In 2009, a stroke victim at a Los Angeles medical center started losing his hair following a CT brain perfusion scan. After some confusion, doctors determined he had been subject to a radiation overdose—a serious accident that might lead to a lifetime of increased risk for cancer.
The center discovered that the error was made a year before when it had reconfigured a scanner to improve doctors’ ability to see blood flow in the brain. More than 200 patients had suffered the same fate, receiving up to eight times the normal dose of radiation.
Publications nationwide reported on the accidents, including a 20-article investigative series in the New York Times that discovered similar cases at other hospitals. The US Food and Drug Administration (FDA) started investigations of the CT devices, dozens of patients joined in a class-action lawsuit, and Congress followed up with a hearing on medical radiation safety.
"The demand shock caused by an accident could actually be good news for companies."
“There was suddenly a huge spike in the public’s attention on medical radiation risk,” says Hong Luo, James Dinan and Elizabeth Miller Associate Professor of Business Administration in the Strategy Unit of Harvard Business School.
The case caught the attention of Luo and Alberto Galasso, a professor at the University of Toronto’s Rotman School of Management, who have together looked at the impact of changes in liability laws on medical device innovation.
The questions they wanted to answer: Would negative publicity from the accident affect future innovation in that segment of the medical device industry? Would people’s perception of safety issues reduce demand for products using radiation?
Do accidents chill innovation?
Conventional wisdom is that accidents can chill innovation by creating uncertainty and convince risk-averse companies to pull back on R&D. When a bicyclist was killed by an Uber autonomous car in Arizona in 2018, for example, the company suspended the program—even though the crash was later judged to be human error. Boeing’s 737 Max has been grounded ever since a malfunction of the plane’s autopilot feature led to the deaths of 348 people last March.
“Many safety-related innovations can only be profitably developed if there is a genuine demand for it,” says Luo. “If there is no demand for it, it’s difficult to recoup your R&D investment.”
In a forthcoming paper in Management Science, Risk-Mitigating Technologies: The Case of Radiation Diagnostic Devices, based on the CT scan case, Luo and Galasso discover that the accidents, while tragic, ultimately increased innovation rather than chilled it, leading to development of both incremental and radical innovations that made the devices safer and better.
How safe a product is, unlike other features, may be difficult to judge before an event occurs. Often, the producer, the consumer, the regulator, and the experts don’t truly know how safe a product is until it fails—and increasing technological complexity only magnifies the challenge.
Many companies, meanwhile, are apt to see investment in safety as the cost of fulfilling regulatory requirements or protecting against lawsuits, rather than as an opportunity to include features for which consumers would be willing to pay.
Luo and Galasso wanted to see if the sustained media attention would have a positive effect on pushing safety-related innovation in diagnostic devices using radiation. To judge that, they examined new patent applications filed after the news of the accidents broke, as well as notifications of new products submitted to the FDA.
Comparing the new patent applications for safety-related features to other features of radiation diagnostic devices, they found an increase of over 100 percent in the five years after the shock relative to before. Comparing FDA product applications in radiation diagnostic devices to medical devices that didn’t use radiation, they also found a 30 percent increase.
“We expected to see something, but we had no idea of the magnitude,” says Galasso.
Moreover, when they began examining detailed FDA filings of CT scanners, they discovered that the innovation was of two types:
- Incremental changes. The first type, which occurred immediately following the accidents, consisted of incremental changes—that is, software fixes targeted at preventing accidental over-radiation. These included dose displays, alerts, and notifications if radiation doses exceeded safe thresholds, and protection mechanisms that could only be overridden by passcodes.
- Radical redesign. A second type of innovation involved a much more radical redesign of the products, using a different technology that could reduce the dosage of radiation needed by 20 to 90 percent, depending on specifications. In talking with industry experts and reading the literature, Luo and Galasso found that companies had known for a long time about the potential for dose-reducing technology, but had not developed it due to lack of demand.
“It seems that the industry didn’t push it partly because, before these accidents, doctors were more concerned about image quality and speed for better and faster diagnosis, whereas radiation concerns, though important, were secondary,” Galasso says.
In addition, he says, the alternative technology was not without its downsides: it was computationally intensive, so substantially slower; image quality was inferior, at least initially for certain applications; and “our understanding is that it also produced unfamiliar image textures that required retraining of radiologists.”
Users will pay you for safety
These accidents and the public attention on radiation led to a substantial demand shift for additional safety features, prompting CT scanner producers to quickly incorporate, commercialize, and continuously improve alternative technology.
Moreover, it was the largest industry incumbents that benefitted most. That, too, runs against conventional thinking that smaller companies can more nimbly pursue innovation.
“In this case, the bigger players were able to integrate this radical technology into their existing machines without being disruptive,” says Galasso. “It didn’t require a substantial redesign of the rest of the products, and large companies can more easily spread the high fixed costs of development and commercialization over their high sales volumes.”
Those larger companies also may have benefitted from brand reputation, which could see customers flock to trusted names in a time of uncertainty over safety.
The researchers speculate that even though not all accidents will lead to sustained innovations, this case isn’t unique. Accidents could play a role in spurring innovation for incumbents, provided that they lead to substantially increased demand for safety and that these new safety features can be developed relatively quickly, and improvement in safety can be demonstrated before consumers leave for safer substitutes.
“The demand shock caused by an accident could actually be good news for companies,” says Luo. “They drive up demand for a product feature that often feels remote in consumers’ minds, making them more willing to experiment with alternative technologies that might be initially inferior to other quality dimensions. This provides an additional dimension along which firms can innovate, differentiate, and compete.”
While companies are used to seeing such accidents as cost-drivers, they may just as well see them as opportunities to improve their products along with safety for their customers.
About the Author
Michael Blanding is a writer based in Boston.
[Image: JohnnyGreig]
Related Reading
- The Hard Work of Failure Analysis
- At Booking.com, Innovation Means Constant Failure
- Crowdsourcing Is Helping Hollywood Reduce the Risk of Movie-Making
How does your company learn from failure?
Share your insights below.