The perception of risky process changes

Now that we have an approved device we're spending at least some time on process improvement instead of all out research and development. Unfortunately, even though everything is fairly new some of it is already difficult to change.

We have a piece of test equipment that displays pass or fail, prints out a page of results, and writes them to a database. The operator runs the test, marks pass or fail in the traveler, then staples the results to it. During a sort of related process change I suggested we take out the print out of the results. This was met with quite the uproar and doubt about how we could ever do it.

Apparently removing a redundant step that no one ever looked at was a big deal. Engineering and complaints thought it might be useful, even though they never used the printouts two years in- but someday they might. The change would save some minimal amount of money on supplies, transferring of materials into and out of the clean room and storage, along with associated labor. I was able to finally make headway when I pointed out that we do plenty of visual inspections that do not have another record and in this case we still had a record in the database. Everyone initially thought it sounded like a risky change based and we could be out of compliance, but it was all our own perception.

While this isn't the most impactful example, I think I may have learned the value of taking time up front to set this all up efficiently, but I'm not sure I'll be able to convince anyone to add extra time for these activities the next time around. "We can take care of that stuff during validation!"

Annual review of received parts' dimensions

Many companies perform an annual review of every dimension or 100% inspection on all received parts. While this may serve as a halfway decent preventive action, I'm not convinced it is worth it. If you have automated inspection methods, maybe then it is not too much effort, but we're still using hand tools or manual adjustments for the majority of our parts.

You could argue that you have a critical device, we have to check for dimension drift. Well why aren't you checking for that anyway? And if it is so critical shouldn't you have insured that the supplier's process was capable before you even signed up with them? If it is truly critical and you're not checking it but once a year, good luck with that.

Additionally, if you do find something out of specification, nine times out of ten it is going to be on a non-critical dimension, or else you would have caught it in receiving inspection or manufacturing anyway. Once you find the out of spec item, you have to go back to your inventory, check it, then justify why it is either okay and change the specification, or send it back, and jump through hoops to justify why it was okay to use on the ones sold, but not okay to use going forward, or whatever other contortions QA wants.

In my experience these generally boil down to unclear or unnecessary specifications or measurement issues, where to start measuring a curve, a hole over specified, etc. Another argument for doing it right the first time and checking your supplier's process.

However, this if often a difficult battle, as new parts are generally part of R&D efforts and the team from R&D is behind schedule and they always try to make up schedule time by trying to get either quality or manufacturing to compromise on one issue or another (I *may* have pointed this out at a meeting once). If this happens and you don't have time to ensure supplier process capability, it is not the worse thing in the world, you can always blame the vendor and shop around for a new one that has no familiarity with the part and uses the same process that can't meet the specification. Sometimes the best you can do is make the risks clear.

Things I never thought I'd have to worry about

A volcano in Iceland slowing shipments to and from Europe, since we ship everything by plane now. Inventory costs money but so do lost sales.

The longshoreman strike on the west coast about five years ago also slowed me down, back when we shipped by boat. Then it was product development and not production so it wasn't as urgent. We just moved to another piece of the project.

Everyone loves the idea of second sources, but small companies don't have time. Each new supplier of a custom part generally requires dozens of hours of training. "If you don't take that burr off, its not going to meet the specification. Please mask those holes next time like it clearly states on the drawing, etc." I can only improve the products or the business at once, take your pick. Improving products improves the business, although that is no good if you're out of business.

So you pick critical parts and go from there. A machine shop, you can get up to speed fairly quickly if you spend the time. Extruded tubing, not so fast, there is a 12 week material lead time, better plan for that. And even better is that some regulators want to be notified of a supplier change or even an alternate. We also implemented a finance check based on the current economy, but that has been a waste of time, our suppliers should do a finance check on us!

Pot and Medical Device Manufacturing

California is set to vote on marijuana legalization later this year. I'm pretty much a live and let live guy, but how is this going to effect California's fairly large drug and medical device industry? I'm sure people will debate this, but lets cut right to how companies will see it:

In California you will have a much higher chance of someone stoned building your product and on average stoned people make worse decisions than non stoned people.

This probably doesn't matter if you're renting movies at BlockBuster, but small mistakes in medical devices can cost lives and / or millions of dollars. Right now most companies mitigate the risk of drug use by screening on hire and periodic screening. How would you mitigate in the future, or is my fear overblown? I don't think legalizing pot will help manufacturing in California, obviously that is not the only consideration, but an important one I think.

Finding Owners

More and more I'm learning one of the keys to making good devices that meet quality and regulatory standards is finding people who will take ownership. Lately I've been working in the packaging area a lot and almost every time I'm in there I find discarded labels without lines drawn through them. Of course these are always discarded by the other shift. Everyone already knows the labels need lined out when questioned, so the basic training is there.

If you're an owner, you walk into your area and you own it top to bottom. You don't not fix things because you weren't around. You correct the issue and take it up with the supervisor on shift change. These are the types of people you want, and why many large medical device companies pay way over what they could for manufacturing labor. Pointing out the same non conformance over and over to the same people is ridiculous.

I've been involved with contract manufacturing in China and they were excellent at following instructions. They picked up GDP about 3 times faster than anyone else and have maintained it. This is not to say everything is perfect, but the basics were taken care of. Some of the more complicated situations that came up with the contract manufacturer were really botched up before we were notified, but at least I know the easy stuff is covered.

I'm not sure that me telling the 19 year old putting labels on boxes that this is the era of global competition and she has to work smarter and harder than they are really sinks in. Hopefully, we have a chance to develop the skills necessary because you only get so many chances.

Labeling for Reuse of Single Use Medical Devices

Another relatively new EU requirement effective March 21, 2010 per the revised Medical Device Directive (MDD) (93/42/EEC) M5 is the following:

Where appropriate, the instructions for use must contain the following particulars:
...
If the device bears an indication that the device is for single use, information on known characteristics and technical factors known to the manufacturer that could pose a risk if the device were to be re-used. If in accordance with Section 13.1 no instructions for use are needed, the information must be made available to the user upon request;

Now it is not good enough (in the EU anyway) to just put single use device in your instructions for use (IFU), but you also must identify the risks associated with reuse (sepsis/infection is probably common). I'm sure these are already documented in your risk analysis already, just cut and paste them into your IFU. Apparently anyone already using the device off label by reusing it will be deterred by the listing of potential risks, a new mitigation is born.

This seems like a potentially slippery slope to me, why not have every warning list potential risks if not followed? Is it okay to exceed the shelf life? After all, there are no risks listed- how bad could it be? If it was really bad wouldn't they list more than just a rule like they do with reuse?

Anyway, it is not hard to comply with this requirement and we like to sell stuff in Europe, so we're in.

See my previous post on the new MDD phthalates labeling requirement.

Phthalate labeling in medical devices

Hello my friends, I am a little late, but a relatively new EU requirement per the revised Medical Device Directive (MDD) (93/42/EEC) is the identification of phthalates used in some common medical devices:

If parts of a device (or a device itself) intended to administer and/or remove medicines, body liquids or other substances to or from the body, or devices intended for transport and storage of such body fluids or substances, contain phthalates which are classified as carcinogenic, mutagenic or toxic to reproduction, of category 1 or 2, in accordance with Annex I to Directive 67/548/EEC, these devices must be labelled on the device itself and/or on the packaging for each unit or, where appropriate, on the sales packaging as a device containing phthalates.

This update was required by March 21, 2010 and your certificate of conformance was probably signed on that date. The main use of a phthalates in medical devices is DEHP in PVC. There are no limitations imposed, but labeling is required (I suggested "Portions of this device contain phthalates").

Since we're involved with tubing and IV bags, we are knee deep in PVC with DEHP. When we started design we had considered trying to go DEHP free, but found the DEHP free PVC alternatives not as robust and as customizable as we wanted. A review of the product risk analysis clearly showed that the risks of a IV bag or tubing leak were many times greater than even long term phthalate exposure (which is now mitigated with labeling...). Maybe the materials will improve in the near future, but right now it doesn't look like it warrants a separate project to go DEHP free, new materials can be worked in as part of other projects as they arise.

I know there was ample warning, but it caught us a bit off guard, luckily we do not have a large inventory to manage or relabel. If there is one lesson I have certainly learned over the past year, it is inventory makes you feel good, but we keep getting burned with costly rework orders due to various changes.