Why is it that we expect, anticipate, and appreciate harnessing electronic medical records to improve the quality of care, but we aren’t talking about it at all from an infrastructure perspective? Isn’t it the same thing?
In an article The Power of Analytics Harnessing big data to improve the quality of care written in Healthcare Executive’s March/April issue the author discussed that Memorial Healthcare System in Broward County, Florida was moving from using clinical data to using analytics in general into their business practices.
“Memorial has become increasingly engaged in clinical analytics to help assess physician practice patterns and outcomes… In addition to accelerating its use of clinical analytics, the health system applied big data to its business practice” in their efforts to manage their vendors to get discounts and they are now using it for their budget and capital planning processes.
So what does that mean in terms of data when it comes to infrastructure data? Traditionally when BMS (Building Monitoring or Building Automation Systems) were first installed in healthcare we would measure the common data, we would log maybe 5 points. For example, with the air handler we would log temperature, humidity, and a few other control points.
Now with Building Analytics it is an entirely new world, it is every boiler, every air handler, the entire central plant, essentially the entire mechanical system. Now all the inputs, all the outputs, all the software set points, valve positions, air volume, etc essentially all information available from every piece of equipment. Every five minutes!
Warren Rosebraugh, Director of Solution Architects for Schneider Electric explains, “ What becomes difficult to manage is the sheer volume of information and the storage of that information. Think about a new integrated, truly integrated hospital. Here is where it gets tricky. Let’s start with an assumption, our experience tells us that 4,500 points would need to be collected per site, per hospital. Each of those points contains 300 bytes per record. So let’s do the math.
- Assuming 4500 points per site to be logged – Each log record is 300 bytes in length 4500 points X 288 entries per day = 1,296,000 records per site/day (every 5 minutes )
- 1,296,000 X * 1825 days = 2,365,200,000 records per site for 5 years history
- @ 300 bytes each = 709 Gigabyte per site for 5 years worth of data
- 100 site = 70.9 Terabyte for 5 years
Now, you need to add the redundancy requirement and it grows from 70.9 to over 150Terabytes of storage requirements. That five years of storage is being driven by the business or the clinical desire to do Analytics, in our case, Building Analytics to the infrastructure of the building.”
How do we integrate it? How do we manage it? How do we analyze it to be useful?
First, we integrate systems under a common suite of software sitting on the healthcare systems network. We are not just talking about the bed tower, a campus, but it is multiple campuses making up the entire healthcare system. Through an enterprise level Integrated Control Platform standards are designed to systematize naming conventions, energy monitoring and regulatory compliance. It is very important that we bring this information together in a central data store, so analytics in different forms can be applied.
“Here is why an Integrated Control Platform is so important”, stated Warren Rosebraugh. “Today big data is available through our technology because Schneider Electric has invested in PLCs and the software that can handle the throughput. This is a term that gets used a lot within healthcare, but usually related to electronic medical records. I was just at an IFMA HCI conference at Harvard Medical School in Boston and the presenter was talking about his healthcare system and how they had 400,000 terabytes of data for patient records in just a few years.” That is really no different than the infrastructure side of the business. Now, with an Integrated Control Platform, what we do every day to maintain a hospital can now be impacted by big data and the analytics that can be ran against it to improve the efficiency of your hospital.”
Tell me in the comments below about how you are managing Infrastructure “Big Data” in your healthcare setting!