# News this Week

Science  19 Aug 2011:
Vol. 333, Issue 6045, pp. 922
1. # Around the World

1 - New Delhi
Himalayan Glaciers: Some Expanding, Most Retreating
2 - Brittany, France
Green Algae Pose Mortal Danger
3 - Washington, D.C.
Shale Gas Needs More Openness, Better Data
4 - London
Research Council Earns Chemists' Wrath, Cuts Ph.D.s
5 - Colombo, Sri Lanka
Elephant Survey Under Fire

## New Delhi

### Himalayan Glaciers: Some Expanding, Most Retreating

Most Himalayan glaciers are retreating, although 21% show no increase in melt rate, according to a new study by the Indian Space Research Organization and the Geological Survey of India in Kolkata.

The pattern is a worldwide phenomenon and part of a natural cyclic process, according to a statement in parliament from India's environment minister, Jayanthi Natarajan. Her statement surprised many observers in that it did not attribute the glaciers' retreat to climate change.

The new results come from the “Snow and Glacier Studies” project, undertaken with government support by the Space Applications Centre in Ahmedabad. Completed in 2010, the satellite-based survey took an inventory of the snow cover and glacier extent across regions of the Indus, Ganges, and Brahmaputra river basins. The 5-year research project monitored 2767 glaciers and found that 2184 were retreating, 435 were advancing, and 148 showed no change.

The melt rate of Himalayan glaciers was highlighted in the 2007 report by the United Nations' Intergovernmental Panel on Climate Change (IPCC), which stated that Himalayan glaciers “are receding faster than in any other part of the world.” IPCC later acknowledged that this was an error.

http://scim.ag/_himalaya

## Brittany, France

### Green Algae Pose Mortal Danger

The death of 36 wild boar at a picturesque beach in Brittany has created a scientific summer mystery—and a fierce political debate. After the carcasses were discovered in July, environmentalists claimed that the boar died from inhaling hydrogen sulfide (H2S) produced by the rotting green algae that blanket many beaches in Brittany every summer. They say the algal blooms, caused by massive runoff of nitrates from Breton farms into the sea, pose a threat to human and animal health. Angry farmers denied culpability and suggested activists might have poisoned the animals themselves.

On 6 August, however, French officials said an autopsy confirmed that H2S was the culprit, and, 3 days later, French environment minister Nathalie Kosciusko-Morizet ordered the closing of any beach where the slimy green carpet can't be removed within 24 hours of arrival. She further enraged farmers by saying the runoff needs to end.

Scientists and environmental activists say the French government has downplayed the risks for decades. In 2009, a horse succumbed to H2S poisoning, and some believe a truck driver who died that year while removing dead algae may have been killed by the gas as well.

## Washington, D.C.

### Shale Gas Needs More Openness, Better Data

A U.S. Department of Energy committee has waded into the fracas over producing natural gas from shale using hydrofracturing, or fracking, technology. The committee released a report 11 August calling for shared data—particularly of the chemical recipes for the fracking fluids—to reduce shale gas's environmental impact.

Shale gas regulation lies primarily with the states, but “we're pointing out what can and should be done,” says geophysicist Mark Zoback of Stanford University in Palo Alto, California, one of seven members of the subcommittee advising Secretary of Energy Steven Chu. To regain public trust, information about shale gas should become readily available to the public, the report says. And much more information should be gathered on the environment before, during, and after drilling.

The report also recommends forming a national organization focused on shale gas, which would include both a database of public information and a list of best practices for industry. http://scim.ag/_shalegas

## London

### Research Council Earns Chemists' Wrath, Cuts Ph.D.s

More than 100 chemists, including six Nobel laureates, have signed a letter criticizing the U.K. Engineering and Physical Science Research Council's (EPSRC's) plans to reduce its research funding for synthetic organic chemistry. The letter, sent to Prime Minister David Cameron, says the cuts will “seriously injure an invaluable section of the UK economy.”

Like other research councils in the United Kingdom, EPSRC is struggling to plan for a future in which its budget, while “ring-fenced” at the current level for the next few years, will slowly shrink in real terms due to inflation. The council also recently revealed that it may fund 1000 fewer new Ph.D.s in the upcoming academic year than in 2010–11. EPSRC disclosed this estimate, in which the number of newly started doctorates would fall to 1900 in 2011–12, in a published answer to a question from a U.K. Member of Parliament. Although an EPSRC official calls the forecast figures a “worst-case scenario,” the council does still anticipate significant cuts—a 20% fall in the total number of Ph.D. students funded over the next 5 years is a more realistic possibility, says Atti Emecz, EPSRC's director for communications, information, and strategy.

http://scim.ag/UKphds

## Colombo, Sri Lanka

### Elephant Survey Under Fire

Sri Lankan wildlife minister S. M. Chandrasena set off a firestorm last week when he announced that following the country's first elephant census, some animals would be captured and sent to Buddhist and Hindu temples.

Chandrasena later backtracked, claiming the survey was instead purely a conservation effort. For Sri Lankan wildlife groups, however, the damage was done. Dozens of groups condemned the remarks and pulled their support from the survey.

A potential ulterior motive is not the survey's only problem, biologists say. Because Asian elephants are elusive and nocturnal, aerial counts don't work as well for them as for African elephants. So during the 3-day census, which began 11 August, government workers and volunteers staked out water holes, tallying elephants as they came to drink. But that indirect approach is “not an acceptable methodology for counting Asian elephants,” wrote Prithiviraj Fernando, head of the Centre for Conservation and Research in Rajagiriya, Sri Lanka, in an e-mail. While potentially useful for looking at population density and range, a water hole count gives an incomplete picture of the country's total elephant population. Even if an accurate count were possible, he added, “I do not see what management decisions can be taken based on the number of elephants.”

2. # Random Sample

## Rover Looks Deep Into Mars

After a 3-year, 21-kilometer journey, the Opportunity rover finally arrived at Mars's Endeavour crater on 9 August. The intrepid explorer has already peeked into and analyzed the rock and soil of 11 craters—and so far, the geologic story Opportunity has read in these rocks has been one of an ancient martian wasteland of windblown dunes pocked by the occasional acid-laced puddle.

## Something in the air

While the prize quarry remains at large, many astronomers are straining for a better look at trophies already in the bag. Ben Oppenheimer, an astrophysicist at the American Museum of Natural History in New York City, started with slightly bigger game. In 1995, shortly after the first exoplanet was announced, a team he led published a paper in Science (1 December 1995, p. 1478) describing the spectrum of the first brown dwarf to be discovered. Brown dwarfs are “failed stars” that have too little mass to sustain the stable burning of hydrogen and constitute a kind of bridge in between large planets and stars in size and temperature. The spectrum Oppenheimer and his colleagues obtained from the object, named Gliese 229B, was similar to Jupiter's spectrum, indicating the presence of methane in the brown dwarf's atmosphere.

As more exoplanets began to be discovered, Oppenheimer decided that these new worlds were far more exciting targets for spectral analysis. “I got really excited about it because you can really understand a lot of the physics and chemistry of these objects from doing spectroscopy,” Oppenheimer says. “I thought, obviously, this is a great field to be getting in to, and I'm getting in at the ground floor.”

Over the past several years, Oppenheimer and others have been marching toward what many see as the next frontier in exoplanetary science: studying the planets' atmospheres. That's a key goal of the Gemini Planet Imager (GPI), an instrument being designed for the 8-meter Gemini telescope on Cerro Pachón in Chile.

The instrument is designed to image and take spectra of star-orbiting exoplanets. In addition to being equipped with an advanced adaptive optics system and a coronagraph, which will block some of the glare from the parent star to bring the planet into relief, GPI will consist of a spectrograph that analyzes 30 different wavelengths of light. From the spectral information, researchers will be able to deduce the temperature, mass, and chemical composition of exoplanets' atmospheres. GPI is expected to begin taking observations starting the middle of 2012, says Oppenheimer, who is a co-PI of the project.

Other groups have already analyzed the spectra of a few exoplanets. In 2008, using instruments on the Hubble Space Telescope, researchers led by Mark Swain, an astronomer at the California Institute of Technology in Pasadena, detected methane and water vapor in the atmosphere of a hot Jupiter, a giant planet orbiting very close to its star. Last year, another group led by Brendan Bowler, a graduate student at the University of Hawaii, Manoa, teased out the chemical fingerprints of the atmosphere from an exoplanet about seven times as massive as Jupiter orbiting a star 130 light-years away. In the light emanating from the object, they saw the imprints of carbon dioxide and carbon monoxide.

GPI and similar instruments won't be powerful enough to probe the atmospheres of smaller exoplanets or of exoplanets of any size more than a few hundred light-years from Earth. But the work done with them will make more-advanced studies possible, Oppenheimer says: “The techniques we're developing, these are exactly the techniques we will be using in the future.”

Eventually, astronomers intend to look for biosignatures in the atmospheres of Earth-like planets. The concept dates back to the 1960s, when James Lovelock, an independent scientist who consulted for NASA, suggested that extraterrestrial life should alter its planet's chemical balance by producing compounds that a dead planet probably would not harbor.

MIT planetary scientist Sara Seager has thought a lot about what kinds of molecules future spectroscopists should look for in distant worlds that fulfill other conditions of habitability. The obvious biosignatures to look for are oxygen, ozone, and nitrous oxide, unique products of life on Earth, which a distant alien civilization could detect in the spectrum of our atmosphere. Another candidate is dimethyl sulfide, which oceanic phytoplankton produce on Earth. Seager says the list she and colleagues have drawn up so far is just a starting point for thinking about the whole slew of chemicals that life might produce in different environments. “We know that biology is much smarter than we are,” she says. “The reality is that we might have a bunch of other gases that we currently don't associate with life.”

For now, exoplanet researchers will stay busy with what is already a full agenda: mapping the atmospheres of giant planets, mining the Kepler data for Earth-like worlds, and gaining insights into the evolution of planetary systems. As habitable planets are found, they will be poised to probe them for life, secure in their science, and comfortably far from the fringe.

4. Infectious Diseases

# Drug Developers Finally Take Aim at a Neglected Disease

1. Mitch Leslie

New and recycled compounds target the parasite responsible for Chagas disease, which afflicts millions in Latin America.

In the mid-1990s, Jim Palmer, a chemist then at Khepri Pharmaceuticals in the San Francisco area, saw an opportunity to prevent months of effort from going to waste. He had synthesized a set of molecules that block cysteine proteases, a class of protein-snipping enzymes. But the blockers weren't panning out for chronic conditions such as psoriasis and osteoporosis, which were Khepri's main targets.

Palmer knew somebody who might be able to use the compounds, however. His friend James McKerrow, a biochemist at the University of California, San Francisco, was hunting for molecules that suppress cruzain, a cysteine protease crucial for the microscopic parasite Trypanosoma cruzi, the cause of Chagas disease. Millions of people in Latin America carry this insect-borne pathogen, often unwittingly because the initial symptoms of Chagas disease are usually mild. Yet the parasite continues to breed inside the body, and years or decades after becoming infected, about 20% to 30% of people begin to show signs of serious illness, most often damage to the heart that can lead to heart failure and death.

Although pharmaceutical companies don't usually share their intellectual property with outsiders, Palmer—with approval from his superiors—handed over his protease inhibitors to McKerrow and colleagues for testing. One was quite good at killing T. cruzi in the lab dish, so Palmer made a more potent version for McKerrow.

Now, after more than a decade of follow-up work, McKerrow and colleagues have received permission from the U.S. Food and Drug Administration (FDA) to launch a phase I safety trial of that compound, dubbed K777, in the United States next year. Palmer, currently director of drug discovery for Biota, a biotech based in Melbourne, Australia, is as pleased as a new parent. “It's like my baby,” he says, but “someone else has been raising it.”

Along with conditions such as African sleeping sickness and leishmaniasis, Chagas disease is one of the tropical diseases that drug companies have traditionally overlooked, even though there's no vaccine against the parasite and efforts to curb the insects that carry it have proved hit and miss. But after years of stagnation, research into new treatments for Chagas disease has picked up. Two other drugs, both converted antifungal compounds, are entering phase II trials to determine whether they clear the parasite from chronically infected people. “It's been decades since there was a clinical trial of a new drug for Chagas disease, so that is very exciting,” says biochemist Frederick Buckner of the University of Washington, Seattle.

Researchers hope it's just the beginning. Some are scanning chemical libraries, reevaluating existing drugs, and probing new molecular classes to identify additional compounds that might combat a disease that kills more than 12,000 people every year. “I'm optimistic that there's going to be a lot of progress in the next decade,” says Peter Hotez, dean of the new National School of Tropical Medicine at Baylor College of Medicine in Houston, Texas. Still, Hotez cautions that this movement “is all relative” because there were so few advances in Chagas disease treatment for decades.

## The cost of neglect

Brazilian physician Carlos Chagas first officially described the disease that now bears his name in 1909. Today, the World Health Organization (WHO) estimates that about 10 million people are infected with T. cruzi, making Chagas disease a bigger health problem in the Americas than malaria. And in recent years, the illness has appeared outside of the tropics more and more frequently (see sidebar).

Although the news offers no consolation to patients, Chagas disease is predominantly an animal disease, says epidemiologist Uriel Kitron of Emory University in Atlanta. The parasite primarily infects wild mammals such as armadillos, opossums, rats, and raccoons, as well as dogs and cats. Spreading the disease are the aptly named assassin bugs, also called cone-nosed or kissing bugs, which feast on blood at night. They usually transmit the parasite to humans through their feces, which people inadvertently rub into a bite or into their eyes or mouth. Like many other neglected tropical diseases, Chagas disease picks on the rural poor. It's prevalent in areas with adobe and thatch homes that offer plenty of crevices where the bugs can conceal themselves during the day.

Chagas disease resembles other neglected tropical diseases in a second way: Current drugs are old, noxious, and impractical. Although not as harsh as the main treatment for African sleeping sickness, which kills 5% of patients who receive it, the only two Chagas medicines, nifurtimox and benznidazole, can trigger serious side effects. Nifurtimox sometimes results in anorexia and nerve damage, whereas benznidazole can spur skin inflammation that may be so severe it kills. Moreover, a course of the drugs lasts 60 to 90 days, and many patients never finish such a lengthy treatment. “The bar has been set pretty low” for any potential new drugs, McKerrow says.

Both compounds can rid a person of the parasite during the early, acute phase of the disease. The catch is that few people are aware they need treatment at this stage. So for most people, “this infection is for life,” says Julio Urbina, director emeritus of the Venezuelan Institute for Scientific Research in Caracas. Whether nifurtimox and benznidazole help the majority of Chagas patients who have chronic infections remains controversial, McKerrow says. A large trial to find the answer for benznidazole is running in several South American countries.

Another motivation for developing new drugs is that parasite transmission continues, despite intensive attempts to curb it. Efforts to control parasite-carrying bugs, involving measures such as spraying infested houses and outbuildings with insecticides, have all but eliminated new cases of the disease in countries such as Chile, Uruguay, and Brazil. Still, as epidemiologist Ricardo Gürtler of the University of Buenos Aires notes, the problem “hasn't been conquered.” He points to the Gran Chaco region of Argentina, Bolivia, and Paraguay, one of the areas hardest hit by Chagas disease, where the insects bounced back after spraying stopped. And in some parts of Latin America, he adds, they have evolved resistance to the insecticides used to control them. WHO estimates that partly because of such setbacks, about 60,000 people still contract Chagas disease every year.

Researchers have also identified a disturbing new way that the parasite can infiltrate the body: through food and drinks, particularly fruit juices. A frightening example came to light in 2007, when more than 100 pupils and staff at a school in Caracas fell ill after they drank guava juice tainted by the bugs' feces.

## Reopening the pipeline

If the prospects for new Chagas drugs are looking up, researchers say that part of the credit belongs to the Drugs for Neglected Diseases initiative (DNDi), a nonprofit organization based in Geneva, Switzerland. DNDi describes its approach as “virtual” because it has no labs and carries out no research. Instead, it serves as a project manager, shepherding promising drug candidates through development by contracting labs, companies, and organizations that can take on different steps in the process, such as screening compounds or running clinical trials. DNDi's funding comes from a variety of donors, including the Gates Foundation, the European Union, and Doctors Without Borders.

DNDi's plan calls for at least one new approved drug to treat chronic Chagas disease by 2014. Along with Palmer and McKerrow's K777, the organization is sponsoring work on E1224, a compound designed by Eisai Pharmaceuticals of Japan. In the body, E1224 transforms into ravuconazole, a molecule currently in phase II trials to determine whether it suppresses fungal infections. Under DNDi's auspices, a phase II trial of E1224's potency against chronic Chagas disease will begin later this year in Bolivia. A non-DNDi project led by the Vall d'Hebron Hospital in Spain is testing the approved fungus-killing drug posaconazole against chronic Chagas disease. Merck, the drug's manufacturer, is also sponsoring a trial on chronic Chagas patients in Argentina.

Why are researchers so keen on antifungal drugs for a parasitic disease? The answer goes back more than a decade, when Urbina and colleagues went searching for existing drugs that could disrupt the synthesis of certain sterols, members of the same chemical family as cholesterol, that the parasite and fungi can't live without. One reason that posaconazole worked particularly well is that it enters the organs where the parasites take refuge, Urbina says. Studies confirmed that posaconazole and ravuconazole, both of which inhibit a key enzyme required for sterol synthesis, combat T. cruzi in animals.

Three new drug candidates might seem like cause for celebration, but Chagas researchers caution against too much optimism. “If you look at the history of drug development, more than 50% of drug candidates fail when they go into people,” McKerrow says. “We need to have the next generation of compounds ready to go.”

Several labs and organizations have started hunting for those compounds. DNDi, for example, has enlisted researchers at institutions such as Murdoch University in Perth, Australia, and the Federal University of Ouro Preto in Brazil to screen new compounds for antiparasite prowess. One project has focused on oxaboroles, boron-containing molecules that have already shown promise against the parasite that causes African sleeping sickness, says molecular biologist Eric Chatelain, head of DNDi's Chagas discovery and preclinical program.

Parasitologist Ana Rodriguez of New York University Langone Medical Center and a team of other scientists also recently completed the first high-throughput screen for Chagas drug candidates, assessing more than 300,000 molecules from a standard National Institutes of Health chemical library. By testing each compound to determine whether it slays the parasite and spares mouse cells, they have tried to weed out obviously toxic molecules. Rodriguez and colleagues eventually whittled the list to seven compounds that killed the parasite in mice.

That was the easy part. “The work really starts now,” Rodriguez says. If they can obtain funding, she and her colleagues hope to partner with the Swiss pharmaceutical company Sanofi-Aventis to further develop these potential drugs. One step they need to perform is what's called drug optimization, tweaking the structures of the molecules to ensure that they can be taken orally, remain stable, and have other favorable qualities.

Posaconazole and ravuconazole suggest that additional compounds that could attack the parasite are already under development for other uses or are even on pharmacy shelves, Buckner says. He and his colleagues have begun to screen FDA-approved drugs for their ability to kill T. cruzi. They've also been modifying tipifarnib, a cancer treatment that reached phase III clinical trials but wasn't successful enough to make it to the market; the compound shows promise against T. cruzi, Buckner says.

Medicinal chemist Michael Pollastri of Northeastern University in Boston says that the trick to faster discovery of drugs reusable against Chagas involves identifying enzymes that we share with the parasite and that have already been targeted by drug designers. Take the phosphodiesterases, enzymes that control communication within cells. A number of medicines that block human phosphodiesterases, including Viagra, are on the market, and drug researchers have scrutinized many other compounds with similar action that never quite made it through development. By retesting those existing medicines and compounds, scientists might uncover parasite killers, he says.

Many factors could still derail progress toward finding new drugs for Chagas disease. One of the biggest obstacles is the lack of a practical way to determine when a drug is successful in humans, says parasite immunologist Rick Tarleton of the University of Georgia, Athens. “We don't have an endpoint indicator that says, ‘Ha, that cures people.’” The gold standard—disappearance from the blood of antibodies against the parasite—can take 10 years or more, which makes it unfeasible for drug trials. One of DNDi's projects aims to find new biomarkers that can more quickly signal when the pathogen has been vanquished.

Despite such concerns, Chagas disease researchers are more upbeat than they have been in years. And yet they are also impatient. “It's like we are going from zero to 20 [miles per hour],” Hotez says. “It's not zero to 60.”

5. Infectious Diseases

# A Tropical Disease Hits the Road

1. Mitch Leslie

Generally considered one of the many neglected tropical diseases, Chagas disease has been drawing increased attention, which some attribute to its growth outside of Latin America.

Generally considered one of the many neglected tropical diseases, Chagas disease has been drawing increased attention (see main text), which some attribute to its growth outside of Latin America. Spain and the United States are seeing the highest incidence among developed countries, but there have been cases as far afield as Switzerland and Japan. The main reason for this rise isn't the spread of insects carrying Trypanosoma cruzi but rather emigration from Latin America of large numbers of people who are already infected.

Because no one has performed comprehensive testing, the number of infected people living in the United States remains unknown, says medical epidemiologist Caryn Bern of the U.S. Centers for Disease Control and Prevention in Atlanta. To improve on previous projections, she and her colleague Susan Montgomery combined new figures on the prevalence of the disease in different Latin American countries with data on immigration into the United States from those nations. Their results, which they published in 2009 in Clinical Infectious Diseases, suggested that about 300,000 immigrants in the United States are likely infected. “It's very much an estimate. We have almost nothing in the way of direct data,” Bern says.

Some small-scale surveys have gathered direct data, however, and they point to a fairly high prevalence in areas with large expatriate populations. Cardiologist Sheba Meymandi of the University of California, Los Angeles, School of Medicine and colleagues have run blood tests on patients with heart failure and other cardiac problems, and on Latin American parishioners from several Los Angeles–area churches. The researchers found that about 1% of the approximately 2000 people they tested harbor T. cruzi. “One in 100, that's pretty substantial,” Meymandi says. Similarly, earlier this year in PLoS Neglected Tropical Diseases, a Spanish group testing Latin American immigrants at a clinic in Barcelona reported that 3% were positive for the Chagas parasite.

An unknown is how many infections in the United States result from native assassin bugs. They live from coast to coast, ranging as far north as Illinois and Pennsylvania, and T. cruzi is prevalent in them. One study found that more than 40% of the assassin bugs in Tucson, Arizona, harbored the parasite. However, since the 1950s, the United States has recorded only a few definite cases of parasite transmission. Most researchers think that the risk from these insects is low, in part because homes in the United States are typically not congenial to the bugs. However, Hotez and some other researchers suspect that such transmission is more common than generally accepted. Meymandi notes that she recently diagnosed two teenage patients who were infected in Los Angeles. “It is here; people do acquire it,” she says.

Whatever its source, the rising incidence of Chagas disease has already required action by health officials in the United States and other developed nations. Although T. cruzi is usually bug-borne, it can also spread through blood transfusions and organ transplants. U.S. blood banks began voluntary testing for the parasite in 2007, and facilities in France and Spain have done the same. And developed countries can expect that the illness's toll will rise. Bern and Montgomery calculated, for example, that if their estimate of the number of infected people in the United States is close to the mark, undiagnosed Chagas disease is responsible for 30,000 to 45,000 cases of cardiomyopathy, or severe heart damage.

6. News

# Past Successes Shape Effort to Expand Early Intervention

1. Jeffrey Mervis

Three longitudinal studies have demonstrated the lasting value of high-quality early education. So why isn't it offered to all children who need it?

It was 1962, and many African-American parents in Ypsilanti, Michigan, didn't see the point of sending their 3-year-old children to a special class at the neighborhood public school rather than just starting them in kindergarten at age 5 or 6. David Weikart, director of special services for this segregated school system in a working-class community near Detroit, had to admit that he didn't really have a convincing argument to overcome their reticence. But he was sure there must be a better way to help at-risk students than simply waiting until they had fallen so far behind their classmates that they were forced to repeat a grade.

Bolstered by the support of a charismatic school principal, Weikart pressed ahead. Over 5 years he enrolled 123 children in what became known as the HighScope/Perry Preschool Project. As researchers have followed the children into adulthood, the findings have shown beyond a doubt that high-quality early education pays off, big time. Two other groundbreaking intervention efforts over the past half-century—the Abecedarian Project in the 1970s and the Chicago Longitudinal Study in the 1980s—have reinforced that message.

The payoff—on everything from better school performance to holding a job, raising a family, staying out of jail, and contributing to, rather than being a burden on, society—can be as much as $16 saved for every dollar spent. That's an impressive return on investment at a time when local, state, and federal officials are desperately trying to reduce large budget deficits and squeeze out the biggest bang for their limited bucks. But as solid as these studies are, they can't provide policymakers with everything they need: a prescription for the best, most cost-effective intervention that can help the largest number of at-risk children. The disconnect comes because studies like the Perry project contain too many important variables to measure. But all of them, from the quality of the teachers to the curriculum to the intensity and duration of the intervention, affect long-term outcomes. Last year W. Steven Barnett, director of the National Institute for Early Education Research at Rutgers University in New Brunswick, New Jersey, spelled out the problem in a meta-analysis of 123 comparative studies over the past 50 years that have tried to enhance cognitive development. “The good news is that a host of original and synthetic studies have found positive effects for a range of outcomes, and this pattern is clearest for outcomes relating to cognitive development,” he wrote. “The bad news is that there is much less empirical information available for designing interventions at multiple levels, with multiple components. … And just as the magnitude of benefit remains somewhat murky, so does the cost [of intervention].” For policymakers, that imprecision can be frustrating. As Senator Barbara Mikulski (D–MD) demanded of one Obama Administration official this summer during a Senate hearing she chaired on the topic: “Everybody talks about high-quality programs. Who doesn't support a high-quality program? High quality is a cliché. We want to know the details.” The landscape of childhood poverty has changed greatly in 50 years. Demographic and geopolitical shifts have widened the gap between the country's haves and have-nots. More women in the workforce has increased demand for child care and early education, but low-income families still have many fewer options than their wealthier counterparts. The federal government is struggling to keep up with those changes. Head Start, begun in 1965, serves nearly 1 million preschoolers from poor families. Even so, that's only 40% of those who are eligible. This summer the White House launched a$500 million Early Learning Challenge, a grants competition for states to improve their network of services to needy preschool children and their families (see p. 957). But early education advocates are worried that pressure to trim spending will bring the budget ax down on the most vulnerable members of society.

Despite all the nonscientific factors weighing on their decisions, government officials still look to these three classic studies for important insights. “What's impressive is how much they got right,” says Joan Lombardi, who oversees early childhood development programs for the U.S. Department of Health and Human Services and who is the person to whom Mikulski put her question. The studies also make clear that the real value of early education is not a higher IQ, or better test scores, but rather the life skills that can help lift someone out of poverty. It may be hard to project such outcomes onto a needy 4-year-old, Barnett says, but they are vital goals in crafting a successful early education program.

## The wonder years

“The economic case for expanding preschool education for disadvantaged children is largely based on evidence from the HighScope/Perry Preschool Program,” says Nobel Prize–winning economist James Heckman of the University of Chicago in Illinois. (Although the intervention itself ended in 1967, in 1970 Weikart formed the HighScope Educational Research Foundation, which has continued follow-up studies.) The project calculated a lifetime benefit of $284,000 from a cost of$17,600 for each child in the preschool program. In fact, Perry's high profile has led Heckman and colleagues at the National Bureau of Economic Research to apply powerful new analytic tools to the original data that clarify how big the long-term payoffs have been to both individuals and society.

The efficacy of intervening before children have a chance to fail was an unproven hypothesis when Weikart launched his study. But it was in step with the growing awareness among educators and child psychologists that early intervention was the key to helping millions of children escape a seemingly endless cycle of poverty.

The school that Weikart chose for his study was the center of the segregated neighborhood's social and cultural life as well as its educational hub. It was led by Charles Beatty Sr., the first African-American principal in Michigan. Beatty also lived in the neighborhood.

“It was truly a community school,” recalls Pat Horne McGee, director of the Head Start program for Washtenaw County, which includes Ypsilanti as well as Ann Arbor. McGee attended the Harriet Street School in the 1950s (its name was later changed to the Perry School), and still lives and works in the community. “There were dances on Friday night, weddings, basketball games, summer activities, you name it,” she recalls. “A lot of the concepts that Dave Weikart was looking for in terms of parental and community involvement were already in place.”

At the time, most mothers stayed at home with their children and didn't see the need for preschool, McGee recalls. “A lot of mothers thought, ‘I'm not sure I want my babies to go with [Weikart].’” Beatty asked his teachers to persuade parents to participate.

Eligible children had a below-average IQ and came from a low-income family whose parents typically had not completed high school. They were randomly as signed to either a half-day preschool program or no intervention. The teachers in the treatment group had advanced degrees, were paid a bonus, and worked in small classes with a child-centered, active learning curriculum.

The study was small, with 58 children in the treatment group and 65 in the control group. The initial results for fourth-grade students were, in a word, disappointing. “They were not very persuasive, to put it bluntly. The achievement test scores showed weak effects,” acknowledges Larry Schweinhart, an education psychologist Weikart hired in 1975 to help with the study and who has since become principal investigator. It wasn't until the two men saw the results from the assessment of 14-year-olds that they realized something special might be happening.

“I had the printouts in my hand and they were just jumping off the page. I thought, ‘This must be a mistake. Why would we get bigger achievement effects at age 14 than earlier?’” In addition to earning higher scores than students in the control group, children from the treatment group required less special education, showed better motivation, did more homework, and placed a higher value on learning. The teenagers also engaged in less delinquent behavior.

“Motivation becomes a much larger factor” in middle school, Schweinhart says, “along with task persistence and other kinds of traits you would associate with being able to do something that's harder. So I think we were tapping into that skill for the first time. And that was pretty exciting.”

The age-14 results “were a real turnaround for us,” says Schweinhart, who succeeded Weikart as president of HighScope after Weikart died in 2003. In fact, the study has shaped the path of his career. Now, at 64, Schweinhart is seeking federal funding to study the cohort at age 50, as well as the impact of the curriculum used.

## From the minds of babes

No study can answer more than a handful of questions. And for the researchers at the University of North Carolina, Chapel Hill, who began enrolling disadvantaged children as young as 6 weeks old in an intensive intervention program called the Abecedarian Project, the biggest one was: How early should you start to make the biggest difference in a child's life?

Starting in 1972, 5 years after the Perry Project ended, the Abecedarian study (“abecedarian” means one who studies the alphabet) was also small, with just 57 children in the treatment group. But the age of the children and the intensity of the intervention set it apart from Perry. “We defined intensity as being in an educational environment 5 days a week, for 5 hours a day, year-round,” ex plains Frances Campbell, the study's principal investigator. The intervention also continued until the children entered kindergarten. By comparison, the Perry preschoolers had attended school for 2.5 hours a day, 7 months a year, for 2 years.

Unlike the Perry study, which took place in a neighborhood school, the preschool portion of the Abecedarian project was conducted at a university-based child-care facility. Participants came from a sprawling rural area that extended far beyond the campus. None of the women (three-quarters of the children came from single-parent families) had cars, and public transportation was nonexistent. Campbell recalls using her own car to bring some of the children and their parents to the center when the regular driver wasn't available.

That personalized attention was a key feature of the study itself. “When the babies in the treatment group were awake and alert, the idea was that someone was always stimulating them, keeping that little mind active and growing,” Campbell says. “There was a heavy emphasis on language development, because studies have shown that's what predicts IQ score. So the idea was to enrich their verbal environment, and we tried hard to do that.”

The boost in academic performance was slightly stronger than in the Perry study. “Our cognitive benefits endured longer than anybody else's,” Campbell explains. And although the biggest difference occurs at age 3, even at 21, “the IQ scores are significantly different for those who attended our preschool.”

The Abecedarian study reports a much lower overall benefit-cost ratio to society, however. That is due largely to a negligible difference in the incidence of crime among treatment and control groups. By comparison, Perry showed a major disparity between its experimental and control groups, which translates into a large societal payoff. “That's where Perry gets the biggest effect, and we don't show the same effect,” Campbell acknowledges. She doesn't know why that is.

Despite the study's acclaim—the ranking Republican on the Senate panel, Richard Burr of North Carolina, gave a shout-out to his home-state institution immediately after Mikulski cited the Perry results in opening the hearing—Campbell readily admits that it still leaves several issues unresolved. Ironically, she wasn't able to answer the question posed by developmental psychologist Craig Ramey when he launched the study: How early should we intervene? (Ramey is now at the Virginia Institute of Technology Carilion Research Institute in Roanoke, the latest stop in a distinguished academic career.) Campbell says the only way to measure that would be a massive study with interventions beginning at different times: 3 months, 6 months, 1 year, 2 years, and so on. But such a study would be prohibitively expensive, she says.

Campbell remains actively involved with Abecedarian, with an upcoming paper in press reporting on the age-30 cohort. But she's not sure how much longer the study will continue to bear fruit, and who will harvest it. “I'm wrestling with whether we really want to see [the results] at 40,” she says. “Part of me wants to know. But I'm 78, and I don't expect to live to find out.”

When politicians think about expanding early education, money is inevitably one of their biggest concerns. Perry cost $17,600 per child, and Abecedarian ran four times that amount. Although the eventual return on investment to society for each child enrolled is hugely positive, most politicians aren't known for their long-term thinking. The Chicago Longitudinal Study (CLS) began in 1985, and some children received services for up to 6 years. It was designed as a low-cost alternative to Perry and Abecedarian. Its overall size—1539 at-risk children—is proving to be big enough to tease out what Arthur Reynolds, a community psychologist at the University of Minnesota, Twin Cities, and principal investigator for CLS, calls “the active ingredients” of the intervention: the components that are most important to its success. CLS's experimental population of 989 children attended preschool at one of 20 Child Parent Education Centers, a program begun by Chicago public schools in 1967 on the heels of Head Start. After either 1 or 2 years of the intervention, a subset of the children continued to receive services through grade 3 at one of several Chicago public schools. Unlike its predecessors, CLS did not assign children randomly to treatment and control groups. Two decades of Head Start had made the value of early intervention so well-known to the public, Reynolds says, that “it would have been unfeasible and unethical to do a randomized trial. Any parent whose child would have been assigned to the control group would have bailed.” Instead, the study follows what is called a quasi-experimental design, with its control group consisting of children attending full-day kindergarten at five randomly chosen schools. Most had previously been in home care and never attended preschool, although 15% had participated in Head Start. Reynolds calls the 6-year course of treatment for the experimental group “a happy integration of basic skills, both for language and numeracy, as well as social-emotional skills, combined with an emphasis on parental involvement.” His results found cognitive benefits from the intervention beginning at an early age. Children entering kindergarten without an early intervention experience, for example, were well below the norm on one widely used test of basic skills, while those with 1 year of treatment had average scores, and those with 2 years were well above the norm. The positive outcomes are sustained as the children move through school: The treatment group is half as likely to need special education classes, has one-third fewer juvenile arrests, and has nearly half the rate of substance misuse. The payoff for the young adult cohort is also impressive, including an 8-to-1 benefit-to-cost ratio at age 27 for those who received the full 6 years of treatment. That positive ratio translates into a higher socioeconomic status, less substance abuse, and a lower rate of incarceration. Reynolds believes the characteristics of CLS make it more relevant than the Perry and Abecedarian studies. And he's not above dispensing what passes for trash talk among early education researchers. “CPC had already become institutionalized within the school system. So in terms of sustainability, it's a much better model than a university-based program like Abecedarian or something run by an outside organization, like HighScope,” Reynolds asserts. Putting four teachers with master's degrees into every classroom, as Perry did, “just isn't going to happen today,” he adds. “Abecedarian generated strong findings, but at a huge cost per child,” Reynolds says, making CLS's price tag much easier for politicians to swallow. ## The way forward There's no question that the children enrolled in the Perry, Abecedarian, and CLS studies benefited from the intervention. The evidence also suggests that those at greatest risk for failure in both school and life received the greatest boost. But can similar intervention programs have the same effect on the country's changing population of at-risk children, in particular a growing Latino population and English-language learners? “Do we need more rigorous long-term studies that address the current nature of poverty? Yes, but they are horribly expensive,” says Grover J. “Russ” Whitehurst, director of the Brown Center for Education Policy at the Brookings Institution in Washington, D.C. “I don't see much prospect of any federal agency launching them anytime soon.” A way around this is to access data already collected for other purposes. One such effort to use so-called administrative records taps into a famous study of 11,500 children in 79 Tennessee elementary schools in the late 1980s. Although Project STAR (Student/Teacher Achievement Ratio) was designed to measure whether students did better in smaller classes, it found instead that teacher quality was a more important factor than class size. A team led by economist Raj Chetty of Harvard University decided to track down the children as adults. Using tax returns from the Internal Revenue Service scrubbed to preserve anonymity, they obtained information such as college attendance, earnings, home ownership, and retirement savings. As with Perry in particular, Chetty found that high-quality kindergarten has minimal long-term impact on reading and math scores. But it notably improves the set of noncognitive skills—think of a teenager who pays attention in class, keeps good notes, and carefully lists his homework assignments, compared with the one with smudged papers stuffed into his backpack, a pattern of disruptive behavior, and no recollection of tomorrow's Spanish test—that persists into adulthood. These skills could explain why some children benefit more than others from the same intervention, as well as the efficacy of the intervention itself, says James Griffin, who heads the early learning and school readiness program at the National Institute of Child Health and Human Development within the National Institutes of Health. Many children who haven't received intensive, high-quality preschool are able to pick up enough academic skills in the early grades to catch up to their peers, says Griffin, whose program has funded Reynolds's CLS work. But they can't always compensate for the intangibles. “For things like better self-regulation, delayed gratification, the ability to concentrate, motivation, and developing a love of learning,” Griffin says, “we want to figure out the best way to teach those skills.” ## Still a special place The wing of the Harriet Street School that housed the Perry study is no more: It was torn down several years back to accommodate a new school, now called the Perry Child Development Center. During a recent visit, McGee recalled that apple trees once encircled the building. And she realizes how much time has passed when Sharine Buddin, the principal, walks her down the hall of the “old portion” of the school that didn't exist when McGee was a student there. But local educators haven't forgotten the seminal role of the Harriet Street School in demonstrating that high-quality early education works. Although the world has changed a lot in 50 years, they still try to honor its memory. When Ypsilanti, with a shrinking school-age population, decided 6 years ago to consolidate all its early education classes into one building, it chose Perry. Buddin, who had been principal of a grade 1 through 5 school that was closed, assumed she would be going to another grade 1 through 5 school. So she asked the school superintendent why he wanted her to move to Perry, now a large, modern brick building that houses 11 kindergarten and 10 first-grade classes along with three preschool classes. “‘This is a special place, and I want you to take care of my babies,’ Dr. Hawkins told me,” Buddin says. “And I said, ‘You're right. It is special.’ You couldn't drag me out of here now.” 7. News # Giving Children a Head Start Is Possible—But It's Not Easy 1. Jeffrey Mervis Begun in 1965, the U.S. Head Start program is still trying to find the best way to prepare poor children for school. Nearly 1 million children aged 3 and 4 will attend Head Start this year. As its name implies, the U.S. government program is meant to give low-income children (and those with disabilities) a boost. In this case, that means helping get them ready for kindergarten by offering the type of educational foundation that children from wealthier families take for granted. Of course, a good preschool education isn't the only thing that these children may be lacking. There are many other federal programs that provide nutritional, health, and child care assistance to the most disadvantaged families in the United States. And despite a budget this year of$7.56 billion, Head Start serves only 40% of the eligible children. (Early Head Start, begun in 1995 for an even younger population, serves fewer than 5% of eligible infants and toddlers.) Even so, many people still consider Head Start, created in 1965 as part of President Lyndon Johnson's Great Society programs, to be the government's chief vehicle for ensuring that those children do well in school and grow up to become productive members of society.

Has Head Start been a success? Unfortunately, that's not an easy question to answer. Overly broad expectations for a program focused on only one consequence of poverty make it hard to evaluate its real impact. On one hand, low-income parents have other publicly funded educational options for their children, notably state prekindergarten programs and subsidized child care, so the existence of good alternatives may narrow the gap in any comparison study with Head Start. On the other hand, any positive effects from Head Start could be suppressed by other aspects of childhood poverty or erased by low-quality education once they enter school.

Nevertheless, even its supporters agree that Head Start is not as good as it could be. The salaries of Head Start teachers are much lower than those offered at regular public schools, for example, and historically the quality of instruction has received much less attention. It also typically runs only during the school year, and few children spend 2 years in the program (4-year-olds receive preference because of the limited number of slots). As a result, Head Start has never racked up anything like the long-term payoffs to individuals and society documented in the Perry, Abecedarian, and Chicago Longitudinal studies that provided high-quality, intensive early education (see main text).

The latest evidence of what some consider Head Start's disappointing track record came last year in a congressionally mandated evaluation of 4667 children at 383 Head Start centers across the country. The first round of results from the so-called Impacts study, issued in 2005, found that the children enrolled in Head Start at either age 3 or 4 in 2002–03 were better prepared for kindergarten than a control group from similarly disadvantaged backgrounds who, by and large, did not attend Head Start before entering kindergarten. But the second round found essentially no difference in academic performance between the two groups of children at the end of first grade. The results of the Impacts study replicated the so-called fade-out effect reported in previous, smaller studies of children in Head Start.

But those results aren't a surprise, say the program's supporters, nor do they mean that Head Start is a failure. Edward Zigler, an eminent child psychologist at Yale University and one of the founders of the Head Start program, says that the confusion over what Head Start is supposed to accomplish makes it easy to misinterpret the results of the Impacts study.

“Over the years, scientists, policymakers, and the public have developed unreasonable expectations that Head Start should raise IQ scores, lift children and families out of poverty, and close the achievement gap between poor and more affluent children,” he wrote in a briefing paper for the National Head Start Association a few months after the study was released. But Zigler and others believe that Head Start does measure up if one applies the more modest yardstick of improved school readiness, which combines health, social, and emotional factors as well as early learning skills. “Although the initial findings were not as robust as hoped,” he wrote, “they clearly indicated that, by the end of the Head Start experience, children were more ready for school entry than those in the control group.”

Recent changes to Head Start, some already on the books and others in the works, are moving the program in the right direction, according to many experts. A 2007 law, for example, requires by 2013 that half of Head Start teachers hold bachelor's degrees, and last year the Obama Administration proposed new rules to increase accountability, including forcing low-performing programs to recompete for funds and grading teachers on their interactions with students. “Head Start has never been subjected to the kind of scrutiny that the Obama evaluation promises,” early childhood advocates W. Steven Barnett of Rutgers University in New Brunswick, New Jersey, and Ron Haskins of the Brookings Institution in Washington, D.C., declared in an op-ed last year in The Washington Post. “Now it seems likely that, within a few years, the worst Head Start programs will be shut down.”

However, others insist that Head Start will never attain the level of quality instruction and care provided children in the Perry, Abecedarian, and Chicago studies. One prominent critic, Douglas Besharov, a public-policy expert at the University of Maryland, College Park, wants the government to declare that Head Start is broken and then experiment with alternative approaches. His choices would be to focus on the neediest children and families—“teen mothers, largely African-American, who have been held back by years of exploitation and discrimination”—and to give parents a bigger role in both policy and management decisions.

But Besharov, who regards the Perry and Abecedarian studies as outliers rather than evidence of what high-quality early intervention can achieve, says he understands why supporters are reluctant to take such a drastic step with a program that serves an important government function. “If the president were to stand up and say, ‘Hey, I've read Besharov's stuff and now I realize that Head Start doesn't work,’ the long knives would come out and Congress would try to defund it.”

Head Start can't be fixed without looking at the larger question of what services these children need from birth through age 8, says J. Lawrence Aber, a professor of psychology and public policy at New York University and a member of an expert panel advising the Department of Health and Human Services on how to follow up the 2010 Impacts study. “The real question is how to provide these children with the most developmentally nutritious experiences,” says Aber, who would like the 2002–03 Head Start cohorts to be tracked through adolescence. The nation's $600 billion investment in elementary and secondary school actually helps low-income children keep up with their peers during their school years, he notes. “But they are at a comparative disadvantage during the first 3 or 4 years, as well as after high school, because their parents can't compensate for what the other children are getting.” Some observers believe that the fadeout effect says more about the quality of the nation's public schools that the children attend once they leave Head Start than about the efficacy of the program. “Perhaps the question should be reversed,” says Lisa Guernsey, who monitors early education efforts for the nonprofit New America Foundation in Washington, D.C. “What can we do to help elementary schools maintain the momentum that children receive from Head Start?” 8. News # A Passion for Early Education 1. Jeffrey Mervis Joan Lombardi brings experience and energy to the Obama Administration's efforts to better coordinate federal children's health and education programs. J. Lawrence Aber means it as a compliment when he calls Joan Lombardi “short and loud.” Aber, a professor of psychology and public policy at New York University, says the 5′1″ (1.5 m) Lombardi has stood tall across three Administrations to improve services for young children and their families. And, he adds, sometimes it helps to be pushy. “She's enormously savvy, both policy-wise and politically,” says Aber, a member of a panel reporting to Lombardi on how to improve ongoing evaluation of the Head Start preschool program. “She's passionate but no wide-eyed idealist. She's profoundly pragmatic.” Lombardi, the deputy assistant secretary of the Administration for Children and Families within the U.S. Department of Health and Human Services (HHS), has served under all three Democratic presidents since Lyndon Johnson, whose domestic policy agenda laid the cornerstone for some of the programs she now oversees, notably Head Start. But she says it's the second part of her title—interdepartmental liaison for early childhood development—that was the biggest drawing card this time around. The phrase, she says, reflects President Barack Obama's endorsement of the idea that “health and education go together”—a concept that took a hit when her first boss, Jimmy Carter, decided to create a separate Department of Education, splitting it off from HHS. The two departments are expected shortly to lay down the rules for the Early Learning Challenge (ELC), a new$500 million competition that is intended to improve the current patchwork of early education services across the country. ELC was carved out of a $700 million allocation in the 2011 budget, approved belatedly by Congress in April. The competition is modeled on the Education Department's$4 billion Race to the Top program that awarded money to 12 states to improve their elementary and secondary schools. States can win up to \$100 million for proposing ways to tackle problems such as tracking and assessing the progress of children from their earliest years through high school and providing parents with more information on the quality of publicly funded child care and early education programs.

ELC aims for results, says Lisa Guernsey, who monitors early education issues for the nonprofit New America Foundation in Washington, D.C. “Joan Lombardi's focus on improving the quality of child care and Head Start programs has been instrumental in pushing the early childhood field to recognize its responsibility to up its game, to focus on teaching, and to professionalize the workforce,” she says.

Once the money was in place—attempts by the Obama Administration to get it funded in 2009 and 2010 came up short in Congress—Lombardi pushed hard to get ELC off the ground. That energy was visible when Science interviewed her on 30 June. Lombardi had plastered on her office door the “visitor” pass she had been issued for every meeting with her Education Department colleagues in the past 6 weeks; the tally stood at 40.

Q:What are the active ingredients in a high-quality early education program?

J.L.:Qualified teachers and a focus on the child and the family are certainly important. The findings say it's better to start early and continue through the early years. We've also learned that standards matter and how the variation in quality affects outcomes. Programs that adhere to the standards were getting better outcomes. … Obviously, the country's demographics have changed a lot [since the Perry, Abecedarian, and Chicago studies; see p. 952], and the conditions have changed a lot. Poverty has changed. We have a diverse population, with more working families. Half-day programs fit into their lives, so hours matter.

J.L.:We're studying them now, and we need to continue to do more research. We're hearing a lot about maternal depression and how that factors in. We're learning more now about the things we can do—and what's impressive about the early studies is how much they did right—and one of them is to make sure that there are enough instructional supports in the classroom, to get more intensive outcomes. … We're putting a lot of focus on the classroom interactions between the teacher and the child, and the social and emotional environment in the classroom, because we know that those things are predictive of later success.

Q:Do we know which factors have the greatest impact?

J.L.:We're looking at all the research. One question is whether there's a threshold effect, that is, a minimal level that you have to reach in order to have a positive impact. I think this is, in many ways, the agenda for the next decade. The program-evaluation data are telling us that, across all the studies—and it's pretty amazing that they are all going in the same direction—that you can intervene and make a difference. … But I think that we need to be realistic, too. One child could be involved in multiple programs over the course of her first 5 years and even within the course of a single day. They could go to a Head Start program in the morning, then go to a child-care program in the afternoon, or go to a pre-K [kindergarten] program in the morning and Head Start in the afternoon. So which is the program that is having the impact?

Q:Why is it important to start young?

J.L.:We have the data about what children look like when they come into Head Start. And they are already behind [developmentally] by as much as a standard deviation. So by age 3 you're already playing catch-up.

We also know that the quality of infant-toddler care requires more attention. So why not build it in? If you look at what we administer here, in addition to Head Start, we do the Child Care and Development Fund. And we serve many more children under 3 in the child-care system than we do in Early Head Start, but with very little focus on standards or on instructional supports or on what the workforce should look like.

Q:Is there a lot of reinventing the wheel?

J.L.:There's so much going on. This is what I'm saying to the research community. Help us sort through what we're learning so that we can translate that into what people should do on the ground. And we need to do a much better job of that.

When I first came to this agency—in 1978, as a HEW fellow during the Carter Administration—I worked on a report on how to put research into action. And I think we need a lot more of that. That's the spirit that we're bringing to the advisory committee: synthesize the research and tell us how to put it into practice. We also do need more longitudinal studies. And we need short-term studies—up to grade 3, perhaps—that tell us what will contribute to a more successful third- or fourth-grader.

Q:What's the biggest political obstacle that you face?

J.L.:I think that people want to invest in something that works. And we think that Head Start works. … Money matters, but so does good management and oversight. … We want to make sure that Head Start works well for all children and that the gains are sustained.