News this Week

Science  11 Jan 2013:
Vol. 339, Issue 6116, pp. 124

You are currently viewing the .

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

  1. Around the World

    1 - Shanghai, China
    Clearing the Air on Particulates
    2 - Shanghai, China
    Toughening Up on Academic Fraud
    3 - Houston, Texas
    Transocean Owes $1 Billion for Clean Water Act Violations
    4 - Washington, D.C.
    Stem Cell Legal Battle Ends
    5 - Mountain View, California
    BGI-Shenzhen Cleared to Take Over Complete Genomics
    6 - Princeton, New Jersey
    Princeton, South Korea Collaborate on Fusion Reactor

    Shanghai, China

    Clearing the Air on Particulates

    Small stuff.

    PM2.5 monitoring equipment in Beijing.


    Following public pressure for greater transparency in pollution reporting, 74 Chinese cities began measuring concentrations of harmful particulate matter less than 2.5 micrometers in diameter (PM2.5) on 1 January. Air pollution has become a touchy subject in China; the Twitter feed @BeijingAir, run by the U.S. Embassy in Beijing, provides hourly PM2.5 readings that provoked protests from Chinese officials in 2009. @BeijingAir helped prompt the Chinese government to release its own PM2.5 readings for Beijing last January, but those often differ from the U.S. Embassy's readings, both in absolute measurements and in how the figures are interpreted.

    Chinese government descriptions "significantly understate the severity of pollution," says Steven Q. Andrews, an independent environmental consultant in Beijing who has studied the issue. Days described by the U.S. Embassy as "very polluted" or "hazardous" have been called "good" or "slightly polluted" by the Chinese government in the past, he says. The new pollution readings appear in Chinese on the China National Environmental Monitoring Center's Web site:

    Shanghai, China

    Toughening Up on Academic Fraud

    A Chinese government regulation toughening penalties for academic fraud took effect on 1 January. Cases of fraud are rife in China and include fabricated data and padding resumes with honors that a scholar never earned. A key provision of the new regulation stipulates that academic degrees will be revoked if misconduct involving graduate or postgraduate thesis work is discovered retroactively.

    Cong Cao, a scholar of Chinese science policy at the University of Nottingham in the United Kingdom, calls the directive a step forward but says implementation could be spotty: "The litmus test is whether high-profile scientists and administrators will be punished for misconduct." He fears that "the regulation will end up being used to punish ordinary students and faculty members," he adds.

    Houston, Texas

    Transocean Owes $1 Billion for Clean Water Act Violations

    Deepwater Horizon


    Gulf of Mexico science and restoration will get another chunk of cash from a record settlement with the U.S. government of charges related to the 2010 Deepwater Horizon oil spill. Transocean Deepwater Inc., the company that owned the doomed drilling rig, agreed on 3 January to pay $1.4 billion in civil and criminal fines and penalties. A record $1 billion is for violations of the Clean Water Act, 80% of which will be dedicated to economic and ecological restoration projects along the Gulf Coast. In addition, the U.S. National Academy of Sciences will get $150 million to help fund a $500 million, 30-year Gulf Coast research program created last November under a criminal settlement with oil giant BP, which was using the rig. Transocean will also provide $150 million to the U.S. National Fish and Wildlife Foundation for ecological restoration projects along the Gulf. NFWF has already received $2.4 billion from the earlier BP settlement. BP still faces pollution penalties that could total $5 billion to more than $20 billion.

    Washington, D.C.

    Stem Cell Legal Battle Ends

    The U.S. Supreme Court has rejected a case challenging government-funded research on human embryonic stem cells (hESCs). The decision ends a court battle that has cast a shadow over hESC studies for more than 3 years.

    In Sherley v. Sebelius, two researchers argued that new National Institutes of Health (NIH) guidelines easing George W. Bush–era restrictions on hESC research violated a law banning federal funds for research on embryos. The plaintiffs won a preliminary injunction in August 2010 that briefly shut down NIH-funded hESC research. A trial court and an appeals court later ruled in favor of NIH. In October 2012, the plaintiffs appealed to the Supreme Court, which this week declined to review the case.

    Mountain View, California

    BGI-Shenzhen Cleared to Take Over Complete Genomics

    A review panel has dismissed concerns that a merger between the Chinese sequencing dynamo BGI-Shenzhen (Science, 3 February 2012, p. 516) and a struggling California company, Complete Genomics Inc., would threaten U.S. security. In September, BGI offered $117.6 million for control of Complete Genomics and promised to invest $30 million in cash to rejuvenate the firm. Founded in 2005, Complete Genomics uses proprietary methods to sequence entire human genomes with accuracy. The company has been losing money since its launch, however, and seemed headed for collapse in 2013. It accepted the lifesaving offer from BGI, but the deal ran into opposition. Several critics—including a U.S. maker of sequencing machines, Illumina of San Diego, California—warned that the deal might give China technology of key economic or military value (such as bioweapons manufacturing). In late December, however, the Committee on Foreign Investment in the United States ruled out such worries. This week, the deal emerged from a second waiting period that allows for an anticompetitive risk review and now seems cleared for takeoff.

    Princeton, New Jersey

    Princeton, South Korea Collaborate on Fusion Reactor

    The Princeton Plasma Physics Laboratory (PPPL), the United States' foremost fusion lab, is to collaborate with researchers in South Korea to design a next-generation fusion power demonstration plant that could generate electricity sometime after 2030. Fusion reactors generate energy by fusing hydrogen isotopes—but no fusion reactor has yet produced more energy than required to keep it running. The United States and South Korea are partners—along with China, the European Union, India, Japan, and Russia—in ITER, an international project to build a reactor that can produce excess energy. ITER's target is to produce 500 megawatts (MW) for 500 seconds at a time. Scientists at PPPL are preparing a work plan this month for the collaboration, which initially will run for 6 months. K-DEMO will be roughly the same size as ITER but it will produce 1000 MW and run for several weeks continuously. Rather than constructing an international demonstration reactor, many of the partners are considering building their own demonstration plants, including China, Japan, and India—and now South Korea looks set to join the race.

  2. Newsmakers

    ‘Third Kingdom’ Discoverer Dies



    On 30 December, microbiologist Carl Woese, who revolutionized microbiology through his studies of single-celled microbes called archaea, died at age 84 of complications from pancreatic cancer. When he began studying archaea, a group that includes methanogens and heat-loving thermophiles, biology textbooks contained only two main branches on the tree of life: bacteria and eukaryotes. Archaea were considered evolutionary offshoots of bacteria. But in 1977, by painstakingly hand-sequencing the ribosomal RNA of microbes, Woese demonstrated that archaea are as genetically distinct from bacteria as plants and animals are and that they merit their own branch. The discovery was controversial, upending traditional views of the evolution of life and taking years to gain widespread acceptance. Today, however, "nearly all biologists and an increasing number of physicians have awarded Woese an honorary Nobel Prize," says Alan Weiner, a professor of biochemistry at the University of Washington, Seattle.

    Pioneering Neurobiologist Dies



    The world's first Nobel Prize–winning centenarian—and one of its great scientific personalities—died on 30 December. Rita Levi-Montalcini, who was 103, was awarded the Nobel Prize in physiology or medicine in 1986 for her discovery of nerve growth factor. She spent the early years of her career in hiding in wartime Italy, studying the development of chick embryos in a makeshift lab set up in her bedroom after the Fascist government of Italy banned Jews from university posts. After the war, she worked for more than 2 decades at Washington University in St. Louis, where she made her seminal discoveries about how growth factors influence development and disease. In the 1960s, she began to divide her time between St. Louis and Rome, where she established a lab as well. After her nominal retirement in 1977, she returned to Italy full time. She remained active in the lab well past her 100th birthday and served until her death as president of the European Brain Research Institute in Rome, which she helped found, and at the Levi-Montalcini Foundation, which supports scientific education for girls and women in Africa.

    Eleven Engineers Win NAE Awards

    The U.S. National Academy of Engineering announced last week that it will honor three teams for notable achievements in telecommunications, bioengineering, and engineering education at a gala in Washington, D.C., on 19 February. Each team will share a $500,000 prize.


    Martin Cooper, Joel S. Engel, Richard H. Frenkiel, Thomas Haug, and Yoshihisa Okumura will receive the Charles Stark Draper Prize "for their pioneering contributions to the world's first cellular telephone networks, systems, and standards." Rangaswamy Srinivasan, James J. Wynne, and Samuel E. Blum will receive the Fritz J. and Dolores H. Russ Prize "for the development of laser ablative photodecomposition, enabling LASIK and PRK eye surgery." And Richard K. Miller, David V. Kerns Jr., and Sherra E. Kerns will receive the Bernard M. Gordon Prize "for guiding the creation of Olin College and its student-centered approach to developing effective engineering leaders."

  3. Random Sample

    Dinos Shook Their Tail Feathers


    A close look at 150-million-year-old fossils of a group of bipedal dinosaurs known as oviraptors suggests that many of them shook their muscular, feather-adorned tails to gain attention during courtship. The oviraptor's relatively stubby tail, scientists found, was muscular and flexible at its base but rigid at its tip, where the last half-dozen or so vertebrae are either fused together into a bladelike structure or so tightly arranged that they're inflexible. Some early fossils from the group included large feathers either attached to the last few tail vertebrae or preserved nearby. Fossils from later species had no preserved feathers, but had the same arrangement of tail vertebrae and the muscles used to control their movements, researchers revealed last week in Acta Palaeontologica Polonica. The most likely use for a feather-tipped tail is for courtship display (pictured: artist's reconstruction of a male oviraptor Ingenia yanshini flirting with a potential mate).

    Not Slippery When Wet


    The pale wrinkles that adorn fingertips after an extended soaking may be unsightly, but they serve a purpose: They help us get a stronger grip on slippery objects.

    Long thought to be caused by osmosis-induced swelling in the outer layer of skin, the wrinkles are in fact produced by the autonomic nervous system, recent experiments indicate. But the purpose of the puckering—which occurs on hair-free skin of the hands, feet, and toes but nowhere else on the body—hasn't been clear, says Tom Smulders, an evolutionary biologist at Newcastle University in the United Kingdom.

    In 2011, a team of neuroscientists suggested that the wrinkles served to enhance our grip on wet or submerged objects, just as treads on tires help improve traction on wet roads. "That seemed like a clever hypothesis that would be easy to test," Smulders says.

    So he and his colleagues designed a test in which volunteers picked up 45 submerged objects—such as glass marbles and lead fishing weights—from a bin one at a time, passed them to their other hand through a postage stamp–sized hole in a barrier, and then dropped them through another hole into a box. When test subjects had wrinkly fingertips—induced by soaking their hands in 40°C water for 30 minutes—they completed the task about 12% faster than they did when their fingers hadn't been soaked, the team reported on 8 January in Biology Letters. When performing the same task with dry objects, wrinkly fingertips didn't provide an advantage.

    By the Numbers

    830 million — Tons of carbon dioxide (2% of global emissions) that the world's telecommunications infrastructure produces each year, according to a study published on 2 January in Environmental Science & Technology.

    100 billion — Planets in our Milky Way galaxy (averaging one per star), according to a new estimate in The Astrophysical Journal based on M dwarf systems, the most common type of system in the Milky Way.


    Join us on Thursday, 17 January, at 3 p.m. EST for a live chat with experts about what it may take to phase out greenhouse gas emissions.

  4. The Children's Study: Unmet Promises

    1. Jocelyn Kaiser

    Having spent about $1 billion over the past decade, a U.S. study of environmental influences on children's health is still trying to define its methods. Critics say it has gone astray; advocates say it is on track at last.

    Up, up we go.

    This family in Queens, New York City, is part of the study's elaborate pilot phase.


    For epidemiologist Maureen Durkin, the end of her dream project came in June, when a moving van pulled up in front of rented office space in the western suburbs of Milwaukee. There, her staff of 15 based at the University of Wisconsin, Madison, had spent 7 years and $16 million in federal dollars gearing up to recruit 1250 pregnant women in Waukesha County whose babies would be among 100,000 followed for the largest ever long-term study of U.S. children's health. Instead, after a pilot phase in which Durkin's group enrolled 200 women, the National Institutes of Health (NIH) told Durkin that a private contractor would be taking over her part of the study. Over 2 hot summer days, her staff members wheeled freezers and centrifuges, computers, and blood tubes into the van for recycling.

    Billion-dollar test run.

    Funding for NIH's children's study has totaled $992 million through 2012, including $114 million for a contract for a coordinating center that was discontinued.


    "We spent millions training and getting the equipment and getting ready to go," Durkin laments. Now, she says, NIH is "just throwing it down the drain."

    The closure of Durkin's study site and 39 others like it reflects an abrupt shift in course for one of the most ambitious U.S. biomedical research projects ever: the National Children's Study (NCS). Twelve years ago, Congress asked NIH to follow the health of a large group of children from birth to adulthood to examine the effects of environmental influences, such as chemicals and education, on children's health and development. Last year, after recruiting about 4000 women for pilot tests, NIH scrapped its plan to have 40 university-led centers carry out the study in a random sample of about 100 U.S. counties. Large contractors are taking over the pilot work. The study's overseers at NIH's National Institute of Child Health and Human Development (NICHD) say that to contain costs they had no choice but to pull the plug and start fresh. But their still-evolving new plan is being challenged.

    Critics note, for example, that about half of the mothers and babies would be recruited at birth, not during pregnancy, as planned, when crucial exposures and developmental events may occur. And although NIH says it wants the study to represent the U.S. population, the agency hasn't said how it will achieve that.

    Onlookers wonder how a seemingly simple epidemiology project—enrolling pregnant women for a long-term observational study—became hopelessly complicated and costly. NCS became "weighted down" with add-ons like "a ship encrusted with barnacles," says pediatrician and epidemiologist Nigel Paneth of Michigan State University in East Lansing. He was an investigator in the original NCS but was dropped, and hasn't yet been invited to join the new one. After investing $1 billion over 12 years, NCS has produced little, critics say, even as similar studies in other countries have surged ahead.

    "I find that so depressing. The missed opportunities, how much could have been accomplished," says Brown University epidemiologist David Savitz, an early adviser to the project. Others go further: NCS has been "a national embarrassment. It is a disaster," says a prominent U.S. epidemiologist who asked not to be named. Former NIH staff members and investigators blame the problems on a confluence of factors: a mandate forced on NIH by Congress, a lack of strong scientific leadership, funding uncertainties, and a plan that embraced too many objectives. Along with bureaucratic hurdles, this led to delays and missteps.

    NIH says it has now figured out how to run a scientifically robust children's health study with a smaller budget. But some of those who envisioned a 20-year survey of children's health now have doubts about whether it can be pulled off. The project's implosion, Paneth says, "is quite an amazing chapter in the history of American science."

    The perfect study

    The idea for NCS arose in the 1990s, during the Clinton administration, amid growing concerns that chemicals in the environment might contribute to childhood diseases such as autism and cancer. The inspiration came in part from an influential 1993 National Research Council report led by Philip Landrigan, a pediatrician at Mount Sinai School of Medicine in New York who had done studies linking low-level lead exposure and lower IQ. The report found that current safety levels for chemicals, based on adult data, might not protect children.

    The study's problems are "an amazing chapter in the history of American science."



    A White House task force on children's health proposed a long-term study to look for links between health, development, and environmental influences—defined broadly to range from chemical exposures to psychosocial factors. In 2000, Congress passed a law directing NICHD to lead the study in cooperation with the Environmental Protection Agency (EPA) and others.

    Diverse picture.

    To represent the U.S. population, the National Children's Study set out to recruit pregnant women living at specific addresses within a random sample of 105 locations, mostly single counties.


    Although some researchers worried that a large study would siphon funds from individual-investigator research grants, others saw it as a successor to the Collaborative Perinatal Project (CPP)—a landmark NIH-funded study of 55,000 children in the 1960s and early 1970s. Among key results, it found that cerebral palsy usually develops long before delivery and that infections are a factor in pregnancy complications.

    Epidemiologist Mark Klebanoff of Ohio State University College of Medicine in Columbus recalls an NCS planning meeting in late 2000 led by then–U.S. Department of Health and Human Services Secretary Donna Shalala with six experts on long-term health studies. "Everybody thought it was a great idea." Says Klebanoff, then at NICHD: "The advice we got was to be bold." Following the lead of a Danish children's health study already under way and another starting in Norway, planners chose a size of 100,000 children, large enough to study rarer diseases such as type 1 diabetes. An early cost estimate was $2.7 billion, spread over 25 years.

    A fundamental problem was that the new administration in 2001, headed by George W. Bush, didn't like the project because of its cost and didn't include it in its 2002 funding request. So with a mandate from Congress and a small budget scraped together from several agencies, NICHD formed a study office in 2002 headed by pediatrician Peter Scheidt.

    Heeding Congress's request for broad input, Scheidt oversaw the formation of dozens of working groups and advisory committees involving more than 2500 experts. They came up with 28 overarching hypotheses exploring, for instance, a potential link between violent video games and gun injuries, and whether genetics make some children more vulnerable to harm from pesticides. Researchers would collect numerous samples, such as placentas and tap water in homes, and follow children through medical exams, home visits, and phone interviews with parents.

    The study "became the vehicle for most hypotheses related to children's health," says Jonathan Samet, an epidemiologist at the University of Southern California in Los Angeles and an adviser at the time. But that was a problem, says epidemiologist Lynn Goldman of George Washington University in Washington, D.C., who, as an EPA official in the 1990s, helped conceive the project: "It was a tabula rasa. Whatever you wanted it to be, it would be that."

    Perfectionism was a drag on the study, critics say: Scientists debated at length how best to recruit pregnant women. CPP had found them through university clinics. But NCS's social scientists successfully argued for a novel strategy: Recruiters would contact or knock on doors of randomly chosen households, much as the census surveyors do. This would ensure a broad sample and reach underrepresented groups such as the uninsured. But the strategy was untested.

    Federal statisticians determined the sampling frame—a randomly chosen set of 105 sites, mostly single counties, from rural Minnesota to Queens, New York City. In September 2005, with a still-modest $11 million annual budget, NICHD funded its first six vanguard centers to begin recruitment. These were to be activated once Congress approved funding for the study.

    All systems go

    Then in 2007, NCS's fortunes changed. Backers had won champions in congressional districts representing some of the 105 counties to be sampled. That year, lawmakers approved $69 million in startup money. With that in hand, NCS leaders announced another 22 centers, saying they would recruit women in waves, and inked a contract with Westat, a private contractor, to centralize data collection.

    A crucial element was missing, however: The study had not yet gone through outside peer review. NIH sent its 700-page research plan to the Institute of Medicine (IOM) in June 2007 for a fast-track review. Eleven months later, an IOM panel endorsed the household sampling plan. But reviewers also warned that NIH needed to test this approach in a pilot before launching the study. Scheidt's office responded that they would add more time to the vanguard recruitment phase, delaying the full study.

    The now-seven vanguard sites began recruiting the first pregnant women early in 2009—2 years later than originally planned. The new administration under Barack Obama supported NCS. But when acting NIH director Raynard Kington asked for updated budget projections, NCS staff members told him the full cost estimate had grown to as high as $6.9 billion, more than twice the original estimate. Kington was outraged, according to a former NCS staffer—and so were Senate staffers. At a Senate spending panel hearing in March, Kington said his staff was working to "assess true costs" and make "adjustments" to the study. In an unusual reprimand, the panel accused NIH in a report of withholding information and "a serious breach of trust."

    "It's hard to argue that our 10 years of planning has been more valuable than the 10 years of progress made by others…."



    NICHD Director Duane Alexander explained that early budget projections were "rough and conceptual" and did not include overhead or inflation, according to a summary of his statement. Behind the scenes, Scheidt's staff members protested that they had followed instructions from the NIH budget office not to update a 2005 cost estimate of $3.1 billion because the White House planned to ask Congress to zero out funding for NCS. In July 2009, the president announced his choice for the new NIH director: Francis Collins, the former chief of NIH's genome institute. That month, Scheidt was removed as NCS director, and 2 months later, Alexander stepped down as NICHD chief to take another position at NIH.

    Course correction

    Steven Hirschfeld, NICHD's clinical director and an expert on childhood cancer, took over as acting chief of NCS. Hirschfeld announced major changes. "Everything got turned upside down," says Goldman, who headed a study site until 2010. If Scheidt gathered too much advice, under the new director the pendulum swung the other way, according to NCS investigators, who say Hirschfeld made decisions internally without consulting them. For example, pilot research at the vanguard centers would be used to work out logistics and costs of data collection, not to recruit the first women for the study.

    To the consternation of many, Hirschfeld also announced that NCS would no longer be a hypothesis-driven study, but a "data collection" platform for exploring hypotheses that would be added later. Hirschfeld's office dissolved protocol committees. Investigators worried that this might yield data that wouldn't be very useful. "How do you characterize a group of women in the absence of hypotheses? What questions do you ask them, what phenotypic measurements do you collect?" asks epidemiologist Roberta Ness of the University of Texas Health Science Center, Houston.

    Meanwhile, as the data came in from the seven vanguard centers, it became clear that it would take 9 years—twice as long as originally thought—to enroll 100,000 pregnant women through contacting households. Some investigators weren't surprised. Because pregnancy is a brief period in a woman's life, Goldman says, "to find women right at that time is not easy." (Others say the strategy didn't get a fair test; it worked "fabulously" at some sites, Durkin says.) Deeming household recruitment a failure, the NCS office decided to designate 30 more study sites as vanguard centers and use all of them to test other recruitment strategies, including recruiting women at doctors' offices.

    But what seemed like a logical plan became a bureaucratic nightmare, study site investigators say. Any significant change to a protocol, such as a reworded questionnaire, had to be cleared through the Office of Management and Budget. Another complication was that Hirschfeld allowed the Westat contract for a central database to expire, after $114 million had been spent on it, because it was not "cost effective." Instead, researchers were told to develop their own data systems, but NCS kept changing the specifications.

    Because of the delays, staff members were hired but had nothing to do for months. Then as the centers got up to speed and began enrolling pregnant women, NCS told investigators to stop in November 2011 so that the program office could evaluate the data.

    As the pilot phase of NCS "lurched" along, as one researcher puts it, some observers sounded an alarm. Savitz and Ness, who both had run vanguard sites for a time, co-authored a lengthy commentary in September 2010 in the journal Epidemiology warning that the study was "still in trouble" and needed "fundamental" changes to improve efficiency, including a prompt switch from knocking on doors to recruiting subjects through doctors' offices.

    The study seemed to veer in an entirely new direction last February. NIH announced that none of the recruitment strategies being tested were feasible. Instead, NIH said it planned to abandon the 105 county sampling plan and switch to a "convenience sample" from health maintenance groups, in order to cut annual costs by 15%, to $165 million in 2013. A total of 40 study site contracts would not be renewed, and the women they had recruited for the vanguard study would be followed by new contractors.

    Two of NIH's advisers on NCS blasted the agency over the sudden change of course and quit. Congressional overseers also found fault. NIH has since moved toward a compromise: Women will be recruited from a random sample of U.S. hospitals and prenatal care providers affiliated with the hospitals. To save money, perhaps half will be identified when their baby is born, and half when they seek prenatal care.

    Righting the ship?

    Now that they've settled on an efficient study design, NIH officials say, the way forward is clear. The goal set by Congress in 2000 hasn't changed, Hirschfeld says: The study will still examine the effects of environmental and biological influences on U.S. children's health and development from before birth until age 21. Guttmacher says the study's pilot stage "is performing as intended, providing needed hard data on which to base the science and logistics of the much larger main study." The $1 billion spent so far was worth it, adds NIH Director Collins: "You would not want to undertake this without being really sure that the model was going to work." Hirschfeld says that the study has also developed "innovative approaches" and "tools" that will be used "by scientists around the world."

    The study's startup phase "is performing as intended, providing needed hard data on which to base the science and logistics of the much larger main study."



    NIH plans to gather more advice on the study design at an National Academy of Sciences workshop scheduled this week and at a meeting in the spring. Recruitment will likely begin in 2014, Hirschfeld says.

    Some former NCS study site investigators support the new strategy. Running the study through 40 contracts with academic centers became "enormously inefficient," says pediatrician Neal Halfon of the University of California, Los Angeles, who ran one of the largest sites. Halfon and Jørn Olsen, director of the Danish National Birth Cohort, say it makes sense to define specific hypotheses later. "Research is moving forward so fast, most of the hypotheses would be outdated" if defined now, Olsen says. Like some others, including Savitz, Olsen feels it's not essential that NCS recruit a representative sample of U.S. women.

    Advocates of the original NCS vision, however, worry that the goals may be lost. The project's credibility is already waning in communities where the academic sites are being terminated, they say. If NIH's plan for sampling hospitals does not recruit a representative sample of U.S. women, it may miss important groups. Recruiting only 50,000 of the women during pregnancy will reduce the statistical power to detect the effects of developmental influences in the womb, others add. Landrigan says he is "very much concerned" that the new, cheaper design "has much lower likelihood … of discovering potentially causal associations between toxic chemicals and disease in children."

    Most of the academic NCS study site leaders, including Paneth, have signed a white paper arguing that the study must return closer to the original 105 county plan. NCS is spending more than $3 million a week, points out Michael Bracken of Yale University, a former NCS researcher—money that could go to investigator-initiated grants. In his opinion: "This thing has either got to be done properly or it should not be done at all."

    Who is to blame for NCS's problems? Congress tops the list, some observers say. Mandating the study without buy-in from top NIH leadership was a recipe for disaster, says a former congressional staffer. But the NIH director played a role, too: Once it became clear that Congress was going to fund it despite NIH's objections, "this project was so big and so important that it should have been watched more closely," says another Washington policymaker who asked not to be named.

    Scientists are to blame as well by designing an overly complex study, many of those involved with NCS say. "It tried to do too much. But if it hadn't, it might have never gotten started. It's a bit of a catch-22," Klebanoff says. "You could say it was born with such ambitions that the grand vision may not have been realizable," Samet says.

    If the redesigned NCS is to succeed, Savitz says, it needs "an experienced, respected, authoritative, politically astute individual to pull all this together." The white paper authors say that this person should be an epidemiologist who has run large cohort studies. But Savitz admits he has "little confidence" that such a person would take command.

    During the decade when U.S. leaders discussed how to recruit pregnant women for NCS, the Danish and Norway children's studies, which recruited for less than $50 million each, have followed children for up to 16 years and have published scores of papers. These studies collected relatively little data on exposures and were able to keep costs down by drawing on national health care records.

    At least a dozen large birth cohort studies are now under way in other countries, Savitz notes. "It's hard to argue that our 10 years of planning has been more valuable than the 10 years of progress made by others with far less money. We've missed the boat and need to have a good look at what is going on to see where we can contribute."

    National Children’s Study Timeline

    October 2000

    Children's Health Act calls for NICHD to lead long-term child development study

    March 2004


    NCS planners decide to recruit 100,000 pregnant women from random sample of U.S. households in about 100 counties. Costs estimated at $2.7 billion over 25 years

    September 2005

    Contracts awarded for first six "vanguard centers" and data coordinating center

    February 2007


    Despite opposition from NIH Director Elias Zerhouni, Congress appropriates $69 million to launch study; number of funded study centers rises later in year

    May 2008

    Institute of Medicine endorses household sampling, but says it must be pilot tested

    January 2009

    New Obama administration backs study, recruitment begins at seven vanguard sites

    March 2009


    NIH acting director Raynard Kington tells Senate panel that "true costs" are higher than expected—up to $6.9 billion

    July 2009

    Pilot data indicate that recruitment by knocking on doors could take 9 years, 5 more than planned

    July 2009

    Scheidt removed as NCS director

    September 2009


    Duane Alexander steps down as director of NICHD. Alan Guttmacher later becomes director

    October 2009


    Acting NCS director Steven Hirschfeld announces study will focus on "data collection," not speciic hypotheses

    Spring 2012

    New contractors begin to take over all 40 academic study sites

    July 2012

    New study proposal would recruit half of women at birth and half during pregnancy by sampling U.S. hospitals, not by door-to-door surveys

    January 2013


    NAS workshop to review new study plan


    Revised children's study to begin recruitment

  5. Consulting

    Expert Firms Play a Hidden Role in Connecting Science and Finance

    1. Jeffrey Mervis

    What every academic and clinical researcher should know before signing up with an expert network firm and interacting with its clients.

    "We have not created a market for consulting. But we like to say that we have professionalized the process."



    On 20 November, the U.S. government unveiled what it called the largest insider-trading scheme in its history. According to documents in the U.S. district court in New York, a hedge fund analyst, Mathew Martoma, generated $276 million in ill-gotten gains for his company, CR Intrinsic Investors, through stock trades based on confidential information about the results of a clinical trial for a new Alzheimer's drug. Prosecutors say that Martoma's inside knowledge of how the drug performed came from a prominent neurologist at the University of Michigan, Sidney Gilman, in a relationship arranged by a third party.

    That third party was the Gerson Lehrman Group (GLG), the dominant company in a cottage industry that it helped to invent. Martoma's company was a client of GLG, and Gilman was a member of GLG's expert network. The two spoke 42 times between 2006 and 2008.

    The case is the latest in a string of high-profile prosecutions that have focused attention on this little-known world of so-called expert network firms. Such firms connect the savviest people in a field with clients willing to pay for the expert's knowledge. Clients, be they financial analysts, venture capitalists, professional consulting firms, or lawyers, submit a request, and GLG sifts through its files to find the right expert in the relevant area. (GLG's networks cover health care, energy, telecommunications, manufacturing, real estate, and a dozen other sectors.) Experts select which offers from GLG to accept, and after the conversation they bill GLG for their time.

    But such expert networks are controversial. Financial managers handling billion-dollar stock portfolios are on the prowl for something their competitors don't know, and some clients may cross the line by prodding an expert to break the law and disclose confidential information.

    To find out how such expert network firms operate, and why a researcher would participate, Science interviewed several of GLG's top-rated experts in health care. None wanted to be named, for reasons that include a desire to avoid being connected in some way to the Martoma case. (Martoma was indicted last month. The 80-year-old Gilman, who retired from Michigan after charges were filed against Martoma, has agreed to cooperate with authorities and has not been charged.) So their stories, which are real, illustrate some of the tough issues facing researchers who consult through these firms.

    Making a match

    A complex network.

    A firm's stable of experts obtain their knowledge from many sources, and its clients put what they've learned to use in many ways.


    Alexander Saint-Amand, the president and CEO of the New York City–based GLG, had just graduated from college and was weighing his career options when he met with a doctor who was treating his mother for early-onset, rapidly progressing dementia. Although the biotech sector was awash in cash, the doctor complained that promising treatments for a host of diseases were languishing while venture capitalists poured money into approaches that any practicing physician could have told them were bad bets. Recalling the conversation, Saint-Amand says he thought: "There must be a better way to get this conversation going between the doctors who know what works and the investors who want to fund it."

    The first version of GLG, launched in 1998, was an online service that allowed experts to connect with those willing to pay for their expertise. "But the best scientists didn't want to work with just anyone who finds them online," Saint-Amand says. "They wanted to talk with the most sophisticated people in their field, whether in industry or investors." That led to the current system, in which GLG constructs both sides of the network, arranges all the interactions between clients and experts, and manages the paperwork.

    "We have not created a market for consulting," Saint-Amand says. "But we like to say that we have professionalized the process." Part of that service, Saint-Amand says, is to educate and repeatedly remind its experts about federal laws prohibiting the disclosure of proprietary or material information about a company or its products.

    Michigan's vice president for research, Stephen Forrest, says the university definitely wants its researchers to "engage" with industry. Those ties are an essential piece of Michigan's attempt to commercialize discoveries by its faculty members, who in 2011 carried out more research than any other U.S. public university, topping $1 billion. Potential financial conflicts of interest stemming from such relationships, he adds, need to be managed, not prohibited. However, Forrest concedes that Michigan's current rules, which emphasize the disclosure of possible conflicts, would not stop a faculty member who decides to break the law.

    In contrast, one of the country's most prominent medical centers, the Cleveland Clinic in Ohio, believes that consulting for financial investment firms like GLG is so fraught with risk that it strongly discourages its physicians from doing so. "We have carved out expert networks from other consulting activities," says Guy Chisolm, a cell biologist who directs the clinic's Innovation Management and Conflict of Interest Program. "We're very skeptical of this activity."

    Chisolm worries that "an expert might know things that could cross the line, or that such firms could use our name to attract business and promote their interests." Fewer than 20 of the clinic's nearly 3000 physician-scientists currently engage in the practice, says Chisolm, who also chairs the Forum on Conflict of Interest in Academe for the Association of American Medical Colleges. That number is dropping, he adds, a trend that pleases him.

    "My colleagues have said there's an altruistic reason to do it: ‘If you know what treatment would help your patients and you see an opportunity to steer investments in that direction, who better than a Cleveland Clinic doctor to play that role?’ But I think the risks outweigh the benefits."

    The personal interaction is one of the reasons that Dr. X, a pulmonary specialist at a prominent medical center in the mid-Atlantic region, signed up with GLG. Some firms feed their experts a steady diet of surveys, says Dr. X, who also consults with MEDACorp, a rival expert network firm owned by Leerink Swann, a Boston investment bank (see sidebar, p. 139). "But GLG seems to like the idea of putting a warm body with another warm body."

    That human connection may have made it easier for Gilman and Martoma to build what federal prosecutors have characterized in court filings as a "personal" relationship, with Gilman "eventually coming to view Martoma as a friend and pupil." But that description doesn't square with what Dr. Y, a research oncologist at one of the top medical schools in the United States, has experienced as a member of GLG's expert network.

    "It's rare to talk to the same person twice," Dr. Y says, "and I've never talked to anyone more than three or four times. Also, the idea that this is a mentor/mentee relationship is just silly. Mentors don't give their students the answer. Rather, they teach the student to think for themselves."

    In fact, the GLG experts who spoke with Science say that they don't spend much time thinking about the person on the other end of the line. "I don't really care who they are," says Dr. Z, a cardiovascular researcher at a major Midwestern medical center. "I can figure out pretty quickly how much they know based on their questions. But my job is simply to provide scientific answers."

    Talk isn't cheap

    "We're very skeptical of this activity. … I think the risks outweigh the benefits."



    Why do these physician-scientists sign up with GLG? Being an expert, they say, forces them to stay current in the f ield, and also builds their reputation. Most also see consulting for GLG as an extension of their role as an educator. And some simply enjoy talking about their work with an attentive listener.

    But first and foremost, they say, is the money. The three experts quoted in this article have been designated as "leaders," meaning that GLG ranks them in the top 5% of its network. Saint-Amand estimates that leaders log anywhere from 100 to 150 calls a year. And he says the average hourly rate for physician-scientists is $250.

    Dr. Y, the oncologist, says he turned his back on the chance to become wealthy when he chose to go into academic medicine. But he believes that society "undervalues" what he and his research colleagues do for a living. Consulting for GLG doesn't balance the scale, he says, but it's still worth doing.

    "I might as well put my medical education to work," he says. "In a good year I clear $12,000," he says. "That's not a lot, but it's better than nothing."

    Consulting also addresses a larger trend in academic medicine, says Dr. Z, the research cardiologist. "Scientists are becoming an endangered species," he says. "We are losing them to clinical care. And the reason is obvious: A classmate of mine [in clinical care] can make $1.3 million a year, while the starting academic salary is $150,000."

    At the same time, all of the experts who spoke with Science say that they would withdraw from the network if they felt that their participation distracted them from doing their academic job. "If I wanted to do more, I'd become a full-time consultant," Dr. Y says. "I'd never put doing this over my research."

    Dr. Z says that he's actually trying to cut back. "I don't want this to interfere with my research and caring for patients. I also get e-mails from another expert network firm, but I don't respond to them."

    How much time experts spend on consulting is important to their institutions, too. Many research universities and medical schools allow faculty members to consult the equivalent of one day a week. And although most institutions consider consulting for an expert network firm to be part of the broader category of outside activities that must be disclosed by any faculty member with federal funding, there are certain aspects of that relationship that could raise a red flag.

    "We do lay out a variety of conditions that need to be followed, including their obligations as a professor, the prohibition against disclosing insider information, protection of intellectual property, and the use of Hopkins' name," says Julie Gottlieb, associate dean for policy coordination for Johns Hopkins University School of Medicine in Baltimore, Maryland. "The amount of money earned could be a trigger to look deeper into the relationship."

    At the same time, Gottlieb says, "faculty consult for a range of entities, and there are risks associated with all of those relationships for which they need to take precautions. Being part of an expert network represents something of a gray zone."

    On a short leash

    The Cleveland Clinic has decided to make doctors who want to enter that gray zone jump through additional hoops. The clinic's law department must first approve all contracts, no matter how much the expected payment. (There's a $10,000 minimum for other consulting.) After the contract passes muster, the physician who signs up with GLG or another firm gets a black-bordered document with 11 bullet points describing practices that must be followed.

    The issue is not consulting per se, Chisolm says, but rather who's listening to the doctor's advice. "When you're talking to Merck or Medtronic, your expertise is going to be used to help the company make decisions on the direction of its research, or improvements to its products," he says. "You know more about the recipient than you do when you're involved with an expert network." The opportunity for insider trading exists in both cases, Chisolm observes, "but the chances are much higher when you're talking to investors."

    Saint-Amand says the firm's compliance practices are designed to eliminate that risk. "We provide this framework with audit trails. And we systemically train and remind doctors of their obligations regarding conflicts of interest," he says. "Because of all that, we like to say our platform is a terrible place to buy insider information. But if someone chooses to do bad things, systems and controls can only mitigate that risk."

    State sunshine laws that apply to his university require him to disclose all relationships with drug companies, Dr. Y says. "But I don't think anybody discloses their nonpharmaceutical contacts," he says, a category that would include payments from GLG.

    Dr. X says that he's not trying to hide any financial activities. In fact, he doesn't even buy stock in individual companies—"the market is illogical, for starters," he explains. But he requested anonymity because he wasn't sure how his university would react if it knew he was working for GLG. And even as a GLG expert, he generally avoids conversations with stock analysts, preferring to talk with a company developing a particular technology or a venture capitalist looking for background information.

    Staying on the right side of the law is easy for him, Dr. Y says, "because I'm risk averse and I'm afraid of getting into trouble." He says his work is preclinical, and he doesn't take money from companies developing drugs in his field, which he says makes him a rarity among researchers he knows.

    His GLG consulting isn't an issue for his institution, he adds: "They just want to be sure what I do doesn't affect my research." Even so, Dr. Y insisted on anonymity because of concern that something he said might offend GLG, which controls the flow of potential clients.

    For Dr. Z, anonymity is all about avoiding publicity. "I'm still young, and I don't want anything to mess up my career," he says.

  6. Consulting

    An Expert's Words Can Move Markets

    1. Jeffrey Mervis

    What happened when a positive drug report led to a declining stock price? How one expert's opinion made an impact

    When scientists reported at a 2005 meeting that an experimental drug for stroke victims had improved short-term functioning and didn't increase bleeding, investors in the small biotech company that had developed the drug expected to see a jump in its stock price. Instead, the stock price of Renovis fell by 23%, shedding almost $100 million in value in less than a week.

    What happened? On the first working day after the meeting, neurologist Marc Fisher of the University of Massachusetts Memorial Medical Center in Worcester spoke to a bevy of stock analysts assembled by MEDACorp, an expert network firm owned by the Boston investment bank Leerink Swann. As one of MEDACorp's experts, Fisher raised questions about how the patients' status had been assessed 3 months after the drug, Cerovive, was administered. Within hours Renovis's stock price started dropping, and it never recovered. A year later, Cerovive flunked a second, larger trial, and in 2008 Renovis went out of business.

    The former CEO of Renovis, Corey Goodman, believes that Leerink Swann analysts were expecting disappointing results from the drug trial and saw a chance to make money by betting that Renovis's stock price would fall. A rise would have meant a big loss for the investment firm, so Goodman surmises that it found an expert who hadn't seen the results but was willing to criticize the study anyway. Goodman first raised those suspicions at a 2006 conference of biotech executives, and he repeated them last month to Science.

    Fisher says that Goodman is wrong. "I attended the presentation [at the European Stroke Conference in Bologna, Italy] and I knew the results," he says. Fisher admits, however, that he didn't initially understand the study's unorthodox use of the Rankin scale used to measure outcomes and that a few days later he modified his criticism. "But the results from the next trial were totally negative, so I guess my initial concerns were valid," he adds.

    Fisher stopped consulting for MEDACorp and for drug companies in 2010 when he became editor of Stroke, an American Heart Association journal that prohibits its editor from engaging in such activities. "I have no conflicts now because I'm not allowed to do it," he says.

    But Fisher agrees with Goodman on one point: "I had no idea my words would have such an impact."