Using analytics to create an ‘Eye in the Sky’ to monitor radiology workflow
Dennis Loucks, CEO of Automated Health Management Services in Houston, a radiology management firm that coordinates well over 100 radiologists, compares Corepoint Integration Engine to the popular Waze highway traffic-monitoring smart phone app that alerts drivers to real-time traffic hazards such as congestion, construction, wrecks, and even police speed traps.
Using Corepoint Integration Engine to facilitate interoperable health data flow, Automated Health Management Services employs a business intelligence analytics dashboard that serves their workflow similar to a helicopter flying over a city traffic grid, with each interface acting as an intersection with the potential to stall radiologists’ workflow.
Dennis was kind enough to explain how his partnership with Corepoint Health has played a key role in Automated Health Management Services’ success.
Q: What were your initial goals, two years ago, when you implemented Corepoint Integration Engine?
Dennis: Initially we teamed with Corepoint Health to deal with billing issues. The integration engine originally was used to perform simple hospital-level report routing that allowed us to manipulate the feed and automate billing functions.
We put Corepoint in front of our different PACS systems to be the sole feed to and from those systems. If a referring partner submitted orders, they landed in the engine. We normalized the data from that order for the server and dropped it in. When the report was dictated by the radiologist, we’d use the engine to grab the ORU and sent it on its way. That process was relatively easy with Corepoint.
It got a little trickier when some referrers required ORUs in a different format such as encoded PDFs, plain vanilla text, or custom-formatted HL7. But not too big of a challenge with the integration engine.
I had plans to implement a real-time monitoring system where i could observe data flow and gain a better picture of our radiologists’ work load. I wanted to know in real time what was going on with radiologists in different locations. unfortunately, there wasn’t anything other than orders and results floating around between PACS systems.
The cool thing about using Corepoint Integration Engine is that i didn’t have to rely on PACS to gain control of the data. i’m sitting on the middle of a data freeway, picking up information that is able to tell me exactly what’s going on at each of our locations.
We started using the integration engine to populate SQL databases that in turn feeds dashboards, which gives us real-time analytics for radiologist workload projections, and much more. it’s extremely beneficial when managing over 100 radiologists in different locations.
During the day, if things are backing up, we know right where to go to find the bottlenecks. if a radiologist decided to leave early or take a long break, I’ll see their orders not going anywhere and be able to take the appropriate action to get things moving again.
What made it even better is that Corepoint Integration Engine is extremely stable – it keeps running and running and running. i never have to worry about it going down or hanging up our dashboards. in fact, if anything gets hung up, it’s usually with one of the other applications.
What advice can you give to other providers looking to get started implementing this type of business intelligence?
To get really good analytics, you’ve got to have a good foundation, and a good foundation for analytics is based upon a consistent flow of information. your analytics are going to be more accurate and tell a more complete story as a result. with Corepoint feeding the data, i know without a doubt that the data will be normalized, regardless of the standard.
True or not, there are a lot of complaints within the industry about the data siloes created by EHR vendors and especially their lack of data interoperability. how have you been able to get past these perceptions to gain these important insights?
Let’s say you’re working with a big hospital system with two different, predominate EMR systems and you want to create some commonality. the key is to not deal with the EMRs – go straight to the hospital to get a copy of their feeds. that way you get all the same data they have and you can create your own solutions.
It just seems like people don’t think that through. they want the EMR providers to be able to do that for them. data sharing is a money-making proposition for EMR providers and they won’t operate outside of their cloud at all, typically. it’s going to be their cloud or the highway.
I consider our operations a cloud as well because we’ve been building this business and data infrastructure for so long. why would i want to drop my data on their cloud? they’re collecting data for the same reasons that i do, the only difference is that it’s not going to benefit me at all.
I’m a big proponent of trying to stay local and independent of the vendor. if you can go straight to the hospital, and they almost always are willing to work with you, then you’ve got a much wider selection of information and metrics to deal with than you ever get through an EMR vendor, who are notorious for filtering out things that they don’t think are important.
In other words, if the data is not important to their product, then it’s not important. i’m not sure i see it that way. i like to know that order was put on the queue by the PACS at a certain time, and then i can see how long it took to actually perform the study. once the study is performed, how long did it take until it landed on the radiologists work list? when did they dictate and when did it get sent back to the end user?
Those kind of metrics are really important in terms of demonstrating turnaround times to the hospital, and you can’t get that from EMR or PACS vendors.
Any specific examples of what you’ve been able to correct with the data?
A hospital administrator was complaining because turnaround time was bad and we find out their night time staff are gathering all the studies every two hours and dumping them on our radiologist, so it looks like the radiologist group is awful. you can’t figure that out unless you have all the different metric points.
It has a big impact especially if you’re doing final reads at night. you can see what’s going on and try to figure out any bottlenecks. when you have the data you can see that the machine’s not down, and you know the network’s not down because everything is still green. so where’s the work? somebody’s sitting on it, and now we can get to a solution.
We wouldn’t know what’s going on if we weren’t able to look at it from a 40,000-foot view, a lot like a helicopter flying over the traffic grid. monitoring traffic patterns and using data analytics have a lot in common in our operations. each interfaces is a new intersection in the highway of data somewhere, and we can’t monitor anything without the data Corepoint provides.
Were you able to perform data analysis prior to partnering with Corepoint Health?
The fastest way we’d get anything from the PACS systems would be the next day. it would take another 48 hours or so to parse the data, which obviously is not real time and gives me no actionable insights.
Now we are able to automate billing workflows using HL7 and the engine – in fact we don’t touch paper anymore, everything’s workflow driven. the engine actually drives each workflow and saves us a lot of time and reduces the chance of order-entry errors that can occur.
Every time a radiologist dictates his report, we have the billing information within seconds. this allows us to know on a daily basis or an hourly basis who’s done the work where, and how. analyzing that data over a longer period of time gives me insights on where the manpower needs are.
How has Corepoint Integration Engine helped improve other areas of workflow?
I picked up a facility that was printing reports all day for billing. we set them up to use HL7 and created some workflows that allowed them to stop printing and even stop handling paper all together. that dramatically improved their daily work – the two people who handled the paper reports could barely keep up and barely had time to sit down and code the reports.
Something that seems minor on the surface can result in a huge improvement in those individuals’ daily work. they had no idea that everything could be automated. they told me, “i don’t know how you did that, but that’s the greatest thing ever happen to us!” that kind of thing goes a long way to improve morale, productivity, and profits.
We’ve done some things with the Corepoint engine that i don’t think i originally imagined doing.
What makes radiology different than other areas of healthcare in regards to technology?
Radiologists are facing technology all day — right in their hands — while other specialties talk to patients more and deal with the technology less, typically. there’s a big chunk of the medical community that doesn’t really have to deal with that much technology unless somebody hands them a hand-held device or they access an EMR.
When you start looking at what it takes to move a DICOM file from one system to another, and in some cases even restrict portions of the DICOM file and create HL7 messages like Corepoint can do, it involves advanced technology. those are things that are taken for granted in the radiology community. in other communities they ask me, “How do you do that?”
A typical radiologist work station has three monitors, it’ll have a work list and then two monitors. that work list gets populated from somewhere, typically through an order. using Corepoint, we auto-create an order from the header information of the DICOM file. Corepoint will unwrap DICOM and use XML and populate that radiologist’s work list. the radiologist can then see they have a ct, but then they might also see they have something else that’s a stat order that needs to be read right away.
When they dictate results using Powerscribe, or whatever, it generates a message out of their PACS system. typically we’re taking all of those messages back and we deal with the distribution and everything else in the Corepoint engine.
You’re the primary user of Corepoint Integration Engine, yet you’re also the CEO – why take such a hands-on approach to your data?
This is so near and dear to our traffic pattern that i really like trying to make sure everything flows nice and tight. it’s not a typical CEO function, but it’s my game, and i’ll play it the way i like (laughs). there’s certain things that i like to do and i don’t want to necessarily give it all up, especially this portion of the business.
There’s nothing better than having Corepoint in place because i really don’t have to worry that the data going from point A to point B like they’re supposed to.
What advice would you give other radiology practices about getting control of their health data?
The Corepoint Health product has really helped me get ahead of the curve and stay there, and that’s because i learned what to do with the data. really figure out how to use corepoint integration engine and you can save yourself a fortune of time.
I feel that if i can get a handle on the data, i will be ahead of the curve. I also feel i get a better result with the people who develop for me because they have a better understanding of the product that they’re working with.
You might also like
Royal Berkshire NHS Foundation Trust
Royal Berkshire Hospital solves staffing shortage with managed services and consulting from Lyniate.
Before deploying Lyniate Rhapsody, there was a growing need for NWAS to share a broad range of information — such as ambulance numbers and locations — with its health service partners. The sharing of information would allow healthcare providers in North West England to better manage patients, draw better analytics, and increase efficiency within the organisations.