A Decision Support System (DSS) is a set of related computer programs and the data required to assist with analysis and decision-making within an organization. Simply put, it is the technology tool and the information that supports your work.
The need for data is crucial as it supports many everyday decisions in today’s healthcare environment and will, in all probability, continue to do so in the future. While the healthcare industry uses a variety of DSS systems such as Clinical Decision Support systems or Pharmaceutical Decision Support systems, this blog is focused on Cost Accounting Decision Support Systems.
Liveware, Hardware, Software – What is the Difference?
Organizations can build their own decision support system, but most choose to purchase a packaged tool. In order for a DSS to be effective, three components are necessary: liveware, hardware and software.
Livewareis simply the people who develop, maintain and use the system. The responsibilities of an organization’s liveware will vary significantly from database administrators to data administrators to end users; however, all of these roles will require domain intelligence. Domain intelligence is the knowledge of how a particular business works, its data and database requirements. This knowledge is crucial for not only employees but also consultants who are supporting an implementation. All participants need to be educated on all business rules, requirements, and unique data elements and layouts.
Hardwareshould consist of four components: a relational database management system (RDBMS), an application or online analytical processing server, a presentation system and an end-user workstation. The RDBMS provides the horsepower necessary to sift through large volumes of data and send the results to the requestor. The processing server or online analytical processing (OLAP) is designed for information retrieval and analysis. The presentation system must be able to handle numerous requests for information. The end user(s) using the DSS software will need a work station to operate from. Files do not need to be stored on the workstation, thus the capacity for this hardware can be lower than the other pieces described.
Software has two components; system software and application software. System software has two primary functions; to move data from a data source to a data warehouse using extraction/transformation/loading (ETL) software and data modeling software to aggregate and summarize data. ETL software transforms the data, making corrections to ensure that entries are consistent with the field requirements of the data modeling software.
Which Data is Required?
The data that feeds into a decision support system comes from your EMR and billing system. Decision support teams require as much data as possible for their analysis. Specific information for each patient encounter is required, including demographic information for the guarantor and the patient, as well as specific transaction information. The data requirements are extensive going into coding detail that will need to match up to specific transactions.
Basic data includes demographics for patient, guarantor, service and billing providers, surgeons and co- or assistant surgeons. Insurance information should match to each account for the patient and the guarantor and include subscriber information and co-pay/co-insurance information and expected payment if this is available from your EMR/Billing system.
Intermediate data begins to become more complex to obtain, especially with integrated systems. This data will generally require linking database items and formatting of the data to enable the information to load into the DSS system without errors.
You will need to work closely with the end users in decision support to develop, test and validate all items to ensure that they will receive the correct data in the required format for analysis. Depending on the size of the facility, you may also need to work closely with them to determine frequency and timing of the extract run.
The Value of Decision Support Extracts
Decision support extracts are detailed and intricate to create. While they require much patience and perseverance, they offer the analyst and/or developer the opportunity to learn where their data lives and how it is linked. This knowledge is invaluable for both future builds and troubleshooting issues. Don’t be afraid to ask questions to hone your support extract skills.