A Pittsburgh-based MD weighs in on an emerging area of focus in the healthcare industry.

EDITOR’S NOTE: The following are remarks made by the author during a recent broadcast of Talk Ten Tuesdays.

Today I would like to focus on outpatient clinical documentation improvement (CDI), often referred to as the lowly stepchild of inpatient CDI efforts. 

Personally, I feel it (outpatient CDI) can take over the spotlight from inpatient CDI, as it encompasses both the inpatient and outpatient world.  Outpatient CDI programs are directed primarily by primary care providers, which are increasingly getting involved in shared savings agreements, Advanced Payment Programs (APPs), and the Merit-Based Incentive Payment System (MIPS). However, many specialists are asking to be involved as well.  

One of the biggest differences between inpatient and outpatient CDI is that in the former arena, providers actually care about what you are teaching them. Providers who are engaged in risk-bearing contracts know that it is up to them to document and code the diagnoses that are relevant to the risk model. What a provider documents and codes will determine how successful they are in these risk-bearing contracts. But how can a provider remember all the things they need to document and code? 

At the University of Pittsburgh Medical Center (UPMC), we have risk-bearing contracts for Medicare Advantage, Medicaid, and the Patient Protection and Affordable Care Act (PPACA) plans, in addition to MIPS. It is impossible for any provider to know which diagnoses are relevant for each model, or to know if the diagnosis has been captured for the calendar year.  

In addition, approximately 80 percent of our outpatient visits are coded by the provider, not a coder. As we all know, providers, especially physicians, are not taught coding and documentation rules in their training. We are trying to remedy that at UPMC, but it would be impossible to rely on providers to remember the relevant diagnoses for each risk-based model, and to ensure that they get on a claim.  

One way to address this is to give the provider a list of diagnoses that have been coded in the past for their patient. This is helpful, but many providers find it annoying to have to deal with paper or electronic alerts for diagnoses that may or may not be relevant for the current visit. 

Most providers get 15-20 minutes to spend with a patient, and we have heard loud and clear that they want an easy-to-use tool to help them document and code their visits appropriately. In response to this, we have created a tool to help our providers. The tool will be embedded in our outpatient electronic medical record (EMR), and when a provider opens the encounter, it will reveal the relevant diagnoses for their patient, show them where the diagnoses came from, and make it easy to document the Monitor, Evaluate, Assess, and/or Treat (MEAT) and put the diagnosis on the claim. We will be going live with this system in the next eight weeks. 

Even with such a tool, it is always important to educate your providers about the risk models. For our doctors, the Centers for Medicare & Medicaid Services-Hierarchical Condition Category (CMS-HCC) model is the most common risk model used. 

I would like to end with an outpatient CDI tip for your providers: CKD 3 is once again an HCC! It has a very small weight, but considering how common CKD 3 is in our population, this is a gift to many providers. 


Adele L. Towers, MD, MPH, FACP

Dr. Towers is the senior clinical advisor for UPMC Enterprises. She is directly involved in the development of healthcare-related technology, with emphasis on use of Natural Language Processing (NLP) for risk adjustment coding and use of clinical analytics to optimize clinical performance. Prior to this role, she has served as the medical director for health information management (HIM) at UPMC, with responsibility for clinical documentation improvement as well as inpatient coding denials and appeals.

You May Also Like

HCCs: The Role of CDI and Risk Scores

HCCs: The Role of CDI and Risk Scores

Predicting coding patterns using the HCC risk scores can be a valuable endeavor. EDITOR’S NOTE: Longtime RACmonitor contributing correspondent Frank Cohen, a senior healthcare analyst,

Read More

Leave a Reply

Your Name(Required)
Your Email(Required)