AI and the development of Real-World Evidence

By Campion Quinn, MD

Aetion, a prominent healthcare analytics company, is pioneering the integration of generative artificial intelligence (AI) to transform how we analyze real-world healthcare data. By partnering with Amazon's Bedrock platform, Aetion makes it easier for healthcare professionals to derive meaningful insights from vast datasets. This advancement holds significant implications for clinical care, administrative efficiency, and patient outcomes.

Understanding Aetion's Role in Healthcare

Aetion specializes in converting real-world data into actionable evidence regarding medical treatments’ safety, effectiveness, and value. Collaborating with leading pharmaceutical companies, insurers, and regulatory bodies, Aetion employs rigorous scientific methods to inform healthcare decisions. Their platform, the Aetion Evidence Platform (AEP), is the foundation for this analytical work.

The Power of Generative AI in Healthcare Analytics

Generative AI refers to systems capable of generating content, such as text or images, based on learned patterns from extensive data. In healthcare, this means AI can assist in interpreting complex datasets by identifying patterns that might elude traditional analysis. Aetion's integration of generative AI allows users to input natural language queries—questions phrased in everyday language—and receive comprehensive analyses in return.

Aetion's Measures Assistant: Simplifying Data Queries

One of Aetion's notable innovations is the Measures Assistant. This tool enables users to express scientific questions in natural language, which the AI translates into complex data queries. For instance, a researcher might ask, "Identify patients diagnosed with diabetes who subsequently received a metformin prescription." The Measures Assistant processes this request and retrieves the relevant data, streamlining the research process. (aws.amazon.com)

Smart Subgroups Interpreter: Uncovering Hidden Patient Patterns

Another advancement is the Smart Subgroups Interpreter. This feature identifies clusters of patients with similar characteristics within a larger population. For example, the AI might detect subgroups with specific comorbidities among patients prescribed a particular medication. Researchers can then interact with these findings using natural language questions, such as, "What are the most common traits of patients in the cataract disorders subgroup?" The AI provides insights that can guide further investigation. (aws.amazon.com)

Impact on Clinical Care

By leveraging these AI tools, healthcare providers can better understand patient populations. For instance, identifying subgroups with unique characteristics can inform personalized treatment plans, leading to more effective interventions. Moreover, quickly analyzing patient data helps clinicians stay informed about emerging trends and potential risks.

Enhancing Administrative Efficiency

Administrative tasks, such as reporting and compliance, often require extensive data analysis. Aetion's AI-powered tools simplify these processes by allowing administrators to pose questions in natural language and receive detailed reports. This efficiency reduces the time spent on data management, enabling healthcare professionals to focus more on patient care.

Improving Patient Outcomes

Ultimately, integrating generative AI into healthcare analytics aims to improve patient outcomes. Treatments can be tailored more effectively to individual needs by providing clinicians with rapid, data-driven insights. Additionally, recognizing patterns across patient populations can lead to the development of better preventive measures and treatment protocols.

Aetion's Role in Managed Care

In managed care, Aetion is primarily used to generate "real-world evidence" (RWE) through its software platform, the Aetion Evidence Platform (AEP). This platform analyzes large datasets from electronic health records (EHRs) to assess the effectiveness, safety, and value of medications and treatments for managed care populations. Doing so allows payers to make informed decisions about coverage, reimbursement, and care management strategies based on real-world data rather than solely clinical trials. Essentially, Aetion helps managed care organizations optimize patient outcomes and cost-effectiveness within their plans.

Key Applications in Managed Care

· Data Analysis: Aetion employs sophisticated algorithms to analyze large volumes of real-world data from managed care organizations, identifying patterns, trends, and treatment effectiveness across different patient populations.

· Value-Based Care: Managed care organizations leverage Aetion to evaluate the impact of different treatment options on key performance indicators (KPIs) like cost per member per month (PMPM) and quality metrics, supporting value-based care initiatives.

· Treatment Selection: By analyzing real-world data, Aetion helps payers identify the most effective treatments for specific patient populations, potentially leading to better clinical outcomes and cost savings.

· Clinical Trial Emulation: Aetion can perform "trial emulation" studies using real-world data to mimic the results of a randomized controlled trial, helping to assess the effectiveness of new treatments in a real-world setting.

· Data Insights for Decision-Making: Managed care organizations use Aetion's analytics to inform formulary decisions, care management strategies, and provider network design.

Specific Applications in Managed Care

· Identifying High-Risk Patients: Analyzing data to identify patients with specific conditions who might benefit from targeted interventions.

· Evaluating Medication Adherence: Assessing how well patients take their prescribed medications and identifying factors influencing adherence.

· Assessing Treatment Effectiveness for Specific Patient Populations: Analyzing outcomes for patient subgroups based on demographics, medical history, and other factors.

· Cost-Effectiveness Analysis: Estimating the cost-effectiveness of different treatment options within a managed care plan.

Accelerating Clinical Trial Recruitment

One of the most critical challenges in medical research is recruiting the right participants for clinical trials. Aetion's AI-powered analytics can expedite this process by analyzing real-world data to identify eligible patients based on specific inclusion and exclusion criteria. By leveraging large datasets from electronic health records (EHRs), insurance claims, and other sources, Aetion can more efficiently match potential participants with ongoing trials. This capability is crucial for accelerating drug development, ensuring diverse and representative study populations, and ultimately bringing new treatments to market faster. By reducing recruitment time, AI-driven solutions help researchers focus on trial execution rather than patient identification.

Actionable Takeaways for Physicians

1. Embrace AI Tools: Consider incorporating AI-powered analytics into your practice to enhance data interpretation and decision-making.

2. Stay Informed: Keep abreast of advancements in AI applications within healthcare to leverage new tools effectively.

3. Collaborate with Data Scientists: Work alongside data professionals to understand and implement AI solutions that can benefit your patients.

4. Advocate for Training: Encourage training programs that familiarize healthcare staff with AI tools to maximize their potential benefits.

In conclusion, Aetion's integration of generative AI represents a significant step forward in healthcare analytics. These tools empower healthcare professionals to make more informed decisions by simplifying complex data analysis through natural language processing, ultimately enhancing patient care.