Physician Data Analytics – Where Do I Start?

Last month, I presented an idea and a scenario on how physician data analytics can be a key to a new payment model. But then comes the next question – “Where do I start?”



Recently, there has been “talk” about data analytics in the healthcare industry. Even in the hospital arena, a recent healthsystemCIO.com Snap Survey indicated only 23 percent of CIOs say their organization is doing some high-level, predictive analytics, while about 52 percent indicate they are performing some analytics – but not at a sophisticated level.

Based on our experience working over the past 3 years, there seems to be almost no data analytics being done at the physician-office level. In general, we have found that the larger the practice group, the higher the chances of some data analytics being done. Very little data analytics are being done in the small private practices with less than 5 physicians.

In a recent book Lean Analytics, by Alistair Croll and Ben Yoskovitz, several ideas are presented to help the “start-up crowd” get in the game of data analytics. A summary of these follows:

• Start with metrics in mind – In order to change any variable in practice, you must be able to measure it and know what “normal” looks like today. If you do not know what “normal” is, then you cannot tell if your efforts to change that metric are paying off.

• Find the one metric that matters – In any practice group, there are many variables that you could focus on. At the beginning, select the most important variable and related data element that will have a major impact on your practice. Focus on this one metric like a laser beam, measure the change, and see what impact this has on your practice.

• Recognize that not everyone will be happy with the results – The focus of data analytics is not to uncover the “norm,” but to discover the abnormal, non-obvious patterns of performance/behavior. In a physician group, you may discover that not all the physicians in the same group are practicing in the same manner with similar patients. The “outlier” physician(s) may, indeed, not be happy with the results of the data analytics, but in the long run, there will be a significant benefit to the practice as a whole.

• Ask good questions – Many of your patients will also be a potential wealth of knowledge about your practice, your procedures, workflow, service levels, etc. Again, by focusing on a particular issue and asking the right questions, you can gain considerable insight into how well your practice group is really performing and what areas would bring some immediate benefit to both your patients and practice.

For healthcare, one of the last frontiers is the use of data analytics in physician practices. Take a simple approach, focus on a key variable, and reap the rewards! Next month, I will discuss in more detail some key benefits of moving down the data analytics path.