Nursing groups respond to demands for quality accountability in critical care

Measure success with focus on process improvements, patient outcomes

Can success be measured? Nursing organizations are trying to determine a meaningful definition of success for the intensive care unit (ICU). So are other groups. The influential Joint Commission on Accreditation of Healthcare Organizations in Oakbrook Terrace, IL, for example, has issued a series of statements in the past year urging hospitals, including critical care units (CCUs), to measure up to quality standards.

Meanwhile, Medicare, Medicaid, and large private employers are contracting heavily with managed care organizations, which are also asking providers for more accountability on quality.

In an unprecedented way, a growing number of outside groups are urging hospitals to determine how they fare on quality standards, says Joanne R. Duffy, DNSc, CCRN, a nursing consultant with AdviCare in Burke, VA. The emphasis on quality isn’t new, Duffy says, but the message is getting louder.

Can success be measured in critical care? Nurse managers and physicians lean in the direction of saying yes — they can measure success in the CCU. Patient acuity outstrips that of almost all other acute care departments except perhaps emergency medicine. And patient mortality rates run higher than elsewhere in the hospital. Nevertheless, nursing administrators say they can measure success in critical care by focusing on patient outcomes, but even more on the effectiveness of their clinical processes.

"Outcomes alone tell only part of the story," says Judith Fisher, RN, MSN, a clinical practice specialist in the coronary care unit of 650-bed Inova Fairfax Hospital in Fairfax, VA. "Your outcomes are really a reflection of how effective you are in implementing high-quality clinical processes," she says.

Her statement takes into account the typical ICU’s record on high patient mortality. "If your nurses are doing everything they’re supposed to do, regardless of whether the patient makes it, your efforts count as a measure of success," Fisher says.

Process improvements in critical care are receiving a lot of attention these days. They were the emphasis of a groundbreaking project a few years ago initiated by the Institute for Healthcare Improvement (IHI), a Boston non-profit group interested in speeding up improvements in health care through integrated strategies.

In 1998, IHI published findings of an exhaustive study of ICU process improvements in a book, Reducing Costs and Improving Outcomes in Adult Intensive Care. This 193-page book is a compilation of internal improvements that focused on positive changes within ICUs.

More than a dozen hospitals participated in testing several "change concepts." One of the concepts included setting up standardized processes that would help to achieve a variety of goals ranging from reducing inappropriate days in the ICU to identifying and reducing bottlenecks in the patient management system.

One such effort involved a method to manage patient ventilator support to reduce the time they spent on mechanical vents and avoid complications. By standardizing various activities, the hospitals attempted to achieve four goals over a 12-month period:

    1. reduce time on mechanical ventilation by 30%;
    2. reduce costs for lab tests and X-rays by 30%;
    3. reduce length of stay in the ICU by 25% while reducing morbidity, mortality, and complications;
    4. reduce complications such as ventilator-associated pneumonia by 40%.

One of the hospitals, 450-bed Phoebe Putney Memorial Hospital in Albany, GA, exceeded the targeted goals. Working with respiratory patients under DRG (diagnosis related group) 475, the medical team reduced the median number of days patients were on mechanical vent to three from a baseline of 5.5 days.

It also was able to reduce average charges for the same cases by about 22% to $28,346 from $36,325 per admission during the study period. Mortality rates also fell to 28% (with ratio of 10/28, or 10 of 28 patients surveyed) from 36% (27/98).

Phoebe Putney "proved that process improvements are actually possible for the ICU when the environment encourages change," says Connie A. Jastremski, RN, MS, MBA, an assistant professor at the State University of New York Health Sciences Center in Syracuse and a member of the IHI planning group. (For a breakdown of the hospital’s performance indicators, see chart on p. 50.)

The IHI Breakthrough Series, the initiative, put much of the theory about benchmarking and quality improvement into direct action, which for Phoebe Putney, led to some striking results, Jastremski says.

How did the hospital achieve its numbers? Bill Brock, MD, director of critical care, says the institution’s three unattached ICUs (medical, surgical, and coronary) began with a historical orientation of adopting process improvements long before the IHI initiative came along.

The practice of using standardized procedures for the ICU has been a function of practical necessity going back several years, Brock says.

It’s been responsible for the medical team’s long-term ability to manage patients under stressful conditions that include running three separate ICUs located on three different floors of the hospital with a total of 34 beds. There is also a nursing staff that is largely shared by the three units, Brock adds.

The process of standardizing certain procedures is based on two key elements of patient care. One involves what goes on at each patient’s bedside. The belief, according to Brock, is "what proved to work best with previous patients is likely to work again with the next."

The second is what the clinical team can do to emulate benchmarked parameters using information provided by bedside nurses. These parameters include everything from gradually increasing ventilator weaning to achieving significant reductions in length of stay for patients with respiratory failure, Brock says.

To do this, clinicians have had to keep a close eye on the patient data, which are obtained from daily nursing records, routine monitoring and nursing consults, then analyzed and compared to previous outcomes data of patients with the same DRG (respiratory failure).

As part of the IHI project, the hospital attempted to standardize the process of weaning the ventilator-dependent patients under DRG 475. By standardizing the management of these patients, the team also believed it could streamline and accelerate the weaning process safely.

The team conducted 100 tests on a designated study group of 98 patients over the 12-month period to achieve its outcomes. The tests included ways to prevent unplanned extubations, deciding on the ideal time for tracheotomy, and the value of screening patients for nutritional needs within 24 hours of admission.

The effort was helped substantially by the addition of a full-time respiratory therapist (RT) to the clinical team, says veteran ICU nurse Debra Williamson, RN, nursing director for critical care. The RTs — one on each shift – were not borrowed from another department on a part-time basis, but assigned to the coronary unit and specially trained to be part of each patient’s care plan, Williamson says.

In a combination of hands-on and classroom training, each RT was integrated into the nursing team. They learned to do conventional ICU nursing tasks as pressure-line set-ups, angioplasty catheter insertions, and intravenous (IV) starts. Some even assisted in performing electrocardiograms.

They also spent most of their time on the floor carefully monitoring each patient’s progress while trying by degrees to take the patient down from full vent dependence, IV and pressure line support, and partial vent status, without trying to reach any predetermined goals or minimum days.

The objective included a regulated reduction in the ventilator IMV (intermediate mandatory ventilation) rate in regular time intervals. Patients were monitored every 15 minutes. The RT, however, worked within the shadows of the attending nurses, who stood by at all times for guidance with a set of physician-endorsed weaning guidelines in tow, says Williamson.

The nurses didn’t directly supervise each ventilator resetting. For example, the RTs didn’t necessarily check with the nurses each time they decreased the CVP (central venous pressure) wedge reading from 18 mm of mercury to 10, or reduced the patient’s oxygenation level from 100% to 92% of FiO2.

But nurses were aware these adjustments were being made, Williamson notes. "There was a certain amount of cross-over between the nurses and the RTs, but they never confused their proper roles," Williamson recalls. (Chart on p. 51 highlights process steps in time on ventilator reduction.)

Without the direct involvement of the RTs and an integrated, multidisciplinary team approach to ventilator weaning, the unit could not have achieved the resulting numbers or established a standardized procedure for those particular patients, Williamson concludes.

Fisher of Fairfax Hospital offers a caveat about using internally collected or imported data as a basis for launching into process improvements. Managers who are cautious about an over-reliance on the quality of data taken from any source would be erring on the safe side, Fisher says.

Evaluate a broad set of factors, she advises. Avoid looking solely at patient outcomes, including mortality figures, whether they are internally generated or taken from outside benchmarks. Don’t rely on the face value of a single set or a limited group of parameters such as morbidity or time-to-medication.

Changing conditions can distort reality

Time intervals and circumstances can greatly alter conditions from one scenario to another, which can end up distorting reality and hurting your process improvements, Fisher says. This is especially true when establishing protocols across a multihospital system.

Fisher recalls an effort at Inova Fairfax to decide whether beginning a standard regimen of aspirin on certain patients within 24 hours of ICU admission was a good idea. The clinical literature had shown that an aspirin regimen initiated on patients suffering from myocardial infarction within 24 hours of an ICU admission reduced mortality risk by 20%.

The clinical staff at Fairfax Hospital did not doubt that the literature was correct. But would it work for Fairfax and its three sister facilities well? Could the overworked nurses administer the aspirin to the MI patients within the prescribed time as a standardized procedure?

After collecting initial data, not on patient outcomes but on whether nurses at the four facilities were able to maintain the aspirin regimen over the course of three months, Fisher discovered that the data obtained from the study were faulty.

Each of the hospitals submitted different success rates. After reviewing the causes, nurses at the poorest-performing hospital of the four could have administered the aspirin by suppository but did not. The alternative would have improved their showing.

Yet, the data suggested that the hospital fared poorly. The nurses couldn’t administer the aspirin regimen consistently due to factors such as the patients’ conditions. The incident reveals the importance of evaluating an entire process rather than focusing solely on outcomes, Fisher explains.


    1. Rainey TG, Kabcenell A, Berwick DM, et al. Reducing Costs and Improving Outcomes in Adult Intensive Care. Boston: Institute for Healthcare Improvement; 1998.