Improving health and healthcare institutions requires understanding of data and creation of interventions at the many levels at which health IT interact and affect the institution. These levels range from the external “world” in which the institution operates down to the specific technologies. Data scientists find that, when they aim at implementing their models in practice, it is the “socio” components that are both novel to them and mission critical to success. At the end of this course, students will be able to make a quick assessment of a health informatics problem—or a proposed solution—and to determine what is missing and what more needs to be learned.

As health informaticists, we need to be very clear in our understanding of the current state (as -is), the future state (to-be) and any unintended consequences that can result from our interventions. Prior to introducing large scale change, we need to assess whether a healthcare organization is truly ready for change. This involves taking into account an organization's current culture and values. Successfully leading change through health informatics also requires strategic planning and careful financial considerations. Proper workflow redesign and a clear change management strategy are of utmost importance when introducing new technologies and in ensuring their successful adoption and proper use.  By the end of this course, students will become familiar with examples of successful and failed attempts at change in health informatics, and the reasons for each. Students will be armed with tools to help optimize their chances for successfully leading change in their respective organizations. 

For clinical data science to be effective in healthcare—to achieve the outcomes desired—it must translate into decision support of some sort, either at the patient, clinician, or manager level. By the end of this course, students will be able to articulate the need for an intervention, to right size it, to choose the appropriate technology, to describe how knowledge should be obtained, and to design a monitoring plan.

Health data are notable for how many types there are, how complex they are, and how serious it is to get them straight. These data are used for treatment of the patient from whom they derive, but also for other uses. Examples of such secondary use of health data include population health (e.g., who requires more attention), research (e.g., which drug is more effective in practice), quality (e.g., is the institution meeting benchmarks), and translational research (e.g., are new technologies being applied appropriately). By the end of this course, students will recognize the different types of health and healthcare data, will articulate a coherent and complete question, will interpret queries designed for secondary use of EHR data, and will interpret the results of those queries.