PHS can divide master data into two types:
· Identity data – such as patient, provider, and location identifiers.
· Reference Data – which includes common linkable vocabulary like ICD, DRG, SNOMED, LOINC, RxNorm, and order sets (give reference)
Based on the above categorization, MDM will be the process of linking identity data and reference data across multiple IT systems into a single, consistent point of reference which could be a patient, or a procedure code.
According to “Master Data Management in Healthcare: 3 Approaches (https://www.healthcatalyst.com/master-data-management-in-healthcare-3-approaches)” MDM is defined as follows: “MDM comprises the processes, governance, policies, standards, and tools that consistently define and manage the critical data of an organization to provide a single point of reference”.
A strategy can be considered an approach using key components and best practices that would lead to a successful Master Data Management. Partners Healthcare System can consider the following key components that should be factored into the organization’s overall MDM strategy:
1) PHS should keep in mind process considerations. This would allow PHS to include crucial capabilities such as data identification, for the MDM framework.
2) Identifying and validating sources of data is crucial. This also includes any external data providers (physicians, nurses, hospitals partnered with PHS, feeding them data).
3) Implementing a data hub that would act as a core to the entire framework. This data hub can be responsible for providing services for data management.
4) The above leads to the necessity of implementing strong data management support within the master data hub.
5) The MDM architecture should be flexible and adaptive to ensure high performance.
6) All employees involved in the process directly and indirectly must be trained and kept up to date on MDM and how to access data.
7) Since PHS may have multiple sources feeding the data to its hub, it should understand the sources and quality of their master data. It will also prevent any inconsistencies with the data stored in different systems.
8) Access should be restricted to only authorized employees and appropriate security measures should be in place to prevent any unauthorized access and to protect the data. This step would also help monitor data quality in the hub.
9) When PHS decides to roll out MDM, they should start small, focusing on smaller data sets that could be the potential pain points.
10) The approach to applying MDM to different domains must remain consistent.
11) In the end, MDM should be updated regularly to ensure optimal performance.
Keeping the above-mentioned strategies in mind, PHS can implement Master Data Management in the following different ways:
1) IT System Consolidation
Abandoning best-of-breed solutions for EMR and ERP solutions can be a recommended way to address some of the MDM challenges. IT system consolidation is not only comprehensive but also allows a patient to be matched the moment they are entered into the system. This is a great example of transactional systems.
This would work best in the case if PHS is trying to solve MDM challenges within an organization. If it outside data to be integrated with the master data, it may require more steps for MDM between those different data sources.
2) Upstream MDM Implementation
Taking advantage of Enterprise Master Patient Index (EMPI), PHS can use the upstream MDM implementation approach to map their master data. In addition, these systems allow more flexibility.
3) Enterprise Data Warehouse (EDW)
PHS can downstream master data reconciliation in an enterprise data warehouse. The mastered data is only good for analysis. This approach will not solve MDM challenges at the transactional systems level but, EDW could be the easiest solution in case the two above mentioned approaches are not useful and there is no solution in place.