Succeeding in Risk-Based Contracts: A Guideline for Healthcare Providers (Part 2)

Author:
Shahyan Currimbhoy
Date:
November 19, 2015

Part 2 – Six Steps to Excel at Healthcare Risk Adjustment

Last week, I discussed the state of healthcare risk adjustment and the risk adjustment challenges facing provider organizations. But an important and looming question remains: if high accuracy coding to support risk adjustment is critical to organizations, and existing processes are either retrospective, labor-intensive, error-prone or challenging to adopt, then what can providers do to effectively address this?

In order to excel at risk-based contracts, providers must leverage data analytics and focus on six key capabilities:

  1. Prospective Risk Assessment – The motto here is “do it right the first time” rather than relying on chart auditors to do laborious chart sweeps retrospectively to identify codes that were either missed or miscoded the first time. This is extremely labor-intensive, inefficient and costly as it entails bringing patients back in to see the doctor. In a lot of cases, patients do not come in for their visit, resulting in these conditions not being reviewed and associated codes not being captured. Prospective risk assessment is the process by which providers leverage analytics applications to identify high-risk conditions before the patient’s visit, optimize the patient’s care plan accordingly, and then accurately code the visit. Prospective risk assessment therefore has a dual benefit of accurate coding as well as improved care delivery.

 

Inaccurate Coding Impact
  1. Point-of-Care Integration – In order for clinicians to be able to use these analytic insights at the point of care, it is important that these be seamlessly integrated into their existing workflow (i.e., their electronic health record). To drive high adoption rates, there should be minimal clicks to access this information with sufficient supporting evidence that helps clinicians determine whether it is still an active condition for the patient and what the appropriate course of action would be.
  2. Self-Sufficiency – Currently, providers rely far too heavily on payers for prospective risk assessment. While claims data is beneficial, there are also limitations with it, including: i) it is retrospective, because it is claims-based; and ii) it is incomplete. If conditions were not coded in the first place, then this would not be found in claims. As providers begin taking on increasingly more risk, it will be important for them to control their own destiny. By leveraging powerful new patient data analytics applications, the treasure chest of clinical data within provider institutions and across their clinically integrated networks can be used to mitigate such limitations, and better identify potential high-risk conditions that their patients may have.
  3. Comprehensive Patient Record Analysis – In addition to structured data found within the electronic health record (EHR) or enterprise data warehouse (EDW), there are a number of conditions often diagnosed and documented in unstructured data such as specialist notes, radiology reports, consult notes, etc. With unstructured data comprising up to 80 percent of all healthcare data, it is important for providers to comprehensively mine this valuable data asset.

 

Mining Unstructured Patient Data
  1. High-Accuracy Condition Detection – To alleviate the manual, time-consuming process of sifting through chart after chart, providers should leverage patient data analytics apps that automate this process. When identifying and selecting such a solution, accuracy – both in terms of precision and recall – is vital so as to limit false positives and misses.
  2. Continuous Improvement – Given the criticality of this, as it directly impacts reimbursement and care delivery, it is important that providers continually look for areas of improvement, whether these are the top diagnosis codes that seem to be missed more than others or documentation gaps and best practices that reduce ambiguity and capture condition specificity.

Stay tuned for upcoming blog posts where I cover each of these core capabilities in more detail as well as key features to look for within each of the capabilities to help risk-bearing providers effectively address these.

Shahyan Currimbhoy is the SVP of Product Management & Engineering at Talix.
View all Blog Posts

1

2