I received an email from the MSOS community today that discussed the process of dual-pharmacist checking on orders for certain high-risk drugs. It reminded me of a concern I have long had that we seem to believe that having a person look at something is more reliable than an automated check (such as scanning the bar code on a container) and that having two people provides more safety than having one person look at it.
Double-checking of an order made some sense when we had to transcribe those orders into a pharmacy system, but, with EMR systems, such transcriptions have largely gone away.
For those reading this blog that employ two-person checks, when was the last time such a check yielded a correction? Do you track such corrections?
We require a pharmacist to visually check a prepared IV. Other than meeting state board requirements, our argument for this often asserts that the pharmacist needs to inspect such doses for particulates. Yet the smallest particles that can do harm are roughly 5 microns in size, and the smallest particle the human eyeball can discern is about 50 microns in size. Do pharmacists have calibrated eyeballs that can see what others cannot?
Other than that, the vast majority of our compounded sterile products (CSP) involve adding a small volume of a clear liquid to a large volume of clear liquid. I would be exceptionally surprised to learn that an inspecting pharmacist can even discern that any clear liquid was added to a 1000 mL IV bag, much less that the correct clear liquid was added to the IV bag. The only tool we have do deal demonstrate that something was added is gravimetrics, and gravimetrics must presume that the correct drug was used to know what density to use to compute the volume injected from change in mass of the IV bag.
Our literature has a instances in which checking by multiple people failed. Perhaps one of the most egregious of those occurred when a pharmacist mistakenly dosed zinc in mg rather than mcg for a neonatal TPN. That order was checked by a second pharmacist, who missed the error. The technician who prepared the dose (which turned out to have a volume of 500 mL rather than the normal 50 mL) didn't notice that they changed the zinc vial on the compounder nearly 20 times. The pharmacist who checked the final dose didn't catch the dosing error or notice that the dose was ten times the size it should have been. Nor did the nurse who eventually administered the entire dose to a neonate who died as a result.
Why does that occur? Sometimes it occurs because people can become task-oriented. Some of it occurs because the person doing the checking implicitly believes that the person whose work they are checking is always flawless and so treats the check as a "checkbox" exercise. By the way, the more often that person's work is flawless, the more that point of view gets reinforced.
But mostly, IMO, it is because we cannot, or choose not to make the process of checking someone's work a separate job. Rather, we require interrupting someone in the middle of doing something else, have them perform the check, and then return to whatever it was that they were doing before. That task-switching is not associate with exceptional performance.
When one looks at federal regulations around current Good Manufacturing Practices (cGMP-21 CFR 210,211) or the medical device Quality Assurance Regulations (21 CFR 820), checking for compliance and quality is a full-time job. Further, it is a job that reports higher in the chain of command than the job of the individual whose work is being checked. And, even then, we still see recalls of both drug products and medical devices.
So, OK... it probably isn't practical to hire someone to just do checking. But, looking at industry, how do they regularly produce medications that are, in general, exactly what they are supposed to be? They certainly don't have a pharmacist standing at the end of an assembly line looking at the contents of each product.
- One of the things they do is that they dedicate production processes to a single preparation. That probably isn't practical for us.
- Another thing they do is to exercise control over their purchasing to ensure that all components that go into a preparation meet certain standards. In theory we do that.
- Another thing they do is that they have a number of both automated and manual checking processes during setup and execution of a run that ensures that the right stuff is being used in the right order.
- Another thing that they do is to have well-defined procedures for each preparation whose proper execution can be checked during manufacture and/or packaging. Many of those checks are automated. Many even have photo-inspection processes that verify that the lot numbers and expiration dates on individual labels are readable and accurate.
How might we apply those principles to our efforts?
Dennis A Tribble, PharmD, FASHP
Retired
Ormond Beach, FL 32174
tribbledennis@gmail.com