Healthcare Issues

  • A common objective in health plans is to analyze levels of activity based on the member-months currently enrolled. Our firm has developed fact tables specifically for this type of analysis which allowed our clients to evaluate their performance independently of any ups or downs in their membership base.
  • Only rarely have we been fortunate enough to work with source systems that had reliable change detection so that the source system could indicate to the data mart when something had changed. Therefore we've gained a lot of experience in using the built-in capabilities of SQL Server to detect changes in source data. This allows the creation of powerful tools such as Slowly Changing Dimensions to track changes to entities (e.g. Providers, Members) over time.
  • In healthcare it is common to encounter a situation where a fact may have more than one relationship to the same dimension at the lowest fact grain. An example would be that a single line on a claim may have more than one diagnosis associated with it (a primary and a secondary.) We've seen this situation in other industries and developed modeling techniques that can deal with the situation without causing problems for the IBM Cognos business intelligence applications.
  • Several of our healthcare clients have wanted to work with large amounts of claims or financial data in a high-speed analytical process (frequently interactively in a meeting.) Our expertise in Online Analytical Processing (OLAP) technologies such as IBM Cognos DMR or PowerPlay as well as Microsoft Analytical Services (MSAS) has been combined with our healthcare specific expertise. This allows us to create systems that can have less than three second response times when working with just under one billion rows of transactional data and navigating healthcare specific hierarchies such as ICD-9 codes.
  • In the Insurance industry a central challenge is to provide information around the entire lifecycle of a claim. A claim goes through various states and frequently some of the most valuable information for analysis is how many claims were in a given state at a given time or how long claims are staying in particular states. This is very difficult to do with just a simple Fact table. One solution we've recommended often is called an accumulating snapshot. This is a table that supplements a Claim fact table by capturing on a regular basis (usually Daily) how many claims are in what state and how long they've been there.
  • Both providers and payers are interested in measuring the positive impact on Outcomes via evidence based practices and formularies. We've worked with some of our clients to link together sets of encounters into episodes in order to better measure the positive health improvement outcomes.
  • Studies have shown that successfully managing the number of providers and services delivered during the treatment of a disease or chronic condition can drastically improve the patient's quality of life as well as controlling cost. We've worked on projects to flag chronic conditions and compare the services delivered to these patients against standard protocols.