American hospitals with a shortage of "intensivists" to treat patients in their intensive care units could benefit from having such experts monitor their patients off site via computer, Johns Hopkins researchers have found.
In a pilot telemedicine study, four of these critical care specialists monitored the 10-bed surgical ICU of an academic-affiliated community hospital without a full-time specialist in house. Death rates dropped up to 68 percent, complications by up to 50 percent, length of stay by up to 34 percent and costs by up to 36 percent. Results were published in a recent issue of the journal Critical Care Medicine.
In the United States, only 50 percent of hospitals have intensivists available, and most are on a limited consulting basis. To staff every ICU sufficiently would require 35,000 to 40,000 of these specialists, says Todd Dorman, an author of the paper and an associate professor of anesthesiology and critical care medicine at Hopkins. Currently, there are less than 10,000.
"This isn't designed to be a replacement of full-time staff but an extension," Dorman says. "It's a way to better leverage critical care expertise among smaller institutions that don't have the resources to hire people.
"Of 5 million adults admitted to ICUs each year, 70 to 90 percent are not being managed by a critical care team," Dorman says. "The problem is twofold: First, people are living longer and therefore have concomitant illnesses. Second, we have less reserve to handle the overflow in the ICUs. Our study shows that remote care programs may provide a means of improving quality of care and reducing costs when on-site intensivist coverage is not available."
Many hospitals can't afford critical care coverage 24 hours a day, seven days a week, Dorman says, and for those with only a handful of ICU beds, it might not be cost-effective. A telemedicine system such as the one described in the study could enable a physician to monitor up to 40 or 50 patients at a time in multiple hospitals, compared to the 10 or 20 patients able to be monitored by a physician working full time in a critical care unit.
For the study, researchers first collected from an urban hospital's ICU outcomes data during two 16-week periods, one in the fall to adjust for seasonal illnesses and one in the spring.
Next, cameras and data transmission equipment used for remote care were installed in the ICU and in the homes of four Hopkins intensivists. During a 16-week period, the intensivists took turns providing from home round-the-clock monitoring of all ICU patients. They interacted with patients and hospital personnel and accessed clinical data via dedicated, computer-based video conferencing and data transmission equipment. Bedside monitoring data were transmitted in nearly real time via a telephone access system. Electrocardiograms, radiographs, consultant notes and bedside data flowsheets were scanned and transmitted digitally; laboratory data were accessed through a telephone terminal-emulation system.
The off-site clinicians participated in rounds via video conference, discussed individual cases with a senior member of the attending staff or bedside nurse at least twice a day and were available by pager to discuss emergencies or patients newly admitted to the unit. Outcomes data during this period were collected.
There were 225 patients in baseline period one, 202 in baseline period two and 201 during the intervention. Death rates decreased during the intervention by 68 percent and 46 percent, compared with the first and second baseline period, respectively. Hospital mortality decreased by 33 percent and 30 percent; incidence of ICU complications decreased by 44 percent and 50 percent; ICU length of stay decreased by 34 percent and 30 percent; and ICU costs decreased by 33 percent and 36 percent.
Hospitals can't afford to be without full-time critical care, Dorman says, as complications can increase a patient's hospital bill by an average of $14,000, and an adverse drug event can add as much as $27,000. Intervening in any way to prevent complications can lower a patient's risk of an adverse event by 2 to 7 percent, he says. In this study, remote monitoring lowered the risk by 10 percent.
"There's almost nothing that helps that much," Dorman says. "It's a real cost savings in the decreased length of stay and mortality, lower hospital and professional fees and controlled complications."
The study's other authors were Peter Pronovost, Mollie Jenckes, Nancy Zhang, Gerard Anderson and Haya Rubin, of Hopkins, and Brian A. Rosenfeld and Michael J. Breslow, of Visicu Inc., who were at Hopkins at the time of the study.