10 Rules for Designing Great Medical Devices

From keeping track of most serious medical device recalls to exploring the sources of medical device failure, Qmed spends plenty of time looking into what goes wrong in the design process. But what are the rules that allow for safe and highly effective design?

Qmed recently asked this question on the Medical Devices Group on LinkedIn, and received a useful response from Eric Claude, vice president of product development at MPR Associates (Alexandria, VA). Claude has been designing medical devices for about 20 years, and he says that he actually uses a top 10 list with such rules while working with clients and design teams.

Here is exactly what Claude then listed off, citing a colleague named Craig for two of the points:

1. Understand all the stakeholders and their needs.

"Today more than ever we have to consider a very broad range of stakeholders--the clinician, patient, regulator, payer, device manufacturer, and more! If we're talking home health care it gets even more complicated with needs imposed by the diverse range of environments."

2. Relate product requirements to clinical needs.

"Understand the clinical and user needs (which are often more qualitative than quantitative) and then translate each of these to quantitative product requirements. Make sure that requirements specify "what" the product must do, not "how" it must be done."

3. Develop a high level architecture for the overall system/product.

"What are the major pieces of the puzzle, and how will they go together? Identify at least one way, preferably several alternative ways, to solve the problem at a "system" level (by brainstorming, own experience, others' experience, similar products)."

4. Understand the risks.

"These could be technical risks for first-of-a-kind products or organizational risks where you're treading in unfamiliar territory. Focus on the high risk elements of the design first. And always have contingency plans for the risky items."

5. Don't ignore the laws of physics.

"They're called "laws" for a reason, and it's critical to understand how they apply to the product you're designing. Develop a science-based "feel" for the problems you have to solve by simplifying and analyzing the effects of key parameters."

6. Design with empathy.

"Ultimately products will be used by people. And people have widely different capabilities and limitations based on their education, backgrounds, working environments, stress level, medical condition, etc. Understand who your product's users will be and design the product with a view from their shoes."

7. Do a "vertical slice."

"Prioritize the design implementation and then analyze, design, build, and test the most important parts of the system first and let those results define constraints for the less important parts."

8. Think about what can go wrong.

"Design the product to cope (i.e., be fault tolerant) when things aren't perfect. For example, think about dimensional tolerances, loose fasteners or connectors, power failure, use errors. Use risk analysis tools to manage this process."

9. Test what you can as soon as you can.

"Test as much of each design element as possible before integrating with the others. And if you find yourself iterating on the design to fix problems, don't try to change more than one thing at a time."

10. Finally, Murphy's law always applies.

"More than ever in the product design process. Be sure to plan for extra time and budget to get things built and tested. Nothing is as easy as it looks, and everything takes longer than you think it should."

Rebecca Herold, owner and CEO of Rebecca Herold & Associates, added an 11th point on top of Claude's list of 10: Make sure to include effective information security and privacy controls.

Is this a good list? Is there something that should also be included? Feel free to chime in on the LinkedIn discussion.

Chris Newmarker is senior editor of MPMN and Qmed. Follow him on Twitter at @newmarker.