The perils of 'dirty data'

With the rising costs of health care and changes brought by reform, health care organizations must find ways to reduce costs. Though the health care supply chain has evolved rapidly over the last 15 years from the traditional “order fulfillment operation” into a “strategic analytics operation,” there’s still a significant opportunity for organizations to reduce costs by gaining more insight into their supply chain.

Supply costs, which account for 20 to 30 percent of operating expenses, frequently are not well-managed because the information needed to guide business decisions is incomplete, incorrect or in some cases missing.

This problem of “dirty data” reduces an organization’s ability to forecast, correlate and analyze supply and service spend which in turn prevents being able to identify and achieve savings.

Weak supply chain data management results in increased labor costs, lack of visibility into and potentially excessive product inventories, increased logistic expenses, potential for maverick spending, and inefficient contract management and accounts payable processes. In fact, industry research by Accenture estimates that through using data cleansing and advanced spend analytics, health care organizations can save 0.5 to 1.5 percent of their annual supply chain spend.

The root cause of dirty source data in health care is the lack of standardized product information. Unlike the retail industry which benefits from the use of universal product codes (UPCs), there is no widely adopted standard product description in health care. Though recent GS1 standards will go a long way in addressing this issue, adoption could take time.

In the meantime, here are some ways to determine if your organization has a problem with dirty data and best practices to mitigate the issue.

How to Identify Dirty Data

Here’s what to look for to identify the depth of dirty data within your supply chain:
•    Product descriptions that lack a consistent and standardized format (e.g. noun, application, attribute, etc.)
•    Products missing vendor or manufacturer information (e.g. name or catalog numbers)
•    Products with incomplete packaging data (e.g. missing unit of measure and conversion factors, as well as standardized to the ANSI code)
•    High rates of EDI transaction errors, and invoice or purchase order discrepancies
•    A high number of manual purchase orders routinely submitted
•    Poor contract pricing utilization with procurement processes

Some organizations instead elect to partner with a data cleansing provider because they offer recommendations on how to best implement corrections and have the scalability and resources needed to isolate and normalize anomalies within supply chain source data.

Four Steps to Clean Data Success

Organizations that have successfully cleaned up their supply chain data, share the following best practices:

1.    View data cleansing as an enterprise-wide endeavor — Buy-in and support are needed to adopt and implement these initiatives effectively. While materials management is the “gatekeeper” for many data management processes, effective data management requires identifying, involving and educating key stakeholders (e.g. Directors of Contracting and Clinical Directors) throughout the supply chain. Communication is critical, including informing stakeholders of both the positive and negative impacts of cleansing process changes.
2.    Establish effective data management policies and processes throughout the supply chain — Data management policies often fail due to lack of process performance measurements. Many health care organizations don’t know the number of duplicate records within their item master; the number of routine, manual record orders that side-step the item master; or the number of manual orders requested within a given month by department and requestor. The ability to report this information to stakeholders and establish attainable goals that are directly aligned with the organization’s data cleansing strategy will facilitate faster supply chain savings.
3.    Implement critical controls at each point where source data enters the supply chain — Critical controls at data entry points across the supply chain are key to managing costs. Adding, deleting and changing product information in the item master is important for successful contracting, procurement, inventory, accounts payable and reimbursement processes as is validating and enriching product information at the initiation of a request to add a record to the item master.
4.    Implement a proactive data cleansing approach — To realize the most value from the supply chain, the process must channel clean, accurate product information at the beginning of the process. Many of today’s item master maintenance processes are reactive in addressing new product validation due to limited staffing. For true value and savings to be realized quickly, this process must be proactive in approach and must begin channeling clean, accurate and enriched product information at the origination of the request.

Tim VanderMolen is a product manager at Novation, LLC, a healthcare supply chain and contracting company. He can be reached at tvanderm@novationco.com.