CASE STUDY CCS
Turning a wealth of information into action for patients
CCS uses predictive AI to deliver comprehensive diabetes care
3-MINUTE READ
CASE STUDY CCS
CCS uses predictive AI to deliver comprehensive diabetes care
3-MINUTE READ
CCS and Accenture are revolutionizing patient care through PropheSee™, an AI-powered predictive model that provides a holistic view of each patient, facilitates more informed decision making and drives effective patient management. To date, the solution has achieved:
50%
Improved adherence in targeted, high-risk patient cohorts
85%
Accuracy predicting patient behavior almost 3 months in advance1
$2,200
Savings of up to this amount per patient per year
CCS, a leading chronic care management company providing clinical solutions and home-delivered medical supplies for those living with chronic conditions—particularly diabetes—had retained their proprietary customer data for over 30 years. Now, they sought a solution that would make their data actionable while addressing a core commitment: reinforcing holistic health and the prevention of co-morbidities so often seen in diabetes patients. Just as importantly, the solution would be a core resource in their journey to reinvent from a medical equipment supplier to a proactive chronic care management organization.
To transform their approach to patient care as well as their personalized outreach strategy, CCS joined forces with Accenture to develop a Customer Analytics Record (CAR), essential for providing insights into customer needs and preferences, and an AI-powered predictive model using advanced analytics. This advanced analytics program, PropheSee™, has been a success—helping to improve adherence among targeted, high-risk patient cohorts by as much as 50%.
The CCS and Accenture team started by analyzing over two decades of the company’s proprietary data. The team combined CCS’s internal data and social determinants of health information, effectively integrating them to create a longitudinal view of patients and ultimately building the CAR.
The team then created and evaluated multiple predictive models, with the 'winning’ model achieving an impressive 85% accuracy in predicting patient behavior almost three months in advance1 and integrating seamlessly into CCS’s existing Azure Data Lake environment.
PropheSee™ has proven its value by enabling specific customer outreach programs, using CAR data to generate intervention messages that encourage patients to stay on track with their treatment plans. The solution can not only enhance patient engagement but is expected to take patients preferred communication methods into account in the future including when, what and how they want to be contacted.
The launch of PropheSee™ reinforces our commitment at CCS to deliver a smarter, more personalized approach to keeping people on therapy—resulting in healthier members and lower cost for the healthcare ecosystem.
Richard Mackey / CTO, CCS
CCS has realized a solution providing a holistic view of each patient and facilitating more informed decision-making and effective patient management, ultimately helping retain customers and drive adherence to essential treatments.
With its ability to track patient behavior, link the impact of patient characteristics to adherence and predict compliance with treatments, weeks in advance, PropheSee™ delivers intervention programs which keep patients on track, and reduce overall costs. In addition to improved clinical outcomes, health plans and risk-bearing providers can also realize savings of up to US$2,200 per patient per year from an increase in continuous glucose monitoring (CGM) adherence and better glycemic control as well as waste avoidance in CGM setup costs.
CCS has made strides in understanding customer needs, enhancing patient care, and proactively identifying and engaging high-risk individuals before adherence issues escalate. This not only delivers better patient outcomes—it emphasizes CCS’s dedication to a holistic health approach that helps prevent ER visits and hospitalization and keeps diabetes patients under the best possible care.
¹ based on the predictive power of the model in a 30-day ordering cycle.