Campus & Community

When problem-solving is a problem:

7 min read

HBS’s Anita Tucker’s troubleshooting addresses root causes

HBS doctoral candidate Anita Tucker: ‘What I saw in my research were all those little things that happen that create the atmosphere that allows all those errors and accidents to occur.’ Staff photo by Rose Lincoln

Observing hospital nurses for 239 hours, Anita Tucker watched them solve problems all day, every day.

If an ill-timed delivery left them short of linens, they found a way to borrow from another unit.

Faced with unclear medication instructions for a new patient, they would track down the physician and confirm directions.

When infant security tags fell off newborns’ ankles, nurses combed the nursery to find them – again and again.

Such initiative taking by empowered frontline workers is a good thing, right?

Well, yes. And no, said Tucker, a doctoral student at Harvard Business School (HBS).

By quickly and effectively solving the many small problems that came up throughout their shifts, the nurses Tucker observed continued to provide excellent care to their patients.

The bad news, said Tucker, is that the root causes of the problems – from annoyances like supply shortages to potential dangers like unclear instructions – went unaddressed and thus continued.

Tucker’s research, much of it in collaboration with HBS Professor Kent Bowen, Associate Professor Amy Edmondson, and Assistant Professor Steven Spear, is attracting attention both for its findings and its methods.

“Health care is a large, growing part of the economy. Getting things done with high quality and low errors and low burnout is really a profoundly important problem as far as I can tell,” said Edmondson, who co-authored three papers with Tucker. “Part of what Anita has learned is how aspects of the job design and aspects of the organizational context make it pretty hard for these nurses to do their jobs effectively.”

Small problems, big consequences

Tucker found that busy hospital nurses, especially the most experienced ones, deftly negotiated the many minor problems that came up during their shifts, solving them quickly and returning to care for their patients with a feeling of satisfaction.

They rarely, however, took action to address the root of the problem: Why are so many security tags falling off infants? Why do we always run short of towels on three-day weekends?

“What we saw is that the nurses focus on the individual patient, getting that patient what that patient needs. It’s all the nurses can do to get that done to the level that they want, so there’s nothing left over,” said Tucker. “And there’s no organizational mechanism to learn from these individual failures.”

Tucker is quick to point out that the nurses themselves are not to blame. “They’re incredibly motivated and talented and smart. It’s got nothing to do with lack of desire or competence on their part,” she said.

Instead, she said, the nurses’ supervisors and hospital management too often dismiss these problems as minor, which, said Tucker, many are. Cumulatively, though, they can pack a big punch.

“In the context of the work of the nurse, they cause a lot of difficulty and stress,” said Tucker, who is completing a survey of 600 nurses that will quantify some of that stress, as well as other factors affecting their performance. She anticipates a link between these daily annoyances and the larger problem of nurse burnout.

And while it’s unlikely that anyone ever died waiting for a clean towel, there is evidence that fatal hospital errors are more likely to be caused by a chain of small missteps rather than by catastrophic negligence.

“What I saw in my research were all those little things that happen that create the atmosphere that allows all those errors and accidents to occur,” said Tucker.

Learning from Betty Crocker

If Tucker’s findings are interesting, her research methods, intense and time-consuming observation of workers backed up by an extensive quantitative survey, are even more remarkable.

“She’s done hours and hours and hours of up-close, unconstrained observation of how the work processes of nurses are constructed … with an open mind, a blank slate, trying to understand these systems,” said Edmondson. “But then she goes a step further by proposing to test it with [quantitative] survey measures.” Edmondson describes this multimethod study as the “gold standard” of research but still relatively rare.

Tucker, trained as an industrial engineer, drew from her own observations of front-line workers at General Mills, where she worked in quality improvement supporting the production of Betty Crocker Ready-to-Spread frosting. Converting recipes into operators’ instructions, she was frustrated by the gap between her view of the process and the operators’. She began to see the disconnect between an organization’s strategic goals and the nuts and bolts of implementation.

“There would be something that would seem logical to me, but I couldn’t get compliance,” she said. Conversely, her operators would suggest a simple, but not a minor, change – increasing the font size on recipe cards, for instance, or reversing the order of two ingredients – that would enhance their production but would require great effort.

When she arrived at HBS, she found that most work at the School emphasized big-picture strategy and assumed that implementation was easy.

Recalling her struggles to implement Betty Crocker’s changes, Tucker was struck by this prevailing assumption. “I thought either I was incredibly dumb or bad at doing this, or there’s something missing here,” she said.

She credits Bowen and Spear and their approach, which she calls “very ethnographic,” with helping her probe what was missing.

“They pushed me farther than I would have gone on my own in terms of going and seeing,” she said, adding that she wanted to observe the world through nurses’ eyes.

Spear, who had already done some research at a local hospital, helped her gain access to nurses; Tucker expanded her field to include nine top hospitals throughout the United States and Canada. Following nurses around for hours, sometimes timing their every chore, Tucker began to understand how minor problems interrupted the rhythm of their busy days.

She learned that nurses’ time ticks by in minutes or fractions of minutes; their average task took just two minutes. They spent an average of 42 minutes per day fixing or circumventing the many small problems they faced; likely not coincidentally, nurses generally remained at work for 45 minutes of unpaid overtime finishing their necessary tasks.

In a day that clips by in two-minute increments, taking time – sometimes a whopping seven minutes – to document a problem and help address its root cause was unrealistic, she found.

“For a nurse to take two minutes to call somebody and say ‘this problem happened’ was a very deliberate, conscious trade-off. It’s not insignificant. It’s a noticeable pain that he or she feels at the end of the shift,” said Tucker.

Hands-on management

Tucker recommends that organizations close this feedback loop by making it very, very easy for their front-line operators to relay – and help solve – problems. Systems for feeding back problems and leading organizational improvement “are designed by people who sit in offices and think that three minutes is not a big deal,” she said.

The physical presence of a manager, someone who roams the hospital floor helping to trouble-shoot, is key, she said, echoing research done by Bowen and Spear.

Tucker, who plans to finish her dissertation this spring, hopes to continue to focus on health care in her research. Yet she sees implications of her work across the range of service industries.

“It’s almost a curse,” she laughed, describing an encounter with a hotel clerk after she discovered that her room key opened a room that was already occupied. The clerk readily replaced her key, but Tucker insisted that he discover how the mix-up happened and what would prevent it from recurring.

Tucker said she might even contact the plant manager at General Mills with some of her research to let him know that a good system does not necessarily run itself.

“There’s this notion that ‘Well, the workers know what they’re doing so the manager doesn’t need to be around,’” she said. “It’s the right idea but the wrong emphasis.”