Boosting Radiology Efficiency: The Power of Non-Monetary Incentives
Below is an article I wrote several years ago about the unexpected success of a project to improve Radiologist productivity using non-monetary incentives. The ideas are still relevant today and nicely emphasize that money is rarely the best motivator for peak performance. The same concepts hold for good financial planning which should not be some pointless pursuit of “more money,” but is instead a method to craft a plan that helps you achieve the goals that matter to you.
Marian Jones had grumbling discomfort after dinner and then woke up at 3AM with excruciating right upper quadrant pain with nausea and vomiting. Her terrified husband rushed her to the Emergency Room (ER). A brief assessment by an experienced nurse ensured that Marian was seen promptly by an ER doctor. After listening to her story and examining her, the physician thought she might have gallstones or a right kidney stone or perhaps an ulcer with bowel perforation. The doctor ordered a Computed Tomography (CT) scan as the single best test to distinguish among those three possibilities. Ten minutes after the test was completed, the radiologist called to say that Marian had a perforated duodenum. A surgeon was consulted and the patient was rushed to the operating room and subsequently had an uneventful complete recovery.
Variations on this story are a daily occurrence in hospitals everywhere in the western world. No incentive is required to ensure the rapid reporting of critical cases which have such an immediate effect on patient care. However, many other studies, particularly those in medical imaging, are more routine, fail to attract attention, and may not be reported for some time. This is obviously not ideal for patient care, but it also causes duplication of work as anxious patients prod their physicians to call for results, generating phone calls to staff and radiologists that could have been avoided with more timely reporting. Faced with this problem of unreported cases, over ten years ago we introduced a quota system using a leaderboard. It transformed our practice in ways we never imagined.
Only a few radiologists have the skills and interest to excel at the administrative responsibilities, but most very much enjoy solving complex diagnostic or therapeutic puzzles. However, radiologists must also give due attention to more routine studies and our quota system was designed to ensure more efficient reporting of all cases. The concept was simple:
Knowing how many computed tomography (CT) and magnetic resonance (MR) scans as well as plain radiographs (such as a chest radiograph or views of a bone or joint) an efficient colleague could report in an average day, we assigned about two thirds of that number to each radiologist as their minimum expected work for a day. (The number varies depending on each radiologist’s assigned tasks for the day)
The radiologist reporting ultrasound at one hospital would be expected to use ultrasound to guide biopsies of the liver or kidneys or various masses anywhere in the body. The same radiologist might also place catheters in the chest or abdomen to drain fluid collections. Adding these procedures to the reporting of other ultrasound studies of the abdomen or pelvis or obstetrics would mean that the radiologist would only be able to report a limited number of CT, MRI and plain radiographs on that day. A different radiologist assigned to cover CT guided procedures would be expected to report a much greater number of CT scans than their colleague on ultrasound. Similarly, the radiologists working in each of the different fields such as breast imaging or angiography or nuclear medicine would have variable quota expectations based on the anticipated volume of work in their specific daily assignment.
The quota is displayed on a webpage open on every radiologist's desktop. Beside the radiologist's name is posted in large yellow font the number of CT and MRI scans as well as plain radiographs they are expected to report that day in addition to whatever other tasks might be assigned. As each report is completed, the numbers decrease almost in real time until they drop to zero; then the color turns white and starts to rise as the radiologist exceeds their quota. Every radiologist is thus fully aware of what each colleague is doing throughout the day, fostering a sense of competition driven by the leaderboard system.
Before we introduced the quota system, we noticed that every afternoon a large number of simple routine chest radiographs or very complex CT scan follow-ups of patients on treatment for cancer would sit unreported. While these examples are studies of great importance to the patient and their referring doctor, an average radiologist will see both in such great volume and limited variability that other activities tend to receive higher priority. Two days after our new system started, at 10:00 o'clock in the morning, I could not find a single plain radiograph or CT to report. Fellow radiologists were so eager to obliterate the yellow color of assigned work that they had come in early, ignored routine email and other distractions and focused on the work until every available study on the list was reported. This dramatic improvement in radiologist productivity was achieved through a system that relied entirely on non-monetary incentives.
“Money is rarely the best motivator for peak performance”
I expected that this initial enthusiasm would fade away, but more than ten years later, the quota system continues to drive efficient work every day. We were concerned that speed would prevail over accuracy and have thus continually tweaked the system to assign more points to tougher cases and fewer points to easy ones. We use structured reports with checklists based on airline pilot practices to ensure important details are not overlooked. Regular peer review is also part of our practice as a quality improvement initiative to raise the standard of all reports. Our peer-review program selects studies reported by radiologists working in one region and has them anonymously reviewed by someone from a different region providing valuable measurable feedback on discrepancies. Regular attendance at rounds to discuss reports with surgical and medical colleagues provides even more valuable data on accuracy and outcomes of importance for patients.
Our most efficient colleagues (known as rabbits) before the introduction of the system continued to be the most productive afterwards. The least efficient (turtles) continued to report at a slower pace, but their average numbers increased as no individual wished to be publicly seen as a free-loader. Some come in early and/or stay longer to make sure they achieve at least minimum expectations. The greatest increase in production came from those who were neither rabbits nor turtles, but somewhere in between. We think the marked gain in efficiency in this group came from avoiding distractions like email or pointless chatter. The best feature has been improved morale as our rabbits no longer feel they are the only ones working hard, but instead most of us are confident that all are doing the best they can.
Although different procedures and tests in medical imaging are variably compensated, we pool these funds and assign payment based on time. A rabbit who does more work in a well-compensated field is thus not paid more than a turtle toiling in a poorly paid portion of the practice. We have found that this does not cause friction as long as the turtles are seen to be working hard. Conversely, the rabbits seem to be happy to generate a very large number on a given day in return for the public recognition of their striking efficiency as there is no monetary incentive.
Quality checks have not shown a significant error difference between turtles and rabbits. Colleagues considered average as opposed to brilliant are not more likely to be fast or slow, suggesting no direct correlation between brain power and speed. Instead, we think the predictor of high efficiency is something we call “processor speed.” We do not know how this works, but believe it is a reflection of confidence and rapid decision-making as might be observed in those who are speedy at cards or ordering dinner.
“More than ten years later, the quota system continues to drive efficient work every day.”
As our practice has evolved, we continuously modify the quota to ensure optimum results in terms of timely reporting of clinically useful reports. The tool is also helpful in workforce planning as we get early and very clear indications of areas where we need new associates and other areas where less emphasis is required. Other medical imaging departments have adopted our system, often to improve efficiency, but mostly to foster a more collegial department.
The daily posted results have had other surprising effects. Blessed with good geography and an interesting practice, we have always been able to attract first-rate radiologists, but we will not consider hiring any new candidate until that individual has worked with us. Despite glowing references, only when a new potential associate sits down at one of our desks do we uncover those who are fast but inaccurate (sloppy rabbits) or slow and indecisive (dithering turtles). The former are more immediately dangerous because they overlook important details; the latter are annoyingly unhelpful to referring doctors since they seem incapable of producing a clear opinion about the most likely diagnosis and the next most logical steps but instead waffle and generate unnecessary extra tests. You might think that both the sloppy and the indecisive radiologists would have been weeded out after at least twelve years of education following high school, but that has not been our observation.
An example may make these descriptions more obvious. A patient seen by an emergency-room doctor is sent for a medical imaging study, a CT scan in this case, to look for kidney stones to explain right back pain. Our visiting fast reader reports no stones or kidney obstruction and moves on to the next case. The locum slow reader reviewing the same study might issue a long report also noting no stones, but commenting on a few large lymph nodes, a slightly big spleen, an adrenal nodule and some abnormalities of the bones. This is then followed by a long rambling discussion with unhelpful phrases like “cannot exclude” or “clinical correlation recommended” accompanied by a host of recommendations for other tests with no clear conclusion. While both correctly conclude that kidney stones are not the problem, neither has significantly helped to improve timely reporting and patient care.
On the other hand, our ideal new colleague would describe the relevant findings giving appropriate weight to important as opposed to less critical data, use older studies or clinical data to make sense of the findings, and then conclude with the following useful comments:
No kidney stones or obstruction
Large spleen and nodes consistent with previously diagnosed chronic lymphocytic lymphoma.
Incidental adrenal nodule requiring no further follow-up.
Probable discitis/osteomyelitis likely accounting for the back pain
This is a concise, clinically useful report telling the referring doctor that kidney stones are not the problem, but there is another unexpected diagnosis which will require urgent treatment. Obviously, this radiologist has the skills of a desirable new associate. They are unlikely to be as efficient as our more experienced rabbits (referring to fast and productive radiologists), but will improve over time and make a positive impact on patient care. We believe that “first you get good, then you get fast.”
Our online recording of volumes continues at night and on the weekends, once again with unexpected useful results. The demand for all imaging, particularly CT scans, has grown after hours driven by both positive and negative factors. On the good side, CT has contributed to a dramatic reduction in exploratory surgery over the past few decades. However, this increased reliance on imaging might be influenced by the obesity epidemic, making it harder to assess patients with a physical exam.
Here's where data-driven workforce planning comes in. By objectively tracking reported volumes by after-hours radiologists, we can rapidly adjust staffing levels to meet demands while maintaining quality. This non-monetary incentive ensures optimal resource allocation and avoids unnecessary overtime.
All imaging tests requiring reporting are placed on a large worklist and color-coded based on urgency:
Emergency (minutes)
Urgent (within an hour)
Important (that day)
Routine (within days)
This prioritization system, similar to a basic quota system, ensures critical cases like those in the emergency room (ER) are reported first. Patients in the ER will often have a typed report (we use voice recognition) of their imaging tests available to their doctor before they return from radiology to the ER.
Radiologist productivity can be further optimized by assigning a dedicated radiologist to the "sweeper" role. This radiologist focuses solely on reporting older cases in the routine list, functioning similarly to a leaderboard system where they "catch up" on outstanding reports. This allows for uninterrupted focus on older and complex studies, freeing up other radiologists to handle daily tasks like consultations, patient care, and administrative duties
We believe that a variation of our radiology department efficiency system based on a leaderboard system and quota might work well in other fields. As an example, a software company ran a trial of our approach to focus energy on fixing bugs that were known to exist, but were considered tedious to address.
The bugs were collected into a worklist and assigned points based on a combination of their importance (critical vs. minor) and the difficulty anticipated in fixing them. Every person's name was posted on a board and moved up or down depending on how many points they accumulated, similar to a leaderboard in our radiology department.
The results were dramatic as each person constantly checked how they were doing compared to others, fostering a sense of healthy competition and increased productivity. Productivity soared as names moved up and down the leaderboard, with a focus on solving important problems rather than just the quickest fixes. Bugs that resisted solutions were assigned more points, further incentivizing collaboration.
Here, again, the motivation was not monetary but public recognition or the admiration of peers.
Traditional quota systems often have negative connotations, with a focus on quantity over quality. A major part of the success of our system, and the software company's adaptation, is that it is not externally imposed. Those whose names are posted on the board have a say in how the system works, fostering a sense of ownership and improved morale. The system can also dynamically respond to changing needs, such as increasing MR requirements and decreasing CT numbers. We can encourage other activities such as attending case reviews with colleagues in other departments by reducing the quota on a day a radiologist attends surgical or medical rounds.
The information about individual performance is displayed only for the radiologists working that day, similar to the software company's approach. This transparency fosters trust and avoids micromanagement by managers or external parties. The limited objectives that inspired the creation of the system over a decade ago were easily met and now we consider the quota system with a focus on timely reporting and quality to be an essential part of our practice, transferable to improve efficiency in software development and potentially other fields.
KEY CONCEPTS RELEVANT TO FINANCIAL PLANNING
Money is rarely the most important motivator - In the case above, the personal satisfaction of a job well done and earning the respect of your peers is more meaningful than a number. Aligning your financial goals with what truly matters to you is a more satisfying and realistic way to commit to a plan.
Track your Progress - In this case, a leaderboard provided immediate feedback to track each radiologist’s output. In financial terms, setting clearly articulated goals and documenting your results is a key to success.