AKASA debuts autonomous AI platform for inpatient medical coding and clinical documentation

Health tech firm AKASA has unveiled an autonomous AI platform designed for inpatient medical coding and clinical documentation accuracy. The company, which currently offers generative AI solutions for hospital revenue cycle management (RCM), is moving from prebill review to a more autonomous mid-cycle approach for the most intricate and resource-heavy hospital workflows.

Inpatient medical coding is a highly complex task due to the extensive clinical documentation required, stringent regulatory rules, and the direct impact on hospital reimbursement. The mid-cycle phase is crucial, as it involves transforming a patient's documentation into codes that influence reimbursement, quality reporting, risk adjustment, and record integrity.

Unlike outpatient coding, which focuses on individual procedures, inpatient coding encompasses the entire patient stay from admission to discharge. This work is also labor-intensive, with many hospitals facing staffing shortages in this area.

AKASA's CEO and co-founder, Malinka Walaliyadde, emphasized the significance of autonomous mid-cycle, stating that it has long been considered the "holy grail" in healthcare due to its complexity. Inpatient stays typically involve around 60 documents and 50,000 words, with medical coders taking 30 to 60 minutes to code a single encounter.

However, with staffing shortages and capacity constraints, health systems often face delays of several days before a coder begins working on a discharged account. The company claims that its AI can complete coding in approximately 90 seconds after discharge, speeding up the billing process and reducing accounts receivable days. AKASA asserts that its autonomous mid-cycle AI is designed to fully code complex inpatient cases across all specialties with no human intervention.

The company conducted third-party, blinded evaluations, comparing AI performance against human coders on key accuracy measures, including MS-DRG assignment, principal diagnosis, clinical quality capture, and present-on-admission accuracy. These evaluations demonstrated that AI performance matched or exceeded that of human coders.

AKASA plans to roll out the autonomous AI platform for inpatient medical coding in the coming months, customizing and fine-tuning the AI for individual health systems to account for differences in patient populations, clinical criteria, documentation practices, and care complexity.