Women Advancing Science

Jo Handelsman is a Howard Hughes Medical Institute Professor at the University of Wisconsin-Madison and Robert Birgeneau is the chancellor of the University of California, Berkeley.

In his much-acclaimed best seller "The World is Flat" Thomas L. Friedman warned that America is slipping further behind China and India in producing scientists and engineers at the peril of the innovation and productivity of our nation. A recent report from the National Academies, "Beyond Bias and Barriers," provides a very direct solution to this conundrum. American science needs more talent and that talent is readily available in a legion of well-trained, but greatly underutilized scientists and engineers who happen to be women. The good news is that a few significant changes in the academic system could stem the loss of these women, thereby fortifying our scientific leadership.

The lack of women among the ranks of the country's university science professors is not due to biological differences between the sexes. There are many genetically determined differences between men and women, but differences in innate ability cannot explain women's low representation in the leadership of science. Women faculty continue to be outnumbered by men even in scientific disciplines in which women have received as many doctoral degrees as men for many years. In some other scientific disciplines, women's representation on the faculty has increased 30-fold in the last 25 years, a rate too rapid to be explained by changes in genetics of the population, illustrating that women have the ability to perform and capacity to endure the rigors of academic science as opportunities open to them. Collectively, the report's findings provide indisputable evidence that social factors, not aptitude or interest, are powerful drivers of the gender composition of the scientific leadership of the academy.

Why are there so many women who excel in science and choose to invest years of their lives in hard, frequently unrewarding work to attain advanced education in science, but so few in academic positions? The Academy committee concludes that a combination of unconscious biases held by both men and women as well as archaic university structures limits the participation of women in academic science.

Unconscious, inadvertent bias often clouds people's perceptions and judgment. Randomized, controlled studies show that when both male and female evaluators are asked to evaluate job applications, they will give the applicant a lower rating and be less likely to hire the person if they are told that the applicant is woman than if they are told the applicant is a man. Similar outcomes are evident when evaluators are asked to review candidates for raises, promotions, or leadership positions. The biases are most evident when the job under consideration is in a male-dominated field, which may explain the slower advancement of women in science than in other academic fields. Studies show the same type of prejudice toward members of certain racial minority groups, which likely contributes to the bleak picture for minority women. For example, in 2002 there were no African American, Hispanic or Native American women in faculty positions in the nation's top 50 computer science departments. Both unconscious and deliberate prejudice also contribute to a chilly climate in many science departments. Women, more often than men, feel excluded from decision-making in their departments and from scientific discourse with colleagues. The cumulative effect of repeated exclusion can reduce productivity and cause women to leave academic positions.

Institutional barriers, such as the tenure system, further augment the difference in men's and women's access to faculty positions. Candidates must establish well-funded innovative research programs, demonstrate their teaching abilities, and contribute to their universities through service, all within seven years. This is a tall order for all scientists, but it is a Promethean task for those who have young children or care of other family members during this same period of life. Since women continue to shoulder a disproportionate share of family responsibilities, the collision between the tenure system and family life affects women far more than men.

The Academy report's recommendations are simple. Educating the academic community about the insidious role of unconscious bias in decision-making could substantially reduce the application of that bias. Department chairs can be trained to improve the climate for women faculty, which may require simply using inclusive practices for decision-making and communicating. Lengthening the time to tenure, ensuring that research funding does not lapse because of child bearing, and making lactation rooms, child care, and flexible work schedules readily available will ensure that both men and women can meld careers with family life.

The report makes the straightforward but, nevertheless, bold assertion that the government should enforce equal protection laws such as Title IX as vigorously for science as it does for intercollegiate athletics. This is not a recommendation about special opportunities or treatment; it is an effort to level the playing field and give women the same access to positions in academic science that men have always enjoyed.

The simplicity of the message in the Academy report is both reassuring and inspiring. American science needs more brainpower and now we just have to make sure that we access that talent. Otherwise, Tom Friedman's ominous warning may well come true.

Higher-Tech Bridge Inspections

The tragic total and instant collapse of the Interstate 35 highway bridge in Minneapolis should force us as a nation to make a careful reassessment of the methods we use for inspecting and repairing our infrastructure.

Material deterioration, fatigue, vibrations, the effects of sustained loads and overloads, problems related to the foundation integrity, design weaknesses--all may be among the factors that contributed to the Minneapolis collapse. Sadly enough, this bridge may have been inspected in recent years, but the conventional technologies commonly used for such inspections generally are not up to the job. Many inspections are performed simply by naked eye or by very simple methods, such as pinging the bridge's surface with a hammer.

We can achieve a more definitive assessment of the behavior and safety of bridge material and structure through the use of systematic and effective nondestructive testing methods that already exist.

Most of the nation's highways and highway bridges were built in the 1950s; our railroad bridges are nearly a century old. Indeed, many of these structures may be in a critical state because we subject them to traffic loads and vibrations greater than their initial design capacity and life span. Today's vehicles are often much heavier than their mid-century counterparts.

Concrete and steel, the most commonly used materials in our infrastructure, naturally deteriorate with age. Freeze-and-thaw mechanisms used in harsh winters are especially detrimental to concrete, as are hot and humid conditions and the application of salt to melt winter ice. The latter conditions can lead to the corrosion of the steel used to reinforce concrete, which expands the volume of the steel. The expanded volume results in internal stress on the structure and can lead to failure of the material or the entire structural system. The columns (piers), deck system, and joints in a bridge structure are all affected.

Research has been done on advanced concepts and laboratory techniques for nondestructive testing using, for example, radar, infrared thermography, and acoustic methods, but usually such developments have not been carried to the implementation phase to produce portable devices applicable in field conditions. The high-tech, nondestructive tools presently available for infrastructure inspection have not been embraced by the construction industry for widespread and systematic use.

We need to carefully evaluate and implement into our engineering and management practices the already existing high-tech methods for monitoring and testing bridges and other structures, and we must encourage the development of new technologies as needed.

We should develop field specifications for nondestructive testing and monitoring technologies and require the construction industry to adopt the use of these technologies. The public authorities responsible for our infrastructure should schedule systematic inspections, require predetermined levels of structural reliability, and demand rapid screening and detailed investigations. A systematic high-tech inspection process of this kind would allow us to rate the condition of bridges and other structures, and prioritize them for repair, retrofit, and upgrade. Decisions as to the options to repair, upgrade, implement a temporary solution, or build a new structure should be made by skilled engineers and based on scientific fact and sound engineering judgment, with consideration paid to the available monetary resources.

Finally, we should incorporate into our engineering curriculum the science of material deterioration, methods for developing advanced technologies for the assessment and repair of infrastructure, and the design of new and innovative materials for infrastructure applications. And we should insist that continuing education for practicing engineers and management staff in this field is essential.

It is possible to avoid another tragedy like that in Minneapolis, but only if we act quickly to make rapid and thorough inspections that utilize high-tech devices developed for the task.

Oral Buyukozturk is a professor of civil and environmental engineering at the Massachusetts Institute of Technology.

Google Health for the Public Good

Last week, when Philipp Lenssen at Google Blogoscoped unveiled screen shots of Google's health-information system, the warning sirens seemed to drown out any praise for the concept.

Code-named Weaver, the system is designed to be a personal repository of all of an individual's medical information, which the patient can selectively share with others (such as a physician offering a second opinion). The system is also a resource for information about illnesses, treatments, drugs, health-care providers, and even community sites where patients can share experiences and rate the quality of care they got from particular physicians.

Comments about Weaver warned that Google "could sell [your medical information] to anyone, including insurers and government," and that insurance providers and employers could hack into your data to deny you coverage or a job.

What gets lost in all this is what Google is trying to accomplish. A careful reading of what Google vice president Adam Bosworth has written and said about his goals paints an impressive picture of what such a service can do.

In a speech at the American Medical Informatics Association Spring Congress, Bosworth outlined three things that we need, which will presumably find their way into a Google Health service.

The first thing we need, Bosworth says, is "discovery." Finding relevant information is critical to good health. He notes in one blog that his mother died of cancer and that "it took her doctors nine months to correctly identify an illness which had classic symptoms."

Doctors don't always pay enough attention and don't always have the latest information at their fingertips. Patients, armed with the appropriate data, could take more responsibility for their own health.

One year ago, my sister, Lindy, died of breast cancer that had metastasized to her brain. I firmly believe that she died, in part, from a lack of information.

Lindy first felt a lump in her breast in 1998 and was relieved when the mammogram finally came back negative. What she didn't know, and might have found out if something like Weaver had been around then, is that mammograms can sometimes give false negative results. She ignored the lump for another year. By then the cancer had spread significantly.

She had a radical mastectomy, chemo and radiation therapy, and a bone-marrow transplant to restore her immune system. She went into remission for five years. She kept seeing her oncologist, who did blood tests which all came back negative. Again, we were all fooled. Blood tests are not sufficient for tracking remission, something I later found out through searching Google.

A couple of years ago, Lindy started having trouble concentrating, was sluggish and forgetful, and lost physical coordination. She also suffered from depression, so she talked to her psychiatrist about these symptoms rather than to her oncologist.

It took at least six months for her symptoms to get bad enough for her daughter, fearing that Lindy was having a stroke, to drag her to the emergency room. A CT scan showed several tumors in her brain. These might have been caught sooner if she had been keeping track of all her symptoms, medications, and test results, and had shared them with all her doctors, as future patients will be able to do with Weaver.

Bosworth's second suggestion is "action." This means seeking out physicians, treatment centers, support groups, and other organizations to treat the patient or help families care for their loved ones. My family went through the difficult process of finding Lindy home hospice care as her condition worsened. Even then, my brother still needed to spend a great deal of time caring for her, dealing with her insurance, tracking her medications, scheduling visits from nurses, and trying to determine what to do when she further deteriorated. Bosworth says that there should be a simple information and management system to do all those things, as well as to help with e-consulting and care-management guidance, and with requesting second opinions and getting online coaching.

Bosworth's final suggestion is "community." From his speech:

Consumers can organize into communities which are protected from inappropriate commercial exploitation but allow consumers to share their experiences with others whose medical situations are similar, to rate the value of resources such as sites and health practitioners, and even to answer questions of each other in real time.

Lindy meant to seek out support groups, and I meant to help her. But we were overwhelmed, physically and mentally. I wish I had managed to overcome those limitations. A system like Weaver would surely have made such groups easier to find and to share information with.

My sister worked for one company for nearly 30 years and had the same health insurance until she died. Her employer and insurer already had all her information, so she had no fear of corporate spies finding out about her condition. Still, Bosworth's promise is that "only consumers, not insurers, not government, not employers, and not even doctors, but only consumers, should have complete control over how [their medical information] is used." That promise is worth more to me than all the speculation of what Google "could" do.

Roni Zeiger, a physician and product manager at Google, notes in a blog entry that "better-informed patients recover faster, manage chronic illnesses better and may even avoid some illnesses altogether."

To that I would add: they may even save their own lives.