Machine learning tool uses social determinants data to predict utilization

machine learning tool

A machine learning algorithm has accurately predicted inpatient and emergency department (ED) utilization by using publicly available social determinants of health (SDOH) data. Now it is possible to determine patients’ risk of utilization without interacting with the patient or collecting information beyond age, gender, race and address.

This study was published in the American Journal of Managed Care.

The researchers note that sociodemographic status, racial and ethnic disparities, and individual behaviors directly correlate with an increase in the incidence of chronic diseases.

Researchers said that at present analysis, predictive models and prevention initiatives focus on addressing SDOH at the population level or the zip code level. But this method has its challenges. There is a gap in addressing the individual patient’s needs, such as defining clinical action steps that are relevant to the patient as opposed to an overall population approach.

Advancements in cognitive science lead to the analysis of individual contributions of social determinants of health at the patient level, informing appropriate interventions that can reduce the risk of negative health outcomes like preventable readmissions and/or hospitalizations.

The study aimed at using machine learning to predict utilization independent of a patient’s clinical condition, while establishing which determinants contribute to the greatest risk of utilization.

The results revealed that the machine learning tool was able to predict utilization with a high degree of discrimination.

It showed that the social determinant most associated with risk was air quality, which had a relative value more than twice that of income, which was the second determinant most associated with risk. Both air quality and incomes are important to the decision making ability of the model as compared to age, gender or ethnicity.

The study shows that it is possible to generate an accurate model to predict inpatient and ED utilization by using decision tree-based machine learning with the help of purchasable and publicly available data on social determinants of health.