Equipment 21 CFR 820.70(g) - User Requirements Document for Off the shelf equipment

[PeterLe]

Hello everyone.

Quick question regarding validation of equipment for manufacturing processes. Is a user requirements documents for purchasing off-the-shelf equipment required per Equipment 21 CFR 820.70(g)? I'm trying to read the regulation to see if it is a requirement or not.

Right now, devise IQ/OQ/PQ protocols based on what specifications our manufacturing process DOPs state. Rather than a user requirements documents. That's how we qualify and ensures the equipment "meets specified requirements and is appropriately designed, constructed, placed and installed to facilate maintenance, adjustment, cleaning and use."

Any advice would be greatly appreciated. Thank you!

 

[Tidge]

Quick question regarding validation of equipment for manufacturing processes. Is a user requirements documents for purchasing off-the-shelf equipment required per Equipment 21 CFR 820.70(g)? I'm trying to read the regulation to see if it is a requirement or not.

Short answer: There is no requirement to have a user requirements doc prior to the purchase of equipment for use in manufacturing (1). A medical device manufacturer will have to have a set of requirements (established by the company with the manufacturing floor) for the equipment in order to do meaningful OQs and PQs. The off-the-shelf documentation's most practical use will be to support the installation and maintenance of the equipment.

Don't forget that for questions about 21 CFR 820 you can consult the Preamble. I believe comment 132 (obliquely) speaks to your question, but it is necessary to recognize that a manufacturer that purchases OTS equipment isn't designing that equipment (and has no control over requirements for the equipment) but is instead designing the process (over which they have full control of requirements).

(1) As a practical matter, you will want to have some idea of the functionality and capability of equipment prior to purchase. No matter how detailed the requirements document (for OTS equipment) and the OTS manuals it is highly unlikely that such things would be found to be sufficient to replace medical device manufacturer's qualifications... so my advice is don't worry about generating user requirements prior to purchase (for regulatory reasons).

 

[PeterLe]

Thank you so much for the info, Tidge! This is super helpful.

Yes, I was concerned about having a user requirement document for regulatory reasons.

When authoring your qualification documents (IQ/OQ/PQ), wouldn't it be necessary to have some User Requirement document to test against to see if the equipment meets pre-defined requirements such as:

Ease of Use
Environmental Conditions
Foot print requirements

Or can those pre-defined requirements be specified in our Manufacturing Process documents or specifications/release criteria?

Again, super appreciate your help on this!

 

[Tidge]

Let's see if I can help... Note that this response in in the context of medical device manufacturing process validation, and NOT pharmaceutical manufacturing process validation, nor product design validation.

When authoring your qualification documents (IQ/OQ/PQ), wouldn't it be necessary to have some User Requirement document to test against to see if the equipment meets pre-defined requirements such as:

Ease of Use
Environmental Conditions
Foot print requirements

Or can those pre-defined requirements be specified in our Manufacturing Process documents or specifications/release criteria?

One approach to validation documentation is to view the protocols as the mechanism by which a manufacturer establishes belief that the process is established as intended, and when executed the protocols contain the objective evidence that the process is working as intended. If you approach process validation in this way, you will likely consider certain things in the appropriate protocol. For example:

IQ (Installation Qualifications) should reference everything necessary to install (and maintain) the equipment at a given spot on the shop floor, such that if the time ever comes to move it... or duplicate it, at the same or other spot (could be in a different country) all of the information exists, along with a checklist of the necessary items to verify. This could be the footprint (for fixed or otherwise immobile EQ), the utility connections (e.g. electricity, compressed air, de-ionized water) as well as plans for waste disposal (old solder, noxious gasses). Consideration of environmental conditions are most likely part of the IQ, with some exceptions such as those related to product sterility requirements (which will get formally challenged in OQ and PQ). A good IQ is one that could be sent along with the equipment to a building on the other side of the planet and be executed in such a way that a person would have confidence that the EQ was installed correctly. If these things are done, a medical device manufacturer will have addressed the necessary elements of 21 CFR 820.70 (g). The IQ should also cover the elements of 820.72 for test and measurement equipment that is part of a process.

It would be unusual for an auditor to question a thorough IQ, but experience among auditors varies and manufacturers should be prepared for 'red-face' questions. For example, an auditor might ask if a PC system has a AC line conditioner, and if it does, where is it specified and documented? If you walk onto the floor and have a hard time maneuvering around a piece of equipment, the auditor might ask about floor space considerations. I can't think of an audit (involving manufacturing) where calibration records were not requested.

While there can be stumbles with the elements of 820.70 (often around "low hanging fruit" like change control and contamination control), it is more typical that manufacturers struggle with are the elements of 820.75 Process Validation. Process Validation requires critical thinking in the areas of process requirements and methods to challenge the process. My personal experience is that it is very rare for an auditor to examine an OQ or PQ with anything but a cursory examination. I can politely ascribe this behavior to the idea that a poorly validated manufacturing process will reveal itself through nonconformances and corrective actions. I can impolitely ascribe such behavior to thoughts like "it's hard, and do I really want to have a shop manager try to explain his vision of both Ohm's law and Study Designs to me?"

The manufacturing process parameters will be established as part of the product development process. The most effective product launches consider manufacturing early in the product development life-cycle, but many manufacturers delay consideration to a specific "design transfer" phase that fits between design validation and product manufacturing. The process parameters are established, and then the OQ and PQ protocols for the manufacturing processes are developed from the established process parameters. The bolded text above is the direct mention in 820.75 (b) and by choosing equipment for your process you end up with some specific equipment parameters in 820.70 (g).

I'll stop here, because you haven't asked any questions about OQ/PQ.

 

[PeterLe]

The manufacturing process parameters will be established as part of the product development process. The most effective product launches consider manufacturing early in the product development life-cycle, but many manufacturers delay consideration to a specific "design transfer" phase that fits between design validation and product manufacturing. The process parameters are established, and then the OQ and PQ protocols for the manufacturing processes are developed from the established process parameters. The bolded text above is the direct mention in 820.75 (b) and by choosing equipment for your process you end up with some specific equipment parameters in 820.70 (g).

I'll stop here, because you haven't asked any questions about OQ/PQ.

First of all, Tidge. Thank you so much for all this information. I'm still relatively new to Quality Engineering, so this excellent reading for me! I much appreciate this!

The bolded and italicized from above "The manufacturing process parameters will be established as part of the product development process." --This is how my original question started with our company since the documentation of the process parameters for this particular equipment is nowhere to be found. Just to give background on our specific situation, we've been using an older equipment that has not been validated as it was introduced to the manufacturing line 15-20 years ago. Its been kind grandfathered in our process. We do have established process parameters but there is no documentation on how we came up with those parameters. Now, the older equipment is now obsolete and we are upgrading to the latest model. This is how the discussion of a User Requirements Documents came up and now i'm here. Since the engineers who developed the process parameters 15-20 years ago are no longer here, it would be a difficult task to reverse engineer how they came up with the process parameters. So the idea of a User Requirement document to at least establish, on a high level, the items manufacturing wants in the new equipment. We can then use the User Requirements document as a guideline for our IQ/OQ/PQ.

Since we are on the subject of OQ/PQ. This is a test equipment for in-line testing and there has been discussion on whether a PQ is required or not. Since it is not a part of the actual manufacturing of product, is a PQ required?

Again, thank you Tidge for all your help! This has been enormously helpful!

 

[Tidge]

It is possible to develop requirements for legacy process requirements, but even the most 'shallow' approach ought to include representatives from responsible product design team (someone must be responsible for the DHF) and manufacturing. Together they ought to be able to:

1. document what the process requirements are,
2. assess the suitability of the requirements (including: are the requirements correct? complete?)

These requirements can certainly inform the purchase of new equipment, but where these are really used is in the OQ and establishing the parameters to be followed in the process work instructions.The situation you describe isn't all that unusual; many medical device manufacturers find themselves doing this sort of investigation when it comes time to replace equipment.

Keep in mind this nugget about 'legacy systems': for the FDA "20 years ago" doesn't mean the same thing it meant 20 years ago.

Since we are on the subject of OQ/PQ. This is a test equipment for in-line testing and there has been discussion on whether a PQ is required or not. Since it is not a part of the actual manufacturing of product, is a PQ required?

I suppose it is possible to (within your process validation plan) to rationalize not doing a PQ, but I'd be skeptical. For medical device manufacturing process validation (not pharma, their vocabulary is slightly different) the PQ includes running the process with the established parameters for typical lot sizes. Assuming that you have a validated test method (again, this means something slightly different than process validation) I would still expect that the equipment used in the test method behaves as expected under typical processes. A PQ can provide evidence that this expectation is met.

Just to provide some off-the-cuff examples where a PQ might turn up an issue when using test equipment:

• A test pin might wear out after so many uses (this could result in items at the end of the lot have an harder/easier time passing!)
• A calibrated caliper might not be the correct size for all dimensions to be checked
• The log used to record test results might not have enough space to record all the entries of a typical lots size

 

[PeterLe]

It is possible to develop requirements for legacy process requirements, but even the most 'shallow' approach ought to include representatives from responsible product design team (someone must be responsible for the DHF) and manufacturing. Together they ought to be able to:

1. document what the process requirements are,
2. assess the suitability of the requirements (including: are the requirements correct? complete?)

These requirements can certainly inform the purchase of new equipment, but where these are really used is in the OQ and establishing the parameters to be followed in the process work instructions.The situation you describe isn't all that unusual; many medical device manufacturers find themselves doing this sort of investigation when it comes time to replace equipment.

Keep in mind this nugget about 'legacy systems': for the FDA "20 years ago" doesn't mean the same thing it meant 20 years ago.

This makes sense. Our company definitely has gaps and the search in the DHF didn't provide any evidence of how these process parameters were developed. We do have a company wide Quality project to find gaps in our system and fix them and documentation of process parameters is definitely a part of that.

We as a team, may have to come up with a plan on how to address documentation of the process parameters.

I suppose it is possible to (within your process validation plan) to rationalize not doing a PQ, but I'd be skeptical. For medical device manufacturing process validation (not pharma, their vocabulary is slightly different) the PQ includes running the process with the established parameters for typical lot sizes. Assuming that you have a validated test method (again, this means something slightly different than process validation) I would still expect that the equipment used in the test method behaves as expected under typical processes. A PQ can provide evidence that this expectation is met.

Just to provide some off-the-cuff examples where a PQ might turn up an issue when using test equipment:

• A test pin might wear out after so many uses (this could result in items at the end of the lot have an harder/easier time passing!)
• A calibrated caliper might not be the correct size for all dimensions to be checked
• The log used to record test results might not have enough space to record all the entries of a typical lots size

This is great information and I appreciate it this. We are going to conduct test method validation on this but again we rationalized not needing a PQ because it does not manufacture product. I see your point regarding doing a PQ as we would conduct a PQ under normal manufacturing conditions with typical lot sizes.

Historically at our company, whenever we do a PQ, it is with final product against our release criteria. But are you saying if we were to run a PQ for this equipment, we would just run it under normal manufacturing conditions and typical lots sizes? For example if a lot size is 10,000 and we do a 2% sampling, then we would do that 3 times (Our normal PQ tests 3 different lots)? Then passing criteria would be what the equipment actually tests but not on final product?

Again, thank you Tidge for the knowledge! This is awesome!

 

[Tidge]

I think in the case you describe (test equipment being applied to 2% of a lot of 10K) it can probably be rationalized that you don't actually need 200 pieces from 3 different lots. The PQ is more about the expected outcome of the process and less about any specific piece of equipment. I suppose you can think about the PQ plans and results as a pre-cursor to process control charts in the manufacturing life-cycle.