Defining Performance Requirements - Jake Cowperthwaite : 18

MedTech Speed to Data

Coagusense developed the first point-of-care prothrombin time/Internationalized Normalized Ratio (PT/INR) monitoring system for cardiac patients to help them maintain warfarin dosage within a therapeutic range. In the latest version of their device, they actually removed connectivity features to accommodate the needs of their older, less-tech-savvy self-testers. Therefore, they had to go back to the FDA with more bench data for re-approval.

Andy Rogers talked with Mike Acosta, EVP/Head of Compliance at Coagusense, and later recapped some of the lessons learned with Senior Electrical Engineer Jake Cowperthwaite. Andy and Jake have an informative discussion about how to define performance requirements when you’re aiming for FDA approval.

Need to know:

• Make sure the performance requirement is objectively verifiable by a measurable test result, functional demonstration of performance, simulation analysis, and/or visual inspection.

• Plan for the number of prototypes you’ll need to create confidence in your statistics. Large companies may have their own internal mechanism to create a plan. Startups can consult with the FDA.

• Understand the regulatory considerations for seamless FDA approval. It’s a good idea to meet with the FDA prior to submission to outline your requirements. The earlier you have regulatory buy-in, the better.

The nitty gritty:

Rule Number One is to make sure that the performance of the device is objectively verifiable. For example, simply stating “the device shall be easy to use” is vague and subjective and won’t cut it with the FDA.

Write your requirement in a way that can be verified through: testing and measuring results, a functional demonstration of performance, analysis via calculations or simulations, or visual inspection. A well-written requirement is specific with clear criteria, for example if your product was a pump, a performance requirement could be: “the aspiration pump shall have X flow rate within Y bounds”. If it meets that requirement, you’re ready to move on. Don’t over-spec.

Three things the FDA is looking for:

1. Above all, efficacy and safety
2. Accuracy
3. Is your device novel technology or based on a predicate device?

At the test bench, start with a good understanding of how many prototypes you’ll need to have statistical confidence in your results. Sample size will depend on the data needed; an on/off switch won’t require a large sample size, but testing with different operators – as with in-home devices – will need a substantial data set.

In some cases, it’s possible to short-cut the process early in development by testing multiple variables at once. This will yield a lot of data, which can then be analyzed. You’ll find some variables meaningful and others not, but understanding these variables and their sensitivity early in product development has great value and can save money in producing fewer prototypes.

USEFUL LINKS

https://coag-sense.com/about-us/

https://www.greenlight.guru/