From the Editor: Mistake Proofing Pharma

How can pharma reach the “Desired State” when so many of its workplaces still encourage technicians and operators to make mistakes?

Remember that 2006 benchmarking report by Jeffrey Macher and Jackson Nickerson, which pegged drug manufacturing waste as over $50 billion per year? One of its key messages was that process complexity breeds waste, noncompliance and quality problems.

Drug making has, arguably, some of the most complex manufacturing processes around. Take vaccine manufacturing. Many of the compliance issues that FDA has cited in recent 483’s stem entirely from insufficient error-proofing, and lack of training. Why is it that more pharma “Lean Six Sigma” programs are not tackling error-proofing, or, if you prefer the sound of the Japanese, poka yoke (PO ka YO kay) as defined by Shigeo Shingo in the 1960’s? He had first called it “foolproofing,” or baka yoke, but a worker found the term offensive.

The key to success in stopping or preventing errors appears to be putting yourself in an operator’s shoes in the gemba (Japanese for “the place where truth is to be found”), or the manufacturing or laboratory floor. Are enough pharma managers walking that floor? Maybe not. Most respondents to this year’s operational excellence (OpEx) survey, results here, agreed that senior or middle managers were the main obstacles to improving processes and quality at their facilities. Search the Internet for examples of poka yoke, and you’ll find many in discrete manufacturing.

There aren’t many published examples for drug manufacturing. For those, you’ll have to work back from FDA 483’s to get some idea. However, change may be coming. This year’s OpEx survey suggests that error-proofing and 5- or 6-S efforts have become important in more drug company strategies than they had been in past surveys.

Lean and OpEx consultant Martin Hinckley suggests that there’s still room for improvement,though. In this article, he describes a recent visit to a drug company client that uncovered over 200 opportunities to reduce error. The facility had only five error proofing devices and procedures in place, but a list of opportunities that reads like a litany of “what not to do.”

Some examples: Controls for two digital scales, located at left and right corners of a room, were placed one above the other at the center of one wall. Controls for three different sources of water, each maintained at a different temperature, were identical and adjacent to each other---even though adding water of the wrong temperature could ruin a batch. Multi-tasking operators were expected to turn off water manually once the correct amount is in the tank, without any alarms or alerts----much less an automatic shut off valve.

At a time when sensors are smarter, less expensive and more plentiful than ever, can’t they be applied to prevent these types of situations from occurring? Ironically, Hinckley says, the typical response to errors is to institute more administrative controls to prevent errors, which only generate more errors!

The Japanese concept of Jidoka calls for empowering the operator by using simple devices, visual and auditory, that will allow him or her to stop a process whenever a quality problems crop up---before they can destroy a batch, or an entire lot, of product. Process Analytical Technologies (PAT), which, like error proofing, ranked higher on more survey respondents’ OpEx priority lists this year, represents an important way to reach Jidoka. But there are also simpler “visual” ways to avoid major quality problems that stem from error.

Have you examined all of the possibilities at your workplace? Error-proofing’s not glamorous. In an industry full of Ph.D’s, it may seem about as interesting as bottle washing, but today’s errors often become tomorrow’s product recalls or quality problems. Please write in and let us know how you’re battling error at your facility.

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments