Health equity is an increasing focus for health services research as well as in ongoing refinements of health care delivery. However, the process of collecting and incorporating patient-level data into models and then translating these findings into the health care delivery processes is fraught with a multitude of potential mechanisms of compounding systematic disparities. Structural discrimination and human biases affect equity in collection of data, which ultimately impacts modeling and resultant findings. Similar factors that influence data gathering may also impact subsequent implementation of derived algorithms and care processes. This paper aims to review mechanisms that introduce disparities in data and modeling and propose potential first steps in addressing these disparities.
Tjoeng et al. (Mon,) studied this question.