How a number one underwriting supplier reworked their doc overview course of

Life insurance coverage corporations depend on correct medical underwriting to find out coverage pricing and danger. These calculations come from specialised underwriting corporations that analyze sufferers’ medical data intimately. As healthcare digitization has surged from 10% in 2010 to 96% in 2023, these corporations now face overwhelming volumes of advanced medical paperwork.

One main life settlement underwriter discovered their course of breaking underneath new pressures. Their two-part workflow — an inner workforce categorized paperwork earlier than docs reviewed them to calculate life expectancy — was struggling to maintain up as their enterprise grew and healthcare documentation grew to become more and more advanced. Medical specialists had been spending extra time sorting via paperwork as an alternative of analyzing medical histories, making a rising backlog and rising prices.

This bottleneck threatened their aggressive place in an trade projected to develop at twice its historic charge. With correct underwriting instantly impacting coverage pricing, even small errors might result in tens of millions in losses. Now, because the medical trade concurrently faces worsening workforce shortages, they wanted an answer that would remodel their doc processing whereas sustaining the precision their enterprise relies on. 

It is a story of how they did it.

When medical report volumes get out of hand

Processing 200+ affected person case information weekly may sound manageable. Nevertheless, every case contained a affected person’s whole medical historical past — from physician visits and lab outcomes to hospital stays and specialist consultations. These information ranged from 400 to 10,000 pages per affected person. However quantity wasn’t the one problem for the medical underwriting supplier.

Their enterprise confronted mounting stress from a number of instructions. Rising trade volumes meant that they had extra circumstances to course of. On the flip facet, the healthcare trade staffing shortages meant they needed to pay docs and different medical specialists prime {dollars}. Their current guide workflow merely could not scale to satisfy these calls for. It was made worse by the truth that they needed to keep near-perfect doc classification accuracy for dependable life expectancy calculations.

The enterprise influence was evident:

• Slower processing occasions meant delayed underwriting selections
• Inaccurate life expectancy calculations resulted in tens of millions in mispriced insurance policies
• Probably shedding enterprise to extra agile rivals
• Greater processing prices instantly affected profitability
• Rising prices as docs frolicked on paperwork as an alternative of research

Their medical specialists’ time was their most precious useful resource. And but, regardless of the 2-step workflow, the sheer quantity of paperwork compelled these extremely skilled professionals to behave as costly doc sorters somewhat than making use of their experience to danger evaluation. 

The mathematics was easy: each hour docs spent organizing papers as an alternative of analyzing medical situations price the corporate considerably. This not solely elevated prices but in addition restricted the variety of circumstances they might deal with, instantly constraining income progress.

What makes healthcare doc processing difficult

Let’s break down their workflow to know why their medical report processing workflow was notably difficult. It started with doc classification — sorting a whole bunch to 1000’s of pages into classes like lab stories, ECG stories, and chart notes. This essential first step was carried out by their six-member workforce.

Every member might course of ~400 digital pages per hour. Which means, a single case file of two,000 pages would take over 5 hours to finish. Additionally, the pace tends to differ closely primarily based on the complexity of the paperwork and the aptitude of the worker.

The method was labor-intensive and time-consuming. With digital medical data coming from over 230 totally different programs, every with its personal codecs and buildings, the workforce needed to cope with plenty of variation. It additionally made automation via conventional template-based information extraction almost inconceivable.

The complexity stemmed from how medical data is structured:

• Important particulars are unfold throughout a number of pages
• Info wants chronological ordering
• Context from earlier pages is commonly required
• Dates are typically lacking or implied
• Duplicate pages with slight variations
• Every healthcare supplier makes use of totally different documentation strategies

After classification, the workforce would manually establish pages containing data related to life expectancy calculation and discard irrelevant ones. This meant their employees wanted to have an understanding of medical terminology and the importance of assorted take a look at outcomes and diagnoses. There was little or no margin for error as a result of even the slightest errors or omissions might result in incorrect calculations downstream.

The paperwork would then be despatched to docs for all times expectancy calculation. Docs principally did this throughout their non-clinical hours, which already made them a scarce useful resource. To make issues worse, regardless of having workers to deal with preliminary classification, docs had been nonetheless compelled to spend important time extracting and verifying information from medical paperwork as a result of solely they possessed the specialised medical information wanted to accurately interpret advanced medical terminology, lab values, and scientific findings.

Some case information had been enormous — reaching past 10,000 pages. Simply think about the sheer endurance and a spotlight to element required from the workforce and docs sifting via all that. That is why when the agency was on the lookout for automation options, there was a robust emphasis on reaching almost 100% classification accuracy, self-learning information extraction, and decreasing person-hours. 

How the underwriter carried out clever doc processing for medical data

Medical report volumes had been rising, and physician overview prices had been mounting. The underwriting workforce knew they wanted to automate their course of. However with life expectancy calculations depending on exact medical particulars, they could not danger any drop in accuracy in the course of the transition.

Their necessities had been particular and demanding:

• Capability to course of 1000’s of pages of medical data every day
• Understanding of advanced medical relationships throughout paperwork
• Classification accuracy needed to be near-perfect
• Fast and safe processing with out compromising high quality
• Combine out-of-the-box with Amazon S3

That’s when their VP of Operations reached out to us at Nanonets. They found that we might assist classify medical data with excessive accuracy, present a filtered view of great pages, extract information key factors, and guarantee seamless information flows throughout the workflow. This satisfied them we might deal with their distinctive challenges.

Here is what the brand new automated medical data automation workflow seemed like:

1. Doc preparation

• The inner employees combines all medical data— lab stories, ECG, chart notes, and different miscellaneous paperwork — for every affected person right into a single file
• Every affected person is assigned a singular quantity
• A folder with this quantity is created within the S3 enter folder
• 7-10 such circumstances are uploaded every day

Be aware: This method ensures safe dealing with of affected person data and maintains clear group all through the method.

2. Doc import

• The system checks for brand spanking new information each hour
• Every case can include 2000-10,000 pages of medical data
• Information are readied for secured processing via our platform

Be aware: This automated monitoring ensures constant processing occasions and helps keep the 24-hour turnaround requirement.

3. Doc classification

Our AI mannequin analyzes every web page primarily based on rigorously drafted pure language prompts that assist establish medical doc varieties. These prompts information the AI in understanding the particular traits of lab stories, ECG stories, and chart notes.

The classification course of includes:

• Figuring out doc varieties primarily based on content material and construction
• Understanding medical context and terminology
• Sustaining doc relationships and chronological order
• Recognizing when context from earlier pages is required

Be aware: The prompts are repeatedly refined primarily based on suggestions and new doc varieties, guaranteeing the system maintains excessive classification accuracy.

4. Information extraction

Our system handles three important doc varieties: lab stories, ECG stories, and chart notes. We now have two specialised extraction fashions to course of these paperwork – one for lab/ECG information and one other for chart notes.

Mannequin 1 extracts roughly 50 fields from lab stories and ECG information, together with affected person title, blood glucose degree, creatinine worth, glomerular filtration charge, hemoglobin worth, prostate particular antigen, white blood cell rely, hepatitis worth, ldl cholesterol worth, and lots of different essential lab measurements. 

Mannequin 2 processes chart notes to extract 13 key fields together with blood stress, heartbeat charge, O2 supply, O2 stream charge, temperature, date of delivery, gender, peak, weight, and smoking standing. Every information level is linked to its supply web page and doc for verification.

5. Information export

The extracted data is exported as three separate CSV information again to the S3 Bucket — one every for doc classification, lab outcomes and ECG, and chart notes.

The classification CSV incorporates file names, web page numbers, classifications, and hyperlinks to entry the unique pages. The lab outcomes and ECG CSV include extracted medical values and measurements, whereas the chart notes CSV incorporates related medical data from docs’ notes.

In every file title, an identifier, like ‘lab outcomes’ and ‘ECG’ or ‘chart notes’, shall be mechanically added to establish the content material kind. And for consistency, CSV information are generated for all classes, even when no related pages are present in a case doc. Every affected person’s information shall be saved within the Export folder on the S3 bucket underneath the identical figuring out quantity.

6. Validation 

The CSV outputs are imported into their inner utility, the place a two-member validation workforce (decreased from the unique six) opinions the automated classifications. Right here, they will evaluate the extracted information in opposition to the unique paperwork, making the verification course of fast and environment friendly.

As soon as the info is validated, the docs are notified. They’ll go forward to investigate medical histories and calculate life expectancy. As an alternative of spending hours organizing and reviewing paperwork, they now work with structured, verified data at their fingertips.

Be aware: For safety and compliance causes, all processed information are mechanically purged from Nanonets servers after 21 days.

The influence of automated medical report processing

With structured information and an environment friendly validation course of, the underwriting supplier has been in a position to decrease the operational bottlenecks concerned within the course of.

Right here’s a fast overview of how a lot they’ve been in a position to obtain inside only a month of implementation:

• 4 members on the info validation workforce had been reassigned to different roles, so validation now runs easily with simply 2 folks
• Classification accuracy maintained at 97-99%
• Automated workflow is dealing with ~20% of the full workload
• Full information classification and extraction for every case file inside 24 hours
• Obtain a 5X discount within the variety of pages docs have to overview per case to compute life expectancy
• Freed medical specialists to concentrate on their core experience

These numbers do not inform the entire story. Earlier than automation, docs needed to sift via 1000’s of pages as a result of they had been the one ones with the required context to know affected person information. Now docs get precisely what they want – detailed medical histories sorted chronologically which are prepared for evaluation. It is a full shift from sorting papers to doing precise medical evaluation. 

This alteration means they will deal with extra circumstances with out having to rent dearer docs. That is an enormous benefit, particularly with healthcare dealing with employees shortages whereas the trade continues to develop.

Trying forward

This profitable implementation has helped the underwriting supplier perceive what’s attainable with clever doc processing. They now need to scale their medical report processing to cowl all ~200 circumstances weekly. That is not all. They’re already exploring easy methods to automate different document-heavy workflows, like belief deed processing.

Fascinated by what this implies in your group? The time to modernize doc processing is now. Healthcare documentation is turning into extra advanced, with a 41% progress in high-acuity care and rising continual situation administration. Add to this the rising staffing challenges in healthcare, and it is clear— in the event you do not modernize, your group will battle to maintain up.

Wish to see comparable outcomes together with your medical report processing? Let’s speak about how Nanonets will help. Schedule a demo now.