Tele-ICU Increases Interhospital Transfers: Does Big Brother Know Better?*

    loading  Checking for direct PDF access through Ovid


The practice of telemedicine in the ICU, either continuously or solely at night, is becoming an increasingly common and important part of modern critical care medicine (1). In fact, the current practice of telemedicine in the ICU (tele-ICU) now covers approximately 11% of nonfederal hospital ICU beds in the United States (2). A single-center study of tele-ICU in an academic center suggests that it may be associated with decreased mortality, increased adherence to best practices, and decreased ICU length of stay (LOS) (3). More recently, however, a large national effectiveness survey of tele-ICU involving over 500 hospitals suggests that there is significant heterogeneity in the benefits gained after implementing a tele-ICU program (4), a finding supported by highly variable results in other studies (5, 6). However, given the national shortage of intensivist physicians, the tele-ICU model is likely to persist in order to provide costeffective, quality ICU services, especially to patients in hospitals without dedicated intensivists.
Another potentially important role for tele-ICU is in the interhospital transfer (IHT) process, both in identifying potential transfers as well as providing supplemental care that would allow for patients who previously would have been transferred to remain in their community facilities. In this issue of Critical Care Medicine, Pannu et al (7) present a retrospective study demonstrating the effects of tele-ICU implementation on IHT and other important ICU outcomes within the Mayo Clinic Health System (MCHS). As background, in August of 2013, the MCHS implemented its tele-ICU system in which intensivists at the quaternary center in Rochester, MN, provide tele-ICU services to the six community-affiliated hospitals that comprise the system. Notably, only transfers within the MCHS were studied. The study defined three 1-year time periods: the preimplementation period, a transitional period (during which tele-ICU was being implemented), and a postimplementation period. Additionally, two tiers of community-affiliated hospitals were identified: those that had access to procedural and subspecialty expertise, including emergency surgical services, cardiac catheterization, an on-site daytime intensivist, and on-site medical subspecialties (“type 1” hospitals), and those that did not (“type 2” hospitals). The quality and quantity of the critical care nursing coverage provided at the two referring types and the receiving hospital was not included as a factor. For the analysis, the transitional period was excluded, and outcomes during the pre- and postimplementation period were compared.
The investigators found that the overall number of transfers significantly increased, driven primarily (but not solely) by transfers from type 2 hospitals; additionally, hospital LOS at the quaternary center increased. At least for the MCHS tele-ICU case, the additional expertise provided by the tele-ICU did not result in fewer transfers from either type of facility. Otherwise, comparing the two periods, there was no significant difference in patient demographics, community hospital or ICU LOS, severity of illness scores at admission to the quaternary center, breakdown of patients according to code status (full, do not resuscitate/intubate, or comfort measures only), or proportion of patients transitioned to comfort measures only. But one crucial piece of information is missing: survival data pretele- and posttele-ICU implementation at the sending hospitals which would give some indication of the outcome impact that the tele-ICU was achieving.
IHT is a complex, difficult-to-study phenomenon. IHT is typically arranged when there is a perceived mismatch between the patient’s care needs and the resources available at the referring hospital. When clinicians choose to pursue IHT, there is an assumption that the risks and costs of transfer are outweighed by the benefits (8, 9).
    loading  Loading Related Articles