Finsights Blog

The Benefits of Data Triangulation

Most manufacturers have an array of data assets at their disposal for deeper analysis of many of the issues facing their business.  However, these assets often remain sequestered within the purview of the various groups that use them for their specific business processes.  Few organizations are aware of the potential that could be unleashed if these resources were fully leveraged across departments.  While it is natural that data stewardship should gravitate to the functional divisions within a manufacturer, organizations taking a holistic view of the data assets that exist across the enterprise can realize significant gains.

The concept of ‘triangulation’ is used in various disciplines to determine the location of a point based on measurements taken from several other known points.  The essential concept is that looking at something from multiple angles imparts more information about an object or idea.  Whether you confirm what you already knew, or are forced to further refine your thinking, both outcomes are equally valuable.  This same notion can apply to data.  To repurpose a term from the social sciences, ‘data triangulation’ can be thought of as combining, or bridging between, two or more distinct data sets in order to impart incremental information not available from any single source.

For example, data triangulation may be applied to the 340B program in a number of ways.  This government program has grown dramatically over the past decade, and requires deep discounts.  Many organizations, including the HHS OIG have published surveys in recent years that highlight the challenges and disconnects ingrained in all aspects of the administration of this program.  While we wait for clarification and regulatory changes to remedy some of these shortcomings, manufacturers should not sit idly by while revenue slips through the cracks.

As a starting point, manufacturers can use channel data to verify that chargeback reversals triggered by end customer returns are issued as expected, and at the correct price.  For certain specialty products the latter is especially important.

An even more ambitious approach would involve multiple data sets.  For example, one might start by examining the contents of the HRSA Medicaid Exclusion File, and consider what can be inferred, not only about the expected purchasing behavior of the covered entities listed, but also about that of the covered entities not listed. If we compare our expectations from this analysis to channel data, does it reveal any unexpected purchasing behavior?  For manufacturers who have purchased or otherwise obtained Medicaid claim-level detail, does this data confirm expectations are being met, or does it raise additional questions?

Many of the data assets required for this type of analysis are already owned by manufacturers.  Others, such as Medicaid claim level detail are available, but are not necessarily easy to access, and can be cumbersome to work with and store.  Finally, some competencies, like the capability to link identifiers used for purchasing with their counterpart on the claims side, reside in areas few manufacturers have yet explored. But no matter what the state of a manufacturer’s current data capabilities, all should take the first step of inventorying these assets across the organization.

Initial focus should be placed on data capturing the movement of product through the channels to the end purchaser.  This data can take several forms, including chargebacks, EDI 867, and specialty pharmacy data, each with their own inherent strengths and weaknesses as determined by their intended purpose.  Data and capabilities around the spectrum of customer identifiers should also be included in this data inventory analysis.  Do you have a cohesive view of your customers spanning from identifiers used for purchasing (DEA, HIN, 340B) to those used for rebate claims (NPI)?  Answering these kinds of questions will be key to understanding the incremental value of coalescing and linking various data sources, and making them readily accessible for multiple groups within the organization. As a starting point, access is of particular importance for groups from which payments originate, mainly because it is easier to quantify the potential return on investment. From there, further analysis can identify root-causes upstream in processes that the organization can target for more proactive actions.

Cross pollinating data, knowledge and capabilities across groups will increase the validation and analytical capabilities of any manufacturer willing to invest the time and effort. Do you know what data and capabilities are available across your organization? The value they represent is likely to surprise you.





Sign up to receive our Finsights Blog recaps by email.