|Title||These Are the Technologies That Try Men's Souls: Common-Sense Health Information Technology [Commentary]|
|Author(s)||Per H. Gesteland, Jonathan R. Nebeker, Reed M. Gardner|
|Source||Pediatrics, Vol. 117, No. 1, Pages 216-217|
|Publication Date||Jan. 2006|
|Abstract||Health information technology (HIT) promises to facilitate improvements in medical care. However, implementation methods are equally important, because both technology and its implementation change the culture and workflow in health care organizations. The report by Han et al provides a powerful example of how HIT implementation may produce tragic results if adequate consideration of its consequences on workflow are not addressed. |
Han et al conducted a retrospective, pre-post study of the connection between computerized physician order entry (CPOE) and mortality among children who transferred to a tertiary care children's hospital for critical care. They reported that CPOE was associated with over a threefold increase in the odds of dying. However, perhaps in contrast to drug trials, the more interesting question for this technologic evaluation was not whether the technology was associated with a mortality difference but why.
Han et al provided abundant anecdotal evidence of potential dangers of CPOE implementation. They described a 6-day, out-of-the-box implementation plan for off-the-shelf software without customization (eg, order sets, or a module specifically developed for a pediatric ICU). They described dramatic workflow changes that were necessary to use CPOE functions for ordering, verification, and dispensing medication. They also reported altered interactions within the care team that reduced beneficial, synchronous communication. These changes allegedly introduced delays into the medication-administration process. Han et al also described a need for increased clinician time at the computer and, consequently, reduced time at the bedside. However, the study did not include measurement or analysis of the described delays, altered work processes, or other potential confounders of a mortality association. Nonetheless, despite the huge potential for confounding in this study and the lack of substantiated causes, the crude mortality risk difference of 3.77% between study periods was real and unfortunate.
Limitations aside, the report is an important contribution to the literature, because it provides a stunning example of how pressure to deploy HIT stifled common sense and had deadly results. Deploying a sophisticated clinical-applications platform including CPOE in 6 days is an audacious task and leaves little margin for error in adapting highly evolved work processes to the new environment. There is a critical need to understand existing culture and processes so that organizational efficiencies may be preserved or effectively transformed.
Although computer-assisted interventions have helped process outcomes such as medication errors, we are a long way from assuming that computerized interventions will improve patient outcomes. A recent review of computerized decision studies that used randomized methods found that only 52 of 100 studies assessed patient outcomes, and only 7 demonstrated a patient benefit. Only 5 of the 100 studies were from a pediatric setting. Other reports have emphasized that merely computerizing order entry may not address the most common and dangerous types of drug harm; accompanying decision support such as order sets and dosing guides are important also.
Because every hospital has a different set of highly refined work practices and because the risks of computerized care are often unanticipated, computer-based interventions require careful planning, pilot-testing, and evaluation. Extra caution and assessments are especially important for computerized interventions in the relatively poorly studied pediatric setting. Although the increased mortality observed in the Han et al implementation may not have been anticipated, the problems with delayed medication delivery should have been. The report by Han et al highlights the pressing need for future investigations to focus on optimal implementation strategies for HIT systems.