ExpressScripts develops technology to predict patient med adherence

by on

Medication complianceExpress Scripts, a drug benefit management company Express Scripts is using a new tool to help lower healthcare costs: predictive computer models. The company announced that it has successfully tested a set of computer models that sift through patient data to predict whether or not a patient will likely adhere to prescribed drug therapies.

According to the company, the models are highly accurate and are aimed at patients dealing with chronic conditions such as diabetes, high blood pressure and high cholesterol. The purpose of these models is to identify patients who may need special assistance in sticking with their medications, intervening before the patient requires more expensive treatment or hospitalization. Assistance may be in the form of auto refills or home delivery of medications. Express Scripts says that over $106 billion is spent annually on treatments that could have been avoided had patients followed the prescribed drug therapy.

The upside of this type of program is clear: helping patients who may not take their medications avoid a more serious, and expensive, course of action. The downside may not be so obvious. The models use data such as past patient behavior to demographics to determine likelihood of adherence. They also use proprietary data that Express Scripts has determined is relevant. However, since the models are proprietary, this additional patient information is not shared. This means patients may not know what part of their medical history is being used to determine how they are assessed by their drug benefit company.

It is likely, as modeling becomes more accurate and cost-effective, that more computer models will be employed by healthcare providers, benefits management companies and insurance carriers to predict patient behaviors and even treatment outcomes. What this will mean for patients, and how their health information will be used, is yet to be seen.

VN:F [1.9.7_1111]
Rating: 0.0/10 (0 votes cast)
VN:F [1.9.7_1111]
Rating: +4 (from 4 votes)

Leave a Comment

Previous post:

Next post: