of prediction model studies, it is recommended that authors include a completed checklist in their submission. The TRIPOD checklist can also be downloaded from www.tripod-statement.org.

/pdf/transparent-reporting-of-a-multivariable-prediction-model-3n8wy14hp9.pdf

Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration.

Clinical prediction rules, sometimes called clinical decision rules, have proliferated in recent years. However, very few have undergone formal impact analysis, the standard of evidence to assess their impact on patient care. Without impact analysis, clinicians cannot know whether using a prediction rule will be beneficial or harmful. This paper reviews standards of evidence for developing and evaluating prediction rules; important differences between prediction rules and decision rules; how to assess the potential clinical impact of a prediction rule before translating it into a decision rule; methodologic issues critical to successful impact analysis, including defining outcome measures and estimating sample size; the importance of close collaboration between clinical investigators and practicing clinicians before, during, and after impact analysis; and the need to measure both efficacy and effectiveness when analyzing a decision rule's clinical impact. These considerations should inform future development, evaluation, and use of all clinical prediction or decision rules.

Translating Clinical Research into Clinical Practice: Impact of Using Prediction Rules To Make Decisions

OBJECTIVE To determine whether an intravenous heparin dosing nomogram based on body weight achieves therapeutic anticoagulation more rapidly than a "standard care" nomogram. DESIGN Randomized controlled trial. SETTING Two community teaching hospitals in Phoenix, Arizona, and Rochester, New York. PARTICIPANTS One hundred fifteen patients requiring intravenous heparin treatment for venous or arterial thromboembolism or for unstable angina. INTERVENTION Patients were randomized to the weight-based nomogram (starting dose, 80 units/kg body weight bolus, 18 units/kg per hour infusion) or the standard care nomogram (starting dose, 5000-unit bolus, 1000 units per hour infusion). Activated partial thromboplastin time (APTT) values were monitored every 6 hours, and heparin dose adjustments were determined by the nomograms. MEASUREMENTS Activated partial thromboplastin times were measured using a widely generalizable laboratory method. The primary outcomes were the time to exceed the therapeutic threshold (APTT > 1.5 times the control) and the time to achieve therapeutic range (APTT, 1.5 to 2.3 times the control). Bleeding complications and recurrent thromboembolism were also compared. RESULTS Kaplan-Meier curves for the primary outcomes favored the weight-based nomogram (P < 0.001 for both). In the weight-based heparin group, 60 of 62 patients (97%) exceeded the therapeutic threshold within 24 hours, compared with 37 of 48 (77%) in the standard care group (P < 0.002). Only one major bleeding complication occurred (in a standard care patient). Recurrent thromboembolism was more frequent in the standard care group; relative risk, 5.0 (95% CI, 1.1 to 21.9). CONCLUSIONS The weight-based heparin nomogram is widely generalizable and has proved to be effective, safe, and superior to one based on standard practice.

The weight-based heparin dosing nomogram compared with a “standard care” nomogram. a randomized controlled trial

Objective: To determine whether an intravenous heparin dosing nomogram based on body weight achieves therapeutic anticoagulation more rapidly than a standard care nomogram. Design: Randomized contr...

The Weight-based Heparin Dosing Nomogram Compared with a Standard Care Nomogram

Physical examination in the care of medical inpatients: an observational study.

Clinicians recognize the potential value of evidence-based medicine (EBM) but many feel ambivalent about its use in patient care. Major barriers to implementation include a perception that the appeal of EBM is more academic than practical, and clinicians cite their own lack of EBM skills as a contributing factor.1

Many postgraduate training programs currently attempt to teach EBM skills, primarily in journal clubs and through didactic lectures. However, published outcomes of these efforts have shown only limited effectiveness.2–8 Most of the studies have focused on critical appraisal while excluding or inadequately measuring other EBM skills.9 Consequently, there are no published studies that assess how best to teach more fundamental EBM skills: the ability to create searchable questions that arise from knowledge gaps in patient care and the ability to perform efficient literature searches to answer those questions.10 The acquisition of these skills, in addition to those of critical appraisal, is essential if EBM is to favorably influence patient outcomes.11–13

We conducted a trial to test whether an educational program could improve four essential EBM skills among first-year internal medicine residents: (1) posing well-structured, searchable questions arising from clinical cases; (2) performing efficient electronic literature searches to answer these questions; (3) understanding quantitative outcomes from published studies about diagnosis and treatment; and (4) evaluating the methodological quality of published studies and their clinical relevance to specific patients. We also carried out repeat assessments to see if these acquired skills were durable over time. These four skills are similar, but not identical, to those espoused by proponents of EBM—namely, question formulation, literature searching, critical appraisal skills, and application of the results to the patient.14

Brendan M. Reilly

Papers

Translating Clinical Research into Clinical Practice: Impact of Using Prediction Rules To Make Decisions

The weight-based heparin dosing nomogram compared with a “standard care” nomogram. a randomized controlled trial

The Weight-based Heparin Dosing Nomogram Compared with a Standard Care Nomogram

Physical examination in the care of medical inpatients: an observational study.

Teaching residents evidence-based medicine skills: a controlled trial of effectiveness and assessment of durability.