CEO SUMMARY: After hospital labs and pathology groups implement Lean and process improvement methods to harvest the easiest cost savings and boost quality, they often take the next step of introducing real-time analytics systems. Access to detailed data about workflows, productivity, and turnaround times then drives continuous improvement projects. At St. Joseph’s Hospital and Medical Center, the lab team used an analytics solution to save $500,000 per year in phlebotomy costs.
WHEN IT COMES TO SOFTWARE FOR CLINICAL LABS, the hottest-selling category in recent years has been real-time analytics solutions. The need among labs to reduce costs and improve the quality of lab testing has fueled sales of these products.
Real-time analytics programs are effective enough to deliver a reliable return on investment, often in a remarkably short time. For Barbara Ballering, the Quality Assurance and Performance Improvement Manager at St. Joseph’s Hospital and Medical Center in Phoenix, that ROI was one of the pivotal issues the lab considered before purchasing such a system.
Two years after installing software to analyze turnaround times in her lab and use that analysis to improve processes, Ballering said the outcome was impressive. In that time, the lab was able to eliminate 12 full-time phlebotomists through attrition, while saving almost $500,000 in annual spending.
In addition, Ballering stated, “We were able to beat our level of phlebotomy service from when we started this project, and this is a big point of pride for our administrative director. We now have a better phlebotomy service with fewer people because we use analytics to know where and when we need those staff. That wasn’t possible before.
Room for Improvement
“Informed by better analytics, we could identify ways to use our phlebotomists more productively,” she noted. “Using attrition, we were able to cut 12 FTEs while improving the key phlebotomy service measures.
“Assume about $20 an hour—including benefits—for each full-time phlebotomist,” she continued. “By using 12 fewer phlebotomists, our lab saved approximately $499,000 per year. That produced an absolutely incredible return on investment of about 467%.
“In addition, we saw other benefits that are more difficult to measure,” stated Ballering. “These include less staff frustration and more productivity, both of which are important aspects of this new system.”
Ballering made her comments during a presentation at THE DARK REPORT’S Lab Quality Confab in New Orleans in October 2017. The title of her presentation was, “Using Real-time Analytics to Improve TAT, Streamline Staff Scheduling and Identify Various Sources of Error, Both in the Lab and in the ED.”
A Need for Useful Data
The 586-bed St. Joseph’s Hospital and Medical Center is a division of Dignity Health, a 42-hospital system in Arizona, California, and Nevada. Although the clinical laboratory has little automation, it still runs more than four million tests each year in anatomic pathology, chemistry, hematology, microbiology, and in molecular testing and cytogenetics. The lab also runs a blood bank and supports point-of-care testing throughout the facility.
As the hospital was becoming part of Dignity Health two years ago, the lab had trouble generating useful and timely information on workflow, productivity, and other key factors.
“Our process for generating reports was cumbersome and sometimes impossible,” recalled Ballering. “That’s because it would often take so long to process the data that the system would time out and the desired data would be lost.
“Also, we couldn’t compare staffing for one time period to another, which meant we couldn’t identify staffing problems. If you can’t identify problems, how can you address them?” she asked.
From Baseball to Labs, Analytics Has Value
ACCESS TO REAL-TIME ANALYTICS in the clinical lab is akin to what the best general managers in baseball do, commented Barbara Ballering, Quality Assurance and Performance Improvement Manager at St. Joseph’s Hospital and Medical Center. As depicted in the best-seller, Money Ball, by Michael Lewis, one of the best general managers in baseball is Billy Beane of the Oakland As.
The book tells the story of the small-market baseball team that lacks the money to overpay for high-priced talent, Ballering explained. “So, they had to figure a different way to find top players and they did so using analytics.
“With analytics, they got useful information on individual players,” she explained. “Team managers used data to identify players who were otherwise overlooked and put them together to make a winning baseball team. That’s what real-time analytics does.
“For labs, analytics deliver information that is available almost immediately after being entered into the system,” Ballering said. “That makes it useful to lab managers.
“This is not like the laboratory information systems of the past, where the lab might wait a month to get performance information,” she explained.
“With real-time analytics, our lab can use up-to-the-minute information to improve performance,” she commented. “And, we don’t just gather data. Rather, we get information that is meaningful to lab managers.
“Another way to think of analytics in the lab is that it’s like having real- time business intelligence available all the time!” concluded Ballering.
“In addition, our lab team needed greater flexibility because we often can’t use just raw data,” she commented. “We wanted data available in a format that allows the lab staff to understand it and put it to use.
“In 2014, our administrative director heard a presentation at the Executive War College from Thomas Joseph about Visiun’s Performance Insight System,” stated Ballering. “One year later, our contract to acquire this system was approved and we implemented the program in January 2016.
“Soon after that, lab staff were asking for a wide variety of specific kinds of data,” Ballering said. “This told me that people recognized how this system can produce a lot of actionable information.
“At first, we focused on the ED, where we needed faster throughput from the time a test was ordered until the results were reported,” explained Ballering. “Physicians in the ED were being pressured to move patients through quicker, to see more patients, and decrease patient wait time; all of which increases patient satisfaction. And of course, when patients are seen more quickly, that improves patient outcomes in many cases.
“In addition, we had complaints from physicians about how long they were waiting for stat lab results,” she said. “We knew that test turnaround times were quite long in some cases.
Getting a Quality Specimen
“We also had specimen quality issues and, like many hospital EDs, nurses collect most of the samples,” Ballering said. “We tried educating nurses about how to do patient draws, but decided it was best to draw our own specimens using phlebotomists managed by our lab. That’s been successful, even though we needed more phlebotomists to do these draws.
“On top of these problems was the issue of poor communication between the lab and the ED,” she added. “As is true in many hospitals, our lab staff tended to think the ED staff simply likes to complain and so the lab staff would reply in a way that didn’t improve the problem.
“Further, when the ED staff called the lab, the lab services staff wouldn’t know where to send these calls,” she said. “That meant the ED staff would get transferred from one area to another without getting an answer.
“Things began to change once we had the data in place from the Visiun system,” continued Ballering. “Now we could identify the source of problems in the lab.
“For example, every morning I now pull up the Turnaround Time Dashboard,” she noted. “Overnight the Performance Insight software extracts the data from the day before, letting me send multiple reports automatically to the proper supervisors in the first seven minutes each morning. That is amazing!
“On the first page of our turnaround time report, we have all the lab tests that we monitor,” she said. “We also get the average turnaround time. Recently, our medical director asked for a report on the percentage completed by a target time. So, we changed from ‘average’ to ‘percentage completed.’
“Lab staff can now measure performance however they want,” she added. “For blood specimens, we use order to verify. For urine specimens and blood gases, we use collect to verify. Also, we can set a target of 80% achieved by 70 minutes, for example, then monitor actual performance against that target.
“By doing so, we get the actual time, how our clinical laboratory did on a particular day, the number that passed, and the number that failed,” Ballering explained. “We also get a list of outliers for each test, which lets our team focus on areas needing attention.
“Also, we get performance trend reports over time, whether by weeks or months,” she said. “All that data is broken down by whatever department we want: such as the ED, the ICU, or microbiology.
Comparison with Peers
“A large number of our reports are related to productivity, which allows us to compare how our phlebotomists perform against their peers,” she explained. “This data is useful because we needed the right mix of phlebotomists in the ED, not just the right number.
“This report gives us the average activity and patient collections per phlebotomist,” Ballering added. “Now we know the number of phlebotomy collections that occur each hour of the day, which phlebotomists worked in those hours, the total phlebotomies performed in each 24-hour period, and what percentage of the total our phlebotomists collected.
“We also can get information on the quality of each specimen broken down by patient location,” she added. “That lets us know if someone needs retraining. Or, if a phlebotomist is new, perhaps that person shouldn’t be in the ED yet.
“With these detailed daily reports, we know that average turnaround time decreased from a low of 6.5% in some areas to a high of 15.1% in other areas,” she reported. “Our percentage of outliers decreased by a low of 11% to a high of 53%.
“Physician complaints also went down,” she said. “At the start of the project, the lab got about 15 complaints a week. Today, we get about two complaints a week.
“That service improvement alone has increased satisfaction in the ED and for the lab staff as well,” she commented.
“As a result of having performance data on so many aspects of lab operations, we saw when the ED is busiest, which was from noon each day until about 8 pm,” Ballering concluded. “Now, our detailed reports deliver hard data that tell us daily when we need to beef up phlebotomy staff in the ED.
“Alternatively, when it’s not so busy, we can recycle those phlebotomists from the ED to the floor to do in-house draws or to work on other projects,” Ballering added. “Clearly, all of this has been positive for us because it allows us to fix problems as they arise.”
Using Detailed Analytics to Improve Lab TAT
EACH MORNING, a full range of reports covering the previous day’s lab operations are delivered to Barbara Ballering, Quality Assurance and Performance Improvement Manager at St. Joseph’s Hospital and Medical Center.
At right is a sample report that shows tropinin turnaround times from the emergency department. These reports combine key statistics and data with charts and graphs that make it easy for lab staff to see the lab’s performance against its goals.
Contact Barbara Ballering at 602-406-2083 or Barbara.Ballering@DignityHealth.org.