On December 16, 2015, the Association for Pathology Informatics and Sunquest Information Systems presented a webinar entitled Informatics & Connectivity to Enhance Point of Care Testing as part of its 2015-2016 series which is devoted to the field of pathology informatics. The lecture was delivered by Dr. David McClintock who is a member of the pathology faculty at the University of Chicago Medicine. The video of the presentation is available for viewing. In this note, I would like to go over some of Dr. McClintock's key points and provide my personal analysis and interpretation of them.
- Rapid growth of POC testing. Point-of-care testing is growing rapidly. This is both an advantage and a problem. Many of the POC instruments in the market are not designed to integrate easily with hospital information systems. One of Dr. McClintock's main responsibilities is to ensure that the quality of point-of-care test results are equivalent to those generated in the central laboratories. This is a challenge for a couple of reasons. First, the classic LIS vendors pay more attention to data in the central labs for historical reasons. They may not provide the means to seamlessly integrate POC test results into the LIS database. Secondly, POC testing may be performed by various hospital personnel who may not always be optimally trained for the testing. Even if it is stipulated that, say, a blood glucose should be performed by nurses, the task may be delegated to other personnel because of the crush of work.
- Integration of POCT results into the EHR database. In some cases, it may be technically easier for pathology IT personnel to integrate POC results into the EHR database than into the LIS database. This may be due, as noted above, to the fact that managing POC results may not always he a high priority for LIS vendors. The LIS database has always been defined as the source of truth regarding lab and pathology test results. This means that the LIS database is the reference database in terms of the completeness and quality of lab data. This perspective will not be be true if the EHR database contains information not be present in the LIS database. This will become even more challenging as the volume of POC test results increases. I think we may be witnessing a widening division between the management of hospital/central lab results and POC results. The latter seems to be more closely aligned with consumer results created and stored in wearable health devices.
- Desirability of wireless POC devices. Dr. McClintock emphasized that the selection of wireless handheld POC devices such as the i-STAT system in hospitals is highly desirable and should now be the first choice. His rationale for this was convincing. Demand for more POC testing is generated mainly by hospital staff and not lab professionals. Pathologists and lab scientists tend to favor central lab testing because it's cheaper, automated, and tests are performed in a controlled environment. Hospital physicians and staff favor POC testing because it provides more flexibility for them and the results are obtained in minutes. The deployment of wireless devices provides even more flexibility such that the tests can be performed truly at the point of patient care (i.e., bedside or patient-side).
- Ability of POC instruments to support entry of "other test results." Not all POC devices enable the input of "other" test results, which is say test results generated by other brands of POC devices. By way of contrast, the Roche Inform II is a hand-held glucometer that supports other test entry (OTE). The deployment of such instruments provides great flexibility to perform POC tests that do not offer wireless data entry.
- Cost of testing. POC testing has historically been much more expensive than central lab testing because there are economies of scale with the latter whereas the former are customized for a single patient. When asked if the cost of POC testing would decrease, Dr. McClintock said that the cost per individual tests would decrease as technology allows a greater number of tests performed per cartridge. However, he also replied that the total cost and savings for the hospital of POC testing needs be taken into consideration. He cited as one example the need for a creatinine level prior to the administration of contrast for radiology imaging studies (see: Creatinine Testing Prior to Contrast Administration). By performing POC creatinine studies in radiology at his hospital, the number of cancelled imaging studies has been greatly reduced and the extra cost of the POC creatinine tests is far outweighed by the reduced number of cancelled studies.