Are RHIOs More Important Than Organization-Wide Allergy Reporting?

I recently wrote that healthcare organizations should attack their internal information problems before investing money and effort into RHIOs. Hospitals that can't share information within their four walls are suddenly aroused at the prospect of sharing it with competitors. I see benefit to the patient, but I'm frustrated that in-house information sharing isn't a high priority. Example: the seemingly simple process of recording and using patient-reported allergies.

Let's say you are being seen at a hospital-owned clinic bearing the suitably impressive Mother Ship General Hospital (MSGH) logo on the standards-compliant sign. You tell the nice folks there, all of whom are wearing MSGH badges, that you are deathly allergic to codeine. You rest easy, knowing that any attention you require from the nurturing MSGH angels of mercy will not involve codeine. You sigh contentedly, happy that your caring HMO has chosen your provider wisely on your behalf, comforted that you are safe in the arms of the benevolent giant and your new bestest buddy MSGH.

A week later, you sprain your ankle and visit a doc-in-the-box urgent care facility owned by the omnipresent MSGH. Same logo, same badges. You're among friends who know you well. You leave with a prescription for Tylenol #3. You take it and you die, the irony striking you as you draw your last breath through your rapidly closing throat, now showing symptoms of deadly anaphylactic shock: those MSGH angels of mercy turned out to be angels of death. They killed you because MSGH hasn't figured out how to spread the word about your allergies electronically throughout their corporate spiderweb of organizations, sites, and information systems.

JCAHO requires organizations to share information like allergies everywhere, even when it's obviously inconvenient (and it usually is.). The IT people will patiently explain with crudely drawn pictures why this cannot be done, citing technical limitations and vendor recalcitrance, hoping you will go away quietly and stop being a pain in their butt so they can go back to working on CPOE. They are unimpressed by your fear of facing a mad-dog malpractice attorney, who asks you under oath why you killed that nice old grandmother by giving her penicillin when your own records show that she was highly allergic to it.

Allergies are reported by patients and family members. You're expected to keep them on file forever, which is a challenge in itself to hospitals with the usual poorly implemented and sloppily integrated electronic medical record system. Some of them are easy to document. Codeine, aspirin, and morphine are common ones. But, what if the patient reports "mycins?" That's not really a category of drug, so how do you interpret it? What about someone who says they're allergic to Tylenol with Codeine? Do you record them as being allergic to both components, or just assume that the far more common codeine allergy is the only concern and the acetaminophen is OK? How do you handle allergies to herbal remedies, foreign products, foods, and natural substances?

One of the biggest problems with allergies is that patients confuse them with side effects or even known drug actions. You are not allergic to a drug that gives you an upset stomach, makes you sleepy, turns your urine brown, or dries your mouth. Allergies involve specific allergic symptoms: hives and throat swelling, for instance. Whatever patients say is taken at face value, so perfectly fine drugs won't be used because the patient is "allergic" to them. From a legal standpoint, that even makes warped sense. Would you want to give penicillin to someone who said they were allergic to it? Even if you are 99% sure they aren't really allergic and simply have had completely normal side effects, most providers won't take the chance. They're picturing that mad-dog malpractice attorney again: you gave my client penicillin when you admit she told you she was allergic to it?

Allergies are often reported to non-clinical, unlicensed personnel who are prohibited by policy from interpreting what they hear. The clerk is required to record exactly what the patient says, taking their best guess at spelling and recording the allergy in free text form (DARBOCETEN 100 instead of Darvocet-N 100, which should be recorded as a propxyphene allergy), rendering it nearly useless. It's the electronic equivalent of a Post-It note. It doesn't do anything unless someone finds and reads it, harking back to the days of plastering bright red stickers on the front of the metal chart cover. Order a med to which the patient is allergic and you will not be warned of impending disaster.

Back in our little automation world, let's assume that a patient properly reports a true allergy that is then recorded correctly in codified form by a nurse who looks up the long code number for penicillin. She works in MSGH's ambulatory practice. Where does it go from there?

Ambulatory practices usually have different software than is used for inpatient units, medical practices, clinics, research, nursing homes, and home health agencies. Allergies can be reported in any of these settings. The only way those systems can exchange information is if a common patient identifier exists, often in the form of a master patient index (MPI.) That provides a common subject, but then what of the codified allergy itself? Systems use different data coding schemes: First Data Bank, Multum, or others. Vendors may tweak the codes for their own purposes, in effect making them proprietary. How, then, can those codes be translated from one system to another? If MSGH is like most of the hospitals I've worked with, they haven't cracked this particular nut yet.

Which MSGH system will be the allergy authority, the one we can always trust to be right? Let's say you record my penicillin allergy. Next time, my wife tells you I don't have any drug allergies (she doesn't really want me dead, she's just traumatized by the the distress of the studly male specimen she worships, at least in my version of the story.) Should the staff eliminate the penicillin allergy permanently, assuming it was incorrectly entered the first time? Suppose I visited one of your doc-in-the-boxes since my last admission and reported a latex allergy. How will you sync up with your other systems? Is the best representation of my allergies the sum of all the ones residing in your various systems, or is any one of them the trusted source?

Even if MSGH manages to do the nearly impossible and shares allergy information in real time among its plethora of systems and locations, providers don't always pay attention. There's a pretty good chance you'll be given a drug before anyone remembers to ask about allergies. Your allergy history may be stored in the clinical system somewhere, but that doesn't stop a nurse from pulling a drug from Pyxis or floor stock without checking allergies first, and pharmacists are completely unprepared to follow the JCAHO mandate that they review all orders before doses are given.

The last chance to not harm someone is how physicians respond to allergy warnings, especially those that are appropriate and significant. Somewhere between 80% and 96% of allergy warnings are ignored. Do providers have reason to suspect the information is inaccurate, or are they overwhelmed by insignificant warnings so that the important ones are missed (aka the Cry Wolf Syndrome?) Surely there is rarely a need to order a drug to which the patient is clearly allergic.

Here's a test. Does your hospital own private physician practices, especially those solo or small group offices that don't bear the MSGH logo? OK, then: if I visit one of them this morning and report an allergy, will it flag new orders entered on me in your ED later today? If not, why in the hell not?

I think we use the information systems and vendors we chose as an excuse. Any fool can plainly see that anything less than instant sharing of codified allergies throughout every part of an organization is unacceptable. On the other hand, that's the situation in what I expect is the vast majority of IDNs like MSGH. Certainly we have high-priority projects that always get the resources, but it's a shame that we accept this embarrassing state of affairs and thereby endanger patients by our lack of ability to share something as simple as their allergies. Based on project funding and planning, this is a medium priority at best for most hospitals. If JCAHO understood IT well enough to figure out how widespread this problem is, the entire healthcare industry would be shamed, and for good reason.