CDS Software Final Guidance Carries Real Consequences for Manufacturers

By Brynn Stanley

The U.S. Food and Drug Administration (FDA) has finalized its Clinical Decision Support (CDS) Software guidance and reinforced its position during the March 2026 Town Hall. For medical device and digital health companies, that combined message is not just a definitional clarification. FDA has drawn a clear distinction between software that will be regulated as a medical device and software that will remain outside FDA oversight.

What FDA Finalized

The guidance interprets Section 520(o)(1)(E) of the Food, Drug, and Cosmetic Act (FD&C Act) and and clarifies FDA's thinking on when CDS software is not a medical device.

To fall outside FDA regulation, CDS software must satisfy all four criteria in the FD&C Act. Specifically, it must:

1. Not be intended to acquire, process, or analyze a medical image or a signal from an in vitro diagnostic device or a pattern or signal from a signal acquisition system (section 520(o)(1)(E) of the FD&C Act); 
2. Be intended for the purpose of displaying, analyzing, or printing medical information about a patient or other medical information (such as peer-reviewed clinical studies and clinical practice guidelines) (section 520(o)(1)(E)(i) of the FD&C Act); 
3. Be intended for the purpose of supporting or providing recommendations to a health care professional about prevention, diagnosis, or treatment of a disease or condition (section 520(o)(1)(E)(ii) of the FD&C Act); and 
4. Be intended for the purpose of enabling such health care professional to independently review the basis for such recommendations that such software presents so that it is not the intent that such health care professional rely primarily on any of such recommendations to make a clinical diagnosis or treatment decision regarding an individual patient (section 520(o)(1)(E)(iii) of the FD&C Act).

The fourth criterion is the critical gating requirement and the focus of FDA’s recent messaging.

FDA’s Focus: Level of Automation and Time-Critical Nature of Decision Making as a Regulatory Threshold

FDA has emphasized that CDS software must enable clinicians to independently evaluate the basis for its recommendation rather than rely primarily on it. The Agency clarified that when evaluating whether CDS allows independent clinician review, it looks at both how automated the software is and how time-sensitive the clinical decision is. In urgent situations, clinicians are more likely to rely on the software due to automation bias, which can increase the risk of errors such as following incorrect recommendations or missing needed actions. As a result, FDA holds that CDS will qualify as non-device software only if it truly enables clinicians to understand and independently assess the basis of the recommendation without primarily relying on it. This has immediate implications for AI-enabled tools. Models that operate as “black boxes” will generally fail to meet this requirement because the user cannot meaningfully assess the underlying rationale. FDA’s position is not that transparency is preferred; it is that transparency is required to remain outside device regulation.

What This Means for Manufacturers

For manufacturers, CDS classification is now a front-end strategic decision rather than a downstream regulatory determination. Companies must decide early whether to design software to meet the CDS exemption or to pursue a regulated Software as a Medical Device (SaMD) pathway. That decision affects product architecture, labeling, and both regulatory and commercialization strategy.

AI-enabled products present a particular challenge. Many will likely not satisfy the independent review standard and will therefore be treated as medical devices. This creates a practical tradeoff between performance and regulatory positioning, especially where model complexity limits explainability.

FDA also continues to evaluate CDS through intended use and product claims. Framing a product as “decision support” will not be sufficient if the design or marketing suggests that clinicians are expected to rely on the output. As a result, claims, labeling, and user interface design function as regulatory controls, not just marketing tools.

Even where software qualifies as non-device CDS, FDA’s expectations are not minimal. Companies should be prepared to support their position with documentation demonstrating how users can independently interpret inputs, logic, and outputs, along with human factors considerations designed to mitigate undue reliance.

“FDA’s CDS guidance is a welcome step toward clarity. It gives manufacturers a real opportunity to be intentional about how they design, position, and scale their products. With the right alignment between transparency, user experience, and regulatory strategy, companies can not only meet these expectations but use them to build stronger, more trusted solutions.” 
Brynn Stanley, Associate Attorney

Practical Next Steps

Manufacturers should conduct a structured assessment of their software against the CDS criteria and evaluate whether users can realistically and independently interpret the basis for recommendations. Claims, labeling, and user workflows should be aligned with the intended regulatory pathway. 

How Gardner Law Can Help

Gardner Law works with medical device and digital health companies to align CDS product design with regulatory strategy from the outset. This includes assessing whether software can qualify as non-device CDS, advising on explainability and user-facing transparency, and ensuring that product claims and interface design support the intended regulatory position. For products that fall within FDA oversight, Gardner Law supports premarket strategy, submission development, and lifecycle compliance, including integration with quality management systems and post-market obligations.

In an environment where small design or wording decisions can determine regulatory status, manufacturers benefit from a coordinated approach that connects product development, regulatory positioning, and commercialization.