Mastering Record Retrieval: Smarter MRR Strategies for Risk Adjustment Success

Risk adjustment professionals are walking a tight rope without a net when it comes to medical record reviews (MRRs) and retrieval.

MRRs play a crucial role in validating medical claims that health plans use for risk adjustment and regulatory compliance. Unfortunately, this process can be expensive, time-consuming, and stressful, as mistakes can lead to steep financial consequences. Despite the industry’s shift to digital solutions for medical record retrieval and data sharing, significant obstacles remain.

With so much at stake, health plans must find a solution that can meet their record retrieval needs, simplify the MRR process, and maintain accuracy under growing regulatory pressure.

The challenges of manual MRRs

Health plans rely on various sources, such as claims, encounters, and medical records, to collect accurate diagnosis data. Without accurate information, plans risk under-compensation for member care or face penalties for overpayments.

Yet, these sources often contain:

• Incomplete or incorrect claims
• Duplicate records
• False positive diagnoses

As a result, there’s still a need for skilled professionals to manually sift through these records and find the clinical data they need to validate members’ condition status. Depending on the size of the health plan, this means manually reviewing tens of thousands to millions of records each year.

These manual processes are not only time-consuming and costly but also frequently yield little to no clinical value for risk adjustment.

The stakes are higher than ever

Scrutiny over risk-adjusted payments is intensifying. In October 2024, the Office of the Inspector General (OIG) released a report on the questionable use of health risk assessments (HRAs) and chart reviews to increase payments to plans. In the report, the OIG found that an estimated $7.5 billion in risk-adjustable payments were issued in 2023 based on diagnoses reported on members’ in-home HRAs and HRA-linked chart reviews, rather than on any other 2022 service records.

This heightened attention means risk adjustment teams must stay vigilant, as regulators closely monitor for misuse or misreporting.

Electronic records are shaping retrieval – but not without issues

The healthcare industry has made major strides in electronic record retrieval and exchange. As of 2021, 88% of hospitals send and obtain patient health information electronically by querying or electronically receiving medical records. This widespread adoption of electronic medical records (EMRs) is transforming risk adjustment processes.

Despite this level of adoption and access, integrating disparate EMR systems remains a significant challenge for health plans seeking the risk adjustment data they need. This involves investing in internal data acquisition teams, including procurement, legal, and development team members, or outsourcing to outside vendors to establish connections to EMRs and gain access to timely data.

Adding to the challenge is the quality of the risk adjustment data. One of the biggest misconceptions about electronic medical records retrieval is that direct access to an EMR is enough – but not all the data is usable.

Some examples of unusable data include:

• No electronic signatures or signature dates
• Missing face-to-face encounters and dates
• Formatting issues with foreign text or symbols

These issues drive health plans to seek new technologies that simplify and enhance data collection for risk adjustment.

How AI and NLP can be helpful tools, not crutches

Given the high cost of manual reviews and the low yield of clinical documentation, health plans are increasingly turning to artificial intelligence (AI) and natural language processing (NLP).

NLP can automate key tasks in the MRR process, such as validating diagnosis codes, removing duplicates, and routing charts for specialized review. These efficiencies can drastically improve productivity.

However, AI and NLP alone aren’t silver bullets for solving health plans’ problems. The OIG and Department of Justice (DOJ) see relying entirely on AI, NLP, and machine learning for reviewing medical records as a potential risk for unchecked errors, so health plans must maintain human oversight when using AI tools.

Choosing the right MRR and retrieval partner

Health plans don’t need another vendor – they need a partner. The right partner provides an easy integration experience and tailored solutions that meet specific operational needs, while also enhancing revenue and efficiency.

Inovalon® delivers this with a dual-solution approach:

Inovalon’s Converged Record Review™ enables medical record reviewers to do their job faster and easier, helping health plans get properly reimbursed for the conditions they manage and avoid compliance risks. The solution’s NLP capabilities help target member records that are highly likely to have a clinical documentation value, highlighting records with potential findings to increase reviewer efficiency.

Inovalon’s Electronic Record On Demand™ ends the disorganization of medical record retrieval with a single, secure source that eliminates inefficiencies, saves time, and improves workflows for data teams. This software takes advantage of Inovalon’s extensive connectivity to electronic health records and health information exchanges, and massive proprietary datasets containing more than 405 million unique lives and 89 billion medical events.¹

Looking to improve your risk adjustment efficiency? Schedule your personalized demo to see why Inovalon is the right partner.

---

¹ Inovalon MORE² Registry® metrics as of April 2025.

Inovalon and design®, Inovalon®, and the Inovalon ONE® Platform are trademarks of Inovalon, Inc.