Structured Chart Review Process in an Abstract

As per an abstract shared by the National Library of Medicine, a chart review is a methodology to determine a retrospective approach and healthcare assessments to conduct quality improvement activities.

Risk Adjustment Vendors

However, the reliability and validity of chart review is depending on medical records. The process of chart review can be challenging as it includes unstructured documentation, which demands a structured procedure for review and assessment to resolve discrepancies among multiple chart reviewers for validity and reliability.

The abstract further described mentioned a common-sense multistep approach in accordance with the original Harvard Medical Malpractice Study principles to ensure a standardized structured chart review process.

The initial goal of the structure chart review approach was to define goals and objectives for the chart review process in order to identify pathophysiological (pathology with physiology) processes resulting in new morbidities and mortalities, including therapeutic advances that could reduce morbidity and mortality.

Next, the approach also defined performance parameters to analyze the time taken for reviews. This criterion is considered to ensure that the chart review process is feasible and that the individual behind the chart review process is qualified.

See the steps taken to create the structured chart review process including the examples of the process used in the abstract:

aa 1

Image source: National Library of Medicine

The study included eight primary reviewers across seven institutions by third-year individuals and professors. 

The primary reviewers performed 327 reviews and confirmed new morbidity or mortality in 292 cases. The average time for the chart reviews was 30.2 (SD 16.7) minutes per case, as seen in the image below.


Image source: National Library of Medicine

The secondary reviews by central reviewers along with the primary reviewers took 4.6 (SD 1.9) minutes per case, as seen in the image below.


Image source: National Library of Medicine

In addition, forty-one reviews underwent 2 independent reviews as described in the image below:


Image source: National Library of Medicine


The abstract concluded that the case reviews resulted in 30 minutes per case and recommended structured chart review with clear objectives by professional primary reviewers followed by a secondary review to be valid and reliable for quality assessment, research, and chart review process improvement activities.

Risk adjustment medical coding professionals are quite aware of the need for having clear and concise medical records documentation before they are submitted as claims requests to the Centers for Medicare & Medical Services (CMS) or Health and Human Services (HSS). And, proper medical record documentation, help, payers to gain correct reimbursements, whilst ensuring smooth revenue cycle management, and physicians to offer value-based care.

However, the individuals behind the risk adjustments are medical coders who perform the chart reviews and audits before the risk adjustment claim data submission deadlines.

In this article, we are presenting best practices of Risk adjustment documentation and coding for the CMS-Hierarchical Condition Category (HCC) and the Department of HHS-HCC models.

With both the models having different applications, one thing in common is the dependency on the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes that are used to risk adjust patients as per their health conditions.

(MAOs), Commercial Affordable Care Act (ACA) plans, Accountable Care Organizations (ACOs), and state Medicaid plans perform Risk Adjustment methodology that equates the health status of an individual to a number called Risk Adjustment Factor (RAF) score.

As a health plan organization, you will have a large team of medical coders who identify International Classification of Diseases, Tenth Revision (ICD-10) codes and validate them with the correct Hierarchical Condition Categories (HCCs) for the derivation of RAF scores.

A manual risk adjustment methodology for a vast amount of unstructured data can be energy-consuming, time-taking, and error-prone too.

To address these risk adjustment gaps, many health plans have increasingly been adopting technology options to streamline the process of chart review and audit in a more cost-effective and efficient manner.

How NLP and AI is helping Health plans in their Risk Adjustments

Natural Language Processing (NLP) is an augmented intelligence (AI) technology that is increasing its popularity amongst health plans in recent years. This latest technology is being used by health plans to analyze unstructured patient data including PDF medical records, call center transcripts, and electronic health record (EHR) exports, to enhance risk adjustment processes where manual review and audit are needed.

What are the best practices for Risk Adjustment Documentation and Coding?

  • Validation of medical record eligibility
  • Assignment of appropriate ICD-10-CM codes
  • Submission of ICD-10-CM codes to CMS or HHS for reporting

First, you will need to ensure the patient medical records include the patient identification number, validation of the provider who has to be a qualified physician and present for a face-to-face encounter, and verify the authenticity of medical records.

Next, to support an HCC, you need to ensure that the corresponding diagnosis must be mentioned in a health encounter.

RAAPID’s HCC Capture is personalized AI-Powered Risk Adjustment Coding and Quality Assurance Solution that is allowing health plans to scale a chart review process and capture complete risk whilst having a greater understanding and visibility into their members’/patients’ clinical conditions, including symptoms, and medications.

HCC Capture uses NLP technology to accurately read unstructured and structured medical records to precisely identify risk, improving the overall efficiency and return on investment ROI for a health plan.

It is built upon state-of-the-art AI, NLP, Machine Learning (ML), and Deep Learning (DL) models that augment the context surrounding identified information to get a better clinical understanding.

The NLP technology will read charts just like any human coder would do and automatically suggests accurate risk adjustment codes (ICD-10 and HCC) along with MEAT (monitored, evaluated, assessed/addressed and treated) evidence and gaps.

The above acronym is used by risk adjustment coders to identify reportable conditions.

A risk adjustment coding professional is recommended to accurately examine all sections of progress notes to determine whether the documentation of the chronic conditions meets the requirement of the risk adjustment models.

Moreover, the specificity of the clinical documentation is pivotal for risk adjustment coding professionals to determine whether the chronic condition is current and active.

Leave a Comment