Is nearly universal prospective order review (NUPOR) really necessary?

Several years ago there was a group of pharmacists taking a serious look at the long-time practice of having pharmacists review virtually all drug orders, sometimes referred to as nearly universal prospective order review (NUPOR). The argument for NUPOR is that it is needed to ensure complete, accurate orders. The argument against NUPOR is that it’s expensive, time-consuming, and unnecessary in many instances. I fall into the latter category of pharmacists, i.e. NUPOR is an antiquated practice that needs to be done away with.

There are those that argue that doing away with NUPOR is dangerous and removes the pharmacist from the medication use process. Nothing could be further from the truth. By making NUPOR a requirement you have taken pharmacists out of the healthcare discussion. NUPOR forces pharmacists to be tied to a terminal when they could be doing other things. 

The introduction of Electronic Health Records (EHRs) and Computerized Provider Order Entry (CPOE) have created a perfect opportunity to change the concept of NUPOR.

Here are some scenarios to think about:

  1. Setting, a large acute care hospital with a busy Emergency Department (ED). Physicians frequently order boluses of NS, or other fluid. Does a 1L bolus of NS ordered for an adult patient in the ED really require verification by a pharmacist prior to administration? Ask yourself, as a pharmacist, what set of circumstances would cause you to reject such an order and call the physician?
  2. Setting, hospital OB-GYN unit. Adult patient comes in for delivery. Physician orders 50 mg of IV meperidine x1 for pain. The patient is in good health, labs are normal, and has no allergies to the medication. Is there any reason that such an order needs verification? What would cause you to reject it and call the physician?
  3. Setting, general medicine floor of a hospital. Elderly patient admitted following a minor surgical procedure. The patient is experiencing constipation secondary to the procedure and medication for pain. Physician orders a bisacodyl suppository or MOM x1 to help get things moving. Labs are ok, the patient has no allergies to the medication, and there are no significant drug interactions. Is there any reason that such an order would require a pharmacist’s blessing?

The list goes on and on. I suppose one could argue that there could be potential for a physician to blow through a catastrophic problem with the drug order that would harm the patient. Sure, that could happen. However, I would argue that a good system would allow the healthcare system to place hard stops in places where there is genuine concern or potential for a problem.

I just don’t see NUPOR as necessary in today’s healthcare environment, especially for those healthcare systems using EHRs and CPOE. Don’t agree? That’s cool. Use the comment section below to convince me I’m wrong, but make sure you have a good reason, because if you don’t, I’ll mock you in front of the other children. 

5 thoughts on “Is nearly universal prospective order review (NUPOR) really necessary?”

  1. I agree with you except we already have “autoverification” set up in our ED so only oddball things have to be verified. The problems are with allergies and with systems issues mostly. In Cerner, you have a breed of orders used for outpatients and a breed used for inpatients. Linking these correctly is one of the primary things pharmacists do. Other things are fixing medication reconciliation mess ups and duplicates. Mainly we have to verify orders because of the clumsy way the systems are designed by Cerner. Maybe EPIC is better but I doubt it is much better.

  2. Thanks for the feedback, Warren. You’re spot on with your view, i.e. “the clumsy way the systems are designed”. I see this as a huge barrier. I believe that a properly designed system – from front to back – would alleviate some of the problems you describe in your comment.

    By the way – EPIC is no better.

  3. First of all, you’re safe with me. I agree with you.

    Maybe this desire for require review of all medication orders by a pharmacist has its roots in ambulatory/retail pharmacy practice (i.e., a pharmacist must review all prescriptions before they can be dispensed to a patient).

    Unfortunately, the circumstances are very different between the ambulatory setting and the hospital setting. In the ambulatory setting, if an incorrect prescription is dispensed, there may not be anyone who will be monitoring that patient for the next 30-90 days. A lot of things could go wrong during that time. On the other hand, a patient in the hospital is continually monitored by nursing, pharmacy, and the medical staff. Problems will be quickly identified.

    Therefore, why have the same standard for both settings? Instead, for routine medication orders for a particular patient type, don’t require pre-dispensing order approval. As long as the drug therapy is part of a routine order set that has been approved, allow the order to be auto approved. Non-routine orders would still have to be routed to the pharmacist for review prior to dispensing.

    The best part of this practice is that it is identical to how it was practiced in hospitals over 50 years ago. Routine drugs were on floor stock in a locked medication room. A pharmacist was only involved in the process when a non-routine drug was requested from the pharmacy for a specific patient.

    If we wanted to add one improvement to the old system, then set up a mechanism to force the review of each patient’s total drug therapy once each day independent of the order process. This would easily catch any problems that might slip through the automated process and it would make much better use of the pharmacist’s time and at the same time improve the monitoring of each patient’s medications. It’s déjà vu all over again…

  4. I agree that reviewing every single order for standard, routine things is a waste of time. And with properly configured CPOE, order sets and decision support, there isn’t much added value to reviewing every single dose of acetaminophen that goes through. You need a pharmacist to validate all that programming more than you need one to okay saline boluses as you describe. And you need a system that is actually reliable enough to operate in these conditions,as other comments state.

    However, there is an increasing number of patients taking complex drug regimens, doctors who prescribe costly drugs, and treatment decisions for which guidelines don’t apply and order sets don’t exist. Those cases definitely need order verification,and there is no shortage of them.

  5. I agree that there are circumstances like those mentioned in your comment that would necessitate a pharmacist order review. In no way am I saying that pharmacists should no longer review orders. However, I believe there is a certain percentage of orders like those mentioned in my post that do not require a pharmacist’s time. How often do those orders occur? 10%? 20% More? It’s difficult to say, but it feels like a lot.

    As systems progress, it would seem only fitting that the technology would allow better automation of such things.

    Thanks for stopping by, and for taking the time to comment.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.