Improved Accuracy & Reduced Man-Hours With Automated NLP Engine for a Healthcare Company in the US Prescience Decision Solutions July 3, 2023

Improved Accuracy & Reduced Man-Hours With Automated NLP Engine for a Healthcare Company in the US

The client is one of the largest Medicaid managed care organizations in the US, that provides a portfolio of services to government-sponsored and privately-insured healthcare programs.


The client works with a large number of providers. For each provider, they have a contract that runs into 60-70 pages. The client’s representatives read through the provider contracts manually and identify the standard and non-standard clauses in each contract. While the deviation was present in less than 10% of the clauses, it was a time-consuming and tedious process. The client wanted to optimize and streamline whole process of identifying the non-standard clauses.


Prescience worked with the client to build an end-to-end Natural Language Processing (NLP) based solution to streamline the process of scanning the documents and identifying the key clauses required to arrive at a decision. An automated solution was deployed to extract, scan, compile, and categorize numerous documents swiftly.

Our solution included the following steps:

  • Data is extracted from printed or written text, from scanned contracts, and the processed text is transformed into a machine-readable form through OCR
  • NLP techniques such as Named Entity Recognition (NER) model, address block identifier, and key-value pair extraction were employed to tag different fields from entities or clauses in the contracts
  • Object detection model is applied to identify tables, headings, sentences, paragraphs etc. present in the contract
  • The rule-based model is employed to identify whether a contract is standard or non-standard, based on the business information provided by the client
  • An algorithm to generate recommendations on whether a contract needs to be sent for manual review or not, was applied

The solution enabled the client to bring in process and time efficiencies in their existing workflow. The solution also helped them identify data in a more accurate and automated manner.

  • Automated data extraction helped save more than 150-man hours per week
  • Reduction in manual intervention resulted in cost effectiveness, bringing down the cost of analyzing the contracts by an estimated 80%
  • 5x more contracts were processed as compared to the existing manual approach, within the same period
  • Improved resource utilization through automation of respective tasks
  • Improved accuracy of decision-making owing to reduction in errors, thus improving the performance

Learn from the Success Stories of Business Transformations Fueled by Data, AI and Machine Learning