Highlight on Utah: New tool cuts costs, improves outcomes

In a study published by the Journal of Hospital Medicine,  investigators from the University of Utah Medical Center (UUMC)  found that a Value Driven Outcomes (VDO) tool reduced the number of unnecessary laboratory tests performed on hospitalized patients and cut costs. Investigators Peter Yarbrough, M.D., Kensaku Kawamoto, M.D., Ph.D.,  and three physician colleagues from UUMC estimated that their multifaceted intervention saved the hospital more than $250,000 the first year. Although patients often need daily laboratory work, Yarbrough said, “it can create a culture where you’re ordering tests without thinking about what you’re going to do with the results.” The goal of the study was to try to create a thoughtful process for deciding what laboratory tests to order for hospitalized patients.

Study design

The study compared the changes in laboratory costs between the hospitalists and other physician providers at UUMC.  The hospitalist services were performed by four teams of Internal Medicine residents and medical students. The control group comprised physicians from the surgical, cardiology, pulmonary, hematology, and oncology departments. Patients admitted to the psychiatric, rehabilitation, or obstetrics units were excluded from the study.  Their use of laboratory tests was measured during a seven-month baseline period, after which the intervention was introduced. The study period continued for 15 months. the design controlled for differences in age and took into account the level of comorbidity.

After an informal review of patient charts, an examination of the physicians’ work flow, and a review of the literature, Yarbrough, Kawamoto, and their coauthors noted that interns ordered the most tests and were responsible for the greatest variation in utilization of laboratory services. They noted that unnecessary testing  poses several risks to patients. False positives results lead to additional unnecessary testing. Over long hospital stays, frequent blood draws increase the risk of anemia. One cardiology study found that cardiac patients lost an average of 454 ml of blood per stay, nearly half a liter. Being awakened early in the morning for daily blood draws can deprive patients of needed rest, negatively affecting their hospitalization experience, as well.

The interventions

The first component of the UUMC intervention was education of the physicians involved on the cost of overuse of laboratory services, the results of previous interventions, and the current intervention and its goals. The residents were issued pocket cards with the most common laboratory tests and the amount charged for each. The rounding process was standardized, incorporating a checklist that required review of previous lab results, pain, telemetry, lines and tubes, nursing presence, and follow-up needed for each patient. All plans for lab testing were to be discussed during rounds. A third year medical student was tasked with making sure that the entire checklist was addressed for every patient.

Each month, the hospitalists reviewed laboratory costs using their VDO tool. The data were presented as a monthly average, though individuals could compare their performance to that of other providers in the group.  There was a financial incentive for the Department of Internal Medicine as a whole, in that 50 percent of the savings would be shared with the department to use for future quality improvement projects, but there was no financial incentive for individual physicians. The results were measured every two weeks.

Results

The number of basic metabolic panels, complete metabolic panels, and complete blood counts dropped significantly. The average daily cost of laboratory services per patient fell from $138 to $123. The laboratory cost per visit dropped by $128.  Length of stay was not significantly affected, but remained constant in both the control group and the intervention group. Readmissions within 30 days fell from 14 percent to 11 percent in the intervention group.

The multifaceted nature of the intervention makes it difficult to tease out the relative effects of the various components. The hospitalists involved, however, believed that the use of the VDO checklist and the feedback at monthly meetings were the most important factors affecting the changes in their behavior.