Welcome back. In our previous lesson, we looked at improving data quality from a baseline. In this lesson, we'll look at dealing with changes, both expected and unexpected, in such a way as to manage data quality. After this lesson, you'll be able to list some expected and unexpected changes that a data analyst may encounter when managing data quality over time. You'll also be able to describe change control processes that are critical for maintaining data quality as things change. Let's get started. As systems evolve, our systems change, our processes change, our workflows change. This is true with regard to our work with clinicians, our work with patients and true of technology in terms of procedures, in terms of drugs, in terms of medical devices. All these technologies evolve as do computer systems. Over time, the data, the quality of the data, the types of data, the way we collect data, and the way we interpret data, necessarily change. In order to work with that, we can think of changes as something we can expect. There are certain kinds of changes we can even predict. When we upgrade a computer system for example, change workflows, change equipment, anywhere we make a change in the system, we might look to expect some sort of change in the data that's coming from that system. Most computer systems use some sort of change control process for their systems. This is true of most large computer systems, and most medical campuses with hospitals use computers to manage health care. For instance, a change control process will be triggered if a computer is going down, if an interface is changing and so on. These things are not left to chance nor handled manually. They are automated to ensure continual and consistent operations of these computer systems that have become essential to ensuring ongoing health care services. There is a process in place, to making sure that all changes are reviewed. There is a quality assessment. You usually go through some sort of acceptance testing, and there is documentation that things are going to change, and how those changes will have impacts on the data you collect, and the care that patients receive. When we look at the culture of data quality we want to encourage in the hospitals we work in, and the medical centers we work in, linking closely to change control in both medical systems and computer systems is critical. This applies when we're working with data, and when we build our processes for maintaining data quality. The change control process is critical in terms of data quality. We can anticipate that the quality of the data will change as the systems' environments and operations change. By embedding the questions of how any expected change will affect data, we can get in front of the changes that will come. When we know that a system will change on a particular date, that knowledge gives us the opportunity to anticipate how to prepare for upcoming changes, while ensuring data quality. This applies to changes that will be made either in work or in the computer systems. Preparing for expected change becomes a way to be prepared to make necessary changes in the ETL processes; that's the extract, transform and load code that feeds our data warehouse. You'll know the ETL code will need to be changed along what ever other changes you'll encounter. What we'll really be able to see, is that the quality of the data will change the system's environment and operations change. Monitoring data quality through these changes, allows us to improve data quality over time. And really, by being involved in the change control process, we can ensure we're moving towards higher and better quality. That our data becomes closer and closer to our ideal, and that we're able to build those processes. That building those processes, as part of the culture of our organization, is key to creating the highest quality data and giving patients the best care possible. In our view, as systems change, we have the opportunity to look at data quality and ask the question, will this change affect our data quality? If this change will affect our data quality, how will we adjust our data through our load processes, and through our data life cycle to manage the quality based on the changes that we will see? In our next lesson, we'll consider monitoring strategies along the data pipeline.