Maximizing Confidence in the Purchase Decision: Assessing Existing Patient Data (Step 3 of 3)

You’ve run test claims, and confirmed medical necessity documentation processes are reliable, but still you’re not sure enough. Some medical or business directors find the statistics on prevalence of allergy hard to believe, clear and compelling as they are. Even accepting the truth of the national statistical prevalence, those stats don’t necessarily hold true in any one particular clinic with its local population and local influences. Clinics in the lush coastal Northwest face a different situation than those in high deserts, for example, even before considering patient population. Luckily, there’s yet another straightforward way to build confidence based on your own clinic’s data alone: see how many of your own actual patients already qualify for allergy testing. That’s as easy as running an EHR report, and it will benefit the practice as much as the patients who show up on that report. Once you have that count, comparing cost to revenue is straightforward.

Checking for Prior Diagnosis Codes

Confirming the (almost certainly large) number of patients in your particular clinic who would already qualify for allergy testing is a simple matter of checking their existing records. Said differently, understanding more precisely how anticipated reimbursement for allergy CPT codes maps to ICD-10 codes already in your patient database is pretty easy. There’s a list of ICD-10 codes associated with potentially allergic symptoms, and the likelihood is quite high that regardless of clinic location, existing records will confirm more than sufficient basis to support an allergy program.

The step-by-step varies by electronic health records system. Epic at one point called the needed functionality Reporting Workbench or Registries. Athenahealth calls it Population Health. Others call it some variation of Report Writer or Custom Report Builder. Every major system has some version of it, because payers and quality programs require one. More important, careful, skilled, and shrewd clinic operations teams require one. Regardless of the system, the broad strokes are the same. They include:

• Open the report or registry builder and start a new patient list, not a billing report. You want patients, not claims.
• Usually it’s best to filter by diagnosis, problem, or condition code, not “reason for visit.” Certainly you don’t want to do this by searching free text. Diagnosis filtering pulls from the coded problem list, which has the kind of data governance that makes this report so valuable.
• Enter some or all of the ICD-10 codes above. Most builders let you paste or import a list rather than typing each code individually. If your system really does take only some small number of codes at one time, though, it’s still not that many to enter (44 total).
• You’ll likely have a choice about how far back to look, including whether to limit to active cases, or, alternatively, to consider any coded encounter within a defined time window (one to three years is common). Active-only gives you the more actionable list, which is more conservative for present, business case evaluation purposes. A key consideration here is that a lookback window can catch patients whose provider coded the diagnosis once, but never carried it forward onto the ongoing problem list. Ongoing problem lists, of whatever definition or variety, will typically miss such cases.
• For conservatism, again, restrict to active patients. Filter out deceased, discharged, transferred, or otherwise inactive records so the list is one your front desk could actually work.
• For present purposes, the code count alone is what you need. If you do contemplate actually acting on the list, though, you’ll need to pull and add to the report fields like patient name, MRN or date of birth, assigned provider, last visit date, phone number, etc.
• Run it, then de-duplicate by patient. Plenty of people on this list will carry two or three of these codes at once (allergic rhinitis and asthma is a common pair), so count patients, not code hits.

That list is your starting roster: patients already in your system, already coded with a condition that supports allergy testing. Multiply that by a suitable number for net reimbursements, compare against program costs, and make a sound decision based on the ground truth reality of your own clinic.