I know this is a long story, but it is a look inside the “Walmart of Pharmaceuticals.” How the everyday products used in hospitals get made...or don’t.
My second GMP laboratory took me to a sterile injectable manufacturing facility. It was a slight pay increase and I had moved near another site the company. I hope that I might be able to transfer closer to home eventually. I was sorely mistaken. It was the worst experience laboratory work experience of my life. But that's another post. Fast forward six years at the facility was to become my worst validation experience of my life.
I was forced into my choice because my two viable opportunities for work were working at this injectable facility close to home or work in downtown Boston in the winter. I chose the former. I also needed to be close to home to go to a training class for a very good cause/family every weekend. When I told my wife about taking work at this facility she said: "Are you sure? You b%#^&*d about going there every day last time you worked there!”
I was getting a healthy raise, over twice what I made when I first worked there. I also had worked with the president/owner of the consulting company I was going to be working for before and like him. So those were also great advantages.
For the first month I was at the parental manufacturing facility, I didn't do much. It was during shutdown and we had no projects to work on until after shutdown. I learned later, my boss paid me out of his own pocket because he didn't have/couldn’t get a PO from the client, just so he could keep me around.
Eventually I get my assignment: Replacement of a part on an injectable line and doing an engineering study. Sounds simple, although the statistical sampling part of the engineering study was intimidating. My dad told me since high school to take statistic, but “Writing the Novel” appealed more to me for an elective class. Turns out I could find nobody in the facility that understood the statistical sampling SOP and the person who wrote it no longer worked at the facility.
So the central question of replacement of the part on the injectable line was: was it a like for like replacement? It had been modified slightly to be made of less parts. Why less parts? It was determined that instability in the part was causing movement on the filling line. And it was filling line of glass containers. These containers were shattering. When they shattered glass when everywhere inside the enclosure. The injectable containers were of a unique design that provided less structural integrity than a usual injectable vial.
Now let’s step back even farther, pre moi arriving on the scene: This facility had been receiving consumer complaints of glass in the injectable vials. They opened a CAPA (corrective and preventive action) these consumer complaint problems of glass in product...and it’s injectable. Very bad problem to have.
The FDA arrives for an inspection and this CAPA is one they were interested in. The CAPA had been open for 5 months. What has the facility done? “The CAPA was overdue and planned actions had not been completed.” It becomes an observation in a very big 483 released by FOIA. (483s are usually not public).
So now they have a cross-functional team CAPA team assembled and meeting (supposedly) that has determined that this part for the filling line being replaced will keep vials from breaking. Since I’m assigned with doing anything engineering or validation to get this part replaced, I started doing some research. I want to see the original validation. They say they have a highest validated speed of, for example, 140 vials per minute on the filling machine. I want to see the original validation to write the new engineering study.
After a week or two of working with document control to get the original qualification, I find out that the qualification can’t be found and won’t be found. They have lost the original validation twenty years ago. They had a memo that was written and signed that said, to the effect: “We have qualified this filling machine at 140 vpm, but have lost the original qualification, but it is qualified at 140 vpm.”
My strategy, of course, is to run empty vials with the new part installed and test vial breakage at 140 vpm. I start working on getting vials and the stoppers. I talked with the purchasing department and learned the news that the vials were part of a drug shortage. Every vial was needed to manufacture the drug on shortage. Even worse, the stoppers were on back order because the composition had been changed and the needed an extended time to off gas.
Being the great validation engineer I am, I try to think of a way to do this without using any more vials. While on the manufacturing floor, outside the clean area as the vials are being inspected as the come out of the isolator and Grade A environment they are inspected by group of about six women, I learn they are also keeping track of how many rejects they take out. After further investigation I learned that one of the managers, we’ll call him Rick, on the filling line has actually been compiling data on the rejects. I say: “Great! I can use the data for an engineering study.” I ask Rick for that data.
I tell my boss about my great solution to executing the engineering study without actually wasting vials using the reject data. I said: :”Yeah, it’s great Rick has been compiling the data already.” My boss tells me that Rick has already performed an engineering study on the reject data and submitted the data to QA in some kind of report. “QA s#&%t all over his report, so he got mad and gave up.”….so now they’re paying me to do something one of their employees has already done.
All throughout this project, a manager, we’ll call Kevin is constantly hounding me: “You’ve got 3 months to finish the project before we respond to the FDA! You’ve got 10 weeks to finish for the FDA!” Non-stop grabbing me in the halls. Meanwhile I am supposed to be leading the CAPA cross-functional team meetings to come to conclusions and agree on an approach to replacing the part. We’ve got QA, regulatory, the line manager above Rick, and maintenance. Most of the time not enough people show up to the meetings to make decisions. They reply they’re coming and then just don’t show up. You would think there’s actually trenches and razor wire separating each department.
In addition, the QA specialist that is my contact, signer, is brand new to the role. She is eager, even helps me meet a deadline on a Saturday, and know the manufacturing facility well. But, she is way in over head and has to get route every decision through her boss the QA manager, who won’t show up to meetings and sits in a glassed-off area that I don’t have access to.
So, I take the fight to them in a traveling show. Tracking them down in person to get them to agree to the approach. Having to get someone to get me into the area where the to see the QA manager that can make decisions.
The head line manager, we’ll call Jose. Jose is really passive-aggressive. I think he was trying to sink my project, so he could blame any failure to finish the project on the contractor, me. I bring him an intensive engineering study draft and he says “That’s too much. We can’t do this.” Weeks later, I bring him a pared down study, he replies: “This is not enough.” Eventually I hound him enough to agree to something.
Time is almost up per Kevin. Kevin then says: “You’ve got another month. To be honest, it was a little white lie. I knew you couldn’t make it on time, so I took a month off the schedule.” So a month later, I get through writing an engineering study, we get the part changed out, and compare reject rates. Everything looks good. I draft language for Kevin to use in the response to the FDA. I am filled with pride, but I do have my doubts that we actually accomplished something. A maintenance technician tells me: “You know, that replacing that part is not going to help vials breaking? They’re just trying to run the line too fast!” Anyway I don’t care. My part is done.
Kevin then comes to me: “When are we going to start requalification?” I’m like WTH? The intent was always to just do the engineering study. I am already sick of this project. Meanwhile all down time is consumed with reading SOPs ad nauseum for training. I could just sit back and stay under the radar, but get frustrated from not getting real work done.
I settle back in and start going back through the sites SOPs to get up to speed which is excruciatingly painful. The have an updated EDMS (electronic document management system), with documents in PDF format, but you cannot search within documents! To search within text you have to go back to their old, non-official, EDMS that is an ancient monochrome AS/400 system.
I am still hooked on the question of how are they justifying using 140 vpm in manufacturing their vials. What periodic testing is performed that they justify since the original qualification was done over 20 years ago. I get their media fill SOP. It states that the media fill is to be run every six months later. One at the slowest speed, then six months later the next media fill at the fastest speed. So it begs the question: “If they’re running the entire media fill at the fastest speed, how is it passing the media fill?!” So, I couldn’t track down the microbiology people that performed the media fill in person, but I did get their name and phone number.
So I called. “How do you perform the media fill at 140 vials per minute?”
The technician replies: “We turn the dial up in the beginning and then turn it back down.”
I was dumbfounded.
I then relate all this to the newbie quality assurance specialist I have been working with. I don't remember her response, but I went back to job of planning a validation. A few days later I got a call on my way home from my boss telling me I was off the project per the QA manager "because I didn't know anything about the validation life cycle."
I believe my getting kicked off the project was not a result of lacking knowledge, it was asking too many questions.
Over the years, about every consulting, engineering, and validation company I know in my state has worked at this facility. Most of the independent, contract-gig validation people I know have worked there. People flow in and and out of the facility and their advice is never taken or implemented or their tired of beating their heads against the wall. The employees know that they can simply bide their time and any consultant haranguing them to institute changes will be gone.
Other contributing factors: The facility is older and designed when things didn't need to be quick and lean. The facility is in a town that does not attract well-educated employees. The town has a crime rate. There is no sharing of beneficial knowledge in the facility. It is more a culture of "I better get you, before you get me." Your best strategy at the facility was to keep your head down, do the bare minimum and get out the door.
So when anyone tells me: "This is the worst project ever!"
I say: "No. I worked the worst facility and project ever."
"[10/365] Injected"by dissolved is licensed under CC BY-NC-SA 2.0
"Steam Stack"by yuan2003 is licensed under CC BY-NC 2.0
I was getting a healthy raise, over twice what I made when I first worked there. I also had worked with the president/owner of the consulting company I was going to be working for before and like him. So those were also great advantages.
For the first month I was at the parental manufacturing facility, I didn't do much. It was during shutdown and we had no projects to work on until after shutdown. I learned later, my boss paid me out of his own pocket because he didn't have/couldn’t get a PO from the client, just so he could keep me around.
Eventually I get my assignment: Replacement of a part on an injectable line and doing an engineering study. Sounds simple, although the statistical sampling part of the engineering study was intimidating. My dad told me since high school to take statistic, but “Writing the Novel” appealed more to me for an elective class. Turns out I could find nobody in the facility that understood the statistical sampling SOP and the person who wrote it no longer worked at the facility.
"20080422__169"by clappstar is licensed under CC BY-NC 2.0
So the central question of replacement of the part on the injectable line was: was it a like for like replacement? It had been modified slightly to be made of less parts. Why less parts? It was determined that instability in the part was causing movement on the filling line. And it was filling line of glass containers. These containers were shattering. When they shattered glass when everywhere inside the enclosure. The injectable containers were of a unique design that provided less structural integrity than a usual injectable vial.
Now let’s step back even farther, pre moi arriving on the scene: This facility had been receiving consumer complaints of glass in the injectable vials. They opened a CAPA (corrective and preventive action) these consumer complaint problems of glass in product...and it’s injectable. Very bad problem to have.
"IMG_1054"by V31S70 is licensed under CC BY 2.0
The FDA arrives for an inspection and this CAPA is one they were interested in. The CAPA had been open for 5 months. What has the facility done? “The CAPA was overdue and planned actions had not been completed.” It becomes an observation in a very big 483 released by FOIA. (483s are usually not public).
So now they have a cross-functional team CAPA team assembled and meeting (supposedly) that has determined that this part for the filling line being replaced will keep vials from breaking. Since I’m assigned with doing anything engineering or validation to get this part replaced, I started doing some research. I want to see the original validation. They say they have a highest validated speed of, for example, 140 vials per minute on the filling machine. I want to see the original validation to write the new engineering study.
After a week or two of working with document control to get the original qualification, I find out that the qualification can’t be found and won’t be found. They have lost the original validation twenty years ago. They had a memo that was written and signed that said, to the effect: “We have qualified this filling machine at 140 vpm, but have lost the original qualification, but it is qualified at 140 vpm.”
My strategy, of course, is to run empty vials with the new part installed and test vial breakage at 140 vpm. I start working on getting vials and the stoppers. I talked with the purchasing department and learned the news that the vials were part of a drug shortage. Every vial was needed to manufacture the drug on shortage. Even worse, the stoppers were on back order because the composition had been changed and the needed an extended time to off gas.
"sharps"by doggo is licensed under CC BY-ND 2.0
Being the great validation engineer I am, I try to think of a way to do this without using any more vials. While on the manufacturing floor, outside the clean area as the vials are being inspected as the come out of the isolator and Grade A environment they are inspected by group of about six women, I learn they are also keeping track of how many rejects they take out. After further investigation I learned that one of the managers, we’ll call him Rick, on the filling line has actually been compiling data on the rejects. I say: “Great! I can use the data for an engineering study.” I ask Rick for that data.
"Clean Room"by OIST (Okinawa Institute of Science and Technology) is licensed under CC BY 2.0
I tell my boss about my great solution to executing the engineering study without actually wasting vials using the reject data. I said: :”Yeah, it’s great Rick has been compiling the data already.” My boss tells me that Rick has already performed an engineering study on the reject data and submitted the data to QA in some kind of report. “QA s#&%t all over his report, so he got mad and gave up.”….so now they’re paying me to do something one of their employees has already done.
All throughout this project, a manager, we’ll call Kevin is constantly hounding me: “You’ve got 3 months to finish the project before we respond to the FDA! You’ve got 10 weeks to finish for the FDA!” Non-stop grabbing me in the halls. Meanwhile I am supposed to be leading the CAPA cross-functional team meetings to come to conclusions and agree on an approach to replacing the part. We’ve got QA, regulatory, the line manager above Rick, and maintenance. Most of the time not enough people show up to the meetings to make decisions. They reply they’re coming and then just don’t show up. You would think there’s actually trenches and razor wire separating each department.
In addition, the QA specialist that is my contact, signer, is brand new to the role. She is eager, even helps me meet a deadline on a Saturday, and know the manufacturing facility well. But, she is way in over head and has to get route every decision through her boss the QA manager, who won’t show up to meetings and sits in a glassed-off area that I don’t have access to.
So, I take the fight to them in a traveling show. Tracking them down in person to get them to agree to the approach. Having to get someone to get me into the area where the to see the QA manager that can make decisions.
The head line manager, we’ll call Jose. Jose is really passive-aggressive. I think he was trying to sink my project, so he could blame any failure to finish the project on the contractor, me. I bring him an intensive engineering study draft and he says “That’s too much. We can’t do this.” Weeks later, I bring him a pared down study, he replies: “This is not enough.” Eventually I hound him enough to agree to something.
Time is almost up per Kevin. Kevin then says: “You’ve got another month. To be honest, it was a little white lie. I knew you couldn’t make it on time, so I took a month off the schedule.” So a month later, I get through writing an engineering study, we get the part changed out, and compare reject rates. Everything looks good. I draft language for Kevin to use in the response to the FDA. I am filled with pride, but I do have my doubts that we actually accomplished something. A maintenance technician tells me: “You know, that replacing that part is not going to help vials breaking? They’re just trying to run the line too fast!” Anyway I don’t care. My part is done.
Kevin then comes to me: “When are we going to start requalification?” I’m like WTH? The intent was always to just do the engineering study. I am already sick of this project. Meanwhile all down time is consumed with reading SOPs ad nauseum for training. I could just sit back and stay under the radar, but get frustrated from not getting real work done.
I settle back in and start going back through the sites SOPs to get up to speed which is excruciatingly painful. The have an updated EDMS (electronic document management system), with documents in PDF format, but you cannot search within documents! To search within text you have to go back to their old, non-official, EDMS that is an ancient monochrome AS/400 system.
I am still hooked on the question of how are they justifying using 140 vpm in manufacturing their vials. What periodic testing is performed that they justify since the original qualification was done over 20 years ago. I get their media fill SOP. It states that the media fill is to be run every six months later. One at the slowest speed, then six months later the next media fill at the fastest speed. So it begs the question: “If they’re running the entire media fill at the fastest speed, how is it passing the media fill?!” So, I couldn’t track down the microbiology people that performed the media fill in person, but I did get their name and phone number.
So I called. “How do you perform the media fill at 140 vials per minute?”
The technician replies: “We turn the dial up in the beginning and then turn it back down.”
I was dumbfounded.
I then relate all this to the newbie quality assurance specialist I have been working with. I don't remember her response, but I went back to job of planning a validation. A few days later I got a call on my way home from my boss telling me I was off the project per the QA manager "because I didn't know anything about the validation life cycle."
I believe my getting kicked off the project was not a result of lacking knowledge, it was asking too many questions.
Over the years, about every consulting, engineering, and validation company I know in my state has worked at this facility. Most of the independent, contract-gig validation people I know have worked there. People flow in and and out of the facility and their advice is never taken or implemented or their tired of beating their heads against the wall. The employees know that they can simply bide their time and any consultant haranguing them to institute changes will be gone.
Other contributing factors: The facility is older and designed when things didn't need to be quick and lean. The facility is in a town that does not attract well-educated employees. The town has a crime rate. There is no sharing of beneficial knowledge in the facility. It is more a culture of "I better get you, before you get me." Your best strategy at the facility was to keep your head down, do the bare minimum and get out the door.
So when anyone tells me: "This is the worst project ever!"
I say: "No. I worked the worst facility and project ever."
Comments
Post a Comment