Going lean so income will grow

A practice takes seven years to alter its nonefficient ways. Result? More patients and revenue, fewer complaints.

About seven years ago ophthalmologists and administrators at the Key-Whitman Eye Center in Dallas realized that the trend toward lower reimbursements for medical procedures was the new normal. Shrinking reimbursements would mean less revenue coming into our practice, while new federal regulations requiring use of electronic medical record (EMR) technology would mean revenue going out. If Key-Whitman didn’t comply, we would be hit with fee schedule penalties.

Our response was to create a business plan to allow us to maintain revenues and address the emerging challenges, all while working with existing resource levels that would not compromise patient care.


We developed a technology-based plan and implemented it in stages over the past seven years. Starting with in-house studies and using a modified, off-the-shelf software program, our practice today serves more patients faster and more efficiently using state-of-the-art programs and apps.

There were the expected twists and turns during the journey, such as EMR implementation and the subsequent impact. And the need for change required everyone — physicians, technicians, staff and patients — to be adaptable and flexible. But seven years later we have improved our processes, upheld our quality of care, decreased our costs, and improved revenue.

Adding technology to our practice hasn’t been cheap. But it has helped us preserve our revenue stream and more. Let’s say a doctor who averaged 40 patients added five new appointments at a per-patient average of $200 or $1,000 a day. Now, multiple that by 200 days. Efficiency drove revenues. No jobs were lost because we added patients to compensate for declining reimbursements.

Sometime around 2010 it was clear that reimbursement rates were never going to rebound. If we wanted to maintain our revenue stream we had to increase patient volume. More patients meant more revenue. The trick was doing it with the current number of exam rooms, technicians, doctors — and purchasing an EMR system.

We already had a serious wait-time issue, replete with patient complaints. We also knew wait time was the number one patient-satisfaction issue for large multi-doctor practices providing a wide variety of services. If we wanted to maintain our revenue stream, let alone grow it, we had to significantly reduce the time patients spent in the clinic.

The riddle we had to solve was how to increase patient volume while reducing wait time and living with resource constraints, all without running our staff into the ground.


My MBA work was focused on operations management, focusing on operational research including queuing models, simulation and optimization. I knew what needed to be done. Still, I found the number of variables to analyze — multiple doctors, various medical services, limited space — daunting. I decided to use the Lean Six Sigma methodology.

Most people have heard of Six Sigma, the set of management techniques used to improve business processes in manufacturing. Lean Six Sigma is sort of the service corollary. It emphasizes teamwork to improve performance and employs a methodical, analytical approach to remove waste, streamline processes, reduce wait times and speed the processing of patients, all of which we wanted to do.

Simple Sx Center design to depict approximately 28 patients during a 7a.m. - 11:30a.m Sx day. The red circled areas highlight the program’s ability to provide accurate wait and patient cycle times, and procedure indicators. This is a real-life example with proven and measured accuracy within the facility.

The first thing we had to do was step up our technological game, analyzing data. We started by studying the time it took to perform certain office functions (your typical “time and motion studies”). It was slow, tedious and necessary to replicate when changes were made. We switched to a new approach that captured ongoing data to help modify our workflow. Using simulation software, the program tested various alternatives before implementing any new idea that reduced patient wait time. Still, we were just scratching the surface.

In 2012 we invested in a technology that gave us constant but passive, real-time information. Some EMR systems could do that, but the data had to be entered into a computer in real time and it had to be accurately entered to be usable. We found major limitations to useful and accurate data when using an EMR system.

We opted for installing radio-frequency identification (RFID) and infrared (IR) sensors (Versus). RFID uses electromagnetic fields to track tags while IR sensors locate the tags. With some sophisticated manipulation you can program the information you want tracked to be displayed on a screen with colors and the names of people or things tracked.

Once the RFID is set up, physicians, staff and patients just have to pin on a badge. Sensors in the ceiling pick up signals located in each badge identifying the wearer’s location and how long the person has been there. The accumulated data, stored in an automated system, generate reliable patient wait times, technician work-up times, doctor-patient times, and cycle times by day and time of day.

RFID works continuously, collecting thousands of highly accurate and reliable data samples measuring components of workflow. Using industrial engineering techniques, the data files can be downloaded and analyzed in concert with the creation of an elaborate flowchart. The chart displays every major step — from arrival to departure — in a patient’s journey through our system. The information includes how long a patient waited to be seen, time spent with a technician, and the time staff spent performing a service like a cataract work-up, consultation or comprehensive eye evaluation. That data can then be applied to the workflow chart to simulate various alternatives to both maximize doctor utilization and minimize patient wait times.

Some employees were initially apprehensive about wearing a badge. They worried the information gleaned would be used to evaluate their work. We made it clear we were only interested in finding and fixing bottlenecks disrupting patient flow.

Even though most ophthalmologists are used to adopting and incorporating new technologies, a few were skeptical until we explained our purpose and process. Patients, on the other hand, rarely balked at wearing a badge, especially when told we were working to reduce their time in the office.

Before we adopted RFID monitoring and simulation software, patients spent several hours completing a cataract evaluation from start to finish. Today they are through the process in about an hour. Our standard cataract surgical times for patients are often less than an hour between arrival and departure time.


A computer-automated design software program (SimCAD, Creative Soft) helped us simulate various workflows and to balance resources like staffing levels, exam room and equipment use. (See “Using Latest Technology to Improve Productivity.”)

We found its greatest valuable in template building patient schedules. Our various doctors average between four and eight minutes with a patient. So, if a doctor sees 50 patients a day, that four-minute per-patient time difference adds up to 200 minutes, almost half a day. Establishing a realistic schedule for each doctor and location is critical and sensitive.

To squeeze the most from the flowchart, we use Monte Carlo simulation techniques. The simulation generates a statistical formula that produces a distribution of possible outcome values by running iterations and simulations of “real life” processes for patient times, evaluation lengths and time with the physician. Our goal, of course, is to replicate the typical experience during the day, not theoretical times.

Customizing practice schedules has increased physician utilization while minimizing patient wait time. Because ophthalmology protocols, equipment and procedures change, we analyze the data several times a year to see if the process can be improved.

Here’s another way to think about it. The airline industry uses the SimCAD model to optimize schedules and revenue stream by planning contingencies for unexpected events. The computerized plan accounts for things like plane size, arrival and departure times, routes and traffic patterns, maintenance, even the average time it takes to load and unload a flight.

But the system can’t anticipate things like changing weather, runway backups, crew sickness or major mechanical issues. That responsibility falls to air traffic controllers who, given the variables confronting them, have to adjust “on the fly” to keep traffic flowing.

Like the air traffic controller, we are constantly adjusting our schedule in real time. If someone calls in sick, a traffic jam delays patient arrivals, or a doctor has an emergency, we have a supervisor adjusting the schedule throughout the day to optimize patient flow so we can reach our goal for that day.


Our technology continues to evolve with the addition of an app called DocAlert (Epocrates). The app uses an indicator light system to sequence workflow for doctors and technicians throughout the day. It is especially helpful at our largest (21 exam rooms) office whose three doctors see 120 to 150 patients a day.

The wireless app, which replaced an older light indicator system that directed physicians to their next stop, connects to iPads placed outside each examination room. The app can be customized and modified, and is viewable on mobile phones. The system also connects with a 55-inch monitor in the doctor’s area.

While technology has clearly helped our practice, there was a side benefit we didn’t anticipate — a greatly improved work environment.

We found that when the work flow moves at a comfortable pace, the staff is happier. Everyone, including physicians, has time for lunch. Even the patients are happier. Well, at least they don’t complain as much. OM

About the Author