Automation Potential in Claims Processing
- Pawan Jadhav
- July 31, 2020
Robotic Process Automation,Intelligent Automation,Data Extraction,Business Process Management
Last week, I explored 8 Challenges Faced by Healthcare Payers in the Pandemic Era. As I indicated in that post, payers in several areas will experience significant pressure to survive COVID 19 and beyond. Of the most impacted areas will be claims processing. In this post, I will address processes that are ripe for automation using RPA, BPM and Data Extraction - the three key ingredients for intelligent automation.
The recommended auto-adjudication rate should be upwards of 85 percent to maintain a solid medical- loss ratio. An auto-adjudicated claim costs health insurers cents on the dollar, while one that needs human intervention costs about $12 dollars or more per claim, as per CAQH. There are also stiff penalties in many states for claims that are not settled in 30 days.
CAQH also provides some insights into the cost that can be avoided in the entire value chain. Figure 1 illustrates the overall savings opportunity. The medical and dental plans can save up to $13.3 Billion in admin costs.
Figure 1: Estimated Medical and Dental Spend Savings
CAQH estimates that the entire medical industry could save $42.5 per patient encounter ($29.27 for providers and $13.18 for plans) and the dental industry could save as much as $30.08 (including $22.99 for providers and $7.09 for plans). The highest cost per transaction being recorded in claim status enquiry, prior authorizations and eligibility and benefit verification.
A typical administrative workflow pre and post claim submission are indicated below.
Figure 2: The Administrative Workflow
Our study and assessment of payers also indicate scope to reduce transaction costs in member submitted claims (highly manual), coordination of benefits, vendor accumulations (both medical and pharmacy) and attachments (medical records).
As we delve deeper into the claims processing area that comprises of several of the processes indicated above and is the major function performed by a payer, we discover several processes that are ripe for automation.
At Lateetud, we analyzed claims processes of major payers and mapped them as per the automation potential using intelligent automation - or Hyperautomation as Gartner calls It. Intelligent Automation is the number one trend to watch for in 2020. We examined it from an end-to-end automation perspective and evaluated the applicability of three major technologies: Robotic Process Automation, Cognitive Data Extraction and Business Process Management (workflow). To find out more about Intelligent Automation and definition of these terms refer to blog post – What is Intelligent Automation?
“As no single tool can replace humans, hyperautomation today involves a combination of tools, including robotic process automation (RPA), intelligent business management software (iBPMS) and AI, with a goal of increasingly AI-driven decision making.” Gartner
Top 6 Areas for Automation in Claims Processing
The following are the top 6 areas of automation in Claims Processing areas using RPA, BPM and Data Extraction technologies.
- Claim Service Inventories: This category of work includes handling enquiries, adjustments and pends and is a large portion of what the claims professionals do in the payer. Automation potential existing from assigning work based on aging and skills, extracting information from documents and attachments, interpreting emails from home/host parties, spreadsheet management for tracking, sending letters, downloading supporting docs, generating metrics, research etc.
- Complex or Unique Claims: This category includes several pending claims because they are special in nature, like Home Infusion Therapy (HIT), fallout from COB, lack of clarity (like the claims for COVID 19 treatment) or high dollar claims where more verification is required. These cannot be automated end-to-end but can be made easier by humans in loop scenarios where certain activities, including research, calculations, and collection of info, is performed by RPA, whereas the rest are performed by the associates.
- Pre-Authorization or Pre-Estimates: This is a major pain point for all parties involved – providers, subscribers, and payers, offering a high potential opportunity to automate using Intelligent Automation. Most requests are submitted via paper or some type of digital document and there is a major scope for standardization and use of Cognitive Data Extraction to provide relief in this area.
- Member Submitted Claims: Majority of the member submitted claims are still manual. If you combine two or more technologies, there is an extremely high scope for automating, accepting, reviewing, and processing claims. Also, there are certain categories within this area that are just ripe for straight through automation.
- Medical Records/Attachments: This includes all requests for medical records – accepting as well as forwarding – can be handled using automation. Automation can also handle extracting data from supporting documents using simple OCR or NLP, reading and interpreting medical terms and mapping that with ICD or CPT codes, using NLP and machine learning, inputting the data in applications., downloading Electronic Claim Attachments and forward the same to answer enquiries.
- Routine and admin: There are several routine level activities that cut across functions and may be straight forward. For instance, taking care of accumulations – both medical and pharmacy related, Stat/Rekey from spreadsheet templates for onboarding new members to a new plan, pricing calculations by reading codes & tallying with pricing database, sending letters or responding to simple queries, updating tracking sheets, and generating status reports based on aging of claims and response times etc.
The sooner healthcare payers act on implementing automation strategies the better customer experience they can provide. More importantly, as indicated above, this is a necessity to survive this pandemic and beyond. This “new normal” is here to stay for the immediate and foreseeable future, creating a clear need for healthcare payers to act quickly.