Overcoming Data Overload with Structured Data and Interoperability

I recently had the privilege of joining a webinar with leading industry experts Dave Rengachary, Maria Beaulieu, Michael Hill and Mark Ma from RGA, Nick Milinovich from Northwestern Mutual and Andy Kramer from M Financial Group. Together, we discussed the challenges underwriters encounter when dealing with extensive volumes of both structured and unstructured data. Our conversation centered around our vision for creating a more efficient data structure for underwriting purposes, while balancing the need for industry integration against the benefits of maintaining proprietary processes. In this article, I’ll share the critical insights and recommendations that emerged from our discussion.

Data Overload Is Challenging Underwriting

In the past ten years, the volume of data for underwriters has doubled, forcing underwriters to work twice as hard when reviewing medical records during the application process. While this data can positively impact the accuracy of risk assessments, it also leads to challenges from an operational and efficiency standpoint, highlighting the need for a next-level of efficiency in the underwriting process.

In addition to the increasing volume of data, underwriters are also experiencing an increase in the complexity of data due to an increase in automation and the need for more complicated rules engines. While these solutions are a step in the right direction, they’ve ultimately neglected APSs and EHRs, leading to inefficient and time-consuming manual work.

Efficient Data Solutions Are Critical

In order to use unstructured data, organizations must first convert it into a usable format. However, the diversity of unstructured formats and structures creates operational inefficiencies and challenges for underwriters who must frequently create rules and rules engines to properly assess the information for a desired outcome. Even structured data is incredibly diverse, with tens of thousands of different codes that need to be distilled into impairments and disease entities to inform underwriting insights. Moreover, data needs to be easily accessible and human-readable to enable underwriters to quickly review information and make informed decisions, empowering efficient risk assessment and enhancing the accuracy of underwriting outcomes.

“We see hundreds of thousands of different codes in EHRs. The dream is to distill those codes into impairments and disease entities, and then add in as much as you can from unstructured data,” said Nick Millinovich, Senior Director of Underwriting Partnerships, at Northwestern Mutual. “It’s even better if we can put those findings into a clean package and use rules to adjudicate the majority of cases. Then our underwriters are free to apply their considerable skills to the most complex cases.”

Digitization is Essential for Future Advancements

Currently, underwriters are struggling to access the vast amount of underwriting data locked in their systems. This has caused a bottleneck of creative ideas, many of which are deemed impossible without sufficient data. Digitizing these files will make the data significantly more accessible, unveiling hidden insights that will significantly enhance the decision-making process. Furthermore, this advancement allows actuaries to shift their approach from case-level reviews to a more analytical approach, allowing for a greater degree of accuracy and confidence.

“Digitizing underwriting files with DigitalOwl enables us to unlock a wealth of viable data that was previously inaccessible.” Mark Ma,  said Mark Ma, VP, Managing Actuary at RGA.

Healthcare-based Formats Are Inadequate

There are significant challenges associated with healthcare-based formats due to the vast complexity of healthcare records which can contain over 200,000 mentions of various elements, such as lab results, APSs and EHRs, each saying something different. Moreover, these existing data types were primarily designed for clinical medicine, with the objective of treating patients.

In the insurance sector, there is a greater focus on understanding how to avoid, maintain and monitor risk, with the end goal of integrating individuals into product offerings. This is a dramatically different use-case from healthcare-based formats. Additionally, the underwriting process involves navigating regulations, transparency and bias issues, and therefore requires a more extensive range of information than what clinical codes or data alone can provide. As such, a deeper, more comprehensive approach to data evaluation is critical for the success of underwriting processes.

Interoperability Is a Shared Industry Goal

Currently, information is often transmitted in non-structured PDF formats, requiring additional steps to convert data into a usable form. Consequently, the process frequently involves unnecessary duplications, leading to inefficiencies and the exchange of data in a less-than-optimal format. In my opinion, this issue is significant enough to warrant an industry-wide solution. Addressing this challenge collectively as an industry will allow us to enhance efficiency and innovation, allowing for configurability remaining with the carriers to balance the need for interoperability with proprietary advantages.

“One of our vendors said that if the industry could coalesce around one standard application, it would reduce the implementation cost of those electronic application systems by about 80%,” said Andy Kramer, VP and Head of Underwriting, Risk & Innovation at M Financial Group.

Enhance Interoperability with DigitalOwl’s Data Format

DigitalOwl’s proprietary data format plays a crucial role in enhancing precision and accuracy in the underwriting process. This format allows for the normalization and de-duplication of data, leading to more reliable medical summaries, audit reports, and data analyses that ensure users receive the most accurate and actionable insights possible.

By standardizing data into a proprietary format, DigitalOwl ensures seamless integration with various systems and platforms. This interoperability is crucial for clients who rely on multiple software solutions for their operations, enabling them to easily incorporate cutting-edge AI solutions into their existing ecosystems without significant adjustments. Furthermore, the ability to structure data into a format that’s optimized for your algorithms means information can be processed more efficiently. It also gives you better control over data security and compliance with regulations, ensuring sensitive medical information is handled with the utmost care to maintain client trust, adhere to industry standards and satisfy legal requirements.

DigitalOwl’s Connect Product, Stands at the Corner of Integration

DigitalOwl’s sophisticated “Connect” API is a critical AI business tool for achieving interoperability. It allows for efficient data retrieval and analysis by integrating DigitalOwl's advanced AI solutions into clients' existing systems. This integration facilitates access to comprehensive medical data in one unified platform, addressing the pain points of dealing with disparate systems and fragmented data that can hinder productivity and lead to incomplete information during critical processes. It also streamlines operations by automating data retrieval, enhancing efficiency and reducing the likelihood of errors.

Contact DigitalOwl to learn how our AI-powered products are transforming medical reviews.

‍