Health Language Blog

A Guide to Different Types of Data that Need to be Normalized

Posted on 01/29/13

Practicing internal medicine in an academic medical center, I’ve found that maintaining a concise and complete active problem list for patients is becoming a mark of “good citizenship.” Trying to maintain a problem-oriented medical record certainly isn’t new, but in the days of shadow charts and siloed systems, updating an institutional problem list for patients was mostly a matter of good intentions and meeting accreditation requirements. Now that my colleagues and I share a truly comprehensive EHR, the problem list is something we rely on every day in providing comprehensive and well-coordinated care. We’re also beginning to see the problem list get even more visibility as it becomes the basis for electronic quality measures on our performance and our institution’s.

As the problem list becomes increasingly important, what will stage 2 meaningful use requirements for encoding problem lists in SNOMED CT mean for my colleagues and me? As an informaticist, I’m sure it’s the right move. ICD-9-CM has always been an awkward fit for documenting problems – a diagnostic round hole for a lot of symptom and clinical findingsquare pegs. (An amusing example is found in online discussions about the difficulty of encoding “buttock pain” in ICD-9-CM.) SNOMED CT is a much better fit for the way clinicians naturally document problems. (It even includes “Pain in buttock” - concept id 279043006). However, as comforting as it will be for the billing tail to stop wagging the documentation dog, the business reality is that we still need to bill for our services. And billing requires ICD, not SNOMED CT. So, will the move to SNOMED CT for problem lists mean higher-quality documentation but a more burdensome billing process?

If properly implemented, I think we can achieve the benefits of SNOMED CT-encoded problem lists with minimal transition costs. Anticipating meaningful use requirements, the industry is creating better mappings from SNOMED CT to ICD and back. These mappings should make the transition in encoding largely invisible for my colleagues. Also, as billing knowledge is increasingly computerized, converting documentation into claims should require less effort from clinicians and ancillary staff, allowing them to spend less time seeking compensation and more time on patient care.

Overall, I think the requirement to encode problem lists in SNOMED CT was judicious, but the challenge is on for the health information technology community. In a time of rapid change in health care systems, I’ve seen how seemingly minor regulatory changes can add up to major irritations and inefficiencies in a busy medical practice. Even so, I think the industry will be ready to assist clinicians in capturing higher-quality medical information and using this information to provide higher-quality care.
data normalization


Topics: LOINC, Coding Challenges, RxNorm

About the Author

Dr. Steve Ross, MD is a physician informaticist with Health Language, part of Wolters Kluwer Health. Dr. Ross joined Health Language after 16 years as faculty in the University of Colorado Division of General Internal Medicine, researching personal health records and health information exchanges.