The University of Texas MD Anderson Cancer Center successfully employed a variety of “nudges” to get surgeons to consider costs when deciding which operating-room supplies to use. The behavioral economics techniques included a dashboard that showed the costs of the discretionary items (e.g., disposable implants, instruments, sutures) to be used in upcoming procedures; comparisons of the supply costs of each surgeon within individual departments; posters placed in dictation rooms and operating suites that displayed low-cost alternatives for common equipment such as staplers, sutures, and thermal cutting devices; and monthly reports that showed individual surgeons how their average costs compared with those of their peers and the overall department for a given type of procedure.
An estimated 50 million inpatient surgical procedures occur each year in the United States, at a cost of $175 billion. Given these striking numbers and the fact that over 40% of costs of the acute care episodes of surgical patients are related to the resource-intensive operating room, it comes as no surprise that the OR has become a locus of cost-containment initiatives for health systems. As a starting point, systems such as the University of California, San Francisco have experimented with deploying cost-transparency tools and financial incentives in the OR. University of Utah Health has gone one step further by bringing surgeons together to review spending data and develop standardized processes for addressing variation in OR supply costs.
These initiatives have yielded savings without negatively affecting care quality. Yet to realize the value of surgical cost-transparency at scale, health systems need to make cost-transparency a daily habit, rather than treating it as an extra action item for surgeons to add to their already busy To Do lists.
Of course, durably changing physician behavior is easier said than done. To accomplish this goal, we believe that hospital leaders need to go beyond just showing surgeons personalized cost information or offering them financial incentives. In addition, health systems should redesign the environmental context (e.g., the OR) where these decisions are taking place and account for the psychological makeup of surgeons (e.g., their innate competitiveness and perfectionism). Examples of ways to encourage surgeons to make more cost-efficient choices include real-time reminders about spending, publicly reporting surgeons’ cost outcomes, and explicitly treating health care production costs as a key performance metric.
At the University of Texas MD Anderson Cancer Center, one of the leading cancer centers in the United States, we used behavioral economics to implement a new approach to OR cost-transparency. Our “Know Your Costs” (KYC) campaign introduced a combination of behavioral “nudges” into surgical workflows, saving nearly $1 million in supply costs during a one-year pilot conducted throughout 2018. Beginning in January 2019, the program was scaled across the entire Division of Surgery, achieving savings of nearly 13% in its first full year. In both cases, there were no adverse effects on the quality of care (surgical complication rates).
KYC was grounded in the principles of behavioral economics, which seeks to create predictable changes in human behavior by selectively introducing information and modifying the presentation of options (“choice architecture”) to steer individuals to higher-value choices. Our thesis was that addressing surgeons’ tendencies to make decisions without regard to their costs (e.g., selecting tools on the basis of the individual’s not-always-accurate subjective view of their quality and ignoring their cost) could encourage better accountability for the costs of care.
A vital element of this strategy was the preference card: an itemized list of instruments, equipment, and supplies that physicians deem requisite to conduct a surgical procedure. Preference cards are ubiquitous across U.S. hospitals, customarily cost-blind, and highly sensitive to physicians’ subjective preferences. Studies show that surgeon preferences contribute to wide variations in costs for the same procedure within the same institution and most surgeons are unaware of the costs associated with commonly used supplies. More important, no relationship has been established between the higher cost of many supplies and outcomes.
During the design of the KYC initiative, a helpful analogy that we kept in mind was the grocery store. For example, surgeons fill out preference cards for tools prior to operations just like shoppers make lists before entering the store. But surgeons normally itemize in a cost-blind manner, even though higher-cost instruments don’t often have added clinical benefits, contributing to high OR expenses. In response, we worked with surgeons in the department to proactively select lower-cost alternatives for intraoperative supplies (those occurring during the course of a surgical operation). Just like shoppers might change purchasing decisions based on weekly sales or store offers, this intervention with surgeons was akin to clinical “coupon cutting” to reduce procedure expenditures based on current supply prices. Our rationale was that if supply selection is a choice with a downstream impact on costs, then perhaps changing the menu of choices for surgeons could create natural, downward pressure on health care spending.
The typical shopper is able to make real-time decisions about purchases because grocery stores typically display substitute items (e.g., organic versus non-organic produce) side by side with prices. However, the hospital setting is often insulated from such informational interventions — after all, tools in the OR don’t carry price tags.
Beginning in January 2018, 26 surgeons in three of MD Anderson’s surgical departments began receiving reports from an “intra-operative supply cost” (ISC) comparison tool embedded within the electronic health record. The ISC was a dashboard that presented individual surgeons with the costs of the discretionary elements in their preference card for an upcoming procedure (e.g., disposable implants, instruments, sutures). This helped filter out the cost inputs that surgeons did not have control over (e.g., salaries, facility fees) and focused their attention on the inputs over which they did have control.
To prod surgeons with higher spending to change their behavior to match that of their lower-spending peers, we generated comparisons of the supply costs of each surgeon within individual departments. These comparisons were done by procedure type (e.g., prostatectomy, thyroidectomy), and we presented the mean cost for the surgery with a line-item list of the disposable supplies used in the procedure. Because the comparisons were available to any physician authorized to use the electronic-health-record system, each surgeon could see others’ information. We also put posters in dictation rooms and operating suites that displayed low-cost alternatives for common equipment such as staplers, sutures, and thermal cutting devices.
The best shoppers check their expenses periodically to make sure they’re not running over their budget. Surgeons, however, aren’t given opportunities to reflect on their own expenditures over time or see how they compare to their colleagues on cost. To address this, in addition to making the cost comparisons available on the ISC dashboard, we emailed each surgeon a personalized monthly report. These monthly reports showed individual surgeons how their average costs compared with those of their peers and the overall department for a given type of procedure.
Social ranking — showing doctors how they compare to their peers in terms of clinical quality and outcomes — has been shown to be a powerful lever for driving behavioral change among surgeons. This effect is most striking when the peers are known to a surgeon or in close proximity. We optimized the design of these comparative reports by using colors to highlight surgeons with the lowest supply costs (green) and highest costs (red) for a given procedure relative to historical spending (defined as the calendar year preceding the pilot). This simple color scheme provided surgeons with an easy-to-understand “nudge” about costs in the same way that menus might be color-coded to help nudge consumers to purchase healthier food options.
The overarching goal of the KYC program was for each participating department to lower its mean ISC (cost of supplies used) per selected procedure by 5% relative to historical spending. The savings target of 5% was deliberate because there is ample evidence that people (a) are discouraged by benchmarks that are unattainable and (b) work harder as they approach a performance metric. Our collective efforts to convert a group of 26 surgeons into a cohort of “smart shoppers” paid off, and the initiative was then scaled to encompass all 159 surgeons in the Division of Surgery’s 10 departments.
We believe this pilot project provides a path forward for incorporating behavioral economics principles into surgical care. To spur others to adopt a similar approach, professional associations such as the American College of Surgeons and the American Hospital Association should work with electronic-health-records vendors to develop these kinds of simple plug-ins and packages to provide cost comparisons and reminders to physicians. Physician leaders should consider initiating “accountability & affordability” tracks to their hospitals’ “Morbidity & Mortality” conferences in order to encourage all clinicians to be more conscious of costs. And residency programs should include cost-transparency initiatives as part of the training of new physicians.
It is indisputable that health care is full of hidden inefficiencies, and within this lies the value proposition of behavioral economics: It offers a toolkit to transform doctors into smart shoppers who consciously try to minimize costs but not at the expense of quality and safety.