In partnership with CHIME, healthsystemCIO.com has developed a blogger series featuring insights from hospital and health system CIOs and other key IT leaders representing organizations from around the country. The blogs focus on the major issues affecting CIOs, including the health IT workforce shortage, mobile device management, and federal regulations.
________________________________________________________________________________________________
George McCulloch, Deputy CIO, Vanderbilt University Medical Center
How and when are we going to get the tools and processes that we need to capture and effectively use the patient physiological data that we have now, and all the new info that is coming?
At the recent CHIME Fall Forum, Dr. Eric Topol described the new and innovative ways in which we will be getting a large and increasingly complex amount of physiological data. I was excited by his vision and how this new data can improve care, but I was also troubled by where we are now. Currently we do not have the tools and processes to manage what we have; getting more data is going to make it worse.
We already have a large amount of physiological data that comes from patient monitoring. Many of these systems run on proprietary networks and hardware and only on certain wireless spectrums. Unfortunately, this captured data is often transcribed into other documentation systems or EMRs. The present designs are complex and fragmented, making systems hard to manage, operationally and clinically inefficient, and maybe even dangerous. We will not be able to hold back, nor do we want to slow efforts to capture new data. Because of factors such as the increase of consumer-oriented tools, the pressure from Meaningful Use, and the uncomfortable state in which we see ourselves, it will require significant effort to create a better environment.
So what are our choices? One is to move these technologies to more common IT transactional systems and technologies.
There are a number of existing efforts geared toward this approach. Standardizing on as few tools as possible is a good strategy, and we have been very successful with IT tools in generalized environments. However, I’d suggest that we have crossed a line for the reliability, availability, and serviceability of general IT tools into areas that are unique to healthcare.
Let’s take wireless as an example. There are a number of wireless spectrums used for physiological monitoring. Adding coverage to these technologies gets incredibly expensive and hard to manage given their lack of compatibility with general IT solutions. This incompatibility often requires duplicate wiring, switches, and separate management tools. If we want to use standard IT designs like 802.11, we are likely to find that it misses some unique healthcare requirements. This problem of poor technology convergence is not unique to healthcare wireless networks; these gaps exist in system monitoring, disaster recovery, and many other areas.
We need to rethink the way in which we adapt these generalized tools to meet healthcare requirements. It may be they can’t be adapted and we need new and different tools. Maybe a series of specialized healthcare specific spectrums are needed. We have not yet created effective data designs or tools for clinical decision support that makes data useful and actionable. There are a number of efforts around these areas and we need to redouble our efforts to catch up and make use of these existing and new data sources.
Today, there are many separate processes for those that are accountable for physiological systems (clinical engineering) and those that manage other application systems (IT). This is evident in the current reporting relationships between IT and clinical engineering (CE), where recent surveys indicate that about half of CE departments report to the CIO. Whereas many IT organizations are ITIL experts, these concepts are often part of the CE culture and process; however, ITIL is not the driving force in process change in CE that it is in IT organizations.
In spite of these separate technologies and process orientations, provider organizations have learned to blend the skills of CE and IT staff to meet needs regardless of their reporting relationships. Vanderbilt Clinical Engineering and IT have a shared technology RFI and require any capital requests to be screened using a common tool for possible impacts to either of these teams.
Accelerating this convergence of new technologies — whether IT oriented, a CE/IT blend, or new designs — is the only way in which we are going to achieve the promise of capturing, managing, and effectively using the data we already have, as well as the data deluge that is already upon us. Developing these new approaches can also help organizations that are working towards a blended CE and IT role where teams can share more than a commitment to their shared customers.
kickabout says
Interesting post. I work in science, and these days one of the big topics, regardless of specific field, is Big Data. With the OSTP white paper focusing on the topic and the sheer volume of data being generated from digital monitoring systems, genomic sequencing, and other new areas, the challenges of everything from storage to analysis to interpretation are being tackled across the board in basic research. It’s interesting to hear how it’s also playing out from an applied perspective.
As a parent with a baby in NICU though, I have an entirely practical concern: EMR compatibility. Our baby is currently in one hospital where she was born prematurely. That hospital uses Epic. The hospital nearest our home though, where she will go in case of emergency after she is discharged, uses Cerner. We’re trying to figure out what information we need to have in paper form to carry with us in case she has a crisis. I gather the two systems aren’t terribly compatible, so the benefits of EMR are largely lost for us, regardless of the data they contain. Please do keep pushing for standardization of data collection, storage, and all the other critical elements to allow for cross-platform compatibility. The thought of time lost tracking down and transferring critical information from one hospital to another during an emergency because of proprietary controls worries me greatly. It’s one thing for my iTunes downloads not to be compatible with my SanDisk mp3 player; it’s entirely another for our hospital not to have previous respiratory treatment records available if our child has difficulty breathing.
George McCulloch says
Thanks for your comment. We’re doing in some cases pretty well in finding ways to store data for later use. Deciding what we want to keep, why we want it and acting on it are different issues.
As to your EMR incompatibility issue, there are formats for EMR interchange (CCA, CCD, etc.) that have been developed and interchange capability has been a requirement for Meaningful Use payments. That doesn’t mean that its easy, useful or pervasive.