Explanatory Predictive Model

From GM-RKB
Jump to navigation Jump to search

An Explanatory Predictive Model is a interpretable predictive model that is an explanatory model.



References

2020

2016

2014

  • (Waljee et al., 2014) ⇒ Akbar K. Waljee, Peter DR Higgins, and Amit G. Singal. (2014). “A Primer on Predictive Models.” Clinical and translational gastroenterology, 5(1). doi:10.1038%2Fctg.2013.19
    • QUOTE: ... Prediction research, which aims to predict future events or outcomes based on patterns within a set of variables, has become increasingly popular in medical research.[1] Accurate predictive models can inform patients and physicians about the future course of an illness or the risk of developing an illness and thereby help guide decisions on screening and/or treatment. For example, predictive models have been developed in gastroenterology to predict the risk of disease flares for inflammatory bowel disease and risk of hepatocellular carcinoma among patients with cirrhosis. [2, 3]

      There are several important differences between traditional explanatory research and prediction research. Explanatory research typically applies statistical methods to test causal hypotheses using a priori theoretical constructs (e.g., hepatocellular carcinoma surveillance underutilization is related to provider-level factors4). In contrast, predictive research applies statistical methods and/or data mining techniques, without preconceived theoretical constructs, to predict future outcomes (e.g., predicting the risk of hospital readmission5).[6] Although predictive models may be used to provide insight into causality of pathophysiology of the outcome, causality is neither a primary aim nor a requirement for variable inclusion.[6] Noncausal predictive factors may be surrogates for other drivers of disease, with tumor markers as predictors of cancer progression or recurrence being the most common example. Unfortunately, a poor understanding of the differences in methodology between explanatory and predictive research has led to a wide variation in the methodologic quality of prediction research.[7] The aim of this primer is to describe basic methods for conducting prediction research, which can be divided into three main steps: developing a predictive model, independently validating its performance, and prospectively studying its clinical impact. …