Revenue leakages are a constant struggle for payers, as they progress through different stages of claims operation. In addition, there are issues around high rework rate, overpayments, and adjustments that contribute to spending effort and cost in administrative tasks. This has created the need to move away from the traditional ‘pay and chase’ approach and prevent incorrect payments, claim frauds, or even other larger issues that arise due to incorrect processing of claims.
The financial burden incurred by payers for claim processing and reprocessing due to incorrectly processed claims, cascades down as additional workloads to downstream processes such as contact centers, collections, clinical investigators, and appeals. Healthcare payers spend around USD 40 billion in administrative costs for processing and reworking on incorrect claims, which includes USD 6-7 per claim to handle claim-related queries by their contact centers.
While the focus is shifting to adopting prepayment initiatives and analytical solutions to handle these challenges proactively, payers need to ensure they are embracing the right methodology.
Minimizing Incorrect Claims Processing
Claims data is an ocean of valuable insights used only for operational purposes so far. Enterprises are struggling to utilize this data effectively due to the fragmented data sources and siloed operations, which makes it difficult to get a holistic view of where the issues lie. It is humanly impossible to gain insights from the vast and complex claims data to find out the reasons for the spike in administrative cost. This is where techniques like pattern detection come in.
Pattern detection techniques, powered by AI and ML, help analyze historical data to identify claim patterns, which have higher propensity of overpayments or chances of incorrect payments and much more. For example, a specific pattern of a primary diagnosis for osteoarthritis which, whenever billed with the procedure codes for knee arthroscopy and a specific modifier 59 performed by provider X always leads to an incorrect denial and thus lands as an appeal resulting in increased rework for a payer. Or, whenever an inpatient stay for a case of child delivery exceeds the length of stay of four days and is associated with normal delivery but diagnosis codes indicating complications has historically been identified as overpayments or frauds.
Identifying such patterns and monitoring them in real time will help handle such exceptions. Moreover, the dynamic trending capability helps make even smarter decisions as the patterns evolve over a period and some become outdated. Continuous monitoring on trends thus helps detect new patterns that are to be added in the watchlist and dropping the patterns that may not be required to be tracked further.
Such solutions that are backed by data, drive transformation for processes of concern by triggering either a bot, workflow creation, next level investigation, training needs analysis, etc. This not only improves the processing accuracy of claims but also reduces manual efforts of downstream processes, thereby enabling reduction in overpayments, late payment interests, call volumes, and rework.
While investing in predictive analytics will allow payers to make the right decisions always, through better insights, its success also depends on adopting data mining solutions to bring the data together. The initial discovery of the dataset and bringing the right influencers in the dataset is equally important. Identifying the right set of attributes in the enormous dataset will also help ensure the success of the analytical model.
Additionally, moving away from a reactive to a proactive approach by leveraging prepayment initiatives like claims pattern detection will not only help bring down costs and administrative efforts, but will also help increase member satisfaction. Predicting incorrect payment, reducing revenue leakages, and getting claims ‘right the first time’ is undoubtedly the way of the future for payers.