HI530 Unit 4 Use Case Analysis Example

Same Drug, Same Strength, Same Form? A Unit 4 Use Case Analysis Fitting a Vocabulary to Medication Reconciliation at Admission and Choosing the Right Level Within RxNorm

[Student Name]

Purdue University Global

HI530: Clinical Terminologies, Vocabularies, and Information Standards

Unit 4 Use Case Analysis

[Instructor Name]

[Date]

The patient, Tolland Ridge Health and the medication details are composite; the vocabulary features and sources cited are real.

The Use Case

Mr. K., a composite 70-year-old, is admitted to Tolland Ridge Health's medical unit with a heart failure exacerbation. Three sources describe his medications. In the emergency department, he tells the nurse he takes "metoprolol, the small white one, twice a day," lisinopril, a water pill and a baby aspirin. The outside pharmacy's fill history, received through the hospital's medication history service, lists the drugs he has picked up in the past six months, each identified by its product code. And the admitting physician writes orders, including metoprolol succinate extended release 25 milligrams once daily. Medication discrepancies at transitions are common, and structured reconciliation is one of the most studied ways to reduce them (Mueller et al., 2012).

The Workflow and Its Questions

The reconciliation has four steps, each defined by the question it must answer.

1. Collect: what medications does each source say the patient takes?

2. Normalize: can each item be expressed in a common form so items from different sources can be compared?

3. Compare: is this the same drug, at the same strength, in the same dose form, across sources?

4. Decide: for each difference, is it intended, such as a deliberate dose change, or a discrepancy to resolve?

Step 3 is where the vocabulary choice succeeds or fails. The comparison needs to recognize two items as the same when they are clinically the same, and as different when they differ in anything that matters to the patient.

Candidate Vocabularies

Three candidates were tested against step 3.

1. Product codes as they arrive on the fill history. Each code identifies a specific product from a specific labeler in a specific package. Mr. K.'s metoprolol tartrate 25 milligram tablets were filled twice from two different manufacturers, so they carry two different product codes for the same clinical drug.

2. Free text with local codes, as the patient interview and older order entries produce them. "Metoprolol" from the interview matches neither of the other sources without a person reading it.

3. RxNorm, the National Library of Medicine's normalized drug vocabulary, which links product codes from many sources to normalized concepts at several levels: the ingredient, the clinical drug with ingredient, strength and dose form, and branded products (Nelson et al., 2011).

Testing the Levels

1. At the product code level, the two metoprolol tartrate fills appear to be two different drugs, and the comparison flags a false discrepancy between them. Every refill from a new manufacturer creates another one.

2. At the ingredient level, all three metoprolol entries, the two tartrate fills and the succinate order, collapse into one concept, metoprolol. The comparison reports a match. But tartrate immediate release twice daily and succinate extended release once daily are not interchangeable at the same dose without a decision, and the admitting order changed both the salt form and the schedule.

3. At the clinical drug level, the two tartrate fills map to the same concept, metoprolol tartrate 25 milligram oral tablet, so they match each other. The admitting order maps to a different concept, metoprolol succinate 25 milligram extended release oral tablet. The comparison correctly flags one real difference for the pharmacist to resolve.

The product code is too fine and the ingredient is too coarse; only the clinical drug level answers the question the comparison step actually asks.

Fit Table

1. Product codes: collect, yes, arrives on the fill feed; normalize, no; compare, fails with false discrepancies; decide, adds noise.

2. Free text and local codes: collect, yes, from the interview; normalize, no; compare, requires manual reading; decide, slow.

3. RxNorm ingredient level: normalize, yes; compare, misses salt, strength, form and schedule differences.

4. RxNorm clinical drug level: normalize, yes, from product codes through RxNorm links and from text through a matching tool with pharmacist confirmation; compare, flags true differences and ignores manufacturer changes; decide, presents a short list of real discrepancies.

Why the Difference Matters Clinically

The metoprolol discrepancy is not a technicality. Metoprolol tartrate is an immediate-release form usually taken twice a day, while metoprolol succinate is an extended-release form taken once a day, and heart failure guidelines favor the succinate form. The admitting physician may well have meant to switch Mr. K. to succinate, which would be an intended change. Or the physician may have chosen the familiar heart failure product without noticing that the patient took tartrate at home, which would leave the change unexplained in the discharge instructions and confuse the patient, who might take both. Only a comparison that sees the two as different concepts puts that question in front of a pharmacist. At the ingredient level, the question never appears, and the discharge medication list could carry the error home.

What the Recommendation Requires

Choosing RxNorm at the clinical drug level is not enough by itself. Three things must also be true. First, the fill feed's product codes must be translated to clinical drug concepts on arrival, using RxNorm's links, and any code that does not map must be shown to the pharmacist rather than dropped. Second, the patient interview must be captured in a way that can be normalized, ideally by a pharmacy technician selecting from a drug list while talking with the patient, so "the small white one" becomes a specific concept or an open question. Third, the comparison display should group items by ingredient but flag differences in strength, salt, form or schedule, so a pharmacist sees at a glance that all three metoprolol entries belong together and that one of them differs.

Remaining Gaps

Some medications in Mr. K.'s list fall outside every source. His baby aspirin was never filled at a pharmacy, so it appears only in the interview. His water pill, which he could not name, matched furosemide in the fill history only because there was one diuretic there. The vocabulary cannot solve incomplete sources; it can only make the comparison of complete ones reliable. Interoperability texts make the same point about every terminology: it standardizes meaning only for the data that are actually captured (Benson & Grieve, 2021).

Recommendation

Adopt RxNorm at the clinical drug level as the comparison vocabulary for admission reconciliation, translate fill history product codes to it on arrival, capture interview medications through structured selection, and display differences grouped by ingredient. Measure the share of reconciliations with at least one flagged false discrepancy before and after the change, with a target of a 75 percent reduction.

References

Benson, T., & Grieve, G. (2021). Principles of health interoperability: FHIR, HL7 and SNOMED CT (4th ed.). Springer.

Mueller, S. K., Sponsler, K. C., Kripalani, S., & Schnipper, J. L. (2012). Hospital-based medication reconciliation practices: A systematic review. Archives of Internal Medicine, 172(14), 1057-1069. https://doi.org/10.1001/archinternmed.2012.2246

Nelson, S. J., Zeng, K., Kilbourne, J., Powell, T., & Moore, R. (2011). Normalized names for clinical drugs: RxNorm at 6 years. Journal of the American Medical Informatics Association, 18(4), 441-448. https://doi.org/10.1136/amiajnl-2011-000116