Which Drugmaker Fails Most FDA Inspections?

do you have a link, Ed?

Hi Harpy,

No, a link will not become available for another 36 hours or so due to the way Bloomberg distributed this story. I have the original version and condensed what stood out as the most pertinent findings. I was provided a graphic, as well.

That said, I believe a link will become available in a day or so and I will link back at that time. Sorry for the inconvenience but I don't have control.

Best ed

ok, thanks!

The article makes one think, what is going into our bodies that might kill us? Very scary!

Also looking forward to link.

It would be interesting to hear precisely what Burgess thinks is "too much paperwork" and not relevant to safety and GMP. Or whether, as I suspect, it's the usual anti-FDA horse hockey.

Abbott is Burgess's top campaign contributor. Pharma and med organizations (mostly surgeons--read, devices) are the top two industries. They don't call it energy and commerce for nothing.

You may find it interesting to compare the chart in the post against the latest chart made available of all warning letters issued, dated DEC 2008, between FDA FY 2004 thru 2008.

http://mgdservices.com/wp-content/uploads/2010/12/fdacloud-graph1.jpg

During the FDA FY 2010, the total number of warning letters posted on their website is currently 524. Please keep in mind that is "posted" as opposed to issued. It also covers ALL warning letters by all Centers, so it goes beyond CDER and CBER.

This is the kind of information that the average consumer should know. What manufacturers are trustworthy, capable and consistent? Those are the ones I want to keep in business. The others should die on the vine like every other poor performer in marketplace.

Can't speak for manufacturing, but in 25 years of monitoring clinical trials and hundreds of investigators, none ever received a Form 483. Why? Because whenever I ran a multicenter trial I insisted that our QA/QC department audit every one of my investigators that was likely to be the subject of an FDA audit. Problems were fixed, and nothing had to be "sanitized", which is a topic for another discussion. My self-inflicted misery with our auditors prevented much larger headaches down the road with FDA.

It will be interesting to see whether or not this article has any impact on share price of any the companies who are mentioned in the article. It goes without saying that J& J has already taken a hit and must be spending a lot of money and expending vast resources on their legal issues. Wouldn;t it be great to see legal fees for compliance issues trend/tracked as an area of potential cost savings/cost cutting?

Thanks to John for graph. As I recall from other charts about warning letters specifically for advertising, the number started going down in late Clinton and bottomed out through 2008. Started to climb again with new administration.

Insider--what you did is the answer.

p.s. Sympathetic as I am to Jaynesday's sentiment, I wonder if such consumer forces could ever impact a company. So many other factors insure pts using, and staying on, a particular med--what they've been using, what doc recommends, what insurance covers, and so on. And, of course, the average person (or HCP) will never hear this story or ignore it.

Good news that the big generic Mylan did relatively well...

Your numbers and the graphic don't add up. The graphic shows 1032 (558 + 474) plant inspections in 2010, or about 0.4 inspections per plant if there are 2567 plants. Also the graphic shows that the most inspections were done in 2005 which was during the presidency of George W. Bush.

Does anyone know how an inspection failure was defined? Was the issuance of a single 483 observation considered a failure?

Hi Alig,

Thanks for the note. As I indicated, the material was generated by Bloomberg, in this case, which explains the attributions.

That said, I believe the number of inspections per plant each year was averaged during the different presidencies - eight years for Bush and two for Obama. It's not a comparison of one year to another. The 0.4 that you cite is only for one year.

As to the other point, some of the 2,567 plants were never inspected, so Bloomberg did not include those in the annual rates, from what I was told. I hope this is helpful.

Regards Ed

Justice - if you click on my name next to this comment, you will find a second chart (labeled Chart 2) that may be the one you remember? It is another FDA ORA summary that covers all letters issued from 1996 through 2004 - the year 2000 is the peak.

Thanks, John. Yes, that is essentially what I have, with the 2004 trend continuing through 2008.

Ed,

Can you please define what is a “failure” in inspection ie is it a 483, a Warning Letter, a non-close out with intent to re-inspect etc etc?

Hi Marcel,

The answer to your question is.... 483s. I checked with Bloomberg, which notes the FDA did not disclose the nature of the violations found or subsequent status.

Hope this helps, ed

Many thanks!

Ed,

I feel the terminology used in the article is a bit misleading. Equating an issuance of a 483 as a "Failure" grade might not be the most accurate statement. A 483 citation is a "significant, inspectional observation" made by an FDA Investigator and presented on a written Form 483 as required by the Act. And, although an issuance of a 483 is not a positive thing, it does not neccesarily mean the manufacturing site is out of a state of compliance. It means that a significant observation was made during the inspection and should be addressed immediately to remain in compliance with cGMPs.

It's a positive sign that the FDA has issued more 483's in line with Dr. Hamburg's push for greater enforcement. Greater enforcement has a ripple effect across the industry, and manufactuers are more likely to invest in technologies which will prevent them from getting a 483. -Al

Skimming the FDA's Enforcement Story Archive it seems a failed inspection must involve far less than a Warning Letter: Year WLs Inspections 97 1,140 15,506 98 905 18,185 99 900 16,920 00 1,154 15,146 01 1,032 18,649 02 755 18,572 03 582 22,543 04 725 21,805 05 535 19,803 06 538 17,641 07 471 15,581 08 445 15,245

I have been in the business for over 20 years and I find that a 483 finding is NOT a "failure". Very often, they are observations based upon a particular inspector's opinion or could be something relatively trivial, such as failure to sign and date an entry. I believe the true score of failure to follow GMPs is(1) a warning letter related to GMP and (2) recalls related to GMP failure. That's where the failures are!

Didn't read Al's comment before posting mine. His is more eloquent.

-GR

Hi Al and George R,

Thanks for the notes and I understand your point. A couple of things in reply...

First, as noted, this story was reported by Bloomberg News. Like newspapers that run 'wire stories' from the Associated Press, for instance, I sometimes also run an item that has been generated by other media. Of course, I am free to edit as I see fit, but do so only if I determine a need to add, subtract or change the copy, as the case may be. In this instance, I condensed what was a longer story but left intact the descriptive language.

As to the specific notion of failure, I follow that a 483 does not mean that a plant is necessarily out of compliance. On the other hand, the plant did not pass the inspection either. As in school, there are degrees of passing, but as you note, a 483 indicates significant problems. I suppose there will be some who may say this is semantics, but to keep it simple, a plant that does not pass inspection and receives a notice of significant problems is tantamount to failing, at least as I understand the interpretation offered by Bloomberg.

Again, I understand and appreciate your thought. And I hope you can do the same.

Regards ed

Aren't 483s basically a single person's opinion? Wouldn't a similar metric be the number of times a company was named in a lawsuit? I can name you in a lawsuit for the cost of a filing fee, never mind if it dismissed outright.

Warning letters would be more analogous to convictions/judgments. They represent the official thinking of the FDA.

I think you also need to look at the nature of the observation. Did somebody forget to sign off on one cleaning form, or was somebody forging cleaning forms and not bothering to clean at all. The first is a paperwork error with little impact on patient safety. The latter is cause to send in marshals before somebody gets killed.

I'm a former FDA investigator and I can tell you that leaving a 483 at a drug company does NOT mean it "failed" the inspection. It means that "in the opinion" of the FDA investigator, a deviation from regulations or current industry practices and expectations was found. Note I mentioned "in the opinion" as many 483 observations are based on a specific investigator's interpretation of regulatory expectations. Issues are not always black and white - most are shades of gray. And all investigators are not created equal. Some are very good and some are not. So the quality of the inspection as well as the observations varies greatly.

In addition, many 483 observations are considered as less significant than others. In Europe, inspectors actually classify their observations as Critical, Major and Other (or Minor). Although FDA does not do the same, people within the agency know which observations are significant and those that are not. Those deemed significant often result in Warning Letters. So if you want to deem an inspection a "failure" or a company as "flunked", you should really consider Warning Letters or other regulatory actions (e.g. consent decrees). Calling a company that has a 483 a "failure" is inaccurate at best and smear tactic journalism at worst.

Dear Peg,

Thanks for sharing your view. And I understand the distinctions you are making. So here is what I will do...

1 - I will send this to the Bloomberg folks to see what, if any, response is made, since this was their piece. As I mentioned before, Pharmalot operates much like newspapers in that some material is generated elsewhere. I take responsibility, of course, for what is on Pharmalot, but I am in a position in this instance to approach the outlet for a reply before I determine what, if anything, to do myself.

2 - I will ask the FDA to provide, if possible, the best agency definition available to describe what a 483 represents - in English - and what, if anything, constitutes failure. In this way, we can all have the same information.

One final note, I appreciate your desire to correct the record as you see it. If Bloomberg was inaccurate, then the record can be corrected. As we all know, mistakes happen. That is one reason that the FDA issues 483 reports. What counts is how mistakes are handled.

Regards ed

Peg, with all due respect, on the clinical side CRA's whose sites get 483's have their performance scrutinized pretty carefully. Unfortunately the 483 doesn't come until WAY after the damage is done. The best and most egregious example was the FDA's continuing support of the drug Ketek after it knowingly approved an NDA containing patently falsified data, as well as its reluctance to withdraw the drug until 54 deaths from hepatotoxicity exposed its incompetence. The 483's turned about to be an after thought. The Ketek debacle, replete with 483's expost facto, as detailed in the excellent investigative piece by Kris Hundley in the St. Petersberg Times clearly illustrates the deadly combination at work at CDER on the medical side: complacency, incompetence and criminal conspiracy. I hope you represented the better side of the agency, and I hope that the agency Ketek medical reviewers are long gone. I hope that Dr. Graham is still there.

I hope that Dr. Graham is still there.

http://www.sptimes.com/2007/05/27/Worldandnation/Drug_s_chilling__path.shtml

BACK IN BLACK!

483's a "single person's opinion????". Are you effing kidding me? Do you work in the industry.

It's appalling, these stats that you've brought to light, Ed, but for those that have been following the compliance of firms over time shouldn't be surprised. Companies hide as often as they can under the cloud of "that's proprietary" and get away with it, when FDA officials should whack these people with billy clubs saying "we're the Federal Government, you nimnut, now STAND ASIDE". That went on just fine during the Bush years, and our own Feds that we pay taxes for, basically pay to allow these sort of warped games to be played in the name of "keeping appearances". Until circumstances turn into the great eye as in Lord of the Rings, when a company is SO INCREDIBLY, HAPHAZARDOUSLY non compliant, that some troops are sent with their longbows, swords, and axes. And even then, some fall to the shadows.

How can this information be disseminated to as many people as possible? I'm all for printing it out and taping it in random isles in drugstores. People need to know this.

BACK IN BLACK and as ready to raise hell as ever.

Agree with BACK IN BLACK. FDA Form 483's have ALWAYS been a big deal, unless I've been living in an alternate universe. One such investigator in the Ketek scandal, Dr. Anne Kirkman Campbell was ALEADY in Federal Prison in Kentucky (see below) by the time FDA finally got around to sending her well deserved NIDPOE letter. By my calculation, Dr. Campbell is out and on supervised release until 12/08. Let's hope she never resurfaces.

http://www.bop.gov/locations/institutions/lex/index.jsp

Do my eyes deceive me, or did someone actually agree with me?

;-)

MB

Peg,

You can certainly make a point that there are severe citations and / or findings documented on a Form 483. And the company does have a fair chance for rebuttal. But if after the dust settles, if a finding such as "flies get into Class 100 areas because there is a hole in the ceiling", you'd be hard pressed to find anyone to not call this a failure that was found during an inspection, that resulted in this failure to cause a Form 483 to be submitted at all. If a plant "passes" with flying colors, no 483. If there are valid failing observations, then yes 483.

MB

Unfortuately Graham only came to work on Ketek after the fact.

The person who tried to prevent it was Dave Ross who was forced to leave FDA.

I agree with Peg. As someone who undergoes regular inspections by the FDA at a non-sterile drug manufacturer, there are various degrees of observations and interpretations by inspectors. We can get one inspector who would look at something, find that someone missed putting a date when they signed a document and not write a 483 because it is not indicative of a system issue. Then there are other inspectors that will write up any single documentation error that is found. As the CFR is very vague at times, the interpretation of those regulations can just be as different from inspector to inspector.

Now I am no way saying that getting a 483 is a good thing. How a company responds and corrects the 483 is more indicative of the company. I would look at warning letters after inspections as a more apt way to look at "failure".

It seems to me that Peg's comments are beside the point. Specially, this point: "And all investigators are not created equal. Some are very good and some are not. So the quality of the inspection as well as the observations varies greatly."

It may be that some 483s are given on different bases than others. But what we're looking at our numbers across companies, and on an "population" level that would wash out whatever may be the idiosyncracies of particular inspectors.

Whether these violations are "serious" I do not know. That they are widespread, and not arbitray as related to a particular individual, seems beyond dispute.

I would have thought someone who actually worked with FDA would know that.

Hello Folks,

While waiting for the FDA to respond, I poked around and was reading about Field Inspection Reports on the FDA web site. Under the section entitled 'Inspection Conclusion,' I noticed the 'Corrections Indicated' definition begins with this sentence...

"Objectionable conditions and practices were found during the inspection, for which the establishment failed to meet either regulatory or administrative requirements."

I am not attempting to take this out of context, so here is the link...

http://www.fda.gov/iceci/inspections/fieldmanagementdirectives/ucm061430.htm

I mention this because the agency uses the word 'failed' to indicate a plant did not meet necessary requirements during an inspection. Given a few previous comments, I suppose some folks will maintain that this definition is not meant to imply a failed inspection, but I offer this since the conversation has been so interesting.

Regards ed

Josh,

I've worked in a non-sterile manufacturing facility. The Feds DO NOT CARE what goes on in these environments. By contrast, I've worked at a large company with multiple sterile class 100 areas, all of which I was responsible for auditing. I've been around the block in my career.

The way the FDA pursues violations in a non sterile facility is vastly different than in a sterile facility. And you're right. Usually the most serious things that make it to a Form 483 from a nonsterile facility are documentation issues, perhaps citing SOP's that are as old as the Bill of Rights, and noting that required GMP training materials are dated from 30 years ago. Yes, that happened. And it was relevant.

MB

Related to the comment from JiM - "Whether these violations are “serious” I do not know. That they are widespread, and not arbitrary as related to a particular individual, seems beyond dispute."

It is common practice in industry, especially concerning safety, to consider the number of minor "infractions" (near misses) to be an indicator of probable future major catastrophic problems. Kind of like, where there’s smoke there’s fire.

Or to put it another way - If you have a company, department or person who is often sited for numerous small infractions, it is likely that this is related to a purposeful disregard for standards or acceptable management practices.

Out of curiosity, which sounds worse:

A) "Company X failed an inspection"

B) "Multiple compliance related failures were noted during an investigation of 'Company X'. These failures were summarized on a Form 483 and submitted to the Company to address and amend as applicable."

Which sounds worse as a headline?

We could play semantics games all week, but the bottom line is, if a company doesn't have it's s&$t together, citations and findings are indicative of failure, and could be construed as a company being out of control, depending on the severity of their failure(s).

MB

MB

Forget about drugs. On the food inspection side, here are the critters the agency allows in YOUR food supply:

fruit flies parasitic cysts insect heads weevils rodent pellets mites insect excreta mold maggots rodent hairs insect fragments mammalian excreta corn lice insect larvae beetle eggs plant lice

Yum Yum, eat em up!

http://www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/GuidanceDocuments/Sanitation/ucm056174.htm

You mean to say that you don't have a healthy spoonful of parasitic cysts in your Boo Berry cereal every morning ? ;-)

As a microbiologist, and in speaking with others that went over to the food industry, I've heard that they do see nasty lil' critters pretty often. But what's worse? Maggots, or the crap they inject into animals (hormones) and put into animal-feed? I don't think maggots cause cancer...

MB

Ed,

Often times the district office is the one that is classifying the inspection as OAI, VAI, or NAI, thus the final EIR tells whether a plant "failed" an inspection. The point of a 483 is to demonstrate objectionable conditions at the plant, and not to "fail" a plant. A better check is, again, warning letters or EIR's in which the inspection was classified as OAI.

M. Black,

The article seems to indicate that this at all types of facilities, not just sterile. I agree that 483's at sterile facilities represent much more than non-sterile but classifying a 483 for all facilities as being a failure is a little sensationalistic.

Josh is right. But is the FDA that shorthanded so that they can stay in just Food or Drugs? It's a bit hard to develop expertise if you're inspecting clinical study sites and grain silos in the same week.

I understand your point Josh, and Warning Letters are more of a "hardcore bar" of a metric. However, as someone mentioned earlier, where there's smoke, there is likely to be fire. Is that the case always? No. But are all fires found by inspectors? HELL NO!

So, we can go back to the semantics game. You don't want to consider the necessity of the FDA filing a 483 Form as a failing inspection? Fine. Then we could be more specific to decloud perceptions, and be more accurate:

"Multiple failures to comply with FDA expectations were found at Company X, during an XYZ Inspection, that resulted in a Form 483 that now must be addressed by Company X before activities 1,2, and 3 can re-commence"

You can say that a missed signature is irrelevant. I've seen a missed signature on a batch record leading to a complete (and probably unnecessary) dumping of a batch of product, and a loss of tens of thousands of dollars. Yes, because of failure to follow cGMP documentation practices. And if there is a little hole in a big ship, that in itself will cause inspectors to put away their little magnifying glass, pull out the big magnifying glass, and rent a room at the Best Western across the street for an extended period of time. Be very careful in blowing off "minor" documentation problems. A missed signature on a training record for an employee can negate the activities that fell under that procedure or procedures performed by that operator. That can cause a HUGE mess to reconcile. It may be easier to recall whatever was manufactured or tested by that employee than get the resources in to prove that no action is necessary.

The regulations are there for a reason, whether one wants to view them as trivial or major. Such as using 30 year old GMP training materials used for federally required GMP training. That just shows that the company does not put high priority on GMP. In fact, by using 30 year old materials, they were blatantly crapping all over them in Macy's window, while laughing. But then again, they were not a public company, so no risk of losing shareholders, right?

Regulations of all perceived severities were not put into place for poops and giggles. The opinions of those that have not seen how something trivial could balloon into a huge mess tend to have tunnel vision when they refer to an action that seems on the surface to seem irrelevant to an outsider layman.

Such as Mike Burgess - the rootin' tootin' Texan cowboy with his head up a place where his head is not intended to be.

MB

By the way Ed, excellent example of GMP documentation practices by putting a single line strikethrough in the statement in this article. It's crossed out, but not obliterated, so we can still read it anyway.

;-)

Insider,

Insects have been found historically in at least one combination drug / med device oral product that people ironically use to "freshen" their mouths.

Did that make it to a 483? Nope. It made its way to the shredder.

MB, that's pretty close to the bioassay used for some FDA approved products. It's where an inspector tastes the product and it tastes "gritty" to use their terminology, there's probably too many bug parts.

But then again, why should we be surprised. During the BP Gulf Oil Spill, government scientists used the Gold Standard-their noses- to test for oil-contaminated seafood. I kid you not. If the inspector could not smell petrol on the fish it was deemed acceptable to eat.

MB,

But Ed did miss his initials, and date. :)

Your example though proves the point. The severity of the error could have a great impact on the quality of the product, and thus just simply getting a 483 is not indicative of the overall compliance of a plant.

During a recent inspection, our company had a product that had an investigation on it. The inspector concluded that at the time of the investigation, based on the information available, we did a thorough investigation. He actually said that he would have made the same conclusion. We were still cited for "a failure to thoroughly investigate a batch", because he felt that based upon the investigation we could have gone one step farther though it was not completely necessary. He said he had absolute confidence in the quality of the batch. Does that sound like the plant "failed" the inspection? He actually reviewed close to 20 investigations and came to that conclusion based on one investigation. I should mention that he had a senior inspector with him from another region as he was trying to move up to an inspector 2...

As a fellow auditor, you should know that you can find errors at any facility and those errors are not all created equal. This is why even the FDA labels then as OAI, VAI, or NAI,

Haha - Bravo guys!

Similar to crunchy fly wings, I've heard a manufacturing consultant "secretly" confiding to a manufacturing supervisor: "everyone in the manufacturing world (pharma parenterals) knows that there are aluminum shards in vaccines. The "trick" is to get them small enough to not be detectable by the human eye."

Then people wonder why vaccines are tied to autism, and have a concern for over-injection. But, oh yeah, go shoot yourself up with Gardacil because there's about a 5% chance that you'd need it to begin with, and a decent chance that it wouldn't work anyway.

And oh yeah, inject infants with four mandatory doses of prevnar. Unless of course the plant is shut down and their stockpile is depleting. Under those circumstances, "two doses are effective enough ... trust us"

Right... Fraudulent scumbags.

Josh, I understand that depending on an auditor's personal ethic, and ability to do their job unbiasedly is not standard across the board.

But this is the twisted world of drug manufacturing. What do you expect, grace and consistiency?

MB