Since the advent of digital healthcare, huge amounts of data have been created and stored. The question is, how do we use it?
Big Data, Big Value
Big data in healthcare is a valuable asset with a wide range of possibilities. The amount of data healthcare churns out each second is astronomical. Before COVID-19 came on the scene, a 2018 Statista estimate predicted that 2,214 exabytes of new data could be generated globally in 2020. An exabyte is one billion gigabytes, or 1018 bytes (one quintillion).
So, how can you visualize something that big?
- If one gigabyte is the size of the earth, then one exabyte is the size of the sun.
- Five exabytes likely equals all of the words ever spoken by humans.
- To audio record 1 exabyte of data, you would have needed to start speaking into the recording 237, 823 years ago.
So – it’s big. Really big. This volume of data includes patient level data, historical trends over time, by diagnosis, by age, and any demographic or segment that you can think of. Even more interesting, this large volume of data can be used as the basis for learning – otherwise known as artificial intelligence (AI), part of predictive analytics.
No Crystal Ball Here
To think of a concrete example of predictive analytics, consider what happens when a house is sold. The real estate agent will look at the appraisal, compare the house to other houses that have recently sold in that neighborhood, and consider the overall housing market. By using all of that data, a price is set that the agent predicts the house will sell at.
Healthcare uses data to predict which patients will be at a higher risk for readmissions, which might be at a high risk for a blood clot or infection, and which interventions have the best outcomes. Armed with this information, clinicians can work to prevent those particular outcomes – whether that might be changing the care plan, adding a specific intervention, or using a specific medicine.
In the world of insurance claims, predictive analytics can be used to head off insurance denials before they have a chance to occur. Why would healthcare organizations want to harness the power of this data?
The AMA’s Administrative Burden Index (ABI) calculates that healthcare providers spend up to 14% of their revenue just to get paid for services.
Consider these stats about claims:
- 9% of hospital claims are initially denied. This equals an average of almost $5 million per hospital system and a national total of $262 billion.
- 90% of all denied claims are preventable! But – only 66% are recoverable. Effort put into the front end to prevent denials is much more efficient than reworking claims and hoping for the best.
- Denials add an extra 16 days in A/R, hindering cash flow.
- Reworking a claim costs $118 – a whopping total of $8.6 billion in administrative costs. This means that only 35% of providers appeal denials.
- Nearly ALL of the top drivers of claims denials are easy to flag in a well-built system:
- Omitted patient information
- Duplications
- Prior adjudication
- No payer coverage
- Late submissions
Look Upstream in the Revenue Cycle
To prevent much of the revenue leakage that healthcare is now plagued with, organizations have to look upstream – to the many processes that lead up to the claim submission. Departments like patient access, HIM, financial services, and clinical staff all have an impact on the data that goes into a clean claim submission. Unfortunately, many times that data is spread out across a few programs. Patients may be registered in one program, clinical staff chart in another, and financial services use yet another. This siloed and disparate data makes it difficult to review and pull together.
Enter technology and analytics. Denials alerting systems take the guesswork and data-crunching out of process improvements.
- Software that automatically detects increases in denials, categorizes them by root causes, and organizes your preventative efforts.
- Go even further by using modeling that examines account attributes and provides targeted account lists within each alert-group. For example, cardio, outpatient, and CPT123 groups.
- Use dynamic alert groups based on all the combinations of CARC, RARC, and Payer that flag at the appropriate granularity of data.
Also, denials analytics software can flag issues that are known to be common denials reasons, sending the chart for correction before the claim is filed.
Contact Us
At Healthrise, Qodex is helping healthcare organizations across the country find meaningful patterns in their denials data and address the root causes – saving time and money. Contact us for a free demo and more information on Denials Navigator.