Artificial intelligence (AI) can revolutionize health care. Imagine an AI-powered diagnostic imaging system that detects diseases early, predictive analytics that identify at-risk patients, and bots that streamline billing and appointments. However, not all technologies are accurate and reliable. Accreditation can help ensure both. With a structured, independent review process, it can verify an AI tool or system against rigorous, predetermined standards to ensure it is safe, unbiased and effective.
What Are the Benefits of Health Care AI Accreditation?
Formal validation elevates patient safety and public trust while creating a framework for responsible innovation. It offers various advantages for patients, health care providers and AI developers.
Validating Performance
Accreditation confirms the validity of an AI model’s claims. For example, if an algorithm says it has a 95% accuracy in detecting a specific health condition, accreditation verifies this through independent assessments. Doing so ensures the percentage is more than a marketing claim.
Mitigating Bias and Promoting Equity
Verification standards mandate that AI systems are tested for and actively mitigate harmful biases, such as those associated with race, gender and socioeconomic status. For example, an AI tool for diagnosing skin cancer must be trained on a diverse range of skin tones.
Strengthening Data Privacy and Security
Accreditation ensures adherence to regulations like the Health Insurance Portability and Accountability Act to protect patient confidentiality. It also determines whether the system features safeguards like end-to-end encryption, a secure cloud infrastructure, strict access control and clear audit trails.
Providing a Clear Market Differentiation
For AI developers, accreditation is a seal of quality that distinguishes their product in a competitive marketplace. It also lets hospitals and other health care organizations know that the tool is responsible, safe and ready for deployment.
Establishing a Roadmap for Excellence
Accreditation provides a clear, step-by-step framework for developers to follow, ensuring they build crucial elements from the ground up. These may include factors like risk management, clinical oversight and ethical considerations.
Key Considerations for Pursuing AI Accreditation
Here are some factors to consider for anyone seeking to have their AI-powered health care solutions accredited:
- Internal readiness assessment: The path differs for AI developers and health care institutions that use AI tools. For example, organizations must review their existing governance structure, risk management protocols and data handling policies to determine how they align with accreditation standards.
- Accreditation process: The steps may vary, but they typically include a self-assessment, a submission of evidence, a review by the accreditation body and continuous monitoring. The independent assessment involves validating the submitted materials and possible interviews with key personnel.
- Continuous improvement and re-accreditation: Formal validation is not a one-time event. It requires ongoing monitoring and a commitment to maintaining standards, with re-accreditation cycles ensuring long-term accountability.
- Time and resource allocation: Accreditation is a significant commitment that typically requires a dedicated team, internal audits and thorough process documentation. However, it is an investment that pays dividends in trust and quality.
Top Organizations Shaping AI in Health Care
While URAC is the first to offer a formal accreditation program, several key institutions are also shaping the standards and ethics of AI in health care. Here is an overview of each one.
1. Utilization Review Accreditation Commission (URAC): The Pioneer in AI Accreditation
URAC is the first organization to launch a full-scale, dedicated Artificial Intelligence in Health Care Accreditation program, which has two pathways. One is for AI users like hospitals and health systems to ensure they are implementing AI responsibly. The other is for AI developers to validate the quality and safety of their technologies before they go to market.
URAC is proactive in addressing the risks and opportunities of AI in medicine, providing a tangible, market-ready solution that moves beyond discussion and into action. Its groundbreaking program offers a clear, verifiable “gold star standard” that demonstrates a commitment to safe and ethical AI. URAC accreditation means patients and partners can trust an AI system.
2. National Committee for Quality Assurance (NCQA): The Measurement and Quality Expert
The National Committee for Quality Assurance is a national leader in health care quality improvement, known for its widely used Healthcare Effectiveness Data and Information Set performance measures. It is actively exploring how to apply its three decades of expertise in measurement to the world of AI.
NCQA is utilizing experts from payers, providers and health IT to establish best practices. It is also looking into risk-based governance, transparency and methods for providing evidence of compliance with standards. This organization’s deep experience in data validation makes it a crucial player in defining how the industry will evaluate whether AI tools are actually improving care and outcomes.
3. Coalition for Health AI (CHAI): The Collaborative Standard-Setter
Coalition for Health AI is a collaborative effort of academic health systems, health care organizations and industry experts. It aims to advance responsible AI through consensus-driven best practices and standards for using AI in health.
This coalition aims to create a gold standard for AI best practice. CHAI includes nearly 3,000 organizations, including regulatory bodies, major health systems and technology companies. Its work is structured around four pillars — convene experts, develop standards and best practices, certify or assure AI is trustworthy, and educate by providing resources. Its strength lies in its diverse ecosystem, which ensures the standards are practical, widely accepted and reflective of the real-world needs of the health care community.
4. National Academy of Medicine (NAM): The Ethical Compass
The National Academy of Medicine is an independent, nonprofit organization that provides nonpartisan, evidence-based guidance on critical health issues. It has developed and published an Artificial Intelligence Code of Conduct for Health and Medicine, which is a unifying framework designed to align the entire field around responsible and ethical AI applications.
NAM’s AI code of conduct can provide the ethical backbone for the entire industry. It centers on principles like ensuring AI is safe, effective, equitable, transparent and accountable. Ultimately, NAM provides an essential ethical foundation upon which all other work — including formal accreditation — can be built.
Embracing a Culture of Continuous Improvement
AI has immense potential, but that promise can only be maximized if it is built on a foundation of safety and accountability. Accreditation is the most powerful tool for building that trust, offering benefits for everyone involved, from the patient receiving care to the developer designing the technology. By embracing formal validation and seeking guidance from industry leaders, the health care community can navigate the future of AI with confidence. Ultimately, accreditation ensures the innovation always serves its core purpose of improving health.
