Close Server: KOPWWW05 | Not logged in


Directions in Analytics

Analytics in Healthcare Payments, Part 2

Editor's note: This is the second of a two-part series on the role analytics plays in identifying and attacking the billions of dollars the U.S. spends each year on erroneous healthcare payments.

Part 1 of the series provided an overview of the industry and cutting edge analytic techniques valuable in uncovering errors in payments, as well as fraud, waste and abuse.  Part 2 serves to provide an overview of data sources and types, and explain the direction the industry is headed in utilizing key tools that can help resolve the epidemic of erroneous healthcare payments.

Data Sources/Types
Our world is exploding with data.  Highlights of recent articles on the topic include items such as: there are twice as many bytes of data in the world as there are liters of water in the Earth's oceans; and Facebook's photo library is 10,000 times the size of the data that exists in the Library of Congress.  

Healthcare is also an area with extensive data -- claims data, medical records, provider information, eligibility files, benefits structures, biometric information, self-monitoring devices (e.g.,Fitbit, Up), data from wellness programs and health risk assessments.  Given legacy systems and imperfect standardization, there is substantial variation in the format and structure in which data is stored.  As healthcare becomes more consumer oriented, with things like high deductible plans, and transparency in cost of various medical services, other sources of data on consumer preferences become relevant such as educational level, financial position, income levels.

Leading edge analytics teams are leveraging these data sources more and more to do a better job of identifying problematic providers or payments. Examples include:  

  • Patient Acuity -- Healthplans frequently look at the acuity of a patient's condition as derived from claims data for various reasons, mostly related to Care Management.  However, most analytics used in the payments space do not currently incorporate this information beyond the basic diagnostic information on the specific claim being analyzed.  Leading teams are using a broader view of how sick a member is (what chronic conditions they have, etc.) to determine whether a specific claim or pattern of billing appears appropriate.  Even if this is primarily lifting member acuity scoring from Care Management and applying it to payments, there is substantial additional value.
  • EMRs/Biometrics/Self-Monitoring/Wellness Program -- One of the key tenants of pushing providers to use electronic medical records and make various data available in electronic form (e.g., biometrics) is that significantly better decisions about care could be made.  Little discussion has occurred about using such information for payment analytics, but there are numerous obvious possibilities, such as: knowing various lab levels that show a drug is unlikely to have been used, comparing a patient's weight with the amount of a drug to see if it fits the dosing regimen when dosing is based on weight, or having data that show a patient was jogging at a time when they claim to have been undergoing  specific physical therapy (e.g., hot/cold packs, Electric Stimulation).
  • Social Media -- From the advent of social media, investigators have been checking social media posts for members suspected of improper conduct (e.g. claiming to be disabled, etc.).  With text mining and other analytic tools, these efforts can become even more robust, flagging new items for investigation based on a combination of claims data and use of key words that conflict with the condition represented therein.

There is still much work to do to address the issue of massive erroneous payments made in healthcare, even as substantial progress has been made.  Two key tools in that effort are sophisticated analytic techniques and bringing newer data sources together with traditional sources (claims data).  Of course, there are numerous complexities in this process, and many new resources and analytic capabilities will be required to succeed.  The battle will not be won overnight.  Succeeding in this area requires sophistication and perseverance, but the rewards are well worth it.

Rodger Smith is senior vice president, Provider and Payment Solutions, SCIOinspire.


Directions in Analytics Archives


     

Email: *

Email, first name, comment and security code are required fields; all other fields are optional. With the exception of email, any information you provide will be displayed with your comment.

First * Last
Name:
Title Field Facility
Work:
City State
Location:

Comments: *
To prevent comment spam, please type the code you see below into the code field before submitting your comment. If you cannot read the numbers in the below image, reload the page to generate a new one.

Captcha
Enter the security code below: *

Fields marked with an * are required.

View New Jobs, Events and More

 
 
 
 
 

Back to Top

© 2017 ADVANCE Healthcare, an Elite CE company

contact@advanceweb.com