Shortly after Paul O'Neill became CEO of Alcoa in 1987, he visited the company's huge aluminum smelter in east Tennessee.
Word had already spread that Mr. O'Neill wanted to make worker safety Alcoa's top goal. In a meeting with management and labor leaders at the plant that day, he recalled recently, he turned to the managers and said, "From now on, we're not going to budget for safety. As soon as anyone identifies anything that could get someone hurt, I want you to fix it and I will figure out how to pay for it."
He then turned to the union leaders and said, "Here's my phone number at home, and if they don't do what I said, I want you to call me."
About three weeks later, Mr. O'Neill got a late night call from a production worker at the plant. "You came down here and you said all this fancy stuff about safety," the worker said, "and I just wanted you to know that for the past two or three days, we've had a broken conveyer belt." The broken belt meant workers had to manually hoist hot 600-pound ingots from one spot to another.
Mr. O'Neill called the plant manager, told him to report to the smelter, get the problem fixed and call him back when the job was done. He got the call-back at 5 that morning. "There was an effective tom-tom network at Alcoa, and so I didn't have to do that at every plant because they soon knew I meant it."
Stories like that are familiar to many Pittsburghers, but they've now been given a global stage by a new book, "The Power of Habit: Why We Do What We Do in Life and Business," by New York Times writer Charles Duhigg. The book, which devotes an entire chapter to Mr. O'Neill and his safety campaign, is currently No. 8 on the Times' nonfiction best-seller list.
Using stories and research from the fields of psychology and neuroscience, Mr. Duhigg argues that much of our lives is ruled by unconscious habits, good and bad, but that by becoming consciously aware of the cues that trigger our habits and the rewards they provide, we can change bad practices into good ones.
Mr. Duhigg also argues that the same patterns and possibilities for change exist in companies and in society as a whole, and he uses Alcoa and Mr. O'Neill as one of his two primary examples of corporate transformation (the other is the story of Starbucks and CEO Howard Schultz).
In successful companies, he writes, leaders often identify "keystone habits" that can be summed up in a simple, clear way and create change throughout the organization. In Alcoa's case, the goal of achieving zero workplace injuries was "a simple phrase that within that culture became a suitcase for something pretty complex," Mr. Duhigg said in a recent interview. In a similar way, he said, Mr. Schultz's emphasis on superior customer service at Starbucks drove change in that culture.
In an extensive interview, Mr. O'Neill, who left Alcoa in 2000, said he wasn't sure how comfortable he was in having his worker safety emphasis described as a method for creating good habits because, "I have a little negative reaction to the idea that human beings are infinitely malleable into habit robots."
He saw his safety goal as part of a broader emphasis on creating "habitual excellence."
"I believe an organization has the potential for greatness if every person can say yes to three questions without reservation," Mr. O'Neill said. "The first is, 'Can I say every day I am treated with dignity and respect by everyone I encounter without respect to my pay grade, or my title, or my race, or ethnicity or religious beliefs or gender?' And you know, there are not a lot of places like that.
"The second question is, 'Am I given the things I need - education, training, tools, encouragement - so I can make a contribution to this organization that gives meaning to my life?' And the third question is, 'Am I recognized for what I do by someone I care about?'
"Every company I know of says somewhere in its annual report, 'People are our most important resource,' but my observation from all these places I had worked was that there was no evidence it was true."
The worker safety campaign at Alcoa, which already had a well-regarded injury rate, was more than mere rhetoric. In his 13 years at the helm, Alcoa went from 1.86 lost work days per 100 workers to 0.2. And the ethos has continued: as of last week, the rate stood at 0.125.
His ideas have also spread to other executives in the region.
David Shapira, board chairman of O'Hara grocer Giant Eagle, said that after he began to meet regularly with Mr. O'Neill and learn from him, "I understood we were not serious about safety in the way he was."
Since then, Mr. Shapira has ordered that reports on the safety of workers and customers be the first item at board and internal management meetings. The company has lowered its lost workday rate from more than four per 100 workers to less than one per 100.
"Paul views safety, and I completely agree, as an ethical issue," Mr. Shapira said. "We believe we have a responsibility to protect the people around us.
"Think about when your children were little and how you felt about their safety; that's a moral obligation you have as a parent. And Paul and I feel we have that obligation in terms of safety to make sure the people who work here and shop here are safe. That doesn't mean we can always make them safe, but we certainly have an obligation to care about it and really try."
Mr. Duhigg said he first got interested in Alcoa when he was taught a case study on Mr. O'Neill's campaign at Harvard Business School, where he got his master's degree.
When he was putting his book together, he said, it seemed like a perfect example of how one clear goal established by a strong leader could drive other changes in a company. In Alcoa's case, that included better worker-management relations, more innovation and a shift away from a hierarchical management style, he said.
This concept of "keystone habits," he said, also works in personal lives, research has shown. "On an individual level, people who start exercising regularly start eating better but also start using their credit cards less and doing the dishes earlier."
For Mr. O'Neill, another central idea during his time at Alcoa was that people ought to be treated as equally as possible, "and so I looked for things that told people they were not as good as anybody else."
One of those things was the physical design of headquarters. In the old Alcoa building Downtown, he said, executives "all had these god-awful offices, and people cared a lot about how many square feet they had." The CEO's office there, he said, was at the end of a long hallway on the 30th floor. "So if you weren't already intimidated, you would be by the time you got there."
His distaste for those trappings drove his ideas for the new headquarters building on the North Shore, which had an open office plan, and where his office was in the middle of the floor, with the same 81 square feet of space as everyone else's.
"One executive told me, 'You at least need to have a different colored strip of carpet so people know you're the CEO.' It's funny how some people want those symbols, but I think they're the enemy of an organization that is habitually excellent."
While Mr. O'Neill had this egalitarian streak, it didn't mean he was afraid to exercise his power.
In fact, Mr. Duhigg said, he believes one reason Mr. O'Neill didn't last long as U.S. Treasury secretary in the George W. Bush administration was not just because of political disagreements he had with the president, but, "He couldn't fire anybody there."
Mr. O'Neill recalled that not long after he took over at Alcoa, he insisted the Laurel Valley Golf Club near Ligonier stipulate it had an open admissions policy regardless of race or ethnicity before he would allow Alcoa executives to remain there as members.
When local CEOs gathered at the Duquesne Club to hear Mr. O'Neill explain his position, "One CEO said to me, 'You're abusing your power.' I said 'Well, I've worked really hard for 30 years or so to have power that was worth abusing.'"
Since he left Alcoa, Mr. O'Neill has focused much of his energy on health care reform, and in particular, on improving safety for health care workers and patients.
It's been a frustrating experience.
The latest figures on hospital central line bloodstream infections, which occur when germs get into incision sites for catheters, show that infection rates dropped by 32 percent in 2010 from the 2006-08 period. "Bloodstream infections are still too common, and the rates vary a lot from hospital to hospital," said John Santa, director of the Consumer Reports Health Ratings Center.
Mr. O'Neill was involved in one of the earliest experiments on reducing those infections, at Allegheny General Hospital. In a 2006 study published in the Joint Commission Journal on Quality and Patient Safety, he and a hospital group led by Richard Shannon showed that a team-based approach to analyzing and reducing bloodstream infections in an AGH intensive care unit lowered infections from 49 to six in a year, and cut deaths from 19 to 1.
Such dramatic improvements occurred in the three intensive care units at Allegheny General overseen by Dr. Shannon, Mr. O'Neill said. "But then, we couldn't even get the then-president of the hospital to adopt practices in the other ICUs that we had proven could save lives. It's showed me that it's really hard to get good, scientifically proven ideas through plasterboard walls."
One major problem with patient safety initiatives in American hospitals, he said, is that "they are projects; they are not efforts to create an organizational culture. Most projects will create incredible results for a short period of time, but there's a wasting away back toward normal because the changes don't belong to the culture, they belong to a project."
Dr. Shannon, now at the University of Pennsylvania, agreed. In an email, he said, "Without the explicit belief and commitment of organizational leaders, improvement efforts such as those in the [intensive care units] at AGH create islands of excellence. It allows organizations to trot out 'examples' without holding them to high performance in all things that we do in health care."
Mr. O'Neill said much of his thinking about the challenges to improving U.S. health care are summed up in a recent presentation by health care reform advocate and Harvard surgeon Atul Gawande.
In a TED Conference speech last month, Dr. Gawande said much of the problem with American medical care is because doctors are trained to be "cowboys - daring, independent and self-sufficient - when what we need is pit crews."
The surgeon has led efforts by the World Health Organization to reduce surgical mistakes by adopting standardized checklists before and during operations. Even though use of the checklists cut complications by 35 percent and deaths by 47 percent, he said, there has been strong resistance by many hospitals to adopting the new approach, "because using a checklist forces us to have a different value system, like humility, discipline, teamwork.. This is the opposite of what we were built on - independence, autonomy, self-sufficiency."
"We've come to a place," Dr. Gawande concluded, "where we have no choice but to recognize that complexity requires group success; we all need to be pit crews now."
Mr. O'Neill, who recently excoriated UPMC and Highmark for their contentious battle to control the Pittsburgh health care market, sees the problem in terms of failed leadership in American medicine.
"I would argue in an awful lot of hospitals and insurance companies, there is nothing I would call leadership," he said. "If you want to see the most incredible hierarchical system in the world, go to a hospital. If you have a white coat, you can kill people and people who know better will watch you do it because it's not OK for people to holler, 'Stop!'
"One thing that's perplexing to me is how difficult it is to get organizations across American medical care to adopt practices that are proven life-savers and money-savers. The things I've been talking about for 15 years would create vast improvements in outcome - and save trillions of dollars a year."
Mark Roth: firstname.lastname@example.org or 412-263-1130. First Published May 13, 2012 4:00 AM