Significance of Data and Data Analytics in Health Insurance Organizations

1/5/20243 min read

The Healthcare industry has been facing a period of transition for some time. The Affordable Care Act, Health Insurance Exchanges, ICD-10 adoption, spurt of ACOs and the changes dictated by CMS mean that Payers have to new technology to accentuate their traditional models of operations. Health Insurance organizations are facing a tremendous pressure to improve patient outcomes and reduce cost. Analyzing and leveraging the large amounts of data at their disposal can go a long way in taking steps in the right direction.

Many organizations have moved in the right directions with Data Warehouse models and Data Analytics solutions. However, they need to adapt to utilize the disparate data from new sources in combination with the variety of old ones.


Let us take a look at some of the traditional and new sources of Data.

Traditional Claims data in payers has been housed in a variety of stores. There is a mix of legacy, client server and internet age. IMS and DB2 databases are still prevalent in home grown systems. Flat files and VSAM files house some information. Much of this data is translated and stored in traditional Relational structures like DB2, Oracle, SQL Server. EDI Data prevalently stored in XML or parsed into databases. Additional complexity might be the proprietary structures of Payer product systems like Trizetto’s FACETS, IKA Systems, NASCO and others.

Membership would be stored in either the main or other platforms like MetaVance, FACETS, or legacy. Feeds would come in from the CMS, Federal Employee Programs (FEP), State Medicaid systems and others. Pharmacy Vendors would have to be harnessed for Drug usage, data and trends. Medco and MedImpact data will have to be integrated for studies.

Provider data is traditionally housed separately. It is imperative to be able to properly understand this data to utilize it for proper analytics. Provider agreements, demographics, the population they provide service to needs to be got. Provider credentialing and network management data along with the Provider contract information for reimbursement models will play a key role for certain outcomes. Physicians fee schedules data can be got.

Benefits information is traditionally complex and difficult to decipher. However, this information, when analyzed, can provide value into a number of areas for the Payer and Provider. Electronic Health Records(EHR) and Electronic Medical Records(EMR) information is also not structured because of a variety of vendors and system formats. The consolidation and standardization that is happening is of definite help here.

Customer Service calls, Case Management recordings, IVR systems contain valuable information that needs to be captured and analyzed. Call center information can provide insight that can be used to produce trends. In spite of all the digitalization, there is still a lot of paper floating around and OCR systems data and digitized images need to be utilized.

New age internet data is a valuable source. Patient trends, reviews and healthcare websites need to be scoured. Ratings, Quality of services, Patient experiences, consumer reports on a variety of healthcare services and organizations are available and need to be utilized. There are a multitude of internet data sources that are available at government and independent consumer websites.

All this data needs to be gathered, cleaned up and organized for analysis. Some will be real-time and some will be in batch modes. Whatever be the case, it is what is done with it that is important. Let us try and determine some of the benefits of descriptive and predictive analytics.


Payer Analytics primarily has the 2 core objectives: Reduce Costs and Improve the Quality of Care

Create Cost savings: Payers can derive information that will help drive down the overhead, improve First Pass rates for claims processing and bring efficiencies in operations. Implement the right physician incentives, Network discounts, optimal Fee schedules and right pricing models. Target the high cost demographic areas and physician groups and focus on improving health to drive costs down.

Improving the Quality of Care: Identifying patient and provide trends to be able to provide the optimal care to consumers. Driving costs down will get more care for the dollar. Use predictive logarithms to direct the patient’s calls to the optimal areas enabling savings in time and efficiencies.

Identify and Eliminate Fraud / Revenue Cycle Management: Utilize all the above information to identify and reduce billing errors, erroneous and duplicate claims. Detect anomalies in claims submission vs patient diagnoses, inaccuracies in drug prescriptions. Identify claims over-payments, predict system and functional areas for potential problems and eliminate these errors proactively.

Consumer Analytics: Target marketing dollars in the right areas. Consumer retention and increased closure rate of New Member acquisition using predictive analytics strategies. Demographic and regions analytics will enable targeted product services, focus on better care giving for severe cases. Utilize predictive modeling in Rating and Underwriting to derive optimal pricing for consumers, ability to engage patients before acute conditions develop and costs spiral.