Since its early beginnings in outpatient specialty areas, Computer-Assisted Coding (CAC) solutions have been adopted to improve medical coding workflows, increase medical coding accuracy, and balance medical coding resources to focus on more volume and complex cases. The recent and significant increase in the adoption of CAC solutions in inpatient environments is linked to the same, compelling, justification benefits. However, it is directed to meeting newer challenges. Such challenges include the eventual conversion from ICD-9 to ICD-10, which increases not only the number of medical codes but the complexity of medical coding, and the ARRA/HITECH Medicare and Medicaid Incentive Program for the adoption and meaningful use of certified electronic health record (EHR) systems by eligible physicians, other clinicians, and hospitals, which exponentially increases the volume of structured clinical documentation and calls for the eventual use of SNOMED-CT for encoding at least the record’s problem list.
In an effort to meet the newer challenges, CAC solutions must be able to fit within larger health information technology (IT) strategies. To accomplish this, healthcare organizations must obtain a working knowledge of key CAC solution features and functions.
Automatic Medical Code Generation / Suggestion consists of the plethora of software functions that automatically generate or suggest medical codes for review, validation, and use based on the clinical documentation provided by healthcare practitioners. Such functions include searches for specific diagnostic statements and unique anatomical site acronym terms and/or abbreviations within any document from any medical record product. Such documents can consist of structured data driven by healthcare providers documenting care in an organization’s EHR system(s) or unstructured data driven by healthcare providers documenting care in an organization’s complementary EHR systems, such as text data generated by an organization’s Dictation / Transcription / Speech Recognition (Voice / Text / Speech) systems.
Also, such functions include identifying both ICD-9-CM and ICD-10-CM/PCS Present on Admission, principal, and secondary diagnoses and procedures for hospital inpatient documents, CPT-4 principal and secondary procedures for ambulatory (office outpatient) documents, and even nomenclature codes, such as SNOMED-CT, RxNorm, and LOINC for clinical, pharmaceutical, and laboratory documents, respectively. Being able to have code suggestions occur at the point-of-care, as documentation is entered into the record, is noteworthy.
Automatic Data Abstraction consists of the software functions that automatically extract coded data elements from the patient record for any number and type of user-defined purposes. Such purposes include performance monitoring, outcome measures, Joint Commission Core Measures, RAC audits, CDC reporting, clinical registries (e.g., cancer, birth), and hospital acquired conditions.
CAC solutions use either structured data input, as discussed above, or NLP engines, or both. NLP engines use artificial intelligence to identify concepts in the unstructured text data and to associate medical codes from controlled vocabularies to relevant phrases in the text. NLP engines must be able to interpret and combine concepts in terms of morphology, syntax, semantics, and real-world knowledge. Consequently, the accuracy of the NLP engines is, perhaps, the most important feature in a CAC solution. The NLP engine must minimize coding errors (the wrong code has been selected), false positives (the code is selected for which there is no documentary evidence), and false negatives (documentary evidence exists but the code was not selected.)
Maximizing the solution’s benefits requires interfaces or integration with systems far beyond the encoder systems that are commonplace in most healthcare organizations. Such related application systems include all systems through which clinical documentation flows, such as Voice/Text/Speech and abstracting systems. Typically, these systems also include clinical documentation improvement (CDI), business intelligence, and reimbursement analysis systems. For example, interfacing CAC solutions with existing CDI tools will help clinical documentation specialists (CDS) more efficiently identify codes and produce a working DRG, which can accelerate the coding process. And, by identifying data that points to the need for more coding specificity, the CAC solution makes it easier for the CDS or the coder to query the clinician.
CAC solutions, like all software solutions, are complex information systems requiring a substantial investment in time, dollars, and resources. Therefore, it is essential for any healthcare organization to develop a strong knowledge-base of key CAC solution features and functions. Only after obtaining such knowledge will healthcare organizations be able to contribute to the newer challenges of today’s healthcare industry.
Deborah Kohn, MPH, CPHIMS, RHIA, FHIMSS, FACHE, is Principal at Dak Systems Consulting. In 1985, Deborah founded Dak Systems Consulting, a national healthcare information technology advisory consultancy. She has over thirty years of healthcare provider organization management and information technology experience. Deborah has a graduate degree from UCLA in Health Services and Hospital Administration and is board certified in healthcare management.