Preparing for CPOE: practical ideas

Computerized Physician Order Entry (CPOE) is a sultry siren. Hospitals with few IT project successes, marginal clinical operations, and spotty physician relations somehow think CPOE is the best way to spend money and energy in advancing patient safety.

The failure rate of CPOE speaks for itself. It's just not working. Blame the software, the hospitals, or the doctors, but it hasn't proved to be worth the investment so far.

In the meantime, patients continue to suffer from medication errors. Many of these could be prevented by projects a lot simpler, faster, and cheaper than CPOE. I'm speaking as both a pharmacist and IT director. In addition, these activities can help prepare you organization for eventual CPOE adoption. I'm not even talking about mid-tech projects such as barcoding. These are non-IT steps that require little investment, just unwavering commitment and willingness to do the right thing.

OK, I came up with this list in literally about two minutes. No rocket science was involved.

1. First and foremost: good pharmacists will prevent just as many errors as extensive and expensive clinical systems such as CPOE, CDR, etc. How are you using your pharmacists? Are they available when orders are entered? Do physicians ask for help when they need it? Do pharmacists actively follow up on questionable orders and reconcile every nursing document such as the Kardex and MAR to avoid transcription and interpretation errors? It doesn't take CPOE to do any of these things. Get pharmacists out of the basement and onto the floors where the decisions are made.
2. Do unit secretaries enter your orders? Before you push docs to enter their own, why not have your nurses involved instead of clerical help? You can't make much of a case for CPOE when you won't even mandate CNOE (N for Nurse.)
3. Be more aggressive in creating and maintaining and using standing orders and order sets. Much of the value of CPOE is in standardizing therapies to reduce variability and error potential. Mistakes often occur with drugs unfamiliar to doctors, nurses, and pharmacists. The same standardization that CPOE encourages can be implemented with paper-based order sets, using best practices, evidence-based medicine, and multidisciplinary care planning.
4. Evaluate the conformance to your hospital's formulary. I'm not a big fan of formularies in general because I really don't think they reduce costs. But, if your docs won't follow the rules on agreed-on drugs, CPOE is a lost cause. Formularies limit the use of unusual drugs that can cause errors due to unfamiliarity.
5. Investigate the capabilities of your existing pharmacy system. Big-ticket CPOE systems talk about alerts, reminders, and other high-tech order guidance, but much of what you need is already present in your pharmacy system. The only difference is that the pharmacist gets the warning, not the doc, so extra communication is needed. However, few hospitals evaluate how well their systems perform clinical checks against standard errors. It all depends on the software, how well the clinical database is maintained, and how well the system is set up. If pharmacy has an even marginally capable system and follows reasonable processes, you should have very few duplicate orders, drug interactions, and preventable allergic reactions.
6. Make sure docs have a comprehensive view of all clinical data from an easy-to-use portal that can be accessed remotely. It's easy to buy or build a browser-based app that displays orders, results, transcriptions, and other data in single view, even if it pulls that data from disparate systems behind the scenes. Many errors (and expensive duplicate or unwarranted therapies) are due to the physician's working blind without critical information.
7. Have all medication transcriptions reviewed by at least one other person.
8. Require doctors to print their orders. It's a pain, but cursive handwriting is horrible to decipher.
9. Develop standard administration schedules and allow few exceptions. CPOE is almost impossible if everyone interprets QID and TID differently.
10. Require placing diagnosis on orders. It will likely be required by JCAHO in 2004 anyway. It's much easier to detect prescribing errors if you know what the doc is trying to treat.
11. Give pharmacists access to physician progress notes on their copy of the orders. It's easier to follow therapy when you see what the doc is trying to accomplish.
12. Have pharmacists take a medication history from each patient upon admission. Home meds create errors and conflicts with inpatient therapy. Who do you really want deciding what "my little blue blood pressure pill" is?
13. Have pharmacists generate the Medication Administration Record from the pharmacy system and reconcile it each day with nurses to uncover discrepancies in order interpretation.
14. Have pharmacists print helpful instructions on the MAR for nurses, such as how to give the med, what signs to look for in an overdose, or what foods to avoid when administering doses.
15. If orders are faxed to the pharmacy, check the legibility. Old fax machines put out crummy images.
16. Limit nurse withdrawals from automated dispensing machines such as Pyxis. They should only be able to withdraw what's been verified as orders by the pharmacist via a interface to your pharmacy system. Overrides mean something's not working: either pharmacy isn't getting the orders entered quickly enough or nurses are working from different interpretation of the order. Without pharmacy supervision, Pyxis is no better than the "drug cabinets" of old.
17. Don't give nurses access to the pharmacy. JCAHO frowns on this because some of the meds there are intended for bulk manufacturing, not administration to patients.
18. Emphasize MAR-quality order entry by pharmacists. Not just what it takes to get charges and labels, but proper dose timing, order duration, nurse instructions, correct ordering doctors, and updating when level of care changes (going to surgery, for example.) Pharmacists are easily the best drug experts you have, but expectations for their involvement in front-line care have historically been low in many hospitals. Make them use what they know.
19. Pay extra attention to high-risk drugs. Identify them, add requirements for their storage and ordering, and monitor them relentlessly. Reducing errors involved in just 10-20 drugs can save a lot of lives.
20. Consider "smart" IV pumps, premixed IV solutions, and pharmacy-only IV compounding since many of the most tragic errors are caused by incorrect IV mixing or rates.

It's just amazing that hospitals don't want to fix bad processes, but they naively expect that CPOE will heal all. Let's face it: if your staff and docs don't follow the rules now, they won't follow it with CPOE either. I'll bet you have bunches of policies on order scheduling, abbreviations, legibility, order dating and signing, restricted drugs, and many others that are completely ignored. How are you going to build this random interpretation and illogical processing into a CPOE-driven automated environment? Answer: you aren't. Before you get your hopes up for CPOE, clean up your existing house first. You just might save a few patients while you wait for that magic wand.