Home / Resources / Articles / Part 3: Innovative Artificial Intelligence (AI) Tool Optimizes Care for T2D Patients

Part 3: Innovative Artificial Intelligence (AI) Tool Optimizes Care for T2D Patients

Oct 13, 2020

Author: Steve Freed, R.PH., CDE, et. al.

Part 3: Optimize Clinical Diabetes Management with AI Technology

A new clinical software using artificial intelligence is being used for those with type 2 diabetes in Kentucky. It shows promising results in improving the Quality of Life for its residents and lowering treatment costs. This feature is a 3 part series.  


Go to Part 1 | Go to Part 2

Or click here to download the complete article as a single PDF. 

In part 3: user testimonials about the experience and exciting findings using the software. A more in-depth examination of the benefits of applying decision-support software in clinical practice.   

GlucosePATH is decision-support software that aids the prescriber in developing patient treatment plans.  One feature of the software is AI that recognizes the importance of using specific standard therapies while evaluating millions of potential treatment regimens.  

The software is user-friendly, collecting a wide range of data and compiling it into a summarized recommendation list highlighting the most critical outputs that need to be considered to make an informed decision. Patient information data include age, sex, A1c level, current treatment regimen, susceptibility to side effects, and comorbidities.  

Providers can also input their own preferences, such as what drug the treatment should contain and what drug interventions to exclude.  The AI uses a Medicare cost formula to generate the medications average cost using the patients insurance information.  

Evaluating drug performance is only part of a good treatment regimen. It is essential for the patient to be a part of the decision-making process and primarily to be heard when expressing concerns about their treatment cost.  It allows the user to set a maximum solution cost that the patient can specify to the prescriber based on their financial capabilities, but this will not affect how the medication ranking process works. 

This software will save physicians time while determining a suitable regimen for a specific patient. It should also decrease the amount of time it takes for the patient to achieve their goals.  

The software makes it easy for the user to factor in cost when computing recommendations for drugs, unique to this tool. It bases its recommendations on various financial variables inputted into the software, including the maximum solution cost, the patients insurance, any coupons available, cost sensitivity, and free medications available. The software creates an estimated figure of what the drug will cost per month based on this information.  

Of the different variables used to get a result, cost sensitivity is the most exciting and valuable factor that many prescribers cannot consider for each patient. The cost sensitivity refers to the financial burden the patient will face if the entire dollar amount in the maximum field is spent using a scale ranging from 0 (no burden at all) to 10 (extreme hardship). From a user perspective, it is comforting that the software intends to make the patient’s treatment successful medically and provides a financially feasible option. 

Also, software users can choose different demo scenarios to show how the software makes recommendations under these conditions. These scenarios help to individualize other medication therapy selection based on what the physician selects as a condition to consider.  

For example, a scenario labeled “demo-GERD” represents a patient with a starting A1c of 8.1, with high LDL, hypertension, hypertriglyceridemia, and susceptibility to gastroesophageal reflux disease.  Another scenario, named “demo-BMI,” shows how the software would evaluate regimens for increased body mass index patients. 

Other demo options represent patients with high insulin resistance, taking into account the patient’s insulin resistance and insulin secretion factors. The patient’s insulin resistance score (HOMA-IR), endogenous insulin production score (HOMA-B), and kidney function (eGFR) must be considered from a user’s perspective to tailor the medication therapy based on those critical factors.  

The software provides the user with up to 10 different treatment regimens, where each regimen can contain up to 5 treatments.  Besides medicines, a “treatment” can include behaviors such as diet and exercise.  The software models both diet and exercise using the same metrics as for drugs, including A1c lowering capability and positive side effects.    

Special optional features are also available to personalize this process, such as allowing prescribers to track the inventory of medicines available to patients once an option is chosen. This feature gives the user a space to record the quantity of the medication selected and replacement costs and alternate doses available so that any changes from standard dosing can be tracked. 

Applying decision support software like GlucosePATH will help patients get to their goals sooner, in an effective way. It allows the user to be in control of patient care while expanding the norm of prescribing habits. Patients are a part of the decision- making process, and tailor a medication to their needs to optimize their goals. The system does the complicated analytical computing for the provider while also considering patient factors and even finances. This will change the future of prescribing and bring a fresh new outlook on patient care using AI technology that can improve a patient’s quality of life in a shorter amount of time and eliminate the frustrations of trying to pick the right medication.  

GlucosePATH uses algorithms based upon treatment standards, clinical data inputs, and other variables such as budget to help individualize treatment to fit the patient. While it is important to consider patient input when making treatment decisions, it is equally essential to ensure that the medication chosen is clinically ideal for an individual patient case. The future of diabetes care needs to consider both – a happy medium where patients can be a part of their drug therapy so that they are more inclined to adhere, and an ideal drug regimen that fits their specific patient profile. GlucosePATH is an AI software that can do this on one user-friendly platform while navigating through millions of therapy options and ultimately yielding the top few choices. Prescribers and patients can confidently use this software, knowing that the options they will get to choose from will keep both of their interests in mind: the clinical factors of an ideal therapy for prescribers, and a cost-effective drug that fits the needs of the patient. 


Overview of how Kentucky Medicaid has used this software:  The Final Report from A Study on Type 2 Diabetes For Patients Among Medicaid Beneficiaries in Kentucky, June 30, 2020.   

More information at www.GlucosePath.com 

For Access and Training for the GlucosePath software at no cost, please email Steve Freed at publisher@diabetesincontrol.com  and request how you might use it, and we will provide you the link and the tools you need to use the software. (For Medical Professionals Only) 




[2]Rowley, William. et al. Diabetes 2030: Insights from Yesterday, Today, and Future Trends. Population health management vol. 20,1 (2017): 6-12. doi:10.1089/pop.2015.0181 

[3]Ozanich, G. et al. A STUDY ON TYPE 2 DIABETES MELLITUS FOR PATIENTS AMONG MEDICAID BENEFICIARIES IN KENTUCKY. https://kyma.org/shared/content/uploads/2019/06/2019-Diabetes-Report-latest11610.pdf 


Steve Freed, R.PH., CDE, with:

Peter Jay Won Pharm.D. Candidate, University of South Florida, Taneja College of Pharmacy 

Aleksandra Kusic, PharmD Candidate, Florida A&M University, College of Pharmacy and Pharmaceutical Sciences  

Joan Prifti, PharmD Candidate, Lake Erie College of Osteopathic Medicine, LECOM School of Pharmacy 


Full article series:

Part 1:   Adherence-related issues are hemorrhaging the costs of diabetes-related treatment in America

Part 2: Collaborative Analysis Concludes that the AI Tool Improves Outcomes for Type 2 Diabetes Patients

Part 3: Optimize Clinical Diabetes Management with AI Technology

 Click here to download the complete article as a single PDF.