The Minnesota Department of Health said Wednesday it has found no clusters of cancer, premature births or low-birthweight babies in parts of Washington County where groundwater was contaminated years ago by a 3M Co. chemical.

The review, prompted by residents’ concerns, flatly contradicts the conclusions of an expert hired by state Attorney General Lori Swanson and comes just days before the start of a long-awaited trial between the state and 3M. The chemicals’ potential health implications are a major component of Swanson’s claims that Minnesota incurred up to $5 billion in damage to its natural resources.

In their report, state health officials said that, while they are concerned about the potential health impacts of the compounds known as PFCs, they employed widely accepted statistical methods used in public health to reach their conclusions. The findings, they noted, are similar to those they reported in 2007 and 2015.

But an internal Health Department document obtained late Wednesday by the Star Tribune indicates that the agency is on a collision course with Swanson’s office over the conflicting public health findings. The document states that underlying data have problems because Health Department scientists were rushed to complete it. In it, a department epidemiologist says that while he is confident in the study’s conclusions, “the cancer portion will be weak; much below our historical standards.”

The Health Department said later Wednesday that the concerns raised in the document “were subsequently addressed before the release of the reports. These concerns did not pertain to the conclusions of the analyses, but rather with a desire to ensure that we could develop additional examples and background information to more easily explain our analyses to the community. The reports had full and proper technical review.”

Said Swanson: “I can only conclude from this that the agency is embarrassed because it is so late to the table in protecting the public health.”

3M officials said they will likely make the disagreement a question for the jury next week. “We do not believe there is a PFC-related public health issue in Minnesota and look forward to discussing the MDH report with the State during trial,” the company said in a statement.

In releasing Wednesday’s report, the Health Department said the state’s long-term efforts to protect Washington County residents from contaminated drinking water have paid off.

The compounds — known as perfluorinated chemicals, or PFCs — were made by 3M at Washington County facilities for decades and, with industrial waste, legally dumped at nearby landfills. Their nonstick, dirt-repellent qualities made them attractive for industrial processes and consumer products.

Over time, however, a series of studies has linked PFCs to various health problems in humans and laboratory animals.

State health officials said they reanalyzed birth and cancer data for the southeast metro communities that have long been affected by a 100-square-mile plume of contaminated groundwater. Their report covered rates of low birthweight babies and premature births in three different time spans from 2001 to 2015, compared to unaffected communities in the rest of Washington County and the metro region. While they found considerable variation, it was within the range that would be expected under ordinary circumstances, said Jessie Shmool, a state epidemiologist.

They also examined rates for 24 types of cancer from 1988 to 2012 and found no significant differences compared to Minnesota’s overall population and residents in unaffected areas.

They added that this doesn’t mean PFCs are harmless. Numerous studies have linked PFCs to higher rates of some cancers, developmental problems in children, high cholesterol and weaker response to immunizations. As a result, in recent years federal and state health officials have drastically lowered the concentrations considered safe in drinking water. Minnesota health officials have led in that effort.

But the report does contradict some of the findings of Swanson’s hired expert, David Sunding, which raised alarms in east metro communities when they were made public in November. For instance, Sunding, a natural resources economist at the University of California, Berkeley, found that from 2001 to 2006, Oakdale mothers exposed to contaminated drinking water were 34 percent more likely to deliver premature babies compared with county mothers who did not have contaminated water.

After 2006, when Oakdale switched to an uncontaminated municipal water system, that rate dropped to 13 percent, which he said fit with a gradual decline in blood concentrations.

While cancer rates are more difficult to assess, Sunding said, he found similar patterns with both adult and childhood cancer cases in Washington County, the smallest geographic unit available. Prostate cancer, for example, was 30 percent higher there than in other Minnesota counties from 2005 to 2012, he said.

Sunding said Wednesday that regardless of the Health Department’s conclusions, some of its data show trends similar to his own. For example, rates of premature births in Oakdale showed a decline after 2006, just as he found. And they did find some unusually high rates of childhood cancer.

“Despite what the report says, they confirm important findings,” he said.