Is your software an information system? This may sound like a strange question, but when you think about it, does your software provide you with clear, concise, decision making information? In most cases, the answer is no. The most common issue users voice regarding their current software, or when seeking a replacement is that they cannot retrieve information from what they are currently running. They often view the lack of canned reports as the problem. Unfortunately with this belief, they will most likely never receive quality information from their software. The true problem is not the reporting, it is the lack of data integrity. Too often software is implemented and used as a transaction processor. Users tactically enter data into the system with little regard for data coding standards in an attempt to complete the transaction as quickly as possible. While this may be quicker on the front-end, it results in data that has no reporting value.

Improving the integrity of your database is a multi-faceted initiative that requires a focus in several areas to ensure long-term sustainability. Without addressing the root cause of why data is inaccurate, the clean up process will improve data quality for a point in time but it will quickly begin to break down. For long-term sustainability, the initiative must focus on policy, procedure, and end user education in conjunction with the actual cleanup of the data. If your plan or service offering is missing any of these elements, your results will be short lived.

The process begins by defining what information you need to access. This is imperative because the Business Intelligence you are seeking will have a direct impact upon database maintenance and transaction processing procedures. Information such as cost per procedure, total purchase volumes, and clinical outcomes requires the procedure to address data coding standards, what data is captured, when is it captured, and who captures it. With this understanding, data coding standards and the data default hierarchy of the system to facilitate data capture are defined.

The physical activity and process for data clean up may include entering data directly into the system or extracting data and updating it outside of the system. The decision should be made based upon the ease of update, the impact upon historical information, and the overall effort for making the change. Improving data management may impact your ability to retrieve historical data but this may be a small issue if either information currently has little value or there are tremendous daily operational issues. Which ever route the organization chooses, the key is attention to detail. The clean up effort must be performed accurately and to the specification outlined in definition stage.

Long-term sustainability of data integrity requires the definition of new policies and procedures for database maintenance and the enforcement of such policies. If data is of value to you, begin by restricting the number of data entry points for your master files. While this may slow down the front-end of the process, the time expended is minimal compared with attempting to create information from bad data. Finally, provide education sessions for both the end users and the impacted “customers.” These training sessions should include an explanation of why the changes are being made as well as an escalation procedure in the event your customers have a transaction slow down.

Your software’s ability to provide quality reports does not happen by default. It occurs through a commitment to data integrity that is recognized both within your department and your customer base. By putting the proper infrastructure in place, you can ensure your software performs as a true information system.