Salinas, California

JOSÉ Camacho once worked the fields here in the Salinas Valley, known as "the Salad Bowl of the World" for its abundance of lettuce and vegetables. His wife still does.

But back in 2000, Camacho, who is 63, got an unusual phone call. He was asked if he wanted to work for a new project studying the effects of pesticides on the children of farmworkers.

The project, run by scientists from the University of California, Berkeley, and funded in part by the Environmental Protection Agency (EPA), is still going all these years later. Known as Chamacos, Spanish for "children", it has linked pesticides sprayed on fruit and vegetable crops with respiratory complications, developmental disorders and lower IQs among children of farm workers.

State and federal regulators have cited its findings to help justify proposed restrictions on everything from insecticides to flame-retardant chemicals. But the Trump administration wants to restrict how human studies like Chamacos are used in rule-making.

A government proposal this year, called Strengthening Transparency in Regulatory Science, could stop them from being used to justify regulating pesticides, lead and pollutants such as soot, and undermine foundational research behind national air-quality rules. The EPA, which has funded these kinds of studies, is now labelling many of them "secret science".

Studying disease trends in specific groups of people - a branch of medicine known as epidemiology - started to gain currency at the EPA in recent years. These studies can be difficult because they require adjusting for all the various substances people are exposed to beyond pesticides.

But researchers had amassed years of data from a wave of compelling chemical studies begun in the 1990s, giving regulators a new body of research to incorporate into their decision-making. Under the Obama administration, the EPA, which had long favoured tests on rats and other laboratory animals in its pesticide regulation, began considering epidemiological studies more seriously.

The agency leaned on this type of research in proposing to ban an insecticide called chlorpyrifos in late 2016, and has been repeatedly prodded to take action on the chemical by federal courts.

But weeks after Donald Trump was elected president, CropLife America, the main agrochemical trade group, petitioned the EPA to "halt regulatory decisions that are highly influenced and/or determined by the results of epidemiological studies" unless universities were forced to share more of their data.

Industry leaders aggressively challenged such studies in high-level meetings and e-mails with EPA leaders, according to thousands of pages of documents obtained through Freedom of Information Act requests.

Gary W Van Sickle, executive director of the California Specialty Crops Council, wrote to the agency last September that "there have been serious flaws with EPA's conclusion to use these data". The council, representing growers of crops as diverse as carrots, garlic, pears and peppers, cited "inappropriate use of the epidemiology". The EPA, whose new leadership is seeded with industry veterans, has responded.

In a mid-July assessment of atrazine, a widely used weed killer long banned in Europe, the agency reviewed and dismissed 12 recent epidemiological studies linking the herbicide to such ailments as childhood leukaemia and Parkinson's disease. It echoed the conclusions of research funded by Syngenta, atrazine's manufacturer, finding the chemical unlikely to cause cancer. Before scandals forced Scott Pruitt out last month as head of the EPA, he proposed the transparency regulation. It would ban many epidemiological studies, and other outside research, unless more data behind the studies was made public.

In doing so, he revived a strategy advanced for years by congressional Republicans and corporate interests like tobacco companies. "The era of secret science at EPA is coming to an end," Mr Pruitt proclaimed at the time. The agency's new acting administrator, Andrew Wheeler, says he's moving forward with the proposal, as the agency re-evaluates a class of widely used insecticides, called organophosphates, that have been the subject of numerous epidemiological studies like Chamacos.

But academics and state health officials say universities are being pressured to release data that would ultimately divulge the identities of study participants, a strategy once used by tobacco companies seeking to undermine research on the dangers of smoking. While participant data is shared with regulators in drug trials, academics fear that the EPA's proposal would additionally require divulging confidential personal information, potentially violating privacy regulations for federally funded research. "It is a naked attempt to use a false claim that something nefarious is going on with these studies in an effort to allow industry to challenge conclusions that are not in their favour," said James Kelly, a manager of environmental surveillance at the Minnesota Department of Health.

At the EPA, the debate swung in favour of epidemiology. While such studies are often complex and can be of varying quality, the agency was reluctant in the past to give them as much weight as lab experiments on animals. But by the Obama administration's final months, the agency moved for the first time to ban a pesticide largely because of epidemiological research. The pesticide, chlorpyrifos, was the same one ingested years earlier by unwitting Nebraskans. It is applied to crops like apples, oranges and strawberries to combat insects like spider mites and sap-sucking bugs. In California alone, chlorpyrifos was sprayed on 640,000 acres in 2016, according to state data. And research from Salinas, and the Chamacos study, became a central element in the EPA's recommendation.

"There is a breadth of information available on the potential adverse neurodevelopmental effects in infants and children as a result of prenatal exposure to chlorpyrifos," the agency concluded in 2016, also citing epidemiological research from Columbia University and the Icahn School of Medicine at Mount Sinai. The pesticide industry's reaction was loud and intense. Monsanto, in emails with the EPA, was dismissive of critical epidemiological research related to Roundup, writing that "such studies are well known to be prone to a number of biases".

Dow Chemical said in reports submitted to the EPA that "the evidence from these studies is insufficient" and called chlorpyrifos a "proven first-line of defence" against new pest outbreaks. A month after taking over the EPA, Mr Pruitt acted. He disregarded agency scientists and rejected the proposed chlorpyrifos ban, later calling for "a new day, a new future, for a common-sense approach to environmental protection".

The EPA and other government agencies have spent millions of dollars funding Chamacos. Half the Chamacos children have been tracked since before birth. Researchers have collected 350,000 samples of blood, urine, breast milk and even household dust and spent nearly two decades studying maturing children. They perform neurodevelopmental and physical assessments and study factors like diet and school performance. After nearly two decades, the study's data appears in more than 160 academic papers.

During a visit to the Chamacos office in Salinas, Brenda Eskenazi, the director of the project and a professor of epidemiology at Berkeley, was testing out brain monitoring equipment, wearing what looked like a black swim cap strewn with knobs and wiring.

Researchers have had wins: This month, a federal appeals court ordered the EPA to ban chlorpyrifos, citing findings from human studies. The Trump administration is mulling whether to appeal. But epidemiologists are unsettled. In mid-July, after nearly two decades of work on Chamacos, the EPA e-mailed Prof Eskenazi requesting "the original data" from her research, citing "uncertainty around neurodevelopmental effects associated" with pesticides she has studied. The agency made a similar request to Columbia. Prof Eskenazi, worried about her study participants' privacy, alerted university lawyers. She is now concerned that the EPA may try to undermine her study's repeated findings that some pesticides may be harming children. NYTIMES