On October 13, 2013, the ICD-10 will go into effect. This code set must be used on all HIPAA transactions as of that date. For physicians, only the ICD-10-CM (diagnoses) will be used. Procedures will continue to be coded under the existing CPT/HCPCS coding systems.

Are you ready for this transition??

The current diagnosis coding system, ICD-9-CM, has approximately 13,000 diagnosis codes. The ICD-10-CM has approximately 67,000. This affords physicians and providers a greater opportunity to provide more detailed information regarding their patient’s conditions. It also presents challenges in documentation, to ensure that more specific codes are supported.

CMS has created an ICD-10 website to help physicians and providers understand and meet the challenges of ICD-10 implementation. It can be found at http://www.cms.gov/ICD10/05a_ProviderResources.asp#TopOfPage.

CMS provides email updates to providers. You can join the email list at http://www.cms.gov/ICD10/02d_CMS_ICD-10_Industry_Email_Updates.asp#TopOfPage

Finally, CMS has conference calls with concerned parties in the industry. Notification, transcripts, slide presentations and mp3 recordings of previous calls are found at http://www.cms.gov/ICD10/02c_CMS_Sponsored_Calls.asp#TopOfPage.

Need to catch up? We’ve posted the mp3 recordings and previous transcripts here on HCC University.

Update on use of the term “diabetes with __________”

As you know, there has been much talk in the industry regarding whether or not a recent (Q3 2008) Coding Clinic had a broader impact than the “diabetes with neuropathy” mentioned in the article.

SCAN was concerned that applying this Coding Clinic article (and a subsequent article on the same topic in Q2 2009) more broadly than neuropathy might lead to errors in a Risk Adjustment Data Validation Study (RADVS) conducted by CMS. Since all health plans and their providers are subject to RADVS, SCAN believes that a cautious approach to coding rules is safest for all parties involved.

Because there were so many questions being raised, SCAN contacted Coding Clinic for their advice regarding conditions other than neuropathy and the term “diabetes with”. We also asked that they publish a clarification of the coding policy, no matter what their determination regarding our inquiry.

Coding Clinic does not allow their ad hoc advice to be published. However, in general terms, we can tell you what we were told by Coding Clinic.

We were informed that the ICD-9-CMS’s alphabetic index subentry term “with” means associated with or due to. So, if the provider indicates diabetes with another condition, it is appropriate to assign the diabetes manifestation codes if it meets these criteria. The letter indicated that the publication of the Second Quarter 2009 advice supercedes advice from Second Quarter 1994 regarding the term “diabetes with.”

We specifically asked if this advice should be considered to apply to services before the publication of the Coding Clinic articles. This part of our question was not answered, so we recommend that the advice only be used going forward, and not in chart review of historical data.

Recently, we received the following question regarding the rules related to diabetes coding:

This is a very difficult question, mostly because a lot of rumors are floating around. We’re hearing on industry calls that there is an absolute change in diabetes coding and that the term “with” now supports a diabetic complication in all cases. We’re also hearing that “someone” from Coding Clinic or AHIMA (which has no authority regarding ICD-9 coding) or CMS has confirmed this AND that it is retroactive. Unfortunately, the “someone” is always unnamed. No health plan has the authority to make rulings on ICD-9 use, that authority rests with Coding Clinic when the Official Coding Guidelines are silent. In this situation, it is important to deal only with the facts.

What is fact is that there is a recent Coding Clinic on Diabetes “with” neuropathy. Basically, the question posed was if the documentation indicates “diabetes with neuropathy”, can it be assumed that the neuropathy is secondary to the diabetes. The answer was yes…but Coding Clinic went on to give a reason, which may be very important. The reason given was that neuropathy secondary to diabetes is so common that the assumption could be made.

This seems very different than “diabetes with CKD”, for example. The National Kidney Foundation doesn’t assume a relationship between diabetes and CKD unless certain conditions exist…eg, diabetic retinopathy, etc. Also, unlike most situations where Coding Clinic has done a complete about face from earlier instructions, there is no acknowledgement that this represents a change in a longstanding rule. Nor does Coding Clinic speak to assuming a causal relationship with any other complication.

We are extremely concerned that these rumors will lead to physicians documenting in a way that CMS will not accept.

We have contacted the CMS and asked them to speak to these issues. CMS has said that they do not know, but will look into this issue.

We believe that if there had already been official confirmation that this ruling applied to all “diabetes with _________” situations, and made it retroactive, CMS staff would have a definitive answer. Since they haven’t given us a definitive answer, we don’t believe it is safe to ignore the previous advice that physician must state a causal relationship just yet. Their lack of definitive response only reinforces our belief that physicians need to be cautious and not accept these rumors as fact.

We will continue to monitor this situation, and await CMS’ response. We’ll post that information here on HCC University as soon as it becomes available.

On January 15, 2009, The Centers for Medicare and Medicaid Services (CMS) issued a final rule in the Federal Register for implementation of ICD-10-CM and ICD-10 –PCS.

The new implementation date is October 1, 2013, giving a 2 year extension for the implementation of these two code sets. The ICD-10-CM (clinical modification) will replace the ICD-9-CM coding system currently in use by physicians, hospitals and clinical entities. The ICD-10-PCS (procedural coding system) will replace the current ICD-9-PCS, which is used only for inpatient hospital services.

The existing ICD-9 code sets are over 30 years old, and can no longer accommodate the expansion required by advances in clinical medicine and causes of injury.

In a related rule, the implementation date for the X12 5010 transaction for claims and encounters is now set for January 2012. The new transaction, required by the Health Insurance Portability and Accountability Act (HIPAA) is required to support the new codes sets. The 5010 transaction rule can be viewed here.

CMS has posted an HHS fact sheet that describes both rules.

Welcome to the newly re-designed HCC University Website. This blog will be updated periodically to help you stay abreast of issues in Risk Adjustment (RA). For our first blog, there is no more pressing issue than CMS Risk Adjustment Data Validation Studies (RADVS) that are currently underway nationwide.

There are two types of RADV- the CY2007 Payment Year CMS RADV at the plan level and the 2007 Payment Year National Sample RADV (National Sample). There are a number of differences between the two samples. In the National Sample, 642 beneficiaries were randomly selected from 158 selected health plans. For each health plan, the maximum number of beneficiaries is 21. In the plan level RADV, CMS will select approximately 200 beneficiaries for each selected health plan.

A more important difference is the purpose of these two samples. The National Sample will be used to estimate a national payment error rate for the Medicare Advantage program to the Office of Management and Budget, as required by the Improper Payments and Information Act of 2002 (IPIA). In addition, it may be used to adjust plan level payment adjustments for the CY2007 CMS RADV.

The CY2007 CMS RADV will be used to calculate a contract level payment adjustment. Although no mechanism for this has been shared with health plans, CMS has made it clear that they intend to use the errors found in this sample to extrapolate an overpayment across all of a health plan’s membership.

What does this mean for an affected health plan and its providers? CMS has given plans a very short turnaround time --12 weeks and no extensions. That means that medical groups, physicians and hospitals may receive two sets of requests from a health plan or medical record retrieval vendor. These are not duplicate requests. Each of these audits is separate, and has a different member list and time frame.

As in years past, CMS requires that plans select the “one best medical record” to support each HCC in the sample. An affected health plan should pursue all possible medical records, and apply both CMS and ICD-9 coding requirements to select the one best record for submission to CMS.

We recognize this is a burden at a very busy time of the year, but failure to provide the needed medical records leads to higher error rates and a larger potential negative impact on revenue for plans and providers. If you have questions about the data validation studies, please contact us at coding@scanhealthplan.com.