Cutting Through the Fog of Guidance for Incident-to Billing

Given the vast quantity of information published by the federal government, Medicare Administrative Contractors (MACs), consultants, and others, it can be easy to lose track of what is binding, what is informational, and what you can safely ignore.

The regulatory hierarchy has always been important, and the existence of artificial intelligence (AI) may make it even more important to understand what sources are authoritative and which are not, because when you perform an online search, you will often be given a purported answer without the actual original source – or the search results may claim to be quoting from an original source, but actually include fabricated information.    

I have noticed that computer searches are far less effective today than they were in the past.

For example, if you perform a search on Medicare’s “incident to” rules, there is a very real possibility you will get a paragraph that tells you that “new problems can’t be treated ‘incident to.’” AI will be relying on the many articles written by consultants and publications from MACs.

But those publications are all wrong. How can I say that so confidently? The answer is all about the regulatory hierarchy.

The most controlling authority is, of course, the U.S. Constitution. But it isn’t relevant in this context, because there is nothing in the Constitution about incident-to billing. Next, consider the statute. Incident-to billing was created by the Social Security Act § 1861(s)(2), which says Medicare covers “services and supplies furnished as an incident to a physician’s professional service.”

That text includes no meaningful detail. 

Next, we consider any National Coverage Determinations (NCDs) and regulations, the final two sources for binding materials. There is no NCD on incident-to billing, but there is a regulation, 42 CFR 410.26. The regulation establishes a requirement that “services and supplies must be an integral, though incidental, part of the service of a physician …in the course of diagnosis or treatment of an injury or illness.” First, I will note that “integral” and “incidental” are literally opposites, so that provision is inherently confusing. But let’s focus on the language relevant to the treatment of new problems. The regulation says that anything in the “course of diagnosis or treatment” can qualify as “incident to.”

We have now completed our review of the binding materials, and there isn’t a single mention of a new problem. Even if there were something specifically discussing the topic in the Medicare manuals, it wouldn’t be binding, but there isn’t.

As you probably know, there are many publications from MACs indicating that new problems can’t be treated “incident-to,” but let’s return to the hierarchy. A MAC publication is no more authoritative than fortune cookies or a horoscope.    

The hierarchy should drive how you conduct research. If your search is limited to the Medicare manuals and MAC newsletters, you will absolutely get things wrong, because you are only looking at information that isn’t binding, and ignoring the real authority. 

You should never refer to anything in the manuals or carrier guidance as a “rule.” At most, it is a policy.

I would strongly discourage you from refunding based on something found in the manuals or carrier guidance. If someone can’t cite a statute, regulation, or NCD, you can hang onto your money.

Manuals, the Medical Learning Network, and MAC newsletters can help you understand how people are interpreting regulations. But the only actual requirements are found in the statutes, regulations and NCDs.