Validating prediction models for use in clinical practice: concept,
steps and procedures
Abstract
Prediction models are extensively used in numerous areas including
clinical settings where a prediction model helps to detect or screen
high-risk subjects for early interventions to prevent an adverse
outcome, assist in medical decision-making to help both doctors and
patients to make an informed choice regarding the treatment, and assist
in healthcare services with planning and quality management. There are
two main components of prediction modeling: model development and model
validation. Once a model is developed using an appropriate modeling
strategy, its utility is assessed through model validation. Model
validation provides a true test of a model’s predictive ability when the
model is applied on an independent data set. A model may show
outstanding predictive accuracy in a dataset that was used to develop
the model, but its predictive accuracy may decline radically when
applied to a different dataset. In the era of precision health where
disease prevention through early detection is highly encouraged,
accurate prediction of a validated model has become even more important
for successful screening. Different clinical practice guidelines also
recommend incorporating only those prediction models in clinical
practice that has demonstrated good predictive accuracy in multiple
validation studies. Our purpose is to introduce the readers with the
basic concept of model validation and illustrate the fundamental steps
and procedures that are necessary to implement model validation.