Skip to content

Automating Pharmacist ‘perfection’ or not discussion


Below is a really good discussion on the use of technology for pharmacist "auto-verifing" with CPOE.  Since this is taken from a list serv that is open to subscribers and this blog is open to all, I have taken the names out to protect the innocent.   I am doing this so this issue is brought to a larger audience.   Perfection, btw, is the term that has been proposed for the order checking and transformation that is done with pharmacist review in CPOE.

These articles are a good back drop for this discussion.

A. J. Flynn
Opportunity cost of pharmacists’ nearly universal prospective order review
(go to the online version NOT the paper AJHP b/c there are some good eLetters)

J. Poikonen
An informatics perspective on nearly universal prospective order review
Am. J. Health Syst. Pharm., April 15, 2009; 66(8): 704 – 705.

D. A. Tribble
Automating order review is delegation, not abdication
Am. J. Health Syst. Pharm., June 15, 2009; 66(12): 1078 – 1079.

P. G. Pierpaoli
Creatively using our intellectual capital
Am. J. Health Syst. Pharm., June 15, 2009; 66(12): 1087 – 1087.

The List Serv discussion (various authors):

At our recent monthly Pharmacy Directors meeting the following question was raised by one of the Directors:
Is anyone aware of any statute, standard or requirement related to CPOE that REQUIRES pharmacist review of the orders that are "written" by a practitioner via CPOE systems?  Are there systems out there where the order that is written can be routinely acted upon by staff prior to pharmacist review. The discussion that took place referenced the standard requirement from JCAHO that wants pharmacist review of all medication orders. I couldn't find anything under the CMS standards similar to the JCAHO standards.
MM05.01.01 – Prior to dispensing medications or removing medications from floor stock, a pharmacist reviews the appropriateness of the medication orders, unless delay would cause patient harm.
For those of you that are either currently utilizing CPOE or a far enough down the implementation curve how are you handling pharmacist review of orders? Is it a) a state requirement b) a requirement of the CPOE system or c) a default that the facility decides upon.?
The discussion that I've seen in the various list-serves has been the issue of pharmacists having to "fix" physician orders before they were "transcribed" into the pharmacy program.
Any comments would be appreciated.



TJC standard requires pharmacist review of orders (except in urgent situations) whether the order is written on paper or via CPOE.  We have a fully integrated system, so there is no "re-entry into the pharmacy system" and pharmacists do verify all orders.  This is set at a system level, but we do have control to not require pharmacist verify if we so desire.

There has been considerable discussion in ASHP on whether the standard as it is currently written make the most sense in the CPOE world.  There is hope that clinical decision support within CPOE and pharmacy systems will make it possible for "autoverify" of some medications while the pharmacist concentrates on reviewing orders that need to be, or other clinical work that may be more important for patient outcomes.

I hope this answers your question.



I’m not aware of any. All of the states in which I’ve consulted (mainly med safety/management engagements in 20, or so, states) require pharmacist review of orders/prescriptions regardless of the “delivery method” (i.e. they don’t make distinctions between CPOE, hand written, fax/scan, etc. for the review portion of the process). Having said that, it’s been several years since I’ve consulted in NH and few others, so things may have changed in the meantime. While TJC’s MM05 is certainly a regulatory “requirement” as it relates to their certification process (and its related ramifications), the State BOP (and FDA, DEA and sometimes the State Dept of Health) is the “higher authority” in terms of legal/statutory requirements related to medication dispensing. Not that you want to invite scrutiny, but you’ll have to go to the Board anyway if you want to do something outside of the statutes (unless you lean toward the “ask for forgiveness” versus “ask for permission” side of the scale).

I’m curious what may have prompted the question. From a broader professional practice standpoint, I’m always concerned about the additional error risk that may result from streamlining certain portions of the medication use process (or any critical process, for that matter). I apologize if I’m reading too much into this, but I’m also concerned if the end result relinquishes any of our responsibility and/or authority as pharmacists. OK, I’ll get off of my soapbox now.


Not clear across the US whether CMS, TJC, state health or BOP's will be the primary decision body for setting the rules of engagement for Autoverify. This creates uncertainty, and conflicting directives. With CPOE, possibility to review orders generated in one state, and have them and verified by another R.Ph., from several cities away, or even different states becomes a real possibility, and also a concern. But, that's another discussion I guess.

Autoverify? Well, it might work on a limited basis, but here's some additional things to think about..Will the "P" in the CPOE be physician, "extender", pharmacist, or an "N" for nurse engaging the computer in wee hours of the morning after having some orders "barked to her" by the doc over phone, or from home, or a remote (undisclosed location) that's not connected to the computer at all??

1. Possibly ok IF docs are ordering off pre-built care sets in their respective systems.
2. Possibly ok if basic demographics, e.g., height, weight, allergy information (including type of reactions people describe to things they state allergy to) are described and available for prescribers to view as they are ordering.
3. Possibly ok if there's a small but important set of pertinent, clinically significant safety alerts for allergy, duplications, interactions, renal screens, lab screens, etc. that fire for prescribers to see AS they are ordering (e.g. not dumped all on pharmacist, or someone else to review and sort through on the back-end).
4. If autoverifying orders for lytes, anticoag agents, or other high alert drugs, will key labs in place to guide the ordering process safely? For example, if no baseline renal function and Lovenox 1mg/kg q12h is ordered, and a subsequent creatinine next AM comes back significantly elevated, or we find out patient is on dialysis? What did auto-verify accomplish besides setting the stage for error and an ADE? Are alert packages really sophisticated enough to provide this safety check up front?  Lots of promises, but I sure haven't seen any that are yet.  How about an order for Coumadin to start, but no baseline INR or HgB at time drug is ordered. If the INR comes back high, and HgB comes back low the next AM, it's too late.

So, in some cases for simple routine orders, auto-verify might work, but the technology solutions will need lots of oversight and regardless of processes, there will be LOTS of potential land mines requiring human oversight. To me, the whole regulatory process seems unclear and prone to confusion.. Needs much more thought and sorting out..     just my thoughts..


We don’t have CPOE yet, but I believe the CMS standards are stricter than the JC standard. The JC standard does allow for exceptions when there is prescriber oversight.

Here is the CMS standard:
Interpretive Guidelines 482.25 (b)  All medication orders (except in emergency situations) should be reviewed for appropriateness by a pharmacist before the first dose is dispensed.”


No regulations or practice standards differentiate between the mode of conveyance of an order.  Thus, it seems clear that CPOE must be treated no differently.  Below is a snippet from a limited analysis of these standards and regulations.

Accreditation Standards and Practice Guideline Statements

The Joint Commission (“TJC”). “All prescriptions or medication orders are reviewed for appropriateness.”  Standard MM 4.10, Comprehensive Accreditation Manual for Hospitals: The Official Handbook (The Joint Commission, Refreshed Core, January 2008). The elements of this standard enumerate what a pharmacist should do for a new medication order but are silent whether a pharmacist must review an order to discontinue a medication.

Institute for Healthcare Improvement. “Every medication order should be reviewed by a pharmacist to ensure appropriate selection of drug, dose, route, and frequency as well as to check for interactions with other medications.”
Accessed May 5, 2009.

Similarly to TJC’s requirement, this practice guideline states that the pharmacist’s review must occur prior to dispensing from a new medication order.

Both of these examples serve to illustrate that there is a clear duty for a pharmacist to review new medication orders against a patient’s medication profile.

Legal Requirements

California pharmacy regulations require a pharmacist to “review a patient's drug therapy and medication record before each prescription drug is delivered. The review shall include screening for severe potential drug therapy problems.” 17 CCR 1707.4. (italics added.)

Kansas pharmacy statutes similarly require a pharmacist to prospectively review the patient’s medication prior to dispensing. “Upon receipt of a prescription order, the pharmacist shall examine the patient's medication profile record before dispensing the medication to determine the possibility of a harmful drug interaction or reaction to medication.” K.S.A 65-1642(c)(2). (italics added.)

Delaware pharmacy statutes require pharmacists to review medication profiles prior to dispensing medications. “Upon receipt of a new prescription, a pharmacist … must examine the patient’s profile record before dispensing the medication to determine the possibility of a harmful drug interaction or reaction.”  24 DE Admin. Code 2500 5.7.4. (italics added.) Delaware regulations specifically address a patient profile review in the hospital inpatient context where the drug is dispensed by the pharmacy. “Prior to administration of the first dose, the pharmacist must examine the profile to determine the possibility of a harmful drug interaction or reaction. Upon recognizing a significant potential for harm, the pharmacist should notify the prescriber and other appropriate persons.” 24 DE Admin. Code 2500 9.13. (italics added.)

The examples above are typical of other states’ pharmacy statutes and regulations and all attach an affirmative duty of the pharmacist to prospectively review a new medication order against the patient’s medication profile, regardless of the way the order is conveyed to the pharmacist.


Implementing technologies such as CPOE, dispensing cabinets, bar-coding, IV compounders, smart infusion pumps, etc should provide us with the time and resources to thoroughly review orders more than we could have without it.  When I entered the profession, we had none of these and I had to rely on my education and memory to catch drug-drug or drug disease state interactions, dose checking, IV incompatiblities and so on.  Despite my best efforts, I wonder how many I missed and if harm ever resulted.
It would be an abdication of our responsibility as pharmacists if we eliminated order review simply because the order is now entered electronically versus using a pen.  The assumption would be that prescribers will note and heed the alerts they receive about the dose, route, frequency, interactions, duplications, contraindications, etc.  Is it realistic to assume that?  I still hear physicians say, if I enter an order incorrectly (with CPOE), pharmacy will catch it.  Maybe?
If we cease to review orders or even some orders from selected care units (which also concerns me), how different is that from the days of floor stock vials and bottles where nurses selected the drug and administered it before the pharmacist reviewed the order and in many cases, the pharmacist never saw the order?  That was not a safe system and the profession invested a lot of effort to change that model.  Why would we go back to it simply because we now have the stock in an electronically controlled cabinet and the order is no longer written on paper?
I also get concerned when I read that we can validate orders in what seems to the the blink of an eye.  Should we not take advantage of the technology we have, to do a more thorough review of the order, rather than a quicker, perhaps less comprehenvise review? To repeat something a posted a few years ago, "Speed Kills" is not just a highway safety slogan.  I suggest we give this the parent or child test.  Would any of us want ours to receive medications based on an order the pharmacist did not review.
Finally as I step of the soap box, although we should do things because they are the correct thing to do for patient safety, rather than because of legislation that (at least for now) requires it, I encourage everyone to look at their state statutes and regulations to see what is expected of the pharmacist.  I have inserted one here.
A pharmacist shall conduct a prospective drug review before each new prescription is dispensed or delivered to a patient or a person acting on behalf of the patient. Such review shall include screening for potential drug therapy problems due to therapeutic duplication, drug-disease contraindications, drug-drug interactions, including serious interactions with nonprescription or over-the-counter drugs, incorrect drug dosage or duration of drug treatment, drug-allergy interactions, and clinical abuse or misuse.

(4+ responses in support of this view)


     One of the most difficult errors to catch is orders entered on the wrong patient. I have yet to see a system that will prevent these. I’ve caught a few of these just because I was familiar with the patient enough to see that things weren’t right. And as Matt says, there are going to be alerts that physicians miss due to alert fatigue or that they are going to minimize when they shouldn’t. Pharmacists still play a valuable role in catching these types of errors.

      However, I can see the other point of view also. Can I prevent more errors by verifying physician orders or being a clinical pharmacist out on the floor looking at the whole picture? I know that a lot of organizations have floor based order entry pharmacists serving a dual rule as clinical pharmacists, but unless the workload is light, you can’t do both adequately. Just food for thought.


I actually see this discussion akin to what FDA does to determine OTC status of a drug.  In that case, they look at the safety profile, intended use, monitoring requirements, etc. to determine if a label on a box can serve as the only "provider." So could we come up with a similar process for saying some drugs are safe for use without a pharmacist review like OTC drugs that do not need a doctor or pharmacist to be involved.  I know OTC drugs are dangerous if mis-used and issues exist with many of them but the concept of FDA transfer to OTC is what I am referring to. I think there are orders we review that provide less value than others.  Why do I need to review a docusate order or eucerin order?  It is important to know that patient has a bowel regimen when on narcotics but I am not sure I need to review that order prior to a pt receiving.  I think that looking at our cognitive abilities as a value added service and not a gate keeper might be a hug
paradigm shift for many.


More to come…….

Posted via email

No comments yet

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: