Healthcare Technology Featured Article

November 14, 2019

Six Ways Artificial Intelligence Is Transforming Medication Management



Over the past few years, artificial intelligence (AI) has been touted as a transformational force in health care. The AI health market has experienced explosive growth, with public- and private-sector investments estimated to reach $6.6 billion in 2021.

Health care is embracing AI for a range of applications, from robotic surgery to drug development to clinical research. In addition, AI is being adopted as a practical tool to reduce costs, improve outcomes and replace labor-intensive, repetitive tasks that are prone to error. Those applications make AI poised to transform medication management. This article takes a brief look at how that is happening today.

What Is AI? Before we look at how AI may be used, let’s level-set. Real-world AI technology may be defined as specialized software being deployed today to improve health care. Two types of AI technology are gaining real traction in health care:

  • Machine learning is the most prevalent type of AI technology. Its algorithms use such heuristic techniques as mixing/matching and trial and error to improve and adapt models automatically (i.e., without programmer intervention). It has applications in areas where traditional statistics start to run out of steam, such as with larger datasets and where it’s not clear what assumptions or models should be used.
  • Natural Language Processing (NLP) uses a combination of expert rules and AI techniques to process unstructured free text, such as clinical notes, and transform it into structured data for more accurate searches/discovery and decision support. Advances in the technology and the explosion of computerized clinical notes are driving NLP into the mainstream of Health information technology (health IT).

Other emerging AI technologies (e.g., artificial neural networks, deep learning) show promise. However, they are probably a few years away from significant, real applications in health care, including in the area of medication management.

Use of AI in Medication Management. We often hear about the use of AI to collect and analyze complex health data and provide previously undiscovered insights through deep learning. Reflecting on persistent problem areas in medication use and therapy, AI is being used in a number of ways to improve medication management. Six examples come to mind. AI is being used today to:

1. Improve medication safety. AI is a complex tool that can reduce diagnostic and therapeutic errors by evaluating large sets of data against complex, multifactorial and dynamically changing criteria (such as drug utilization review). The value of such approaches already is being demonstrated. A team at Brigham and Women’s Hospital evaluated a probabilistic, machine-learning approach — based on statistically derived outliers — to detect medication errors. The study found the screening system could generate alerts that might otherwise be missed with existing clinical decision support systems.

2. Reliably and cost effectively predict health risks and outcomes across large populations. Currently, AI is being used to prevent drug overdosing. A group in Michigan is using patients’ medication histories from various sources, including electronic health records (EHRs) and state-run prescription drug monitoring programs, in algorithms that calculate “overdose risk scores” and predict risks of overdosing from a prescribed opioid. Interventions based on outcomes of high-risk patients are then recommended. This example exemplifies AI’s value in medication management by augmenting clinical decision making.

3. Reduce time and expense. AI shows promise for improving accuracy in, and thereby reducing costs of, medication-related tasks. For example, AI-enabled dosage error reduction is among the top 10 AI applications in health care with the greatest cost-saving potential. This application of AI could save $16 billion annually.

4. Streamline the prior authorization process. Despite the growing use of electronic prior authorization (ePA), submission and review of clinical documentation for drugs requiring approval — especially therapies involving specialty drugs — is plagued with duplicate data entry, delays and rework. The required documentation is often located in the clinical notes narrative of the (EHR. Required clinical data can be extracted and transformed into structured data using NLP. This information — along with other necessary data — forms the clinical documentation to send at the same time as an ePA request is submitted.

How do ePA programs ascertain which data are required? This is where evidence-based algorithms and machine learning come into play. With AI, the adjudication process is transformed. ePA becomes a decision support tool for the prescriber. The evidence-based algorithmic programs factor in the therapy policy and restrictions (including formulary) and gather patient-specific data (e.g., the patient’s health status and medical history) and outcomes of patient populations with similar characteristics to present the prescriber with therapy recommendations. Selection of one of the recommendations triggers automatic approval of the prior authorization (PA) request in real time.

This “next generation” ePA will be prevalent in the not-too-distant future. In fact, it’s here today. One company has implemented an AI engine that uses NLP to extract data needed to adjudicate a PA. Another is working on increasing the level of automatic adjudication of PAs by using algorithms to predict which PAs should be approved. As a result, we’ll see more consistent and accurate PA adjudications, faster turnaround for patients to get onto therapy and reduced administrative costs for both payers and providers.

5. Monitor medication adherence. AI technology can be used to monitor patient adherence and notify clinicians when intervention may be necessary. The application of NLP and machine learning turns what would otherwise be a labor-intensive and cost-prohibitive activity into one in which clinicians devote time only when needed. This has the potential to save money and lives. Medication nonadherence is estimated to be responsible for $100 billion to $300 billion annually in excess health care costs, a quarter of hospitalizations and about 125,000 deaths.

As a result, medication compliance is of great interest to payers. A Medicare Advantage plan identified members at risk of medication nonadherence who are receptive to such digital interventions as personalized text messaging about refills and remote patient monitoring instead of follow-up by letters or phone. By focusing on patients at greatest risk and communicating in the most effective way, adherence rates are rising.

This strategy could be extended to predict specific difficulties a patient will likely experience in adhering to a prescribed medication regimen and determine the most appropriate intervention (i.e., right level of intensity and approach for patient compliance, optimal effectiveness and lowest cost). Drawing on the patient’s EHR (both structured and unstructured data) and interactions with a virtual assistant, multiple sets of factors can continuously be monitored by analytics that both predict risks of failure and prescribe actions. For example, factors such as the number of medications a patient must take, involvement of multiple prescribing physicians, whether a spouse or caretaker resides in the home, and travel distance to a drugstore can all be considered simultaneously when determining the need for and frequency of follow-up consultations.

6. Schedule follow-up visits to assess therapy progress. The scheduling of follow-up appointments with a patient’s doctor or pharmacist to assess a medication’s progress often doesn’t happen because of the time required and inconvenience. AI can help with a scheduling application that reduces patient wait times. It dynamically predicts expected wait times and sends a text message to patients two hours before their scheduled appointments with updates on when they can expect to be seen. This program uses machine learning to detect patterns in more than a million past appointment wait times and hundreds of related factors, including attributes of patients scheduled to be seen on a given day, the day of the week and the doctor’s past performance on a given day.

Looking ahead. In just a few short years, machine learning and NLP will have profound and positive impacts on the delivery of health care and medication management.




Edited by Maurice Nagle
Get stories like this delivered straight to your inbox. [Free eNews Subscription]




SHARE THIS ARTICLE



FREE eNewsletter

Click here to receive your targeted Healthcare Technology Community eNewsletter.
[Subscribe Now]