FDA Regulatory Environment Index for Tracking Changes

The U.S. Food and Drug Administration’s (“FDA”) regulatory environment has a significant impact on how companies operate, so data about that environment is very useful for business planning. In keeping with the theme of these averages posts, drill down enough to get a sense of the regulatory environment in which a particular company operates, rather than relying on more global averages across the medical device industry. On the other hand, making the data too specific and focusing on a specific product category prevents companies from seeing the forest for the trees.

I was recently asked by a company interested in the field of digital medical devices used in the care of people with diabetes to help assess trends in the regulatory environment. To that end, we decided to create an index that captures the regulatory environment for medium-risk digital diabetes devices. We tried to avoid global data on all medical devices without being too specific. In this sense, the index is similar to other indices such as the Standard & Poor 500, which are used to measure the economic performance of large companies in terms of market capitalization. My plan is to first define an index of product codes for these intermediate-risk digital diabetes products and then use that index to assess the regulatory environment in both premarket and postmarket regulatory requirements. was.

Create index

A product code is the language of FDA regulation. All FDA regulated medical devices are incorporated into the product code. If you don’t have a product code for your new device, the FDA often has to create one through a de novo process. The approximately 2,100 device classifications in the regulation are further subdivided into over 6,700 product codes, loosely organized by medical specialties such as anesthesiology, clinical chemistry, and general surgery.

Medical specialties can be useful groupings, but each specialty is very broad and not confined to a particular disease. As a result, for example, if a company is focused on products for diabetes, diagnostic codes may belong to one or two different medical specialties, while therapeutic products may belong to yet another medical specialty. may belong to the group of Even more problematic, when focusing on diabetes, those medical specialties are too broad to be useful in assessing the relevant regulatory environment.

For previous submissions, we have created a database of 510(k) abstracts submitted to the FDA. Using that database, I started by searching for the word “diabetes”, but words and phrases like “insulin”, “diabetic retinopathy”, “diabetic nephropathy”, “diabetic neuropathy”, etc. I also searched. My search used the PubMed MeSH (Medical Subject Headings) framework to identify synonyms and related terms, only to ensure that the more exotic terms were comprehensive. also included.

I focused on digital diabetes products, so I searched for the word “software” separately. I reasoned that any digital product would very likely mention at least the word “software” in its abstract, but technically, someone said, “Our product does not contain software. ” may also be written. Given the trend towards digital, I doubt that very many submissions include the phrase “no software” as if to brag about being low-tech, but it’s possible. There is

Finally, I chose a product code that is at the intersection of two different searches. That process produced a list of about 50 product codes.

I wanted to narrow the focus of the list so that it didn’t include products that just touched digital devices for diabetes, so I took two more steps. First, we divided the number of submissions for each product code that mentioned diabetes by the total number of submissions. I wondered if this would give me the percentage of submissions mentioning diabetes in a particular product code. This would be a good precursor to the relevance of production code for my purposes. Second, I consulted several clinical experts and asked from my list what they considered to be the most relevant product codes for this purpose. Although the treatment product referred to diabetes, experts did not believe that wound care was really a diabetes product.

At the end of this process, we created an index based on the following 30 product codes representing medium risk digital diabetes medical devices.

Frankly, Class I, Class II, and Class III medical devices (from products exempted from premarket review to products that require premarket approval) and those I was talking to are mostly intermediate risk. was involved in the business of

Using indexes

By defining an index for each product code, the index allowed us to see broadly how that category performed in nearly every dimension in which FDA generated data. For example, I first wanted to see how these medium-risk digital diabetes medical devices were fair in terms of FDA review time.

Black lines are 30 digital diabetes product codes. What looks like a blue area is actually 6600 vertical blue lines sorted from shortest to longest average review time within each product code. The shortest average review time for a single product code is just over 50 days, and the longest average review time for a specific product code is over 500 days.

Digital diabetes products are clustered towards the end of a lengthy review. The mean is shown by the dotted line, and all digital diabetes products are well above the mean. You can see why I do this because there really isn’t such a thing as a typical medical device review time. Instead, they vary greatly by product code.

Seeing this, I was curious if this pattern would change over time. The chart above is just an average of about 12 years of data (2010-2022). In the chart below, I’ve broken it down by year.

These are calendar years and we did our analysis in October 2022, so that year is a partial year.

Since I only used two colors, I need a little explanation about the colors. We used pink to represent the Digital Diabetes Index review time and chose blue to represent the average for all medical devices. It is easy to see that the average for all medical devices is usually significantly exceeded over the period of . However, in his two years of 2019 and his 2020, the average of all medical devices surpassed digital diabetes product review time. I honestly don’t understand why 2019 and 2020 are different from the overall trend.

Looking at the data, I understandably wondered why the FDA is reviewing digital diabetes products so slowly compared to the overall average. To find answers, we looked at data on post-market experience with these products. Specifically, we looked at the number of medical device reports submitted against indexed digital diabetes product codes and all other product codes.

Here, we focused only on the 20 worst medical device product codes to make the visualization easier to read. “Worst” means the highest number of Medical Device Reports (“MDRs”) filed in a particular year. So these are the things that the FDA has the greatest interest in when it comes to product safety and efficacy I chose his 2021 because that’s the most recent year’s worth of him I’ve had because it was the data of

You can see that the number of MDRs drops sharply by product code. The black bar is the digital diabetes index product code. There are 5 product codes in the top 20 product codes included in the digital diabetes space.

I don’t want to oversimplify this complex issue. Analyzing why these data are the way they are is beyond the scope of this article. Most people with diabetes are aware that many of them are elderly and face many challenges, such as difficulty using various medical devices due to their poor eyesight and handicap. think. Also, obviously, the sheer number of diabetics means that many of these products are in circulation at once.

We do not deny the need for a more nuanced evaluation of the data, but from a big picture perspective, these data are what FDA sees in postmarket experience, and this experience is likely a reflection of FDA’s conservatism. I would like to point out that I am contributing. About how we review new product submissions in this category.

Conclusion

In many cases, when looking at the data, it is important not to treat the medical device industry as a whole as if it were homogeneous, or to focus too much on individual products or product codes when trying to discern FDA trends. is. Indexing is one way companies can track the relevant regulatory environment in which they operate. And that tracking gives us insight into what the industry generally needs to do at a high level and what we can expect from FDA regulation.