So Lucy (@wizardofobjects) and me were finalists in the Saving Lives at Birth funding competition. We were one of the top 50 out of 750 applicants, which is pretty alright, and were invited to a pitch competition/conference in DC last week. We’re currently working on a blog post for the CDP website which should be out later this week, which will dive into much greater depths about the content and takeaways from the event. In the mean time, you get to read conjecture and personal opinion that isn’t appropriate for an organization’s website. The leftovers, if you will, of what keeps me up at night.

DSC00164 (2)

Part of this meeting was what they called the DevelopmentXChange conference, which gives attendees networking time and workshops on various topics related to scaling global health technology, which despite losing out on funding, made the trip somewhat worthwhile but also changed my perspectives on a few things.


I had just finished hosting an informal session on Risk and Ethics associated with Global Health Technology, so I had risk management and risk acceptance fresh on my mind when I walked into the session pictured above. I felt I HAD to attend since I’ve been eating and sleeping medical device regulation for the past year and I was curious to see where my opinions lined up with others in the room. Spoiler alert: They didn’t.

The speaker had finished remarking on the importance of recognizing the differences between a 510(K) and CE marked product when I made what I would immediately recognize as an under-calculated remark in a room full of people, who weren’t necessarily engineers, with vested financial interests in their global health innovations.

“I appreciate your differentiation between 510K and CE Marks, and their impact on whether they indicate that your device is safe and effective for use in a specific market. I’m of the opinion that they should have little to no impact on whether your device is marketable in a low resource setting.”

Cue someone who won’t be named associated with a medical device being marketed in both America and certain countries in Africa grabbing the mic and lecturing me on consensus standards for what seemed like ages in a room of 40 people (Yes, I had to lick some wounds following that encounter). She’s absolutely not wrong in this instance and I should have clarified that the class and risk profile of your medical device should play a large role in the decision to market in LRS. But I disagree with her assessment that I was completely out to lunch(obviously. Or else I’d have nothing to write about) and that’s what’s keeping me up at night.

The idea of consensus standards and regulatory approval is a double-edged sword. On one hand, obtaining a CE mark or a 510(K) demonstrates that your device is safe and effective for use in a specific market. Hence why a 510(K) doesn’t have jurisdiction in Europe and a CE mark means diddly-squat in America. A group of people somewhere have decided that there’s enough cultural, demographic, and healthcare environment differences between the two to decide that their regulatory pathways will have different demands. So pardon my naivety, but if you have a CE mark for Europe, why should that apply in Senegal? (Again with the inflammatory, flippant remarks! And yes I’m aware of the GHTF, NMRO, ADNI, ARSO etc. This article can be considered to apply to a broader audience)

Getting to the point, the crux of what keeps me up at night has to do with my favourite of medical device issues – Risk Acceptance. Anyone working with medical devices will be familiar with ISO 14971 – Application of Risk Management to Medical Devices, and how it takes a semi-quantitative approach to quantifying what makes for an acceptable risk. ISO 14971 forms a critical component of both CE marking and 510(K) applications, the risk management plan, where the manufacturer identifies and controls the risks associated with their medical device from design through to manufacture and post-market surveillance.


The ISO consensus standard empowers the manufacturer to define what an acceptable level of risk is (although deviations from the above table towards something less conservative would be at your own regulatory peril), by attempting to rank the severity of the harm and the probability that that harm may occur. And now is where I make my point:

I take issue with the idea that the risk acceptance criteria of a medical device manufacturer in the developed world would be the same as that of a healthcare professional in a low resource setting, someone who’s obviously willing to accept a much greater level of risk for the same level of benefit. That manufacturer needs to accept the realities of where they’re shipping their device and TRULY understands the use environment they’re entering.

Without even belabouring the point of dumping of developed world medical devices in low resource settings, devices that are apparently safe and effective for use in these settings, there’s an even more egregious example of abuse: Single Use Medical Devices.

Reliable sterilization remains a challenge in low resource settings, with many healthcare settings utilizing a bleach cleaning program in lieu of a functioning autoclave. This is far from perfect and reusing a device that hasn’t been properly cleaned can lead to fairly severe contamination events. In the developed world, you can eliminate the risk of contamination between patients by making the device single-use, throwing it away when you’re done with the patient. But in settings where reliable distribution networks are hard to come by and the difference between using a contaminated device and not using it could result in death or disability, the choice to reuse becomes pretty clear. (Autoclaving non-autoclave compatible devices also applies here).

So when I speak with the manufacturer of a rather elaborate SL@B, WHO, PATH, and GCC endorsed medical device (again, not taking shots here) what their plans are for cleaning and sterilization are, and their response is “Oh we’ll make it single use to avoid those issues”, you’ve completely missed the fucking point. People are going to reuse your device because they’re willing to accept a greater risk for the same level of benefit than what you’ve accounted for in the design of your medical device, and they’re going to reuse it until it fall apart.

Design it for autoclavability. Design it for bleach cleaning. Design it for reuse in the environment you’re actually selling it in, not some swanky lab in Switzerland.

Don’t do the Ostrich thing and assume that you’re willing to accept the same level of risk as everyone else.