Structured Data in Medical Transcription
Has Structured Data Become Healthcare’s Monster?
Increased Use of Unstructured Data Could Reduce Physician Frustration
JOHNSON CITY, TN – One of Meaningful Use’s most misunderstood requirements is the extent to which data needs to be captured in a structured format. Does a physician really need to document the entire patient encounter as structured data in order to meet Meaningful Use?
The answer is no, according to The Center for Medicaid and Medicare Services (CMS) – providing the physician is capturing a select number of data points the CMS requires in a structured format within an ONC-certified EHR.
What patient data does Meaningful Use require to be captured in a structured format? Of the twenty-three Objectives of Meaningful Use, only eight pertain to structured data entry into the EHR. Those are:
1. Patient demographics
2. Problem list
3. Medication list
4. Medication allergy list
5. Patient Vitals
6. Smoking status
7. Family health history
8. Lab Results (LOINC format)
(One additional Meaningful Use Objective) – “Record electronic notes in patient records”, specifically states that the patient note can be dictated and transcribed, providing the document is in a searchable format. See list of all 23 Meaningful Use Objectives at conclusion of this article). For the above eight Objectives, those data must be captured within the structured format of the EHR in order to meet Meaningful Use requirements. But all other patient information routinely documented as part of the patient encounter – such as: History of Present Illness, Subjective, Objective, Review of Systems, Social History, Assessment, and Plan, to name a few – can be dictated and transcribed without in any way preventing the physician and clinic from meeting Meaningful Use. The bottom line is that there is nothing in Meaningful Use that restricts healthcare providers from using dictation and transcription to document those sections of the patient encounter not specifically cited as needing to be structured.
Structured vs. Unstructured Data
What is the difference between structured and unstructured data? Simply put, structured data is information captured within a field or format that can be automatically identified by the EHR. The CMS requires certain data to be structured for two key reasons: first, to make it portable to other EHRs or electronic applications; and secondly, to enable it to be associated with standardized code sets and clinical terminologies like SNOMED CT. Structured data is recorded within EHRs via documentation tools, including but not limited to, drop-downs, check boxes, radial buttons, and in limited cases via text entry (such as the entry of like blood pressure measures or patient weight, height, age, etc.).
The other type of data found within EHRs is “unstructured data”, so named because it is not entered in a field or format automatically recognized or identified by the EHR. Examples of unstructured data are the free text notes typed into a text box by a healthcare professional, and transcribed patient notes which are interfaced into the patient record. Unstructured data is often, but not limited to, qualitative information about the patient’s health history or health context that provides additional decision-making support. The Structured Data Monster Since Meaningful Use requires only a limited subset of patient data to be structured, could it be that EHRs are placing an over-emphasis on structured data at the expense of physician efficiency and patient care? Have we created a monster out of the EHR-based clinical documentation workflow, placing unnecessary demands on physicians to structure data that in many ways is better captured in an unstructured, or narrative, format?
Critics of unstructured data would argue that it impedes our ability to collect and analyze the data needed to move our nation toward a more evidence-based approach to healthcare. Data is indeed the engine for driving improvements in healthcare, but wouldn’t it be far easier and faster for physicians to narrate the details of the patient encounter, and then use technology to index and structure the free text for analytics and reporting purposes? A large population of physicians – as many as 30% or more – express on-going frustration with their EHR-based clinical documentation workflow. For many of those physicians, a greater use of dictation and transcription – provided it is in a searchable text format – could represent a faster, easier and less frustrating means of documenting their patient encounters.
In sum, data remains the key to improving our healthcare system. The current emphasis on structuring all data generated via the patient encounter instead of just those data points mandated by Meaningful Use, however, may not be the optimal experience for many physicians or patients. For those physicians experiencing high levels of frustration with their EHR-based clinical documentation tools, dictation and transcription could provide an effective alternative for documenting those parts of the patient encounter not specifically mandated for capture via the EHR’s structured data capture tools.
About the author:
Mark Christensen is CEO of WebChartMD (www.webchartmd.com), a healthcare company specializing in software applications that manage the clinical documentation workflow.