Cleaning power: Clean data is a powerful tool - Surgical Business

Healthcare Purchasing News, April, 2002 by Suzanne Noble

Cleaning and backing up your data is probably a chore you find about as urgent and exciting as watching the grass grow. Why clean data? Because clean data is the foundation of a successful e-commerce initiative. Clean data improves the ability of a hospital and a department to provide the best service at the most reasonable cost.

Clean data can help answer these important questions: Where does my department stand regarding contracts? What are the best ways of taking advantage of contracts? Typically there is enough bad data in any given database to prevent managers from making business decisions with confidence. At its worst, flawed information can lead managers to make ill-informed decisions. This is especially true when it comes to contracts with GPOs and other suppliers. Compounding these frequently overlooked pitfalls is the general confusion that goes with locating and cleaning data.

"Dirty" data comprises the weakest link in the supply chain. The maxim GIGO (garbage in, garbage out) -- rings as true as ever. "Do you want to move garbage at the speed of sound?" asks Scott Minniear, founder, president and CEO of AristoMed, a data management solutions and purchasing process improvement firm based in Salt Lake City that focuses on e-commerce readiness.

Minniear, a familiar name and face to many in the GPO and vendor communities, began as a medical equipment sales rep for Midmark Corp., Versailles, OH, then moved to become vice president and a partner with National Contracts Inc., a healthcare consulting firm now based in Palm Harbor, FL that represents medical manufacturers to GPOs and multi-hospital systems. He has also spent time as vice president of sales for Barriers For Diseases Inc., a manufacturer of medical disposable products. He left the manufacturing side of the business to co-found Medical Companies Alliance, UT-based think tank partially funded by Premier Health Alliance and Child Health Corporation of America that explored methods for small- to medium-size. companies to more effectively compete in the healthcare industry. Minniear also consulted for Columbia/HCA Healthcare during that chain's heyday.

Bad data, he says, is both contagious and wasteful. Bad data doesn't just evaporate into thin air; it can continue to spread by being merged with other good data or repeatedly used within various summary reports. It's wasteful because managers and employees are frequently forced to filter out inaccurate data by manually creating new spreadsheet versions of existing database reports. Catalog number changes can become insidious, literally targets that move alongside the rhythm of supplier and provider consolidation. Print catalogs are almost always a step behind; electronic databases are dynamic, but they too must be kept current, consistent and correct, says Minniear.

The cleanup process

"Data cleaning your files and system may seem insurmountable, but it is a manageable project if you approach it as a process," says Edward E. Carty, Jr., vice-president of operations for vGPO in Dallas, a company that serves as an online solution for managing GPO purchasing contracts and seeks to uncover savings from those contracts. "Large-scale cleanup is generally achieved by either filing manual correction transactions on a case-by-case basis until the problem is resolved, or by 'scrubbing' submissions with predetermined rules or algorithms and executing routines that follow those rules against all the submitted data."

Carty suggests ways of determining purchasing volume through a particular vendor. When requesting volume from J&J, for example, returned data might not be accurate unless every item from J&J was coded with the same identity. J&J can be spelled in several ways. J and J, Johnson and Johnson, and even JNJ, the company's stock ticker symbol, are just a few examples. Unless you identify each and every label, meaningful data might not be returned. This issue is compounded even further when trying to clean up data involving a vendor that has merged with another.

A statute of limitations usually applies for retrospectively adjusting for errors in the terms of a contract. Carty recommends reviewing the previous 12 months to examine data for errors, reconciling line item details of the errors and getting the system back in synch. He says that vGPO's "scrubber" solution, known as vScrub, contains a process orientation for cleaning data in a fluid landscape by cleansing and enriching purchasing data into meaningful and actionable information that is critical to any supply chain strategy.

Moving forward after the initial data cleanup, quality materials management information systems incorporate algorithms to prevent future errors by grouping aliases together. The program learns and gets smarter as it gets cleaner.

Cleaning up helps pinpoint the right product at the right time at the right price. Apply this principle to all other contract opportunities, says Carty. Here's just one scenario that shows some of the worst that can happen with bad data: An essential product for a 7 a.m. surgical procedure has been ordered, the patient is prepped and under anesthesia, the surgeon is poised with scalpel in hand. But the supplier has sent the wrong item and the case cannot begin. Of course, this inevitably produces an unhappy patient and an even unhappier surgeon who could leave the hospital, not to mention a range of unanticipated costs. It is helpful to look at the total cost concept of a wrong item. The patient is unhappy and must be rescheduled; the hospital has lost the case and may lose the unhappy surgeon. And, the O.R. must be prepped for the next case, even though it wasn't really used. A costly mistake indeed.