What many of us are learning lately is that no news is not necessarily good news. It seems that no news about ICD-10 has simply been disguising some of the lurking problems just waiting to surface.
Problems are just starting to appear as payers add edits and issues with electronic health record (EHR) functionality become more and more evident as we try to help providers perform audits. Even more issues jump out when an outside reviewer requests records. In addition, some codes that may appear to be correct to the provider ultimately are shown to be incorrect due to mapping errors in the products upon which they rely. In PCS, proving that the correct code was selected for certain procedures has turned out to be essentially impossible in some cases.
The Medicare Administrative Contractors (MACs) are still correcting and updating local coverage policies. While that is good news, it has added an ongoing operational burden because the implementation is flawed. For example, the official published policy is updated, but all claims in this area were still denied. Several HBMA members reported that their MAC explained that the policy was updated, but the system programming to match had not been completed. We have also been informed that at least one MAC has updated its system, but will not reprocess any claims denied incorrectly. Those must all be resubmitted if any payment is to be received, thus transferring the burden and cost to fix errors to the providers.
Intensive looks into why so many providers are still reporting unlisted procedures have demonstrated over and over that the pick lists that they must choose from in their EHRs are simply too massive to be useable. Quite a few of us have had the experience of sitting down with our providers while they select the diagnosis code for the encounter. In spite of the fact that they know exactly what the diagnosis is, it is a very frustrating process to try to find it. Exactly what word or words do they need to enter to narrow the search or hit the exact code they want? How long does it take? Even when it appears to be correct, is it? Do they receive prompts that they must code something else first, or in addition to the code they select? How would they know if the mapping program had errors than results in an incorrect code?
There can easily be cases of providers not knowing what they don’t know. Obviously, not all EHRs have the same issues, and functionality varies widely. However, a significant number of physicians, coders, and reviewers have reached the same conclusions. Much work remains to correct and improve usability and accuracy. In addition, providers need to have far more coding knowledge than most of them currently possess. What all of these issues mean is that it this process will take time – time that some providers don’t have. Physicians are already incredibly busy. The realities of patient care simply do not allow even more time for administrative functions, i.e. finding the most specific diagnosis code. How much more difficult will this become in October, when the next surge of ICD-10 codes are added?
HBMA members and others recently reported that some Medicare Part C plans are now taking payments back on services previously pre-authorized and allowed, based on the premise that the services were not covered by Medicare. This is directly related to the NCD errors and the huge delay in the correction of implementation dates. Whether these plans will retroactively reprocess the covered services for payment remains to be seen. We are hopeful that CMS will mandate that they must do so, but we have not been made aware of any direction to the Part C plans at this time.
We have also received reports that some commercial plans that follow Medicare guidelines are incorrectly denying claims based on the NCD and LCD errors. Not only are legitimate services being denied based on the errors, the incredible amount of work and expense to the providers to appeal and/or wait for payment at some possible future date is simply unreasonable. There is no doubt that by the time all of the CMS fixes are implemented, the timely filing deadlines will have been long exceeded for many plans.
On the PCS side, we are learning of significant challenges for procedures that have explicit product type information required, for example, prosthetic hip joints. One IRO auditor opined that it would be a Herculean task to validate the correct code, if it was even possible. Although they know the product name/number (i.e. Stryker 12345), knowing exactly what that means is an unknown. Is it porcelain and synthetic? Is it a synthetic substitute? Is it ceramic or ceramic synthetic? The physician operative reports do not include that level of nuance, and the exact prosthetic information is often difficult to find.
In summary, we’ve got some experience now. We know what is working well, but we are identifying more issues that need to be addressed. I believe we all agree that the correct and most accurate coding is the goal. We need to work together to make that a doable outcome in the realities of clinical practice.
Until we work together to resolve the issues identified, we are far from the end game in ICD-10 implementation.