How much is it worth to reduce deaths and complications in the notoriously error-prone intensive-care unit of any hospital?
For David Bernd, president and chief executive officer of six-hospital Sentara Healthcare, it was at the very best a break-even proposition.
If the Norfolk, Va.-based healthcare system could cover its costs on any investment aimed at reducing mortality in the ICU-plus increase doctor satisfaction and improve efforts to recruit and retain ICU nurses-then "it would be a home run," he says he told his medical team.
To that end, Sentara began wiring its ICU beds to a remote office in an industrial park eight miles away from 478-bed Sentara Norfolk General Hospital. The idea was to add an extra layer of around-the-clock ICU vigilance through a combination of software, hardware and a critical-care specialist working the remote controls (Sept. 4, 2000, p. 62).
The results of that effort, which began 18 months ago, are just now coming in. An analysis by Cap Gemini Ernst & Young found a 25% reduction in the hospital mortality rate for Sentara Norfolk's ICU population. That translates to approximately 60 lives saved per year, meaning at least one person a week who might otherwise have died at Norfolk General returns home safely.
And that ultimate achievement in quality of care not only broke even but also paid dividends: an annualized net financial benefit of $3 million after subtracting all program costs.
The report was commissioned by VISICU, the Baltimore-based company that developed the so-called eVantage system installed at Sentara. The analysis covered more than 600 patients discharged from 16 ICU beds during the first half of 2001. Those results were annualized and compared with data on patients who were discharged during the 12 months before the program's implementation.
The eye-popping decline in mortality rates sent everyone back to check their math.
"These were real saved lives, so it was pretty powerful," says Rod Hochman, M.D., Sentara's senior vice president and chief medical officer. "For each week, someone was walking out of the Sentara system who wouldn't have without the (remotely run) system."
In addition, the average length of stay for critically ill patients, both in the ICU and in subsequent recovery on a nursing floor, was reduced by 17%. As a result bed turnover increased by 20% in the overtaxed ICU.
The study covered 10 beds in the hospital's general ICU and another six beds in a vascular ICU. In all, Sentara has 50 beds wired into the remote monitoring site-called an e-ICU-at four ICUs in three of its six hospitals.
The average stay in Norfolk General's ICU declined to 4.36 days from 5.19 days. In the vascular ICU, stays declined to 2.43 from 2.92 days.
The improved efficiency in the ICU had a domino effect on the general hospital floor as well. Once patients were moved out of the general ICU, they stayed in the hospital an average 9.1 days, down from the previous 10 days. For vascular patients, the stays were trimmed a full two days on the floor, to 6.5 days from 8.5 days. There was no change in the average severity of illness in the unit.
Quantifiable benefits of efficiency
Although some experts for years have insisted that improved quality would reduce costs, did anyone truly believe that saving lives with cutting-edge technology could really save money?
Long before the Institute of Medicine's 1999 damning report on hospital errors and even longer before the powerful Leapfrog Group set three gold standards for patient safety, Bernd says he proselytized that "the best potential for prosperity and improvement in healthcare was through clinical intervention."
But even Bernd was somewhat disbelieving at the start of the e-ICU project, promising Hochman that he would put him on a pedestal if it didn't cost anything, Hochman says.
The e-ICU's operation met that mark and more, delivering a 150% return on the system's annual investment of $2 million, Hochman says.
The $3 million net savings is the result of a 26% reduction in hospital costs for ICU patients. A shorter length of stay; a lower use of supplies, laboratory tests, therapies and medications; and a 4% decrease in nursing hours worked per patient day helped to reduce costs, according to the Cap Gemini Ernst & Young analysis. The savings worked out to $2,150 for each patient.
The additional ICU cases that came in as a result of reducing the length of stay brought in an extra $460,000 monthly in gross revenue, and a $274,000 bonus contribution to the bottom line after subtracting costs.
The $3 million in savings accounts for about 7% of Sentara's $40 million profit margin on $1.3 billion in revenue, Bernd notes.
Sentara is practically giddy over the results.
"For our organization, this is our biggest home run-almost ever," Hochman says.
"Everybody assumes you pay for quality when it comes to healthcare, but it turns out to be less expensive," says Brian Rosenfeld, M.D., VISICU's chief medical officer, executive vice president and co-founder. "I don't think people in Washington appreciate that. That to me is the 12-second sound bite for this whole analysis."
As the alpha site testing the VISICU system, Sentara is nearly two years into a five-year contract. The hospital system won't disclose details of the agreement, but it's more of a partnership than a relationship between vendor and customer, Hochman says.
Rosenfeld says depending on volume discounts, the system can cost anywhere from $30,000 to $50,000 per bed. The company is aiming to go public in 2003, he says.
Spreading intensivists around
Bernd says it was the problem posed by cash-burning ICU operations in general, not any particular problem at Sentara, that brought Sentara and VISICU together. He says he was acutely aware that ICU days represent a small amount of a hospital's admissions but an alarming majority of a hospital's operating costs.
"I intuitively figured there was great potential to improve clinical outcomes through improving ICU care," Bernd says.
So Bernd was game when a retired colleague called to tell him about VISICU's fledgling inroads into managing ICU care. This was some time before the Leapfrog Group identified the ICU as one of three hospital venues for patient-safety standards that could offer the basis for provider performance comparisons.
According to the daunting standard, based on research showing a direct correlation between the level of training of ICU personnel and the quality of patient care, board-certified critical-care physicians should work exclusively in a hospital ICU a minimum of eight hours per day.
At other times, a critical-care specialist, known also as an intensivist, should be available to return more than 95% of ICU pages within five minutes of being called.
As additional backup, intensivists also should be able to rely on a hospital-based doctor who can reach 95% of the ICU cases within five minutes, according to the standard.
Critics charge the Leapfrog criteria would require more intensivists nationwide than there are to go around. Gene Burke, M.D., Sentara's e-ICU medical director, says a nationwide supply of 30,000 intensivists would be needed-five times the 6,000 certified intensivists.
Despite the shortage, VISICU's aim never has been to reduce ICU staffing but to supply an extra level of care, Rosenfeld says. It is simply the difference between reactive and proactive care.
"This is supplemental care, but right now most hospitals don't have intensivists on site 24 hours a day, seven days a week. Even those that do are covering multiple ICUs, so they are running from one to the next putting out fires or trying to sleep, whereas the e-ICU is totally designed like air traffic control," Rosenfeld says. "With one click (an intensivist) can go from hospital to hospital." That leverages the scarce physician commodity throughout multiple ICUs, he says.
Sentara's ICU operating costs in 2001 totaled $28.5 million. Its six hospitals operate 13 ICUs with about 150 beds. The exact staffing needs and accompanying costs are difficult to pin down because the ICUs are open to almost any community physician with hospital privileges who would like to treat patients there, Burke says.
Staffing and coordination issues
The open access and latitude granted to attending physicians in the ICU raised some sensitive political issues at first, as does any new technology when a diverse group of private-practice physicians is involved.
Some doctors were comfortable with allowing e-ICU physicians to make decisions and take actions on their behalf, but others preferred to continue making some or most patient-care decisions themselves.
To ease community physicians into the concept of the e-ICU, Burke says they could opt for four levels of e-ICU care depending on the extent to which they agreed to put the care of their patients in the hands of other hospital-based doctors.
The community doctors included the usual distribution of early adopters and laggards associated with all new technologies, he notes. Initially the bulk of the community physicians chose the lower two levels of care, but now most are going with the higher levels, he says.
"Doctors always go to bed at night thinking they provided great care, and we were going to show them a way to give even better care," Burke says.
Before the e-ICU was wired to four ICUs at three Sentara hospitals, Burke counted six intensivists in his group. To staff the e-ICU, the system needed to hire four more intensivists, and working remotely was a condition of employment. But to date only two new intensivists have been recruited, a fact Burke attributes to the workforce shortage and Sentara's exacting standards.
The group of eight split the e-ICU shifts with a dozen or so community intensivists who were interested in the program and, of course, are paid by Sentara to cover the shifts.
The number of additional nurses needed was nominal, Rosenfeld says. A staff of 12 nurses has been trained to work in the e-ICU, with four of them working exclusively at the remote location and eight dividing their time between the hospital and the remote location, he says. As part of its agreement with Sentara, VISICU has borne the additional nursing costs, but the system will assume those expenses in July, he says.
The e-ICU, whose neighbors include a software company, an insurance office and a state agency, is staffed with three people: an intensivist, a nurse and a clerical support person. The team covers two shifts over 19-hour days, shutting down the remote operation from 7 a.m. to noon when the doctors are routinely in the ICU making rounds.
Each of the e-ICU staff divide time between the hospital and the remote location. For example, Burke could work four 12-hour overnight shifts in one week. In another week, he might work three seven-hour day shifts in the e-ICU and two days in his office seeing patients. In the meantime, his partners will be in the hospital doing bedside consultations, he says.
Backing up the front lines
The e-ICU workstation offers real-time video and audio of each of the 50 ICU beds covered. The e-ICU doctors and nurses are tuned in to bedside monitors, which supply every detail of patient information-including oxygenation, pulse, blood pressure, heart rhythms-that a doctor has at bedside, Burke says. "Smart alarms" incorporated into the software also alert physicians and nurses to troubling or out-of-the-ordinary disturbances.
"It's the virtual equivalent of walking in to a patient's bedside with my hands in my pocket," Burke says. "I simply can't put my hand on the patient."
But Burke can watch as he asks a bedside nurse to, for example, touch a patient's abdomen or shine a light in the eyes. He also works from an electronic medical chart-the only area systemwide where the medical charts are electronic. Cameras and microphones at the nurse's station also offer an opportunity for family conferences.
Perhaps the biggest difference between then and now is that the e-ICU physicians have the time and ability to look for trouble before it happens.
"The whole point of this thing is, we don't come to the bedside and change the diagnosis and plan; we simply work it in frequent, little intervals," Burke says. "We don't let little things become big things. That's where we save lives, complications and money."
As for how the intensivists view the e-ICU, Burke says if you ask the eight members of his group how they feel about working in it, you will get eight different answers.
Burke says he really likes it: It has all the mental challenges of critical-care medicine without the physical challenges.
"I get to address questions sitting there and (information on) 50 patients is electronically funneled to me, and I don't have to run around a nine-story building with an ICU on four different floors," Burke says. "I just sit there and everything comes to me."
Burke says he's been told that the average burnout age for an intensivist is 47 years old. "I'm 52, and I've been doing this for 22 years, and in my group there are five over the age of 40. I see this as an important tool that is going to keep us from burning out," Burke says.
Rosenfeld says the same principle holds true for ICU nurses, whose job demands are perhaps even more physically challenging.
Impact of closer vigilance
The medical team has no lack of anecdotal stories about lives saved and costs reduced by the e-ICU.
Burke relates the case of a woman in her late 30s who came to the ICU from the emergency room with a diagnosis of toxic shock syndrome. She immediately was started on treatment in the ICU, but Burke, monitoring her from the e-ICU, noticed she was not responding as she should have.
He suspected she was suffering from an adrenaline deficiency and immediately started tests and treatment for that. She responded and returned home a week later.
Had he not been in the e-ICU, the doctor making morning rounds in the ICU would have found her the way she was when he left, Burke says, likely with significant lung, kidney, liver or brain damage.
"The main thing was I was awake; I had the data in front of me. The (ICU) nurse was just following instructions," Burke says.
Hochman, Sentara's medical chief, notes that like every other health system, there's typically a backup of ICU beds. The overnight intensivist can keep things moving along to help save costs and reduce complications.
The e-ICU physician can, if appropriate, begin weaning a patient off a ventilator at 2 a.m., Hochman says; reducing the length of time patients are on ventilators has been proven to improve the patient's survivability and reduce the length of stay, he says. Before the e-ICU, even if an intensivist were in the hospital, weaning a patient off the ventilator would have required a nurse finding the doctor. Usually the task was postponed until morning.
"The e-ICU doesn't wait for someone to say there is a problem. That's the difference," Burke says.
Acceptance of the e-ICU has boosted revenue for Burke's group of intensivists, but considering the new physicians he's had to hire, it's been a wash, he says. The real financial impact has been in savings.
"The e-ICU stops the cash hemorrhage," Burke says.
Besides reducing length of stay and increasing availability of beds for new patients, the e-ICU presents other unanticipated opportunities, Hochman says.
For example, Sentara believes it will go a long way toward keeping ICU nurses satisfied and will help in recruitment efforts.
The e-ICU has burnished the reputation of all the hospitals in the system as well. In the past doctors were not so confident about sending patients to the ICU at 100-bed Sentara Bayside Hospital, Virginia Beach, Va., because they did not feel it was "comprehensive enough," Hochman says. But Bayside, which went live with the e-ICU enhancement about six months ago, just had its best year ever: Admissions rose 2.8% to 5,227 and surgeries, including outpatient procedures, increased 8.3% to 6,049.
"It's no coincidence to us," Hochman says.
Hochman also gives the e-ICU credit for helping to standardize practices systemwide. Physicians working in all the ICUs are comparing notes now. It has "put a light on critical-care medicine for us," Hochman says.
Sentara is next considering opportunities in outlying hospitals that are not part of the system but refer seriously ill patients. "This could be a win-win for rural Virginia hospitals that have three or four beds hooked up so we can co-manage patients to prevent having to ship them to Sentara," Hochman says.
As to the $3 million question: Is the e-ICU Leapfrog-compatible?
The standard was expanded last fall "to allow for the presence (of a critical-care physician) to be achieved through an e-ICU," says Suzanne Delbanco, executive director of the coalition led by a group of Fortune 500 companies.
Delbanco would not comment on how Sentara's e-ICU complies with Leapfrog's standards. However, the 10 key features of Leapfrog's ICU-tele-monitoring guidelines were based partly on studies conducted by researchers led by VISICU co-founder Rosenfeld when he was an intensivist with Johns Hopkins Hospital in Baltimore (See chart, this page).
Bernd says he believes the technology has great potential in other areas, such as helping with overflow in the emergency department and even in primary-care offices. The focus right now, though, is to just "make sure it works."
When all is said and done, the real return on the investment is coming from patient satisfaction, Bernd says. "The key to success is continuous quality improvement of clinical processes and driving out variations of outcomes in clinical care," he says.
"This will do more for quality patient care and financial viability than anything we can do on our business side. I think the e-ICU is one of the breakthrough technological advances that can put us where we need to be in our industry."