Published online 4 November 2009 | Nature 462, 20-21 (2009) | doi:10.1038/462020a


Children's study fights to survive

US politicians, once supportive of a massive research project on childhood health, are now criticizing it.

North Carolina

A study of 100,000 US children should provide insights on how the environment affects their health.A study of 100,000 US children should provide insights on how the environment affects their health.J.-F. BOURKE/CORBIS

Late last month, in a former Blockbuster Video store in tiny Kenansville, North Carolina, an insulated cardboard box awaited a placenta. Collected after a nearby birth the night before, the placenta would be shipped to the University of Rochester Medical Center, home of the central histological lab for the US National Children's Study. There, pathologists would catalogue the placenta's weight, size and shape; note any abnormal features; fix it in formalin and embed it in paraffin; then make slides and tissue blocks to preserve it for future use by researchers studying major childhood ailments from asthma to obesity.

Many more specimens may soon be coming from rural Duplin County, North Carolina. Over the next several years, investigators in this land of hog farms and cotton fields hope to enrol half the new pregnancies in what is arguably the most ambitious study ever of how the environment affects children's health.

Authorized by Congress in 2000, the study has ramped up only recently, with funding rising from US$69 million in 2007 to $180 million this year (see graph). It aims to study more than 100,000 children at 105 locations around the country, following them from before birth to age 21. Its 28 proposed hypotheses intend to explore, among other things, whether prenatal infection is a risk factor for autism; how media exposure influences neurological, cognitive and social development; and how recurrent, low-level pesticide exposure might affect cognitive skills.


The study is now enrolling patients at seven 'vanguard' centres, including the one in Duplin County. But just as it is getting off the ground, it is running into political tumult. In August, angry senators on the Senate appropriations committee declared that the National Institutes of Health (NIH), the study's principal overseer, had committed a "serious breach of trust" by withholding from Congress news that estimates of the study's cost — initially around $3 billion over 25 years — had grown to as much as $7 billion. Among other things, officials at the NIH's National Institute of Child Health and Human Development (NICHD) had failed to include estimates of the indirect costs routinely paid to institutions.

Until the NIH provides "the most up to date information possible", the senators wrote, they would reserve a decision on how many dollars — "if any" — they would allocate to the study in 2010. (The number will be settled during upcoming negotiations between appropriators in the Senate and the House of Representatives; lawmakers in the latter have agreed to fund the study at $194 million in 2010.)

In the staid parlance of Capitol Hill reports, the words "breach of trust" landed with the impact of a hand grenade. "I have never seen report language referring to anything that the NIH has ever done as a 'serious breach of trust'. I think that is telling," says a seasoned observer of biomedical politics in Washington DC who asked not to be named.

The NIH had already responded by shaking up its staff, removing study director Peter Scheidt in July. By early October, NICHD director Duane Alexander was out of his job, too. (Alexander, now at the NIH's Fogarty International Center, has said that he was already considering leaving the director's job after 23 years.)

The NICHD's new acting director, Susan Shurin, says that "we obviously take the Senate's concerns very seriously." Some of the 28 hypotheses will "probably" be dropped, she says; "we are going to be managing to a budget." The NIH's deputy director Raynard Kington says that when a pilot phase finishes in spring 2010, the agency will embark on a "rigorous evaluation" to determine the study's future trajectory.

The Senate rebuke was far from the study's first labour pain. In the mid-2000s, the White House under George W. Bush had repeatedly zeroed out its budget — with the support of the NIH's then-director Elias Zerhouni, who argued that its long-term costs were prohibitive. But members of Congress, many of whom had potential study locations in their districts, repeatedly reinserted the funding.

Power of the people

Study leaders say that big is the way to go with this study — that the power of 100,000 participants and the extensive amount of information they will provide, from house-dust samples to parental toenail clippings, will produce a data bank that detectives of disease causation can mine for decades to come. Other major longitudinal studies, such as the Framingham Heart Study, have provided this kind of trove — but for adults, notes Barbara Entwisle, director of the Carolina Population Center in Chapel Hill and the principal investigator at the Duplin County centre. "All of the recent science shows the importance of what happens early on in people's lives for health all the way through their lives," she says. "This is the future of our country."

In 2004, after long and controversial consultations, study leaders adopted a household-based sampling strategy, a costly and labour-intensive decision meaning that participants from the 105 far-flung study locations will be more representative of the broad population than, say, those recruited from health facilities and doctor's offices, a more common approach.

In May 2008, in a 140-page report, the National Research Council and the Institute of Medicine praised the study's household-based sampling approach, as well as the statistical power rendered by its 100,000 participants. But it also identified a laundry list of weaknesses in the study's design. The first was the absence of a pilot phase; because the vanguard centres' data were to be part of the final data set, investigators would be less likely to experiment with methods and change things that weren't working. That concern is being addressed, says Shurin, by turning the vanguard centres into long-term development platforms for the main study's methodology and feasibility, and not, as initially intended, folding them into the main study. Study leaders have also pushed back the launch of the study's main phase until late 2011, giving investigators time to tweak the methodology according to what's learned at the vanguard centres.

Despite the rough waters, advocates remain passionate about the study's potential, arguing that it is an investment that will more than pay for itself. Leonardo Trasande, a study investigator at the Mount Sinai School of Medicine in New York, notes that just six of the conditions the study explores — asthma, autism, diabetes, injury, obesity and schizophrenia — cost the United States at least $650 billion a year (L. Trasande and P. J. Landrigan Environ. Health Persp. 112, A789–A790; 2004). "If the study identifies preventive interventions that can reduce those annual costs by even 1%, [a $3-billion] study repays itself twofold in one year," he says.


Meanwhile, Duplin County continues to enrol participants. While recruiters work church dinners and organize soccer tournaments, data collectors recently finished enumerating 10,800 households — roughly half of those in the county — aiming to locate reproductive-age women.

"I have two daughters. Both have asthma. I have lived this," says Entwisle. "But the scientific reason for my involvement has to do with the fact that there are so many questions that need answers. This design is exactly what needs to be done to begin to develop some of those answers." 

Commenting is now closed.