Category Archives: Validation

Basic Components of a Test Form

Hello good people of the world! Today’s post is a short one on what are the basic components of a test form. In Validation, you’re going to record a lot of data, and you want this data to be well organized and easily understood. Here are the basic components I think every test form should have:

  1. Numbering! Each step should have a unique number so that it is easily identifiable and easy to reference elsewhere.
  2. A Title! What is this test all about? A short description should be provide.
  3. Purpose! What is the purpose of the test? Make it clear.
  4. Verification Steps! Clearly define what steps need to be performed.
  5. Expected Results! Clearly define what the expected results are. Does every step need an expected result? Every step can have one, so include it.
  6. Actual Results! This is where the actual data is collected. The actual results can be recorded exactly as the expected results are stated to avoid any confusion.
  7. Pass/Fail! Did the step pass or fail? This will quickly tell you. Also a good place to reference any comments.
  8. Initials/Date! In order for the data to be attributable, initial/date uniquely identifying the test executer must be included for each step.

What basic components do you include on your test forms? Comment below.

Like this MWV (Mike Williamson Validation) blog post? Be sure to like, share, and subscribe!

What NOT to Re-qualify

autoclave
Hello good people of the world! Today’s post is on what NOT to include in requalification/revalidation. I was recently on a site that had a five (5) year requalification requirement for sterilizers per a site SOP, which sounds reasonable (continuous monitoring would be better). But then I noted they included in their requalification requirements a re-execution of the entire initial controls system IOQ! The requalification included verification of:

  • Hardware/software installation
  • E-stop, guarding, and door interlocks
  • Restart and recovery
  • Recipe management
  • Temperature, pressure, and time control
  • Communication
  • Security

And it was expected that this would be done every five years! It just so happened that in 2014 they paid a contractor to do the work, who sadly did not help the site out by letting them know the wastefulness of such an endeavor. This is an egregious example of resource misuse and not understanding the expectations of a validation program/taking a risk-based approach. The point of requalification/revalidation is to look for drift in processes, not blindly repeat testing already performed.

What misunderstanding-of-validation-expectation horror stories do you have? Comment below.

Like this MWV (Mike Williamson Validation) blog post? Be sure to like, share, and subscribe!

Serialization Basics

vials
Hello good people of the world! Today’s post is high-level regarding serialization. Serialization is a process mandated by the world’s regulatory agencies to reduce counterfeit drug products in the market. Besides being costly to drug companies, counterfeit drug products are often less efficacious and less safe than the real drug they are purporting to be. Additionally, counterfeit drug products can be contaminated with other APIs and/or toxic excipients.
Continue reading Serialization Basics

Validation of Legacy Processes and Equipment

abandoned factory
Hello good people of the world! Today’s post is about dealing with legacy processes and equipment from a validation perspective.

I’ve worked in two plants that have legacy processes/equipment. By legacy, I mean processes/equipment that can be decades old. These processes have been producing products for the market for a long time. Initial validations will not meet the standards of today, if they exist at all. It is typical that documentation for legacy processes/equipment will not meet the today’s standards as well, and may also be nonexistent.

So how do you handle legacy processes/equipment from a validation point-of-view? First it is important to consider that because the product has been on the market for some time, there is a proven track-record. This must be leveraged. Also, legacy processes have been audited internally and externally many times. It is important to look at those audits to see what gaps and/or risks have been identified.

The next challenge is how to document changes to legacy processes/equipment. It can be difficult to track changes to equipment that, for example, does not have up-to-date P&IDs (Piping and Instrumentation Diagrams). Because creating missing documentation after installation can have a large resource cost, it will be up to the organization to determine their documentation needs and what risks they are willing to assume. This can lead to challenging conversations within the organization and unique validation programs.

What have you learned working with legacy processes/equipment? Comment below.

Like this MWV (Mike Williamson Validation) blog post? Be sure to like, share, and subscribe!

Warning Letter: Daito Kasei (Japan)

WarningLetterFDA

Hello good people of the world! Today’s post is about a recent (January 2018) warning letter issued by the FDA to a Japanese cosmetics and drug manufacturing company called Daito Kasei. The warning letter, which can be found here, underscores the need maintain focus on data integrity requirements in our modern age. This warning letter includes explicitly the following data integrity remediation steps: Continue reading Warning Letter: Daito Kasei (Japan)

Start-Up Tip: Task Duration Buy-In

Buy In.jpg

Hello good people of the world! Today’s post might be a bit of a rant, but this is something I’m seeing lately and is a good way to kill your project schedule. What I’m talking about is task duration buy-in. It is critical in a start-up project to get all stakeholders and especially responsible parties to buy in to the amount of time they are allotted for each task in the schedule.

And I mean formal buy in. Get it in writing. Have them sign their name to it.

Too often I have seen a schedule suffer due to this gross oversight. Yes, a schedule may say you have three days to complete a task, but if that duration feels arbitrary or not “owned”, it is not going to be met when other priorities come in to play. This is especially true with resources that are not used to working in start-up environments/come from steady-state situations.

Get buy in, have consequences for failing to meet dates, force resources to delegate when they are overwhelmed.

What’s your go-to method for getting buy-in on task durations and keeping that schedule moving along?

Like this MWV (Mike Williamson Validation) blog post? Be sure to like, share, and subscribe!