How did the use of middleware in the lab first arise?
A: Fifty years ago or more, automated instruments for the clinical laboratory had primitive software capabilities. These instruments were directly cabled to a laboratory information system or laboratory information management system (LIS/LIMS) port and used vendors’ custom-designed software on the instrument and LIS/LIMS (also known as a vendor-specific interface) to enable connectivity. Usually, this software transmitted test orders from the LIS/LIMS to the automated instrument while also transmitting interfaced laboratory results from the instrument back to the LIS/LIMS, all with no middleware software involved whatsoever.
In those days, it was very expensive and time-consuming to get a working and reliable interface, mostly because of the variety of different instrument and LIS/LIMS vendors involved. Also, there were essentially no uniform standards for communication between various instruments and LIS/LIMS computers for data transmission protocols, data record formats, and the like.
AACC’s Informatics Division was formed in part to address this connectivity challenge. As chairman and founder of the division, I was involved early on in writing the first prototype specification for the now universally accepted ASTM interface standard for medical laboratory instrument interfaces. Shortly after we developed this standard, several vendors began to provide flexible middleware solutions to make the ASTM instrument-LIS/LIMS connectivity less vendor-specific and more flexible. This was essential because many LIS/LIMS vendors were quick to embrace ASTM connectivity standards while instrument vendors lagged far behind, with many older, non-ASTM compliant instruments still deployed in clinical laboratories.
However, these early middleware technologies were not designed to handle the highly diverse and emerging connectivity challenges of point-of-care testing (POCT).
Why do we need middleware designed specifically for POCT interfaces?
When the use of bedside POCT first took off, interface challenges for this service rapidly became a major problem. This was due to several factors, including: 1) the large number of hand-held devices being deployed, 2) compliance issues with training and quality control for the large number of nonlaboratory personnel performing POCT, 3) support issues involved in remotely monitoring devices, testing personnel, quality control, and patient results in real-time, 5) specific and unique POC CLIA compliance issues and monitoring, and 6) billing and other resource management issues.
Additionally, the need for interfaced POC patient demographic information prior to testing made the ASTM interface standard a poor option for POCT. Fortunately, this particular issue at least was solved early on by conforming POC interface protocols to the HL7 interface standards instead.
As for all of the other connectivity challenges associated with POCT, vendors of POC devices initially addressed these issues with vendor-specific interface solutions. However, these soon became a significant problem in their own right as laboratories deployed different POC vendor solutions throughout the hospital.
It was middleware vendors who came to the rescue by developing hospital-wide interface solutions with an open system approach. This enabled: 1) multiple vendor POC solutions to be interfaced and managed with a standard middleware turnkey solution, 2) speedier deployment of best-of-breed POC devices, since laboratories were no longer locked into particular vendor solutions, and 3) better overall control and compliance of the entire POCT program throughout the hospital using a lean- management approach.
These days, middleware dominates as the preferred connectivity solution for the clinical laboratory. Even though connectivity standards exist, subtle connectivity issues remain between different instruments and LIS/LIMS vendors. Middleware’s flexibility enables it to handle these issues effectively and reliably, thus providing a stable interface connectivity solution.
Kenneth E. Blick, PhD, ABCC, ACS, is an emeritus professor of pathology at the University of Oklahoma Health Sciences Center.+Email: [email protected]