Clinical Prediction Models and Decision-making in FPMRS(On-Demand)-April14WebinarRecording

Video Transcription

Video Summary

Dr. Eric Jelocic, an Associate Professor of OB-GYN at Duke University, presented a webinar on Health Services Research at the FPRMS. He discussed the development and validation of patient-centered prediction tools for various women's health conditions, including pelvic floor disorders and urinary incontinence. Dr. Jelocic highlighted the importance of clinical prediction models in improving patient and clinician decision-making and emphasized the need for accurate and well-calibrated models. He also discussed the challenges and limitations of different types of prediction models, such as risk factor identification, counting risk factors, decision trees, and nomograms. Dr. Jelocic stressed the significance of validation in assessing the performance of prediction models and shared the Tri-Pod statement, which provides guidelines for reporting a model's development and validation. He also touched on emerging issues related to the regulation of prediction algorithms and the potential for algorithmic bias and unfairness in decision-making. Dr. Jelocic concluded the webinar by encouraging clinicians to consider incorporating prediction models into their practice to improve patient care and shared decision-making.

Meta Tag

Concept Prediction Science

Concept Predictive model

Concept Probabilistic Prediction

Concept Calibration

Concept Individualized Risk

Keywords

Dr. Eric Jelocic

patient-centered prediction tools

women's health conditions

clinical prediction models

validation

Tri-Pod statement

algorithmic bias

decision-making

patient care

Prediction Science

Predictive model

Probabilistic Prediction

Calibration

Individualized Risk