So, you’ve decided that you want to use data from the Colorado All Payer Claims Database (CO APCD) for your project. Brilliant! You’ve come to terms with the sheer magnitude of the database and narrowed down the mindboggling array of potential analyses and have honed in on the question(s) you want to answer. Now you get to choose the format you want the data in and how much info you want it to contain.

At the most fundamental level, non-public releases of CO APCD come in two formats, a data set or a report. A data set is just that, a big set of data elements that have been chosen and quality checked based on the questions you want to answer. But that’s all, it’s a lot of data and nothing fancy – you do the analysis. For reports, the CIVHC team performs analyses on the data so findings are more readily apparent, and, if requested, they can also create visualizations to help you understand the results.

Additionally, there are two kinds of data sets and reports – custom or standard. These are exactly what they sound like. In a custom release, the user chooses what data elements to include and, if a report, what analytics are performed. Standard releases have specific data elements and analytics already incorporated into their design. The benefit of standard data sets and reports are 1) you know exactly what you’re getting, 2) they are more cost effective than custom data, and 3) they can be delivered quicker than custom work.

Last, but certainly not least, in some cases, stakeholders need access to information in the claims data that contains protected health information (PHI) like date of birth or zip codes in order to conduct research that will ultimately improve the lives of Coloradans. For these requests, there are two levels of PHI for non-public CO APCD data releases: fully identifiable, and limited. * The amount of PHI in each breakdowns like this:

  • Fully Identifiable
    • Identifiable data sets provide enough detail to differentiate between patients. This data is extremely difficult to obtain; the CO APCD Data Release Review Committee (DRRC) evaluates each project to ensure that it meets all state and federal criteria for release and must recommend each request for approval. These types of requests also generally require review and approval by an independent Institutional Review Board (IRB) to ensure that patient privacy will be fully protected.
  • Limited Data
    • Limited data does not have anything that might directly identify a patient, such as names or medical record numbers. The PHI that is included in a limited data set can allow requestors to perform analysis by birth date, zip code, admission, discharge dates, etc. Because some PHI is present, the DRRC must review and recommend approval of all limited data set requests.

Requestors who don’t need any PHI receive de-identified data. Nothing in a de-identified data set can be used to recognize an individual, and therefore it doesn’t have to go through the same DRRC review process as requests with PHI. Instead, the DRRC has established criteria for the release of this type of CO APCD data and CIVHC evaluates and reports on each one to make sure that it meets the statutory requirement of benefitting Colorado.

Sound confusing? Don’t worry, you won’t be on your own sorting this out. The CIVHC team will help you to make each of these decisions based on your comfort level with data, your analytic requirements, and how much information is required to answer your question(s). If you’re curious about what type of release might be your match, check out our CO APCD Data Release Decision Tool. While each project is different, answering these questions can give a very rough idea of what type of release may be most helpful for your needs.

*Limited and Fully Identifiable CO APCD data is be available with all reports or data sets. Please email ColoradoAPCD@civhc.org for more information.