Coding and Billing for Risk Adjustment

Risk adjustment has been used to entice payers and providers to accept patients with multiple chronic conditions along with those patients who are relatively healthy.

Is coding and billing for risk adjustment really any different than what we have been doing all along? It is different in that the codes we assign project how much it will cost us to provide care in the upcoming year, rather than coding for services that have already been provided. 

As part of the value-based purchasing agenda put forth by the Patient Protection and Affordable Care Act (“PPACA”), risk adjustment has been used to entice payers and providers to accept patients with multiple chronic conditions along with those patients who are relatively healthy. The process helps to assure that all parties are adequately reimbursed for the services that might be required by sicker patients.

We have been repeatedly told that with ICD-10 we have the ability to provide greater specificity in acuity reflection through coding. However, since not every code is assigned to a Hierarchical Condition Category (HCC), the documentation should demonstrate the specificity of the condition, as well as the services provided to address this condition. An example of this could be diabetes with complications versus diabetes without complications. The limited number of HCCs are affected by the specificity of the codes assigned, thus, if the patient had diabetic complications we would want to include that specificity in the documentation. The problem is that in many cases coders are held captive by the available documentation, or lack thereof. The lack of quality or specificity in the documentation is what is not new but is an ongoing problem.

Data has become the elephant in the room. We accept that information is being collected and used to perpetuate programs such as value-based purchasing but are not clear how this is achieved nor what part we play in the megadata world in which we live. We are told repeatedly that the codes assigned are aggregated to improve healthcare to the general populace. How does this happen? The co-operating parties are the following:

• American Health Information Management Association (AHIMA)
• American Hospital Association (AHA)
• Centers for Medicare & Medicaid Services (CMS)
• National Center for Health Statistics (NCHS)

These entities use data submitted on claims to identify trends as well as make recommend changes, therefore we constantly strive to assure that the data is correct. Additionally, the data submitted on the claims is used to assign the provider a risk score, or Risk Adjustment Factor (RAF), by the payer. The RAF determines the payment to the physician to provide services in the upcoming year.  Codes qualifying as HCCs must be documented in a face-to-face visit. Diagnoses documented only in the problem list are not considered to be eligible for inclusion on the claim unless the physician has addressed the problem in some manner. Regardless of whether you use the MEAT (Monitor, Evaluated, Assess, Treat) criteria or another method to assure that the documentation supports the diagnosis, accuracy is important. CMS and other auditing groups also uses this data to identify trends among physician groups. These trends can include the RAF score submitted by physician A as compared to peers in the same group, same geographical area, and/or same specialty to decide whose documentation should receive a deeper investigation.

More than just diagnoses are considered when CMS conducts Risk Adjusted Data Validation (RADV) audits, including such items as checking that the record is for the correct enrollee and the correct calendar year, with a documented date of service for the face-to-face visit. If handwritten, can the documentation be read easily. Was the service provided by a valid provider type with documented credentials or documented physician specialty, and is it signed by the provider? All of this is in addition to the documented diagnosis which supports a valid HCC.

A recent report from the US Department of Health and Human Services (HHS) Office of Inspector General (OIG) indicated that 28 percent of Medicare Advantage (MA) encounters presented with at least one “potential error.” The report goes on to say that CMS fixed many of the errors, thus the overall error rate is about five percent. Some of the remaining data errors were indicated to have the potential to raise concerns about services documented in the data.  The errors were noted to include claims that lacked a beneficiary last name or a valid identifier for the billing provider. When one considers that the encounter data is used by CMS to calculate MA plans’ risk scores, this could have a significant effect on the provider RAF and payment. For the MA plan this can affect their reimbursement from the government. These types of errors are where billing plays an important role. Are all required fields completed?

Have you, as a coder, ever been asked to review a physician’s documentation to add diagnoses that may qualify as an HCC but have yet to be submitted on a claim? This appears to be happening with increasing frequency. 

You may think, “Well, isn’t that what documentation improvement in the outpatient setting is all about?” The answer would be yes and no. Documentation improvement is about clarifying inconsistent or unclear documentation to obtain the best reflection of the patient’s acuity and resource utilization in the documentation, whereas the practice of reviewing chronic conditions to add the appropriate ICD-10 diagnosis code may occur in the absence of clarifying physician documentation. 

We also know that there is an increasing focus from the auditing groups on assuring that codes qualifying as HCCs are adequately documented. If you are adding diagnosis codes based on a chronic diagnosis previously submitted I encourage you to assure that not only is the documentation present to support this addition but those codes already submitted are supported.

Some organizations have been known to send notices to the physician that certain diagnoses have yet to be added to the documentation for a given year. While this may sound like the physician is being queried for documentation clarification, the forms I have seen are usually much more direct than would be allowed in a typical physician query. A more acceptable way to gather this information might be to ask the physician to review the problem list and either confirm or delete diagnoses by including the documentation during the next encounter. We do want the physician to get credit, in terms of having an accurate risk score or RAF (Risk Adjusted Factor) for all services provided, but do not want to encourage the addition of diagnoses that have not been addressed during the face-to-face encounter. 

It is the need for accurate documentation and coding that has led to the increasing movement to outpatient documentation improvement programs through the recognition that the need for accurate data has moved beyond the hospital into to the outpatient arena. Early review of documentation allows for time to clarify documentation to obtain the highest appropriate RAF.

It is my opinion that as the focus on value-based purchasing and quality indicators expands in both inpatients and the outpatients we will continue to see a growth in clinical documentation improvement (CDI).

Comment on this article