TRIUMPH: Best Practices in Inpatient Glucose Monitoring. UCLA Clinicians Use IT to Facilitate Innovations in Hyperglycemia Care


March / April 2008

TRIUMPH: Best Practices in Inpatient Glucose Monitoring

UCLA Clinicians Use IT to Facilitate Innovations in Hyperglycemia Care

Patient Safety and Quality Healthcare

Tracking glucose levels in ICUs at UCLA Medical Center.
Photo courtesy of Global Care Quest

Clinicians long have known that hospital inpatients who have diabetes tend to have far poorer outcomes, require longer hospitalizations, and develop more complications, than do inpatients who are not diabetic. Clinicians also know that non-diabetic patients admitted for a range of conditions and injuries experience an elevation in their blood sugars (hyperglycemia) while hospitalized. In fact, studies performed in the past several years indicate that non-diabetic patients whose blood sugars surge during inpatient hospitalization actually experience worse outcomes than do diabetics.

Yet these findings are often ignored by clinicians because of the knowledge that illness or trauma to the body induces physiological stress, which may result in elevated blood sugar. This was previously thought to be a transient phenomenon that was harmless. However, a landmark clinical trial performed in 2001 changed clinicians’ views about stress hyperglycemia in the inpatient setting. The authors of that New England Journal of Medicine study, concluded that “Intensive insulin therapy to maintain blood glucose at or below 110 mg per deciliter reduces morbidity and mortality among critically ill patients in the surgical intensive care unit” (Van Den Berghe, et al., 2001).

Still, few patient care organizations have followed through on these findings in the day-to-day world of the ICU. One that has, however, is UCLA Medical Center in Los Angeles. There, clinicians have established and evolved a program that could potentially serve as a model for inpatient glucose monitoring, clinical management, and facilitation of critical information that technology has provided to UCLA Medical Center’s program.

“This is still a relatively new area of focus, and we don’t know everything yet about the underlying physiological dynamics, but by and large, clinicians understand that the development of hyperglycemia in inpatients is something that should be addressed,” says Archana Sadhu, MD, the UCLA endocrinologist who established the hospital’s program, called Targeted Insulin Therapy to Improve Hospital Outcomes, loosely acronymed “TRIUMPH.”

“We started the program in January 2005 with a small pilot program in the medical ICU and in the coronary care ICU,” Dr. Sadhu continues. “First, we had to develop the actual clinical protocols, which included specific insulin delivery order sets. Then we trained the nurses.” After developing the program’s protocols and training all the nurses working in the medical and coronary care ICUs, Sadhu and her colleagues presented their recommendations to the hospital’s pharmacy and therapeutics committee, which gave its approval for implementation hospital wide. In April 2005, the TRIUMPH program started live in two ICUs at UCLA; and, over time, it has expanded, and currently, it is live in five of the medical center’s nine ICUs.

Essentially, the program involves the daily monitoring, tracking, and recording of blood glucose levels in surgical and medical inpatients in UCLA ICUs by Dr. Sadhu and her team, with expedited clinical interventions, as appropriate, for patients exhibiting signs of hyperglycemia.

Cost Savings through Reduced ALOS
The result? “We compared the ICUs in which this program has been implemented, and those in which it hasn’t, and found that patients in the program had ICU lengths of stay that were 1.2 days shorter,” based on a baseline ICU average length of stay (ALOS) of approximately 9 days prior to the program implementation. That reduction of ICU ALOS “was an amazing, exciting result,” Sadhu says. What’s more, she says, “We found a trend toward cost savings of about $5,000 per admission” in the group of patients treated under the TRIUMPH program. Since there were about 1,000 patients in the program in its first year, the total cost savings was a potential $5 million for that first year alone. That figure is especially significant, she notes, because the majority of UCLA’s ICU patients are non-diabetic; thus, the potential for additional savings, as well as improved outcomes, as more of the hospital’s ICUs go live with the program, is considerable.

IT: A Key Facilitator
The use of information technology has been a critical success factor, Dr. Sadhu notes. She and her colleagues early on chose to implement a software program from the Aliso Viejo, Calif.-based Global Care Quest (GCQ) in order to facilitate all the data entry and documentation required to make the TRIUMPH program successful, she reports. She and her colleagues had seen a presentation of the GCQ tool, ICIS (Inpatient Clinical Information System) by Neil Martin, MD, the UCLA neurosurgeon who with other surgeon colleagues helped to develop ICIS at UCLA Medical Center. Some version of the tool, Martin confirms, is now being used in nearly 30 different departments and services across the UCLA organization.

Dr. Sadhu recalls, “Early on in our development of the program, Dr. Martin had given us a demonstration of how GCQ was able to pull up glucose values for patients, and also how the GCQ tool was able to stream together data from several different sources at once, an important capability with regard to what we do here.”

Indeed, Dr. Sadhu notes, she and her team were able to determine that their program would end up treating large numbers of patients every day. Given the small size of her clinical team (which encompasses just three people — Dr. Sadhu herself, one physician assistant, and a hospital diabetes educator), IT-facilitated efficiencies have been not just “nice,” but vital to the ongoing success of the program.

For Dr. Martin, the success of the tool as an IT facilitator for the hyperglycemia program mirrors and exemplifies its successful use in departments across UCLA Medical Center. “When specialists like Dr. Sadhu have access to these tools,” Martin says, “they can achieve several things at once. They can optimize their workflow and their effectiveness and reach as physicians; they can help support the educational element of UCLA’s medical mission; and most of all, they can improve the care delivered to patients.”

What’s more, though Dr. Martin and his colleagues in the neurosurgery department first created the core capabilities of the GCQ tool to assist in their neurosurgery work, they all recognized the potential for the tool to facilitate improved patient care and clinician workflow in many areas. Just as the tool is used in a number of intensive care units (ICUs) by residents and interns to help them assess patients’ status during their roundings and to compile data for the clinicians on the next shift, the tool is now being used in a variety of areas, including assessment of dialysis patients in the nephrology department, and in the hyperglycemia program in endocrinology.

“Almost every service or department is finding ways to customize the core GCQ solution, and to apply it to specific needs and circumstances,” Martin adds. “To me, this simply reinforces what we already know — that physicians and other clinicians very much need facilitative IT tools in order to optimize their workflow and improve patient care quality. If a tool like GCQ’s can work for neurosurgeons, ICU residents, nephrologists, and endocrinologists, who knows what is the potential for improving patient care at teaching hospitals using such solutions?”

For Dr. Sadhu, the impact of IT facilitation for her program has been very straightforward. “Without GCQ, we couldn’t do half the volume that we do now, because it takes a lot of time for us to gather this information, and we would be more limited in the number of patients we can treat at a give time,” she says.

Indeed, it is in the mundane process of data-gathering and data analysis that it becomes clear how much IT facilitation matters to the success of a lightly staffed program like hers. Every morning, when Dr. Sadhu and her assistants come in, they electronically review their lists of medical ICU, coronary care ICU, and cardiothoracic ICU patients, first determining which patients have hyperglycemia (usually 80 to 90% of the census, she notes), then evaluating those patients and beginning their clinical interventions.

The ability to quickly identify which ICU patients need the hyperglycemia program’s interventions, and to produce a daily note on each patient electronically, makes all the difference in being able to facilitate the running of the program. Dr. Sadhu and her team intervene on an average of 40 to 50 patients a day and gather information on an additional 15 patients a day who it turns out do not require intervention. Given that she and her team save approximately 10 minutes per patient in terms of gathering and then documenting data, the IT facilitation saves about 400 minutes a day — a huge impact in terms of physician and clinician workflow and efficiency.

What does all this say? In the end, Dr. Sadhu says, the lesson is clear:

The way medicine traditionally has been practiced is redundant and inefficient. If there are four doctors treating a patient, and each one comes in and looks at the same set of information, and then copies that information into paper documents in the same way, that process is using the doctors’ time and without adding any value to the care of the patient. It’s always bothered me, especially at a training institution, that you often have multiple clinicians and specialists doing the same thing every day such as documenting the same body temperature values and the same blood pressure values. Replicating this basic information is to no one’s benefit.

In my mind, our medical school education is wasted in our obligations to hand-write vital signs or lab information in a duplicative and repetitive way. Because GCQ saves me from doing this kind of busy work, I can spend more time with the patient to develop the best treatment plan for him or her. This is a much more valuable use of my resources.

Future Directions: Contributing to the Evolution of Glucometrics
Of course, there are always future horizons to look toward. “We’ve been working on a comprehensive database for GCQ that will permanently record any information we want on patients for future studies. We can go back and say, OK, let’s pull up all these patients in terms of their diagnoses and see what we did right or wrong and do a comprehensive analysis,” Dr. Sadhu reports. “We have also started a project to do various glucose analyses on patients in different units of the hospital, for safety and efficacy. So how did the neurosurgery ICU do last week with their glucose control? How many were at goal, and so forth? This will help to facilitate quality assurance throughout the hospital.” In fact, she says, such work ties into a new field called glucometrics which focuses on different approaches towards analyzing blood glucose levels in patients and optimizing them.

In the end, all those involved in this kind of work say the frontiers of patient care continue to be pushed forward in exciting ways. Using leading-edge IT tools to facilitate pioneering work in patient care at organizations like UCLA Medical Center can only encourage innovative work in many different clinical areas going forward, to the benefit of hospital patients nationwide.

Mitch Work is president of The Work Group, Inc; a Chicago-area based healthcare technology marketing firm specializing in emerging technologies. He has 25 years experience in healthcare technology and is a HIMSS Fellow. He may be contacted at


Van Den Berghe, G, Wouters, P, Weekers F, et al. (2001, November 8). Intensive insulin therapy in critically ill patients. New England Journal of Medicine, 345(19).