After two years at the top, the U. slipped to a distant second in AUTM's 2011 rankings for startups, behind MIT, which reported 25 startups.
And that's fine for U. technology commercialization director Bryan Ritchie. The U.'s dropping startup numbers remain three or four times larger than most similar-size universities.
"If we have 25 or 17 startups [a year], I'm OK. If we have 10, I'm not OK," Ritchie said. "Starting companies is in our DNA. We know how to do it. We need to do it better. I want to drive value into companies and make them viable."
The university systems of Illinois, California and Texas reported more startups that the U., but did not break out their numbers by institution. Brigham Young University and Utah State University, which have much smaller research footprints than the U., logged four and three startups, respectively.
In the AUTM survey, 186 responding institutions reported winning 4,700 U.S. patents in 2011, maintaining more than 38,000 active licenses and earning $2.5 billion off these licenses.
"These data are a testament to the large and critical role universities play in the innovation economy," AUTM vice president Richard Kordal said in a press statement.
Assessing the impact • For the 2011 survey, AUTM wanted a more meaningful picture of startups' economic impact, so it asked for full-time employment numbers and product sales.
The U. was among the majority that did not provide this information, saying it would be too difficult to gather. Startups might not willingly provide their employment numbers, and such data would reflect only a snapshot in time, Ritchie explains.
In a March 2011 report, however, the technology commercialization office reported 98 U. startups employed nearly 6,000 in Utah in 2009, reflecting a payroll of almost $360 million and contributing $1.2 billion to the state's economy. That's an average of 60 employees per U. startup still operating.
Ritchie, who arrived at the U. months after that report was released, said the employment figure was based on estimates.
"It's a pretty solid number. You have to remember that 90 percent of the employment happens in 5 to 10 percent of the companies. Perhaps instead of averages, we should think in terms of median, which is probably around 10," he said.
The latest AUTM data is a year old and the U. has already released its numbers for fiscal year 2012, indicating a further slowdown. Ritchie's office reported 17 startups, 275 invention disclosures and 80 licenses executed for the fiscal year that ended June 30.
Questioning the numbers • Since 2005, the U. has spun off about 150 companies through a pipeline engineered by Ritchie's predecessor, Brian Cummings, and then-vice president for technology venture development Jack Brittain.
But some of this success appears to have been a mirage. Alongside Xandem, Domain Surgical and other tangible firms with real products in the works, the U. has counted companies that had no leadership, offices, employees or outside investment or grants.
Several exist only on paper, still listing associate technology commercialization director Zach Miles as a principal and his office as headquarters. The U.'s goal is to land permanent management and small-business grants for these firms, but many remain inactive shells.
Meanwhile, the nation's perennial commercialization leader and the U.'s main rival, MIT, only lists companies as university startups once they have garnered at least $500,000 in outside funding.
In other instances, listed U. companies have been based on intellectual property that was developed elsewhere, then "spun in" to the U. For example, 2011 startup U.S. BioRemediation Inc., launched in 2009, is based on patents licensed from Montana Tech, whose scientists developed a processes for using extremophilic microbes to combat industrial pollution.
The research began on Butte's notorious Berkeley Pit, a former copper mine that has become a toxic lake on the edge of Montana's most historic city.
The company, based in Salt Lake City, is adding to the intellectual property by evaluating microbes that adhere to toxic metals in highly acidic environments, then testing them on acid mine discharge. The technology could speed up and reduce the costs of cleaning up this legacy of hard-rock mining, according to the firm's website.
Quality over quantity • In recent months, U. officials have signaled a retooling of technology commercialization to emphasize "quality over quantity." Officials want to see more successful start-ups like Myriad Genetics and BioFire, formerly Idaho Technology.
These Research Park-based firms, which were launched in the early 1990s, employ hundreds and illustrate how technology commercialization can drive economic development.
"We won't throw out the baby with the bath water," Ritchie said. "Starting companies is an important prerequisite, and it's not easy. Looking at the numbers, most places don't know how to do it."
One 2012 startup Ritchie is particularly excited about is Vaporsens, which uses nanofibers that act like "a dog's nose" to detect explosives. The technology is the brainchild of Ling Zang, a professor of materials science and engineering who is part of the nanotechnology team of USTAR, a statewide initiative to promote university technology.
Xandem's Patwari conceived his sensing technology by arranging several transceivers units that both transmit and receive radio signals around the outside of a room. The radio waves leave "shadows" as they pass through moving objects and are picked up by the various units.
His team developed computer algorithms to process the strength of these signals in a way that enables an unseen person's movements to be tracked on a computer screen. Xandem now sells a burglar alarm based on the technology, but the firm is interested in developing products for new uses in an assisted-living environment. The idea would be to continually track an elderly resident's movement, so help can be sent if something is amiss.
"You want to make sure they don't do anything really weird, like sleeping on the floor or staying in bed for a long time," Patwari said. But care providers also want to respect the person's dignity and privacy, which is why cameras are not a good solution.
The U. is helping Xandem secure patents on a wireless breath-monitor application, which has been the subject of further U. research with promising results.
Last year, Patwari and Wilson's team showed a network of 20 wireless transceivers placed around a hospital bed could identify a patient's breathing rate to within two-fifths of a breath per minute, based on 30 seconds of data. But Wilson cautioned that much work needs to be done before a product is developed and cleared by federal regulators for the medical market.
"A patent is a piece of paper that doesn't really do something," said Wilson, who earned his doctorate last year and devotes himself full-time to Xandem along with another employee. "What does something is a piece of plastic. The plastic represents an embodied product and that's the world needs."
University of Utah technology commercialization, 2011
Start-up companies: 19
Patent applications: 125
Patents issued: 47
Licenses executed: 81
Pre-patent invention disclosures: 237
Licensing income from royalties: $11 million
Gross income: $36 million
Source: Association of University Technology Managers, FY 2011 survey.