ICU telemedicine shows improvements in survival, other quality outcomes (finally)
Neither a large 2009 multicenter study nor a 2011 meta-analysis showed any clinical benefit from the use of ICU telemedicine. Lilly et al report the results of a large single-center study in which they progressively implemented ICU telemedicine among 6,290 patients in 7 ICUs (a stepped-wedge design), with non-telemedicine groups acting as controls at each step. Patients in ICUs with active telemedicine had lower mortality (a 1.8% absolute reduction, an odds ratio of 0.40 for death) and shorter durations of ICU stay and mechanical ventilation. They received more evidence-based preventive care and had lower rates of ICU complications. Benefits were seen preferentially in telemedicine ICUs at each time point, weakening criticism that improvements were due to secular trends. The telemedicine MDs were more muscular in this than in previous studies -- actively promoting adherence to guidelines, daily goal sheets, etc., rather than simply responding to alarms. JAMA 2011;305:2175-2183. FREE FULL TEXT
Jeremy Kahn, a respected voice in the field, gives a cautious thumbs-up and his acceptance of these findings in his editorial, although he calls the magnitude of the reported mortality benefit "implausibly large." He urges caution and further study of ICU telemedicine, rather than accepting these findings as proof that the technology has now arrived and will be universally beneficial. Kahn points out that telemedicine is effectively a bundled intervention, and that as in other quality improvement bundles, it's unclear which components are responsible for measured quality improvements; that local cultural & system factors (e.g., all the ICU tele-doctors also rotating as attendings in the ICUs as in this study) may make a large impact on effectiveness and outcomes; and that implementing quality improvement programs that increase intensity of efforts at the bedside, rather than launching an ICU telemedicine system, might produce the same results. JAMA 2011;305:2227-2228. FREE FULL TEXT