To prepare confidently for the transition to ICD-10, an organization should undertake a self-analysis that includes modeling actual claims data and drilling down to assess potential risk by physician provider.


At a Glance
  • Rice Memorial Hospital assessed its financial risk posed by the transition to ICD-10 by reviewing records, analyzing subsets of claims in all risk categories, and ten-coding at-risk claims to identify documentation deficiencies.
  • An important step was an in-depth review and validation of the original ICD-9 coding and resulting DRG assignment for each of the 250 inpatient medical records included in the assessment.
  • Through the assessment, Rice Memorial identified opportunities for clinical documentation improvement (CDI) and developed a CDI road map.

Imagine you are sitting in a canoe on a peacefully flowing river. Looking ahead, you see the froth of white water and you hear a low roar. You paddle on cautiously, committed to face whatever perils await you. Suddenly, you are plunged into the rapids—and find yourself faced not with moderately difficult class II rapids, as you had hoped, but with highly difficult class V rapids, which are only for the brave of heart. You could react in one of two ways: Let yourself be swept helplessly along, or set your mind with determination to paddle against the current and navigate through the cataract, reacting quickly to its obstacles and perils. 

The second response is the same kind of mindset that is required to navigate the substantial changes occurring in health care today—particularly those associated with the transition to ICD-10.a

Consider the case of Rice Memorial Hospital in Wilmar, Minn. The revenue cycle leaders at this Level 3 trauma center, the largest municipally owned hospital in the state of Minnesota, were concerned about the hospital’s risk of being submerged by the ICD-10 “rapids” that lay ahead. Would the organization have the determination and resources to navigate this fast-approaching transition safely? The revenue cycle leaders understood that fundamental change was needed in Rice Memorial’s revenue cycle—and specifically in clinical documentation—to effectively accomplish the transition and stay on course after Oct. 1, 2014. But they needed to show executive leaders proof that this need was real.

The revenue cycle leaders had considered predictive modeling as an option for dealing with the challenges that ICD-10 would bring. Predictive modeling generally uses mapping tools to develop a crosswalk to the ICD-10 codes for one year’s worth of ICD-9 claims to determine what the claims would look like under ICD-10. Depending on the circumstances, a facility can use data from the Medicare Provider Analysis and Review (MedPAR) file administered by the Centers for Medicare & Medicaid Services (CMS) for this purpose, or it can use its own previous year’s claims data for mapping and analysis. 

This mapping process allows claims to be created, grouped, and analyzed against the original ICD-9 claims.

However, to make Rice Memorial’s executive leaders aware of the specific financial risk the organization faced in the move to ICD-10, it was necessary to move beyond the limitations of MedPAR data. Modeling actual claims data and drilling down to assess potential risk by physician provider as well as by diagnosis and procedure code was the only way to expose the true financial risk.

Because speed was also a critical concern, Rice Memorial’s revenue cycle leaders enlisted the help of a coding and clinical documentation technology company to assemble a project team that would perform the necessary analysis.

Understanding that the highest-risk DRGs may not be the top 10 or even the top 20 grouped DRGs, the project team decided to take a nontraditional approach, which involved reviewing actual records, analyzing subsets of claims in all risk categories, and ten-coding at-risk claims (i.e., coding from medical record documentation using ICD-10) to identify documentation deficiencies that occurred as a result of the additional specificity and granularity required in ICD-10. 

The team analyzed more than 250 of Rice Memorial’s clinical documentation records to determine inefficiencies under ICD-10 over the course of six weeks. During this time, the team identified MS-DRG shifts with potential impacts on the financial health of Rice Memorial, as well as problem claims that illustrated documentation gaps due to unspecified codes or insufficient documentation.

The assessment provided Rice Memorial’s revenue cycle leaders with a means to obtain compelling data—to be shared with the executive team—regarding not only the magnitude of the organization’s financial risk, but also the opportunity for mitigating this risk.

Documentation-Focused Risk and Opportunity Assessment

The project team selected its target sample of 250 inpatient medical records from a base of 1,114 inpatient records. The team employed a proprietary risk identification process that blends predictive modeling using general equivalency mappings (GEMs) from ICD-9 to ICD-10 with results from a repository of more than 3,000 cases natively ten-coded by ICD-10-CM/PCS trainers approved by the American Health Information Management Association (AHIMA). The results of the assessment provided a basis for developing a clinical documentation improvement (CDI) road map.

Exhibit 1

Hinderks_Exhibit1

Sample demographics. Roughly 46 percent (108) of the 235 MS-DRGs represented in Rice Memorial’s inpatient record base were represented in the target sample. When the case mix index (CMI) of the record base was compared with the CMI of the target sample (1.3493 versus 1.6047), the risk stratification was apparent: The most financially significant MS-DRGs were reflected in the target sample. Rice Memorial’s project team recognized that clinical documentation in these cases must be adequate to support ICD-10 coding that would result in a similar or better CMI, so the recommended target sample was approved as the focus for the project.

The ten-coding process. Rice Memorial recognized that clinical documentation is not the only factor that affects coding and DRG-assignment accuracy, and that the ten-coding process would need to systematically categorize all other factors, including coding issues and differences between ICD-9 and ICD-10, that created variances between claims coded under the two systems. Looking at all factors enabled Rice Memorial to identify and focus on specific CDI opportunities while simultaneously gaining insight into other ICD-10 risk-mitigation opportunities, such as targeted and structured ICD-10 education for the coding team.

As part of this process, original claims data with ICD-9 codes and resulting MS-DRG assignments, along with complete medical records, were made available to the project’s coding team so that team members could review and validate the original ICD-9 coding and resulting DRG assignment. This step was deemed critical in that it enabled the team to properly categorize opportunities for mitigating risk as being related to ICD-9 versus ICD-10. 

Exhibit 2

Hinderks_Exhibit2

For example, if a case had not been completely and accurately coded in ICD-9 and the ten-coding process resulted in a different DRG assignment, there was no way of knowing whether the DRG shift was directly related to ICD-10 without first assessing the ICD-9 coding. It was critical to know for certain whether ICD-10 presented clear CDI opportunities with financial implications for the present (i.e., in correcting existing coding issues) and future (i.e., based on the changes inherent in the new system) versus only the future. To get the most bang for its CDI buck, Rice Memorial was intent on prioritizing its CDI efforts to take advantage of opportunities with both present and future financial significance. 

Results analysis. To achieve the project’s primary objective, analysis of results focused on coding or DRG variances that were categorized as offering some type of CDI opportunity. In particular, the team invariably performed detailed reviews of all cases where incomplete, imprecise, inconsistent, unclear, or illegible documentation had made it impossible to assign diagnosis or procedure codes in ICD-10—either at all or to their highest level of available specificity.

The analysis also included a statistical assessment of Rice Memorial’s overall assignment of nonspecific codes, also known as not-otherwise-specified (NOS) codes. Drilling deeper into these statistics, Rice Memorial was able to identify its top-10 conditions where CDI could result in code assignments that would be not only more specific (the primary reason for the transition to ICD-10), but also, in some instances, significant from a grouper and financial standpoint. 

Armed with case-specific and provider-specific examples of CDI opportunities and baseline performance benchmarks, the project team was able to develop a CDI road map for consideration and approval by Rice Memorial’s executive team.

CDI Road Map

Building upon what was learned from the documentation-focused risk and opportunity assessment, Rice Memorial’s CDI road map focused first on implementation of a collaborative process between the coding and clinical teams. The hospital initiated weekly “coding and clinical documentation huddles,” where assessment findings could be explored and examined for coding and CDI opportunities.

Rice Memorial’s chief medical officer was brought on board as the physician champion to participate in the weekly huddles and help shape physician education for ICD-10. By expanding the “huddle” concept, which is traditionally an approach used by clinicians, to coding and clinical teams, Rice Memorial signaled the importance of broader team involvement for navigating the complex ICD-10 transition. The weekly coding and clinical documentation huddles helped break down silos within Rice Memorial. And with physician engagement from the start, the combined coding and clinical meetings provided a means for improved understanding of how best to communicate with physician groups while being sensitive to clinician time constraints. 

Another element of the CDI road map was the creation of a common vision: to optimize coding and reimbursement by ensuring accuracy of clinical documentation on the front end to better support a complete chart on the back end. To realize this vision, Rice Memorial focused on integration of tools, people, and additional processes to complement the weekly huddles.

Tools. Rice Memorial deployed a clinical “point of care” physician documentation support tool designed to help bridge the gap between the clinical information that is typically documented and the clinical documentation needed by coders for complete and accurate coding. This physician “ontology” tool identifies where gaps exist and guides the clinician by providing options for closing the gaps. 

During multidisciplinary rounding sessions, a member of the CDI team accesses the web-based ontology engine and enters the diagnostic statement found documented in the record—hip fracture, for example. The return display shows the additional elements of information that must be added to the documentation before the proper code can be assigned. These “missing” elements are displayed by category—e.g., laterality, anatomic geometry, etc.—and can be added to the diagnostic statement at the click of a button. Upon completion of the process, the physician is then able to enter a comprehensive diagnostic statement in the patient’s medical record to support the coding process.

Rice Memorial recognized that implementation of this tool not only would help improve clinical documentation, but also would provide a focused approach to ICD-10 training and education for its clinical staff. As clinicians are exposed to or use the tool, they learn the documentation requirements of ICD-10 in the context of their own cases, which helps them to minimize and sometimes even eliminate these gaps in their future documentation.

People. Also critical to realizing a common vision was the need to build a more robust team of clinical documentation specialists. Rice Memorial determined it would be best to capitalize on its current case management structure to build up its CDI staff. Four RN case managers were trained and transitioned to assume a concurrent role as clinical documentation specialists. This approach made the most sense because the case managers were already working with charts daily, reviewing an average house census of approximately 42 charts each, including six to 10 admissions per day. Taking on the expanded role meant reviewing every chart for present-on-admission (POA) conditions, admission and ongoing medical necessity, and adherence to core measures for quality while also looking for documentation improvement opportunities.

Utilizing current case manager staff also would minimize physician frustration because physicians would have only one individual contacting them regarding multiple aspects of care (e.g., medical necessity, quality core measures, and clinical documentation).

Training RN case managers in CDI was accomplished through online learning modules, a CDI tool using dynamic query and tracking, and the weekly coding and clinical documentation huddles, which promote team learning and collaboration.

Processes. Once training of clinical documentation specialists was under way and physician engagement was in full swing, Rice Memorial was able to focus on other clinical areas and DRGs with the greatest chance for improvement, as indicated by the audit, including many unspecified codes around anemia, heart failure, respiratory failure, hemorrhage, laterality, psychosis, and major depressive disorder.

Coding staff and the trained documentation specialists have expanded work beyond the audit and assessment findings to understand essential elements of clinical documentation importance. These efforts have included focusing on body-system groupings and systematically planning meetings around particular groupings, such as cardiovascular or neurologic. 

The first physician group to become engaged in Rice Memorial’s newly refocused ICD-10 transition is a group of five internal medicine hospitalists. Based on the assessment, it was possible to develop customized, preliminary education for each hospitalist based on the hospitalist’s individual clinical documentation performance. Drawing on specific examples of how these hospitalists have improved their documentation, Rice Memorial has been able to highlight real-world opportunities for improved documentation.

Next Steps

Rice Memorial remains committed to the continued rollout of its tools-people-processes approach. Nonhospitalist physician groups will be educated and provided with unique documentation opportunities based on the audit findings. These steps will complete Phase I of the CDI implementation road map. Looking ahead to Phase II, Rice Memorial plans to roll out a fully integrated, electronic CDI query and tracking process, thereby expanding its focus on the use of tools and technology for clinical documentation improvement. With that, Rice Memorial expects to leave the class V rapids behind for much smoother sailing.


Jackie Hinderks, FHFMA, MBA, is director of revenue cycle, Rice Memorial Hospital, Willmar, Minn., and a member of HFMA’s Minnesota Chapter.

Jessica Vagle, MA, APRN, CNS-BC, FNP-BC, is director of adult health and care management services, Rice Memorial Hospital, Willmar, Minn..

Jill Wolf, RHIT, CCSis vice president of compliance, VitalWare, Yakima, Wash., and a member of HFMA’s Southern California Chapter.


footnotes

a. The International Classification of Diseases, 10th Edition, Clinical Modification/Procedure Coding System (ICD-10-CM/PCS).

Publication Date: Monday, February 03, 2014

Login Required

If you are an existing member, please log in below. Username and password are required.

Username:

Password:

Forgot User Name?
Forgot Password?

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:

Email:

   Become an HFMA member instead