Lean Labs Reduce Blunders

Until recently, Medical Center Laboratory in Jackson, Tenn, took a batch approach to specimens. The phlebotomist dealt with a large number of tubes and labels, and, because the team did not use the tube system, he or she would bring all of the samples to the laboratory at once. This collection process alone took 20 minutes, says Debbie Robinson, a BSMP and specialist in chemistry, and assistant director for laboratory services in the core laboratory.

The lab would then have to deal with the influx of tubes, creating a rush that increased the risk of errors. In this way, the facility processed about 14,000 sticks and 7,000 tests per day; full-time employees (FTEs) handled about 13.3 tubes per hour, or 100 tubes per day. The process wasn’t seen as problematic, but just a regular day at work.

“Because of the way we work, we look at problems as natural, when in fact, they are really problems,” says Stephen S. Raab, MD, professor of pathology and director for the Center for Pathology Quality and Healthcare Research at the University of Pittsburgh Medical Center. This has often left laboratorians with the inability to fix the process. But that is changing.

Before (top) and after (bottom) ValuMetrix, the consulting service of Ortho-Clinical Diagnostics, put in place the principles of Lean and Six Sigma in operating rooms at Le Bonheur Children’s Medical Center in Memphis.

Automation has helped to reduce the opportunity for manual errors, and new software tools have expanded those capabilities. Automated order entry, bedside bar code labeling, and one-batch processing can help to reduce errors in the preanalytic stage. Automatic rules making, results delivery, and regulatory-compliance assistance help in the analytical and postanalytical stages.

Medical Center Lab combined these tools along with a process-improvement system to improve its performance by 32%: after implementing Lean and Six Sigma, FTEs were able to handle an average of 19.3 tubes per hour, or 147 tubes per day. The laboratory managed this in collaboration with ValuMetrix, the consulting service of Ortho-Clinical Diagnostics, a Johnson & Johnson company located in Raritan, NJ. Its Process Excellence methodology uses principles of Lean and Six Sigma as a quality process-improvement approach.

The laboratory team has been examining and reworking one process at a time, tailoring the improvements according to its needs. “You cannot fix everything simultaneously,” Raab says. Because of the many options to improve processes and reduce errors, laboratories must evaluate products and services according to their specifics. “A lot of the scope of errors depends on what methods laboratories have to detect them and what one classifies as an error,” Raab says.

If limited to “errors” that produce serious patient consequences, then the occurrence is rare, Raab says. A broader definition of error would incorporate deficiencies, such as specimens submitted without complete information or suboptimal samples. “Some people have estimated the frequency of these problems to be as high as 30% to 40%. But most of these do not lead to serious patient consequences,” Raab says.

Unfortunately, many opportunities for error occur outside of the laboratory’s area of control. “Lab error is the smallest component of the whole testing cycle,” Raab says. But it is still worth targeting for reduction, not only to reduce any small amount of patient risk, but also to reduce economic impact.

“In this day and time—when we are not graduating MTs like we used to and have reduced budgets—if you don’t learn to lean out your waste and think smarter, then you are cutting your own throat,” says Vickie Mayo, MPASCP, ACSPMT, process excellence manager with Medical Center Laboratory.

It’s the Process…

Mayo’s colleague, Robinson, thinks a learning process should be undertaken before automation is brought in. “If you automate a bad process, you’re not helping yourself,” she says. But the process needs the commitment of management and the team, who may be skeptical and/or withhold resources. “They are going to have to invest time and money, but in the end, it makes life easier and reduces overhead tremendously,” Robinson says.

Jamie Miles, a senior consultant with the Johnson & Johnson Co, notes there are two types of investment: the financial price and the administrative time. “From a cost standpoint, it can run about $200,000,” Miles says. The Medical Center Laboratory’s first project took about 15 to 20 weeks and required four employees’ full attention. Miles estimates a typical return on the investment is seen within 16 to 20 weeks.

Earl Buck, MT(ASCP), a Six Sigma black belt and VP of operations management at Chi Solutions Inc, Ann Arbor, Mich, suggests that the investment could rise even higher. “Lean and Six Sigma require up-front investment, and the cost can run as high as $1.5 million over a 18-month period. A single Lean project does not fix everything, so it must be repeated. But those who do this will see a 3% to 15% cost reduction over the life cycle of conversion,” Buck says.

The process-improvement methods require such large investments because there must be a formal evaluation of the lab. “The first thing you must do is evaluate the process flow,” Mayo says. To begin its first project, the Medical Center Laboratory team videotaped a tube of blood from collection through to results reporting and broke the process down into steps.

According to Miles, applying Lean helps to reduce waste, while Six Sigma targets variability, and design excellence builds a process from scratch.

“The best way to reduce error is by coming up with a standard procedure,” he says. In this way, everyone follows the same protocols and wasted processes are eliminated.

The use of smaller volumes creates more opportunity to catch errors; hence, the end to batch processing at Medical Center Laboratory with such successful results. Miles says that phlebotomists and laboratorians work with three to 10 tubes at a time, as opposed to 20 or 40.

The principle can and should be applied throughout the process. “If you’re good at it, you take what you learned and apply it to new processes,” Miles says.

Buck concurs. “Lean is a culture, and the way we approach implementation is to begin the process of changing the culture.” Comparing the process to Six Sigma, he notes the latter has a greater focus on value. “The first generation of Six Sigma targeted error and defect reduction. The second looked at reduction of costs. The current, third generation is focused on the creation of values,” Buck says. The process, when conducted singly, uses more quantitative methodology than Lean but can produce greater returns.

Failure Mode and Effects Analysis (FMEA) is a process-improvement method that falls under the Six Sigma umbrella, according to Jan Krouwer, PhD, Krouwer Consulting, Sherborn, Mass. FMEA is proactive, focusing on the prevention of errors not the recurrence of observed errors, Krouwer suggests. Using another industry as an example, he equates the effort to prevention of a meltdown at a nuclear power plant. A medical example would be avoiding the transplantation of organs of the wrong blood type. “The risk never goes to zero, but we can make sure it remains low enough with the right controls in place,” Krouwer says.

FMEA can be completed on developing processes as well as existing ones. It is designed to be specific, so it targets one process at a time. Krouwer suggests that shared laboratory processes, such as labeling, could be targeted with one FMEA while specific assays may require their own.

He offers a software product that assists with the process, including a flowcharting capability that facilitates better control. With the process mapped, a team can evaluate potential errors for consequence, severity, and quality control. “The software provides a framework [for process improvement],” Krouwer says.

In general, process control—whether FMEA, Six Sigma, Lean, or others—provides a framework. Medical Center Laboratory’s Robinson suggests that process excellence requires an organization to evaluate operations, expectations, clinical lab tests (including complexity and volume), staff specification levels, employee work styles, and software. “You must have LIS connectivity. Any lab looking at process excellence needs to look at these items first to figure out where they are and where they want to move to,” Robinson says.

Mechanics Over Man

The LIS itself helps to reduce errors in a number of ways through automated specimen handling and results reporting. These can include: electronic ordering; specimen identification (bar coding); quality control; regulatory compliance; notification of critical values; results entry and delivery; trends reporting; autofiling and data storage; validation of diagnosis codes; and interfacing with systems such as accounting, HIS/EMRs, and practice-management systems. The ability to communicate with other systems improves the sharing of accurate data.

“If your lab does not have an LIS and is paperless, you are probably doing a lot more manual work than a lab using an LIS—for example, searching for previous patient results in a filing cabinet or faxing and calling results to physicians,” says Kerry Foster, director of marketing with Orchard Software Corp, Carmel, Ind. Automating these tasks eliminates the possibility for human transcription error and frees the technologist to perform other jobs.

Before (top) and after (bottom) photos illustrate the effectiveness of the Visual Management and Control Tool Organization of Lean and Six Sigma used by ValuMetrix at Le Bonheur Children’s Medical Center in Memphis.

The benefit can begin with collection. Handheld technology for phlebotomists allows electronic orders to be placed and patient identification to be automated through the use of bar codes. Portable printers enable patient labels to be printed bedside, allowing less room for error with single-specimen processing. “When specimens don’t need to be relabeled, it reduces the possibility for error,” says Debbie Tillman, MT(ASCP) senior product manager, (diagnostic enterprise) with Misys Healthcare, Raleigh, NC.

Similarly, reducing the need to transcribe analyzer results into another system can improve results reporting and technologist efficiency. Advanced systems can also implement rules-making capabilities that monitor the analyzer results for action. “Technicians can review results in question and not have to waste time on those that fall within preset parameters,” Foster says. It simplifies technologist training and standardizes the process, thereby reducing the potential for error. How specific the rules can be varies with the system. Some allow users to get as specific as physician and patient.

“The LIS will compare a result to the rules, and if the technologist needs to call the physician, it will display a message, including the phone number, for the end user. JCAHO requires documentation of this call, and the LIS provides the opportunity to complete that immediately,” Tillman says.

“A good QC package can also help to determine if the analyzer is generating accurate, precise results,” says Jim Kasoff, VP of Antek HealthWare, Reistertown, Md. Having these transferred electronically also results in fewer transcription errors.

Storing the information digitally provides easy access, not only for compliance materials but also for patient histories and billing. “Cumulative reporting over time, whether for a patient or test, allows the lab to review trends,” Kasoff says.

Similar accuracy in billing can also help the bottom line. “There are a lot of errors that can be made in diagnosis codes; frequency testing; and limitations related to CMS, diagnosis, and insurance policies,” Foster says. By improving reimbursements, the laboratory can sometimes improve its revenue.

The investment will vary for each lab, depending on its size and complexity. Kasoff suggests the range is anywhere from a few thousand to $1 million. Foster agrees, estimating that a small POL, with a hematology and chemistry analyzer, could run approximately $25,000. “There are lab systems out there that are more like pure data managers with limited rules capabilities,” Foster says.

He suggests that laboratories with more staff and complexities would require a more complex LIS. “In a large hospital, the price of a LIS could potentially be $1 million. However, there are a lot fewer large installations, and the average price is just under $100,000,” Foster says.

Enterprises that want to incorporate handhelds can expect to invest anywhere from $75,000 to $300,000, depending on the number of devices needed and the vendor selected, Tillman estimates.

Kasoff suggests that institutions can see a return on the investment in as few as 2 to 3 years. This includes time savings by the technologists. “This allows for additional testing to be performed in the lab,” Kasoff says.

The LIS can even offer workflow-management tools. Kasoff notes that a new dashboard feature provides a real-time view of workflow, complete with statistics. “It can track workflow in a number of ways, including calendars, task lists, pending requisitions, samples not yet received, pending tests, and volume,” Kasoff says.

The tool may help to gather the information laboratories need to improve their processes, including determining which processes to improve first. Knowing where to trim the fat will certainly help a laboratory to lean down.

Renee DiIulio is a contributing writer for  CLP. For more information, contact .