July 2009
Feature Story

Karen Lusky

Laugh and the world laughs with you. Share a tale of professional failure with a group of fellow laboratorians and the room tends to become unusually quiet.

Richard Zarbo, MD, DMD, senior VP of pathology and laboratory medicine at Henry Ford Health System, Detroit, does both to engage people in learning about Lean. He wryly pokes a little fun at himself at times and shows hilarious training videos, such as the one of someone stuffing too much tissue into a cassette and trying to pound it shut with a hammer. And he rivets trainees’ attention by sharing how he initially couldn’t get off the ground what has since become a top-notch Lean quality management system in all of the Henry Ford labs.

The false start was in 2004, shortly after Dr. Zarbo had taken a course on Lean and the Toyota Production System. Excited about what he’d learned, he came back and educated the surgical pathology staff about the key principles, including the need for standardized work to eliminate variation. Then he spent time observing how the pathology lab staff did their work and wrote standardized work for them to follow, posting it on the wall in the lab.

A week later, no one was paying attention to what he’d posted. In fact, the staff never did follow what he had formulated.

“I failed because I started as a top-down leader,” Dr. Zarbo said in a presentation in April at the two-day, jam-packed Lean training event he and his leadership team host five times a year for Henry Ford employees and a growing number of laboratorians from around the U.S. and other countries.

A truly Lean lab, Dr. Zarbo says, harvests the creativity and efforts of the entire workforce.

Rita D’Angelo, MS, ASQ CQE, SSBB, quality improvement specialist at HFHS, notes that people love to hear about Dr. Zarbo’s early attempt to be a Lean leader because “we all fail at times.”

Not that Dr. Zarbo stayed down for the count. Instead, he obtained a grant, hired D’Angelo to help him execute his Lean vision, and took his executive leadership team to a weeklong Lean training course provided by the Pittsburgh Regional Healthcare Initiative.

“Once I got their buy-in, I knew I could actually start a process of enculturating Lean,” he told CAP TODAY.

You could say the rest is history, but Dr. Zarbo and his team and the rest of the laboratory staff continue to break new ground by producing hundreds of quality improvements a year. And they have key lessons to impart about how they rolled out and continue to maintain and expand what has become known there as the Henry Ford Production System, a compilation of their founder Henry Ford’s original production system, Deming, and the Toyota Production System principles.

The first lesson, one Dr. Zarbo says he learned the hard way, is that you have to find a way to engage staff at every level in the quality improvement and management process.

Thus, in 2005, after getting the executive leadership on board, Dr. Zarbo retackled the surgical pathology laboratory as the initial Lean pilot project. He identified seven work cell leaders in the lab—people who handed off work to each other—and took them to the Lean training in Pittsburgh, securing their buy-in.

Despite his early failure, Dr. Zarbo was determined to start out with Lean in the surgical path lab because, as the longstanding chairman of the department, he had direct control over it and knew it well. In addition, the work there was done manually, and manual work processes are easier to Lean out than already highly automated processes. And the pathologists were frustrated with how things were going. The lab at the time was a “scowling place” to work, he says, reminiscent of the auto factories where people dashed out of the gates at quitting time.

“One major problem,” D’Angelo says, “was that people did not get along—that was a huge factor. People did not speak to each other, had a silo mentality, and were unaware of the issues the other side faced.”

So Dr. Zarbo and D’Angelo decided to open the floodgates, as D’Angelo puts it. They handed out a survey asking people in the surgical pathology lab to identify the top 10 things about their work that didn’t go well daily—the “defects,” if you will, Dr. Zarbo says. “We defined a defect as anything that requires your work to stop: a pause to fix, a pause to return it to the sender, or delays.”

Based on the employee survey, 100 different defects were identified that could take place in the lab. Then “we created a data collection tool we called a visual data display poster,” he says. The lab staff members were instructed to get up and write on the posters every time they saw the defect or waste occur, and for how long.

An analysis of the data provided a “wealth of information about what didn’t go right in the shop,” Dr. Zarbo says. Then the team could begin to target those defects with root-cause analysis to determine what was causing them.

The data collection also allowed the Lean management team to work with staff to establish a “baseline state of in-process waste, so we could subsequently quantitate our progress from numerous process improvements,” Dr. Zarbo says. “If you can’t measure it, you can’t improve it.”

After everyone in the lab who hadn’t gone to Pittsburgh for training received in-house training on Lean, the process improvements began in earnest.

“The first task,” D’Angelo says, “was getting people into the same room for dialogue and sticking to the task. We said, ‘Forget about what you feel about him or her—it doesn’t really matter. Let’s stick to the process.’ That was our standard line. We asked people to look at the process and how we can do things better for everyone,” she says.

Once the dialogue started to flow, the arguments escalated. People were willing to participate at that point “but disagreed on how to proceed or on which improvement to focus,” D’Angelo says. “Many times Dr. Zarbo would attend the meeting to act as a referee,” she adds. Or people refused to attend the meetings, “so we would search them out and bring them to the meeting. It was painful.” But since the chairman of the department said it had to be done—there was no way out.

Once quality improvements started occurring —“and people saw they were contributing to the success”—the tide began to turn, D’Angelo says. The lab racked up 42 quality improvements in the first month. “Every month, we showcased our improvements at our Share the Gain meeting where the expectation became one process improvement per month per team.”

The pilot effort reflects a number of key Lean and Toyota Production System tenets. For one, it involved rapid process improvements that produced quick wins, which got everyone excited about the prospects for Lean.

And Dr. Zarbo and his leadership team took steps to ease fears. For starters, Dr. Zarbo promised that no jobs would be lost as the lab began “Leaning out” inefficiencies. Without that promise, “no one in their right mind would participate,” Dr. Zarbo says. And “this was before the economic crisis when jobs became a much-valued commodity.”

The leadership team also created a “blameless” environment where people didn’t have to worry about finger pointing when examining what didn’t go well in the lab. “We rewarded people for identifying a defect and assisting us to figure out how it was arising and using that as an opportunity” for improvement, Dr. Zarbo explains.

At the outset of the Lean pilot in surgical pathology, one of the largest defects passed on to pathologists involved cassettes that didn’t properly specify the part type. So the accessioners looked at all of the part types in the dictionary and worked with the pathologists and clinicians to develop a “clean, Lean part-type dictionary” that everyone involved in that process was trained to use, says Dr. Zarbo. “What you call [the specimen] up front is tied to what happens with the work down the line,” Dr. Zarbo says in explaining this step’s importance. “We have a lab information system-encoded work specification. So we had to have the [part] name right up front as it [determines] how many cassettes and slides and stains [are produced]. We started at the beginning to clean up defects down the line.”

The lab also had “many, many recuts,” he says, “and we tried to understand why.” Using root-cause analysis, the team traced that defect to how the specimen was dissected, oriented, embedded, and fixed. That discovery made it possible to change the specifications for processes and to reduce the number of recuts by 97 percent in less than one year.

Pathology assistants came up with a way to shave a day off the fixation time for prostate specimens by injecting them with formalin during grossing, D’Angelo said in a Lean presentation at the Dark Report’s Executive War College, also in April.

Pathology residents eliminated the time required for them or the pathology assistants to take breast specimens to radiology for x-raying to pinpoint the location of tumor or calcifications before cutting the tissue blocks. Once in the mammography center, the PAs or residents had to wait in line just like patients, D’Angelo told her audience. So the team did a cost-benefit analysis and successfully made a case for the pathology lab to buy its own digital mammography machine, which cost $100,000. But the resulting improvement in turnaround time (1.5 days compared with five to seven days pre-improvement) made the investment worthwhile, D’Angelo said.

Meantime, Mark Tuthill, MD, head of pathology informatics at HFHS, and others were working on a bar-coding system for the lab that would prevent patient identification errors and allow the bar code to interact with the LIS from accessioning to sign out. (See “Using Lean to end labeling errors”). Dr. Zarbo says bar coding “was developed in parallel” with the Lean initiative in pathology, and brought in later, because it required using technology in a new way. And “you don’t want to automate bad processes,” he notes.

When the surgical pathology pilot began, the lab had 50 cases a day that were defective. When you have that many defects, “you can’t fix them on the fly,” Dr. Zarbo told the Henry Ford training session attendees. “You’re drowning in them.”

Within a year of starting Lean, the 50 defects per day were whittled to 30, and within two years, to five a day. That’s a 91 percent reduction, and a small enough number that people can stop the line and fix them in real time—another Lean and Toyota Production System concept.

Amended pathology reports have dropped from about 25 out of 50,000 cases a year pre-Lean to about nine a year. “And we keep trying to lower that,” Dr. Zarbo says.

The goal for the Lean effort in any Henry Ford lab, he says, is to reduce defects to zero. “So we have zero-defect performance goals. This goes beyond what we used to be hung up on, which is benchmarking [using a mean or median] for performance. When you do that, you’re just trying to hit the goal without understanding” what factors are causing people to perform at that level.

“Here you are your own benchmark. What you did yesterday, you can do better tomorrow.”

Under Dr. Zarbo’s direction, Lean teams in the clinical laboratories have also scored numerous Lean quality successes. The microbiology lab, for example, tackled a high rate of breakage of blood culture tubes by working with the vendor that makes the bottles and instrument. “The vendor is working on a more robust bottle, which is a long-term project that will take years,” says John L. Carey III, MD, vice chair of pathology and director of clinical laboratories at HFHS, who also spoke at the training session. Meanwhile, the vendor supplied the lab with a plastic insert in which to securely hold the bottles during transport in the pneumatic tube. Nurses, too, were trained to use the plastic insert. As a visual control, a one-page laminated sheet of instructions on how to use the insert was placed at the pneumatic tube stations where the nurses send the specimens to the lab.

In another project, a team sought to eliminate the time involved in having the microbiology staff resort aerobic and anaerobic blood culture bottles into pairs for each patient, as required to put them on the blood culture instrument.

“Instead of just bunching all the bottles into one large bag for transport, we designed a rack that maintains the pairing for each patient,” Dr. Carey explains. The approach also led to the microbiology lab receiving smaller, more frequent deliveries of blood culture bottles from the core laboratory.

As a result of the changes, breakage of blood culture tubes during transport went from 30 per month before the project to less than one per month, and these remaining breakages occur when the nurses don’t put the tubes in the plastic carrier. The second project eliminated 1.5 hours a day of staff time seven days per week, annualized to .20 FTE in terms of time saved, Dr. Carey says. “While the smaller, more frequent deliveries of sorted racked bottles did improve the preanalytic TAT, this was not formally measured,” he adds.

Most recently, Dr. Zarbo hired several engineers, who had previously been with Chrysler, to further refine production flow and inventory management in the labs. One of those engineers, Yad Sidhu, MSME, MBA, is working on a pilot project to implement a new inventory management model and the Kanban system (whereby cards or containers are pulled for reordering at a certain point in the inventory). Using the new system, “we are able to reduce lab inventory up to 70 percent for pretty much everything” to free up space and capital, Sidhu says.

To identify the best time for reordering a particular supply, “we calculate the point at which the quantity and frequency of the orders gives us the lowest total cost. ... Once we find that point, then we implement the Kanban system.”

For example, the Henry Ford labs are now ordering a six-month to one-year supply of some re­agents and quality control materials to avoid having to do QC or cal­ibration when they use a new lot. Upon closer an­alysis, how­ever, what sounds like a smart strategy actually costs the lab more in most cases.

Using cost-allocation methods for employee hours and machine hours, Sidhu and his team figured out that the optimal frequency for reordering is somewhere within the three- to six-month time frame. “There are a few exceptions, depending on the price and how much work we have to do for quality control and calibration for a new lot,” he says.

Not every Lean project is a slam dunk or even one for which a consensus is reached about what to do. Discussions still get heated at times, and, in some cases, people have to vote on a solution for eliminating a defect. Sometimes the selected solution doesn’t eliminate the defect, or it helps but not enough. Then it’s back to the drawing board with Dr. Zarbo’s anchoring mantra in the background, reminding people to stick with the scientific process to define the problem and come up with the answer.

What Lean isn’t about at Henry Ford is a rigid insistence on small things, such as always putting the phone or stapler in the same place marked off by tape on the desk, as some Lean instructors advise.

“If that’s the way workers want to design their work cells, fine,” Dr. Zarbo says. “If they have an issue where they can’t find the stapler and need it for work processes, they should address it. But I don’t care if the stapler is hanging from a rope.” He adds, “If you don’t want to make your bed, I don’t care.”

“There is no cookbook approach to Lean,” he says. Leaders must create the program and tailor it to their needs. “The tools you can read about, the results you can read about, but how Lean is applied is different [in each environment] because Lean is a living thing—it’s the people,” Dr. Zarbo says.

For example, at Henry Ford, where teamwork and professional development are important, the Lean training sessions for employees are designed, in part, to help the group begin to coalesce.

“When you train an entire team,” D’Angelo said at the War College, “they learn how to work together and start to like each other.” During training, “we take people to the ... Ford truck plant to show them the Lean ideas there. We take them to the bar and out to eat to bond as a team. We want them to see their work differently and improve daily.”

Dr. Zarbo and his executive team have also come up with a systematic way to keep the Lean stream of initiatives flowing. He provides the visionary goals, while D’Angelo teaches the Lean theory and tools. Ruan Varney, CT, ASQ, CQE, SSBB, quality improvement coordinator, supports the Lean teams by convening weekly 20-minute meetings at the work sites to discuss and document projects, using what’s called an A3 report format. That’s a problem-solving tool backed by data and used to size up the magnitude and root causes of problems and whether the interventions are working. These reports are shared at the monthly departmental Share the Gain meetings.

When a team member in a work cell asks why something is done a certain way, an “observation exercise” is done. At one such exercise, histotechnologists observed pathologists as they signed out the slides of a previous day’s biopsy case. “The histotechs were amazed to learn how the quality of slides affected the pathologists’ decisionmaking process,” Varney says. This gave them what they needed to improve how they prepare slides.

In the microbiology lab, a similar exercise helped solve a problem at the setup window, where specimens are received, plated, and incubated. Varney worked with setup area team members to translate their complex problems into simple process flow maps so they could see where the delays and waste were, level out the incoming work, and create continuous process flow.

Lean at HFHS has come a long way from the day that Dr. Zarbo posted on the surgical pathology lab wall his version of standard work that everyone ignored. Today, there’s a wall of fame that people can use to recognize their co­workers’ contributions. “It is amazing,” Varney says, “to see this wall filling up month after month.”

As for what’s next on the Lean map, Dr. Zarbo says he and his team are “edging their way” into doing Lean in the hospital system itself, “which requires a conversation with the people who hold the purse strings.”

As a step in that direction, Dr. Zarbo and his leadership team reformatted the two-day HFHS Lean training session, starting with the one held last month, so it applies to all health care, including inpatient and outpatient, rather than just laboratories. The April session, in fact, already had some non-lab HFHS employees in attendance, including behavioral health, cardiology, dermatology, and surgery staff and even kitchen staff.

“What Dr. Zarbo has done,” Dr. Tuthill sums up, is to come up with a vision and marketing that frames quality improvements “under the single moniker of Lean or the Henry Ford Production System.” That approach “allows us to lump [the quality improvements] together, galvanizing our effort and allowing us to take credit for the outcomes.”

It also makes the improvements more visible to the organization’s leaders, who, as a result, are now considering workflow solutions from end-to-end rather than in silos.

“Whether we will have the ability to apply these principles across the institution to ultimately impact patient care to the nth degree remains to be seen,” Dr. Tuthill says. But that goal “is in Dr. Zarbo’s cross­hairs.”

Related Links

• Using Lean to end labeling errors