ICD-10: Why and What Matters to Long-Term Care Providers

LTC providers have never had to be very knowledgeable about the International Classification of Diseases, 9th Edition—or, as it is better known, ICD-9. This is for a host of reasons, but primarily because provider reimbursement historically has not been tied to ICD coding. There was also a belief that the 10th Edition of International Classification of Diseases (ICD-10) would never happen, given that it’s been long in the making as a result of many delays. But, not only implementation of ICD-10 but also a direct link of coding to reimbursement is really coming on October 1, 2015.¹

ICD-10 has been in the works for years. ICD-10 is a clinical modification of the World Health Organization’s ICD classification system. The ICD-10 code sets are not a simple update of the ICD-9 code set but rather have fundamental changes in structure and concepts. For example, while ICD-9 used a maximum of 5 characters creating some 13,000 codes; ICD-10 now uses a maximum of 7 characters included not only numbers but letters as well resulting in some 68,000 codes.

Of course, it is not just reimbursement that is driving ICD-10; the major drivers are improvement in identification of clinical issues and treatment best practices. This is possible through a more detailed and specific labeling system that is a shared language among all developed countries throughout the world, which already have been using ICD-10 for many years.

As the timeline in Table 1 illustrates, the US ICD-10 started more than 2 decades ago, with four proposed implementation dates leading up to a kick-off date of October 1, 2015. This is truly a drop-dead date, as all services rendered after October 1 must use ICD-10.

In July, Centers for Medicare and Medicaid Services (CMS) announced that they would not deny reimbursement claims for the first year of ICD-10 due to lack of specificity. In conjunction with the American Medical Association (AMA) agreement with CMS to educate AMA members on ICD-10, this policy has made it even more likely that the implementation of ICD-10 will not face additional delays.

As reimbursement becomes increasing tied to performance, the ICD diagnostic coding will contain information about who is being treated and how well they are being treated. As a result, the timing of the move to ICD-10 on October 1, 2015 and the move by Medicare and all payers toward pay-for-performance is truly a perfect storm for those providers who are not prepared—those locked in the fee-for-service model, in which procedure volume is all that matters rather than the quality of treatment being provided.

While providers may not be directly involved in coding, the codes are driven by the provider descriptions in the chart. As such, all providers will need to have a clear understanding of the basis of the ICD-10 codes.

ICD-10 Link to Reimbursement

Several years ago, Medicare changed how managed care organizations were paid. To assure that managed care organizations were reimbursed based on anticipated costs and to eliminate incentives for care providers to avoid caring for the most frail older adults, Medicare moved to a risk-adjusted payment system. This risk adjustment is individual for each plan member and based on submitted ICD codes. These ICD codes are tied to the Hierarchical Condition Category (HCC) system. As a result, the HCC payment system means that two patients within the same community can have a different payment rate based on several factors relating primarily to the amount of risk—or work—it takes to maintain the health of a patient.

To improve care delivery, reimbursement will be based increasingly on the severity of illness and the outcomes of care, and these will be measured by the ICD-10 codes used. These codes will drive reimbursement for not only managed care plans but also innovative provider groups such as Accountable Care Organizations, Bundled Payments, and Skilled Nursing Facilities.  As a result, LTC providers will benefit greatly from assuring appropriate coding occurs.

As Medicare and other payers shift providers from volume procedure–based reimbursement to paying for performance, accurate information regarding who and how well we treat individuals will be critical for ensuring appropriate reimbursement. Understanding the codes is important even for those that do not believe they are involved in coding. Even providers utilizing professional coders must understand the code descriptions so that they can describe their patients’ conditions more accurately in these terms. If providers gloss over details, or provide only the most basic notes such as congestive heart failure or hypertension, they will be missing out on opportunities for more appropriate reimbursement for the higher acuity individuals they successfully treat.

The most efficient and effective approach for providers will be to identify their most prevalent disease states that they manage, ordering these by HCC risk factors will provide an ordered priority list. For each of these items, providers should know the description, or at least the common elements, such that they can document appropriately; therefore, providing the needed direction and support for their coders. This is the only way for LTC providers to assure they are compensated for the work and outcomes that they are achieving. But these results will only be possible for LTC providers that speak the new language of ICD-10. 

Preparing for ICD-10

As LTC providers move into ICD-10, much greater specificity in describing issues is required. This includes describing conditions in the following manner:

1. Laterality (side; i.e., left or right)

2. Stage of Care (i.e., initial, subsequent, sequelae)

3. Specific Diagnosis

4. Specific Anatomy

5. Associated and/or Related Conditions

6. Cause of Injury

7. Documentation of Additional Symptoms or Conditions

8. Dominant vs. Non-dominant Side

9. Tobacco Exposure or Use

Take an examination of a stroke description under ICD-10. ICD-10-CM stroke codes are more specific than their ICD-9-CM counterparts. First, codes specify the location or source of a hemorrhage as well as its laterality. Second, codes specify the type of stroke that caused the sequelae as well as the residual condition itself. For example, code I69.01 denotes cognitive deficits after nontraumatic subarachnoid hemorrhage. In ICD-9-CM, code 438.xx simply denotes the residual condition-not the type of stroke that caused the condition. Also the details regarding the events leading up to the stroke are needed. For example, unlike ICD-9-CM, ICD-10-CM distinguishes intraoperative stroke during cardiac surgery (I97.810) or during other surgery (I97.811). 

Another medical condition that illustrates the specificity of ICD-10 coding is coma. ICD-10 uses the Glasgow Coma Scale (GCS), a neurological scale that captures a patient’s conscious state for initial and subsequent assessment. Here, the description must include the level of responsiveness, such as: 

R40.21: Eye response (eyes never open or eyes open to pain, sound, or spontaneously)

R40.22: Best verbal response (clarity of words: incomprehensible, inappropriate, confused, oriented)

R40.23: Best motor response (voluntary and involuntary responses [extension, flexion, abnormal, obeys commands])

If a physician doesn’t document the GCS—or documents only a portion of it—coders must report R40.244 (other coma, without documented Glasgow coma scale score, or with partial score reported).

Again, these descriptions drive coders’ activity; coders are required to code to the highest degree of specificity, but the specificity must exist in the physician’s documentation in the medical record in order for them to do so. This specificity is essential, as provider reimbursement increasingly is based on outcomes. 

1.    Centers for Medicare and Medicaid Services. ICD-10. https://www.cms.gov/Medicare/Coding/ICD10/index.html. Accessed August 24, 2015.