The Agriculture Department has dropped its demand that staff scientists label peer-reviewed research as “preliminary,” after angry protests followed a Washington Post story disclosing the policy.

But the latest guidelines, released on Wednesday, for internally reviewing science within the department raise additional questions about scientific integrity, said non-USDA researchers who inspected the guide.

Since July, the department required peer-reviewed studies to include a disclaimer, The Post reported last month. Some finalized reports came with a caveat — these findings and conclusions are “preliminary” and have “not been formally disseminated by the U.S. Department of Agriculture.”

USDA employees, editors of scientific journals and science advocates worried the disclaimer might be used to undercut scientists whose research was at odds with Trump administration policies. They also were concerned the language was confusing and potentially misleading.

The scientific community, generally speaking, does not expect peer-reviewed journals to publish preliminary results. The disclaimer, though enacted only as a temporary policy, appeared in several published articles, including a report on the best practices for capturing wild pigs.

This week, acting USDA chief scientist Chavonda Jacobs-Young released a memo that replaced the July policy. It requires the following language when disclaimers are necessary: “The findings and conclusions in this [publication/presentation/blog/report] are those of the author(s) and should not be construed to represent any official USDA or U.S. Government determination or policy.”

Brooks Hanson, vice president of science at the American Geophysical Union, an organization that includes soil and agricultural researchers, said the updated disclaimer is a “welcome” change.

“The previous wording that required scientists to label their research submitted for publication as ‘preliminary’ would downplay the significance and importance of their work and cause confusion among readers,” Hanson said. The new language is closer to other disclaimers required by federal departments. “It is a positive development that USDA seemed to listen to the criticism and feedback.”

Still, he added, even this language may not be needed: “Many journals have this statement on their mastheads. This expectation, that an article represents the views of the authors only, is indeed the standard.”

Rebecca Boehm, an economist at the Union of Concerned Scientists, a D.C.-based organization that advocates for scientists, said that “removing ‘preliminary’ from the disclaimer is a step in the right direction, but there still may be unnecessary obstacles preventing agency researchers from publishing their work in peer-reviewed journals.”

Not every study published by a USDA scientist is required to have this disclaimer. Some research agencies at USDA, including the Agricultural Research Service, the Economic Research Service, the National Agricultural Statistics Service and the Forest Service have “agency-specific policies” that determine when a disclaimer is appropriate, said William Trenkle, the USDA scientific integrity officer.

Others, such as the National Institute of Food and Agriculture and the Animal and Plant Health Inspection Service, have been encouraged to follow the new guidelines.

Those outline when disclaimers are needed and also describe the department’s internal review process before scientists can publish results in journals. It lists several “flags” that may trigger additional scrutiny. Some flags, under the umbrella of “prominent issues,” include “significant” scientific advancements, the potential to attract media attention, and results that could influence trade or change USDA policy.

Members of other USDA agencies, including the office of the chief scientist, the communications office and the office of Agriculture Secretary Sonny Perdue, can request “corrections” and “changes,” if the scientific publication falls under the category of “prominent issues” or policy implications.

“It is the responsibility of reviewers at all management levels to identify whether the manuscript contains prominent issues,” Trenkle said.

If USDA or other federal “stakeholders” request alterations based on these issues, then study authors should return “the REVISED PUBLICATION” for “ACCEPTANCE of the CHANGES and/or RESPONSE to comments,” the guideline says [capitalization is in the original].

Trenkle said the policy conforms with the USDA’s scientific integrity rules and “helps to ensure that USDA scientists communicate their scientific findings objectively without political interference or inappropriate influence.”

But Susan Offutt, who was the administrator of the Economic Research Service under presidents Bill Clinton and George W. Bush, said the guide twists internal review “into a process by which policy officials get the final say on content.” Because researchers at the Economic Research Service publish statistics to aid policymakers, “just about any output” from that agency could be flagged, she said.

USDA’s “interests apparently concern consistency with prevailing policy,” Offutt said, “not the public’s access to the best, unbiased science and analysis.”

Read more: