Most everyone is familiar with the famous nursery rhyme line “Eeny, meeny, miney, moe/Catch a tiger by the toe,” recited by children looking for a way to select who will be “it” for a game. The undergirding principle of this ditty is selecting one over another, whether that be “eeny, miny, or moe.” This process is not unlike what can occur in the classification assignment when converting from the ICD-9 to ICD-10 code set. Important details can be lost in translation.
Consider the above example of catching a tiger by the toe. If we were to select the details using ICD-9 to document this dangerous encounter today, we would accurately note that we caught an orange and black-striped, furry mammal, with four legs, a long tail, and whiskers – by the toe. However, based on this description, that tiger could also be interpreted as a kitten. While we would all certainly rather try to catch a kitten, it doesn’t convey the same sense of severity or urgency.
Specificity distinguishes meows from roars
A recent study published in Pediatrics reports that 26 percent of ICD-9 codes are convoluted when mapped to ICD-10, which can have a substantial negative impact on pediatricians’ bottom line. It is this level of detail required in ICD-10 that makes general equivalence mappings (GEMS) an unreliable way to prepare fully for the transition. While GEMS does allow coders to see how most general ICD-9 codes will translate to ICD-10, this practice does not account for the level of specificity required under ICD-10, which will require that clinicians record additional identifiers to more fully describe care being provided. For instance in the case of the mistaken tiger/kitten, details around whether the furry mammal is domestic or feral, meows or roars.
Although the ICD-10 transition and the level of associated granularity has caused a lot of anxiety for providers and health organizations alike, the new level of detail captured in the documentation will help to improve quality patient care and enable providers to better manage the health needs of their specific patient populations. Additionally, the lack of specificity available in ICD-9 codes (in relation to that provided by ICD-10) can lead to diagnosis confusion or the misidentification of important patient information.
The ICD-10 impact on technology
The level of specificity doesn’t just hold true for physicians and coders, these details also require that technology keep pace with the level of sophistication required under new coding standards. Natural Language Processing (NLP) engines able to understand clinical narrative, and parse the right level of details using ICD-10 specificity, can be leveraged to meet coding and quality measures accurately. However, if health IT can only recognize and manage those details recorded by clinicians using an ICD-9 lens and discards additional information, the picture changes dramatically from a savage beast to a house pet. Using intelligent systems and tools that accommodate specificity and “learn” which details are needed to ensure that the appropriate information is captured can help ease the burden being placed on both the care teams documenting their patients’ conditions, as well as the coders and clinical documentation specialists who are working to ensure a patient’s story is complete, accurate, and compliant.
Although the reset button has been hit on the ICD-10 countdown clock, keeping the momentum going is important. Strategies that take into account the increased levels of clinical documentation specificity will ensure health care organizations are paid for the high quality care they are providing to their patients and to their community. The danger of not doing so places providers at risk for RAC denials, and in the dangerous position of misrepresenting a tiger as kitten.