The Healthcare Hub

Clean Up Dirty Data in the Item Master: A long-term plan

Thursday, December 7, 2017

While every healthcare organization has its own unique challenges when it comes to item master data management, many of the problems are the same – inaccuracies, duplicates, missing information – in essence “dirty data” that causes errors and rework throughout a number of operational and clinical processes (e.g. procurement, inventory management, patient billing).

Both Catholic Health and Oregon Health & Science University have implemented master data management strategies that leverage the NuVia® content management solution. While these organizations have derived significant value from their work, they can tell you that best-in-class data management doesn’t happen by accident. In this Q&A, Ashleigh E. Gross, supervisor for Supply Chain Analytics at Catholic Health, and Dina Walden, Content Manager for Oregon Health & Science University, share tips on how they developed sustainable processes to achieve and maintain quality data.

Q. What prompted you to embark on your master data management journey?

A. Ashleigh E. Gross
We are a little unique in that we have a larger item master – over 230,000 records, including pharmacy items. We process approximately 30 item adds daily, mostly because we don't allow special items to flow through our system.

We wanted to achieve GHX Best 50 status in 2016, so we needed to lower our price, SKU and unit of measure (UOM) exceptions. In order to do so we had to focus on cleansing our existing item master data, while at the same time minimizing the noise that comes with adding so many new items per day. Furthermore when adding new items, we had to avoid reintroducing the same bad data habits that we had prior to our item master clean up (e.g. errors, missing information, duplicates).

A. Dina Walden:
A few years ago our perioperative department conducted a Kaizen event to address missing, inconsistent and incorrect information in our item master. These issues were preventing us from fully documenting consumption, billing for it and replenishing items. This led to rework, adjustments, inaccurate billing of patients and increased operational costs.

Following the Kaizen event we brought our goals to the table. For billing metrics we wanted to identify all possible billable items. We discovered that poor item descriptions were the main reason why coders were unable to quickly identify items that were billable to patients. We also wanted to reduce miscellaneous billing and on-contract items with incorrect pricing. In addition, we wanted to align our item file to industry standards, including descriptions, manufacturer numbers, vendor numbers and Healthcare Common Procedure Coding System (HCPCS) codes.

Q. Standardization of product descriptions can be a major undertaking since it requires support from both operational and clinical stakeholders. Dina, how did you achieve this?

A. Dina Walden:
There were vast differences between the descriptions in our manufacturers’ catalogs versus the hospital’s idea of what those descriptions should be — we were excited to get everyone on the same page. NuVia offers a standardized description with options to abbreviate all or part of it, which came in handy because of downstream system character limitations.

We collaborated with nursing staff, re-stockers, the perioperative team and representatives from each of the downstream systems to come up with a 100-character, unabbreviated, NuVia optimized description for each of the products in our item master. The NuVia full description is always available to users as well. We really like how NuVia breaks out every one of the 22 attributes for us to see. It is wonderful to have such enriched data — we didn't have that level of insight before.

Q. Ashleigh, you said that your team has been both reactive and proactive in your data management approach. Please explain.

A. Ashleigh E. Gross:

We’ve taken a two-staged approach with our use of NuVia. During the first stage we were more reactive in making correction actions in our item master based on fatal exceptions that we see on a daily basis. For example, a manufacturer had hyphens in its part numbers within its catalog but for electronic data interchange (EDI) purposes the company had removed the hyphens. After we came across this problem, we did a mass processing to clean up all of those exceptions in our system. The second stage of our implementation has been more proactive where we have applied what we learned in the reactive phase. Now we are taking a more global approach to normalizing supplier catalog numbers on a vendor basis to prevent those fatal exceptions from happening in the first place.

It’s also important to mention that with any implementation, especially with the size of our item master, we needed to pick a focus point and set small milestones to reach our end goal. If we had tried to process 200,000+ changes at the same time it would have posed far more challenges than addressing the fatal exceptions first. It was easier to get that out of the way and then focus on the bigger picture rather than having to constantly deal with these exceptions on a day-to-day basis.

Q. Dina you mention having to take into consideration the impact on downstream systems when developing your master data management strategy. Could you please elaborate on this? 

A. Dina Walden:
In order to allow NuVia to push information throughout our supply chain we first had to identify any system limitations. Our Oracle item file feeds many downstream systems, including our Epic electronic health record (EHR) system, as well as our point of use, warehouse, sterile processing and patient billing systems. As I previously mentioned, we had to look at each of those systems to determine where we had character limits so we could address them prior to implementation.

We also had to understand the details of how the data would flow from NuVia to our Oracle materials management information system (MMIS). It is a loop from our Oracle system to GHX Connect Plus to NuVia. We also stream data from our group purchasing organization (GPO) into GHX Contract Center, which flows into NuVia so we have access to contract pricing. Once the item file is synchronized and approved it comes back into our MMIS from GHX Connect.

Q. What advice do you have to other healthcare organizations that want to implement a master data management strategy?

A. Ashleigh E. Gross:
Good quality data is achievable regardless of the size of your item master – we had over 200,000 items and we were able to achieve GHX Best 50 status with very strict data requirements so you can achieve it as well. It is important to stress that data management is an ongoing process where you are constantly looking at different segments and always working data to ensure it stays clean. There will always be bumps in the road but lessons learned along the way will pave the path to your long-term success.

A. Dina Walden:
We went to NuVia for item description clean up and came away with a lot more. We’ve reduced billing errors and aligned with industry standards (e.g. HCPCS codes), while improving the efficiency of our data management processes. 

Executive buy-in was critical to our success. It was the only way that we could manage this level of change within our organization and across so many stakeholders. Communication from the top down was another success factor. Individuals learned about the initiative from their leaders and kept hearing about it as the work progressed so there were no surprises.

Listen to the GHX Summit Series webinar recording - Data Management: Operationalizing Best Practices and Driving Value from Data – where Gross and Walden share their data management journeys, including actionable insights and takeaways that you can leverage to create value in your own organization. The webinar recording can be accessed here: