Data is important. Using data to measure company performance and understand customer tendencies has shown to have a positive financial impact in many industries. While healthcare systems may have gotten off to late start collecting and turning data into information, there have been significant improvements in the last decade. This can be seen in the increasing adoption rate of ambulatory electronic heath records in US healthcare systems (Timothy Ferris, Partners Healthcare). However, in order for healthcare systems to truly benefit from this information we need to take data analysis one step further. There needs to be a defined methodology for how this information should be used to make decisions and a defined appointment of authority to execute these decisions.
My intent is not to discredit data analysis. The process of turning data into information is no small feat – it requires a complicated, yet well defined, methodology for “governing” data from definition through analysis. Rather, my intent is to bring awareness to the need for an INTEGRATED governance model in healthcare systems. A governance model that provides standardization at the data level and empowers smart decision-making at the operational level. For example, there is a huge difference between having the knowledge about how many patients occupied your hospitals beds last month and having the capability to schedule hospital staffing around the number and type of patients predicted to be admitted in the next month. How can we get to this level of decision-making?
Let’s start with something familiar… during my experience as a management consultant, I learned that the process of governing data was implemented by performing three main tasks:
- Defining a clear set of standards for how data is defined and collected (aka data standardization)
- Selecting the location for and method of data storage (e.g. data warehouse)
- Determining the tools used to analyze and make sense of the data (e.g. queries and formulae that generate dashboards using historical information)
As far as the resources needed for implementation – an organization should complete the first step internally in order to ensure the data standards align with the objectives of all its stakeholders. Steps two and three can be handled both internally and/or externally as there are many established practices and vendors that offer solutions.
Is it too optimistic to assume process scalability? Perhaps. But I guess the real question isn’t whether the data governance process is scalable, but whether it can be applied at all levels of an enterprise, and, whether it can be applied in a way that catalyzes enterprise integration. If we, as a nation, are committed to improving the health of our population while decreasing the cost to care for individuals, we cannot stop at merely creating information. We need to combine data analytics with professional experience to make and execute insight-driven decisions – decisions that will increase the quality of care delivered while decreasing the cost to deliver it.
I look forward to discussing this and the many other hackable areas of medicine this weekend!