AI-powered precision drug dosing can boost outcomes and cost efficiency
Shivrat Chhabra is CEO of Dosis, a company that offers an artificial intelligence-powered personalized dosing platform used by dialysis providers to manage chronic drug regimens.
Dosis developed the Strategic Anemia Advisor, an AI-based dosing platform for patients with chronic anemia. Chhabra provides the following statistics for provider organizations using his company's anemia advisor:
- In dialysis clinics across the country, thousands of patients who suffer from chronic anemia now are receiving AI-based medication dosing that is personalized and precisely based on the patient's own demonstrated response to medication.
- As a result, dialysis centers have been able to reduce medication usage by 25%, maintain or improve patient outcomes, and reduce time spent managing anemia by 75%.
- To date, hundreds of thousands of dosing recommendations have been made using AI for chronic anemia alone.
Healthcare IT News interviewed Chhabra to get his thoughts on whether AI precision dosing can be the standard of care for chronic disease management, why AI dosing is particularly impactful in the dosing of drugs used to manage chronic conditions, and the downsides of relying on one-size-fits-all drug dose recommendations from manufacturers and the FDA.
Q: Why should AI-powered precision drug dosing be the standard of care for chronic disease management?
A: Drug dosing driven by AI is becoming more prevalent in many areas of medicine, such as dialysis, cancer and transplant medicine. It is in these areas that dosing plays a critical role in achieving favorable outcomes.
Today, AI-powered dosing is particularly impactful in the dosing of drugs used to manage chronic conditions, as the potential for adverse events, as well as costs, increases over the months and years that patients are on these drugs.
What's more, AI can help determine the minimum dose required to achieve the desired therapeutic outcome. In the case of dosing medications used in chronic disease management, inefficient dosing can result in significantly higher than necessary drug exposure for the patient and cost of care. At the same time, dose adjustments must be made frequently, as patients' drug response changes over time.
AI-powered drug dosing empowers clinicians to deliver personalized care to their patients, maximizing therapeutic safety and efficacy while minimizing cost of care. It allows for a practical and consistent approach to solving the complex, multifactorial problems associated with dosing.
It also allows large quantities of data to be processed quickly, and algorithms to be improved over time with relative ease as more data becomes available. The advantages to this approach are clear and significant for patients, providers and payers.
Q: Why is AI dosing particularly impactful in the dosing of drugs used to manage chronic conditions?
A: When patients are on medication regimens to manage chronic conditions for months and years, the consequences of getting the dosing wrong build significantly over time. Conversely, the benefits of getting the dosing right and managing the condition well also grow over time. These increased stakes make artificial intelligence an especially important tool in the management of chronic conditions.
Ultimately, I believe AI-powered dosing algorithms will be accepted as the standard of care. In five to 10 years, we have good reason to believe that the majority of therapeutic drugs for chronic diseases will be dosed using AI-enabled precision techniques. Precision dosing already is being proven to be able to provide significant improvements in dose efficiencies and ease of dosing.
Q: What, in your opinion, are the downsides of relying on one-size-fits-all drug dose recommendations from manufacturers and regulatory agencies?
A: One-size-fits-all recommendations are typically imprecise because they tend to be based on data from clinical trials that may or may not accurately reflect the results that a particular patient may experience. They also cannot easily be adapted to reflect a patient's actual response.
There is therefore an upper limit to how precise these recommendations can be. AI-based precision dosing approaches, on the other hand, use an individual's actual demonstrated response history to calculate optimal dosing, which allows for much greater precision.
Precision-dosing techniques today focus on increasing and decreasing dosage depending on a patient's actual response to a drug. In most instances, it is in a patient's best interest to receive the smallest amount of a drug required to achieve the desired therapeutic effect, as greater exposure to some drugs may be associated with higher risk of serious adverse events.
Q: How does Dosis's Strategic Anemia Advisor work? What have been some results?
A: To give some context, there are more than 550,000 End-Stage Renal Disease (ESRD) patients in the U.S. currently undergoing dialysis treatment, and more than 90% of these patients experience chronic anemia.
Their anemia is managed by a class of drugs called Erythropoiesis Stimulating Agents. Our anemia advisor is an AI-based drug dosing recommendation system that is used by dialysis providers across the country to help personalize the treatment of ESAs, and therefore help manage their patients' chronic anemia more effectively and efficiently.
The anemia advisor works by first evaluating a patient's historical response to ESA. It uses that information to calculate that patient's place on the spectrum of dose response, from hypo-responder (very low sensitivity) to hyper-responder (very high sensitivity).
It then calculates the optimal dose to drive patients as close to target as possible over the next several months – in other words, not necessarily to get to target as quickly as possible, which often results in overshooting, but to achieve the target and remain there in the long run.
We have found, through both research and through commercial use, that a personalized approach to chronic anemia management results in increased dose efficiency. We help clinicians use on average 25% less medication to achieve equivalent or improved patient outcomes.
This has both a clinical benefit and an economic benefit. On the clinical side, patients are exposed to less drug, which means a lower likelihood of serious adverse events like heart attacks or strokes.
On the economic side, this means a correspondingly lowered drug usage, which directly translates to lowered cost of care. Since its inception as a working platform, the anemia advisor has been able to generate more than two million dosing recommendations.