Leading with Lean Principles
How implementing Lean measures revolutionized one hospital system's lab operations and culture.By: Richard Zarbo, MD, DMD, Gaurav Sharma, MD, and Rita D'Angelo, MS (from: Advance/Laboratory, pp 71-72, April 2010)
After developing and implementing the Henry Ford. Production System (HFPS) quality management system, our integrated laboratory system at the Henry Ford Health System transformed its laboratory work operations and culture. We emulated the successful Deming-based management principles of the Toyota Motor Corp. known as Lean to obtain an empowered workforce using defined work rules that drive standardization, eliminate in-process variation and waste, and produce process improvements.
Pathology and Laboratory Medicine is composed of 775 employees handling over 11 million clinical diagnostic tests for the six acute care hospitals and 30 regional medical centers in the system. Four years ago, we implemented HFPS at the main teaching hospital, then expanded to our ancillary labs. This transformed the performance of our lab services and enabled our workforce to react flexibly and quickly to external challenges. In 2009, the human talent in the anatomic and clinical pathology labs on the Henry Ford Hospital main campus alone achieved 536 process improvements.
Defects and Waste
During the identification phase of a defective process, little is known about in-process variation leading 10 waste and poor quality. As part of the learning process, defective hand-offs need to be observed to identify 1) if a process is in place, 2) customers/suppliers involved and 3) defects in the process. We defined a defect as a deviation from a predetermined outcome in specimen processing, which led to work being delayed, stopped or returned to the sender. Waste is represented in a number of forms: defective products, overproduction, unnecessary processing, time wasting, excessive product/material transportation, excessive human movement and excessive stock. In pursuit of a zero-defects performance goal, we designed novel data collection tools to assess sources of defects.
A Lean workflow is a continual flow process composed of smaller evenly paced batches, introducing raw materials and utilizing workers at the right time and place with the right tools to produce a finished product that can be distributed quickly. This can be challenging, as a lab can not schedule deliveries but must accept uneven batches of specimens from different locations, perform multiple analytical processes and report results to a diverse group of customers, each with their own set of expectations. In pursuing just-in-time workflow, waste is eliminated and time is compressed. Lean moves away from batch mode production wherein defects are addressed at the end of a production cycle to a continuous mode wherein each step detects and resolves defects in real time, minimizing downtime and re-work.
Standard work output is a goal, yet not easy to achieve with diverse workers performing the same task. Despite standard operating procedures (SOPs), workers can get creative when carrying out tasks. Standard work output also requires a production cycle with predefined connections between customers and suppliers and a predefined pathway for product/services.
Through the HFPS culture, we've created a novel means to discuss a no-blame defective handoff by meeting with "'customers"- employees- to discuss processes not people. Each customer receives a derivative of specimens and a supplier processes the derivative and forwards it for the next analytical step. Customer teams are encouraged to document defects and communicate them during customer-supplier meetings. The. supplier team is expected to understand the customer's expectation and re-evaluate and improve their processes to satisfy that requirement.
For problem resolution, HFPS depends on a worker-driven, bottom-up approach rather than a management-driven top-down approach. Our approach is not slavish to data collection and statistical analysis; rather, it uses data for root cause analysis and to assess interventions. We have created a quality improvement organizational structure by aligning lab personnel to team leaders in work cells of three to five people that identify the nature and scope of defects, stimulate and guide discussion on possible solutions and implement/test a solution until it addresses the scope of a defect. This cooperative approach hinges on a no-blame-but-all-accountable sense of process ownership. The work expectation is to never pass along a defect.
HFPS promotes presentation of defect resolution experiences- successes and failures- by workers in monthly "Share the Gain" meetings (separated into anatomic and clinical pathology areas). The goal is to inform team members about lessons learned to sustain process improvements. These forums disseminate successful defect solutions and best practices. Further, the open discussion reinforces the no-blame-but-all-accountable culture, and it fosters group learning and recognition of team and individual achievements.
HFPS encourages use of simple and prominent visual cues in the workplace that identify predefined workstreams and guide movement of a specimen (or its derivative) as it moves through the laboratory. Wall posters and white boards at the workbench give instructions and record defects. Order inventory "kanbans" define, locate and re-order supplies. Multiple staff members are trained to pull a card at predetermined order points and pass it to inventory staff, reducing time lost from unavailable resources. Inventory stock is kept at minimum, and former stockroom space is reclaimed for other use.
HFPS reinforces its high standards, work rules and quality tools through its informatics infrastructure by introducing technologies
such as barcode-driven histology workflow and digital photography. We have also launched system-wide online availability of ISO-compliant manuals, policy/procedure documents and work guidelines that support our paperless operations. We have one common laboratory information system to map movement/processing of specimens and electronically deliver standardized reports across locations and physician offices. These reports use synoptic menus to eliminate omission of essential diagnostic information and facilitate clarity for the end user.
In one year, the surgical pathology division leveraged roughly 100 process improvements. The number of surgical cases with defects of any type (e.g., specimen receipt, specimen accessioning, grossing, histology slides, slide recuts, etc.) was reduced by 55% from one in three cases to one in eight. This dropped the following year to one in 40 (91% from baseline), achieving the goal of fixing defects in real time. Through process redesign and barcode-specified histology work processes, internal in-process misidentification defects were reduced by 62% overall with a 95% reduction in slide misidentification defects and 125% increase in technical throughput at the microtomy station.
After several years of our learning experiences in surgica1 pathology, we introduced HFPS principles to our clinical pathology core lab and have been rewarded by similar success. Baseline performance was 98% of all clinical lab test results reported by 6 a.m. the next morning in the electronic medical record. In less than a year, that turnaround time improved to 98% results reported by midnight of specimens received by 5 p.m. These labs were early adopters of technological automation, autovalidation and standardization, leaving greater opportunities for pre-analytic process improvements from specimen collection through transport, core lab receipt and in-lab delivery to testing platforms.
Lean is the continual driver of better quality, higher productivity and worker satisfaction. Positive outcomes have the potential for laboratories to gain recognition for their role in strengthening the economic future of the hospital systems they serve.Rita D'Angelo is manager; Quality Systems Division; Dr. Sharma is chief resident; and Dr. Zarbo is senior vice president and chair, Pathology and Laboratory Medicine, Henry Ford Health System, Detroit.