Our nation’s healthcare system is currently undergoing a metamorphosis. The industry is shifting to quality motivated care delivery, patient-centered care, and performance driven payment. One of the many steps in the overall transition is our change to the ICD-10-CM coding system, set for implementation on October 1, 2015. This up-to-date coding set will provide a significant improvement over the ICD-9-CM set, which is unable to accurately describe the diseases and diagnoses of the 21st century.
ICD-10-CM will allow for greater clinical specificity which translates into improved ability to measure healthcare services, increased understanding when refining groupers and reimbursement structures, the ability to monitor health trends, and less need to include supporting documentation with claims. The codes will also provide better data for measuring patient care, conducting medical research, making clinical decisions, designing pay schedules, processing claims, and identifying fraud and abuse.
As the implementation of ICD-10-CM draws closer, healthcare providers are looking for solutions to transition as seamlessly and painlessly as possible. One of the common misconceptions providers make when preparing for the roll out of ICD-10-CM is that mappings or crosswalks will be a viable shortcut tool for translation of ICD-9-CM codes to ICD-10-CM codes post implementation. In order to understand the practicality, or lack thereof, of mappings and crosswalks we must first understand their purpose and scope.
The currently developed ICD-10-CM General Equivalence Mappings (GEMs) were created by the Centers for Medicare & Medicaid Services (CMS) and the Centers for Disease Control and Prevention (CDC). The GEMs act as a translation dictionary to bridge the language gap between ICD-9-CM and ICD-10-CM, primarily in settings where large data exchange occurs such as research, informatics, and internally with data analysis. Because the GEMs are more complex than a simple one-to-one code crosswalk, they are not a viable solution for coding home health or hospice claims or the Outcome and Assessment Information Set (OASIS) post ICD-10-CM implementation.
Below are some of the primary reasons the home care and hospice field should not rely on crosswalks or GEMs:
1. GEMs were developed to meet a specific, limited, short term need. This includes translating code lists or tables used by an application or other coded data. They are also primarily intended to be used when a code in one code set is the ONLY source of information you have. In home health and hospice care, we code from patient records that describe the illnesses, complications, manifestations, and injuries. GEMs cannot take all of this information into consideration when pointing to a diagnosis code.
2. GEMs simply provide a linkage between a code in one set and its closest equivalent in the other code set, without consideration of context or patient encounter information. In ICD-10-CM, we will have to select a 7th character that represents the initial or subsequent encounter, or perhaps the sequela of certain diagnoses. GEMs would simply point to a cluster of diagnoses and would not have the ability to choose the appropriate diagnosis.
3. There are not always one-to-one code translations from ICD-9-CM to ICD-10-CM in some instances, including: when there are new concepts in ICD-10-CM, when there is no matching code in the GEMS, or when there may be multiple codes in one set for a single code in the other.
For example, ICD-9-CM code 820.02, Fracture of midcervical section of femur, closed has no direct one-to-one code translation to ICD-10-CM. There are actually 30 different choices in ICD-10-CM for this same fracture, and the correct code depends on whether:
- The fracture is on the right or left side
- It is displaced or not
- The fracture is healing routinely or normally versus delayed healing
- There is a non-union or malunion of the bone
- This is an initial or subsequent encounter (subsequent encounter replaces the aftercare codes from ICD-9-CM)
Sometimes one ICD-10-CM code is represented by multiple ICD-9-CM codes. In these cases, there is no one-to-one translation in the GEMs. For example, the ICD-10-CM code E11.41, Type 2 diabetes mellitus with severe nonproliferative diabetic retinopathy with macular edema is represented by three ICD-9-CM codes:
- 250.50: Diabetes with ophthalmic manifestations, type II or specified type, not stated as uncontrolled
- 326.06: Severe nonproliferative diabetic retinopathy
- 326.07: Diabetic macular edema
4. The GEMs are not intended to be a substitute for using ICD-10-CM directly. The code sets, whether you are using a coding book or an online look up system, should be used to look up the correct code based on the information you are given in the health records you receive from the physician or discharging facility. We are to use the patient’s health records to choose codes that represent the primary focus of care as well as to the highest level of specificity known. GEMs simply cannot translate documentation into codes. Additionally, the GEMS cannot be used as a substitute for learning how to use the ICD-10-CM code set. Like the rules that govern advanced math, ICD-10-CM has certain rules, called conventions and guidelines, which must be followed in order to arrive at the correct answer. Coding professionals must be able to review documentation, and assign codes within the confines of these guidelines and conventions in order to accurately code for reimbursement and data reporting, such as OASIS data.
5. GEMs cannot be used as simple crosswalks to get from one ICD-9-CM code to one ICD-10-CM code.
6. Using GEMs would be more time consuming and less accurate than simply looking up the codes in an ICD-10-CM code book. The lessons we learned from the Canadian implementation in 2001 indicate that we can expect an approximate 50 percent decline in productivity when the transition occurs; so any timesaving measures that can offset this backlog will be critical. And since accuracy is equally important, using GEMs is generally not a practical solution.
7. CMS directs coders to use coding books or encoder systems rather than using the GEMs when ICD-10-CM is implemented on October 1, 2015. CMS has indicated that GEMs may be useful in some scenarios, such as early exposure to the new coding set; however, because of the possible inaccuracies in translation, CMS directs us to use coding books or electronic encoding systems beginning October 1, 2015. CMS also recommends that agencies use dual coding, or look up diagnoses both in ICD-9-CM and ICD-10-CM now through ICD implementation. This will give coders the practice they need to increase productivity and become proficient in their most commonly used codes.
In closing, the best strategy for agency operators to take in preparing and planning for the change to ICD-10-CM is to invest the time and resources required to learn to code properly, ensure enough coding staff is available to prevent backlogs in OASIS and claims processing post implementation, focus on accurate and timely documentation from referral sources and clinicians, and practice dual coding until implementation on October 1, 2015. These strategies will directly translate into a successful transition for your agency.
REFERENCES:
“General Equivalence Mappings Frequently Asked Questions,” Department of Health & Human Services, Centers for Medicare & Medicaid Services, Medicare Learning Network article, ICN 901743, Published August 2014, downloaded May 2015.
“General Equivalence Mappings ICD-9-CM to and from ICD-10-CM and ICD-10-PCS,” Second in Series, Official CMS Industry Resources for the ICD-10 Transition, www.cms.gov/ICD10. Downloaded May 2015.
“Putting the ICD-10-CM/PCS GEMs into Practice”, Sue Bowman, Rhonda Butler, Kathy Giannangelo, and Nelly Leon-Chisen, Rita Scichilone,
library.ahima.org/xpedio/groups/public/documents/ahima/bok1_050190.hcsp?dDocName=bok1_050190. Downloaded May 2015.
“ICD-10-CM/PCS An Introduction”, Official CMS Industry Resources for the ICD-10 Transition, www.uth.edu/dotAsset/2409977.pdf, downloaded May 2015.
