Meet the new all-in-one platform for medical record reviews! Learn more

AI in Insurance: How DigitalOwl saved $270,000 on an improperly-paid disability claim

Published On
May 7, 2020
Share this post

When it comes to disability claims, accuracy is essential.  Claim analysts will not make a decision without reading the claimant's full medical record first.

The problem is that finding crucial medical information in the medical record can be very tedious and time-consuming work.  Reviewing hundreds of medical pages a day is exhausting, and this inevitably leads to errors.

Test yourself:  How many medical conditions and procedures can you find in this puzzle?

*The answer is at the end of the post.

Now, imagine that you need to do this all day long. How many conditions and procedures will you miss?

Using cutting-edge AI technology to review and summarize medical records can save time and money for insurance companies, TPAs, and others.

Case Study

Jane Doe is a 50-year-old female with an annual income of $90K.  Jane applied for an Income Protection (disability) policy in January 2019.

She completed the application, stating that she is smoking five cigarettes a day, but has had no disabilities, prior hospitalizations, heart disease, diabetes, tumors/lumps, or any other significant medical problems.

The insurance company, relying on her statements, approved her policy with no exclusions or extra premiums.

A year later, Jane submitted a claim when she was diagnosed with breast cancer.

During the claim adjudication process, the insurance company collected all of Jane's medical records (~800 pages), and one of the insurance company's physicians reviewed the entire medical record to ensure there were no relevant pre-existing conditions.  The physician found nothing relevant, and Jane's claim was approved.

A few months later, Jane's claim was analyzed by DigitalOwl's AI system (as part of a POC).  The review took just a few minutes.  Within the 800 pages of medical records, a "breast lump" diagnosis from 2012 was discovered.  DigitalOwl's system instantly extracted the medical finding and flagged it.  This medical history had not been stated on her application even though the question had been asked.

Using DigitalOwl's system at the time of the claim adjudication would have uncovered the misrepresentation of medical history and led to an accurate decision on the claim, saving $270,000 in improperly-paid benefits.

To learn more about DigitalOwl's solution visit us at

Puzzle solution:  16 medical conditions and procedures: Arterial Bleed, Biopsy, Cancer, Cellulitis, Dyspnea, EMG, Hypertension, Laceration, Nerve Repair, Obesity, Osteoarthritis, Physiotherapy, Strain, Stroke, Surgery, and Tendinitis

Yuval Man
Co-Founder & CEO
About the author

As the Co-Founder & CEO of DigitalOwl, Yuval Man empowers insurance companies to unlock the full potential of their medical data for better outcomes by harnessing the transformative powers of AI to streamline and elevate the review of medical data.