Washington, Dakota Counties Report On Cancer Occurrence

Armen Hareyan's picture

Overall cancer rates in Washington and Dakota counties are very similar to the rest of the state, even slightly lower.

In addition, the rates and types of cancers that occurred within specific communities in those two counties were generally on a par with other communities in the metropolitan area, the report says.

The analysis of data from the Minnesota Cancer Surveillance System (MCSS) was undertaken in response to concerns expressed by area residents and elected officials over people's exposure to perfluorochemicals (PFCs) in their drinking water. Some communities in the area have groundwater concentrations that exceed MDH health based values for certain PFC-family compounds, while other communities have levels that are below HBV but still high enough that some residents are concerned.

Studies in laboratory animals indicate that at higher doses, PFCs may interfere with liver and thyroid function and may cause developmental effects. Research to date has shown no direct evidence that PFCs cause health problems in humans. At public meetings hosted by MDH and other state agencies, some area residents have questioned whether specific incidents of cancer or other illnesses are related to their exposures to PFCs. In general, the presence of PFCs has led to concerns among residents of Dakota and Washington counties about whether community cancer rates are unusually high in communities where PFCs have been detected.

"Given the level of concern, we felt it was important to address public concerns by examining cancer rates in these counties," said Dr. Alan Bender, manager of the Chronic Disease and Environmental Epidemiology Section of MDH. "While this type of analysis of cancer data cannot and does not address the question of whether PFCs can be linked to cancer or other health effects in humans, this report can be very helpful in informing the public as to the actual occurrence of cancers in their communities."

The report provides detailed profiles of cancer rates among residents of Dakota and Washington Counties. Using MCSS data for the 15-year period 1988-2002, it examined county-wide cancer rates for all cancers combined and for each of about 25 of the most frequent types of cancer, including liver and thyroid. In addition, analyses were also conducted to examine incidence rates for selected (16) cancers for specific communities, (by zip code), within each county. For that analysis, data from 1996-2004 was used, largely due to population growth and limitations on community census data.


In Washington County, there were 4,397 new cancers diagnosed among males, compared to 4,549 that would be expected for this group, a significant 3 percent deficit. Among females, there were 4,263 cancers diagnosed, nearly identical to the 4,261 expected.

In Dakota County, there were 7,479 new cancers diagnosed in males, compared to 7,702 that would be expected for this group, a significant 3 percent deficit. Among females, 7,583 cancers were diagnosed and 7,440 cancers were expected, an excess that was not statistically significant, according to analysts.

When the rates of specific types of cancer were examined, they were comparable to the statewide averages also. In those few instances where rates for certain types of cancer were noticeably higher or lower, the difference is most likely due to the natural, random variability in cancer occurrence, Bender said.

For the eight communities where PFC contamination has raised health concerns (Cottage Grove, Hastings, Lake Elmo, Newport, Oakdale, South St. Paul, St. Paul Park and Woodbury), the distribution of 16 cancer types was similar to the metro-wide pattern. For the most part, the observed and expected numbers of cancers were similar. There were some rates that were significantly high, some that were significantly low, and these varied according to gender and occurred over a number of communities. No pattern emerged.

"It is often very difficult to look at cancer rates in small populations and discern any kind of useful or informative pattern or attribute any kind of pattern to low-level environmental exposures," Bender said. "There are many factors that need to be accounted for: many types of cancer are generally very prevalent in our society as a whole, so pockets of excess will naturally appear from time to time and place to place; risk factors for cancer can include age, gender, race and family history, as well as lifestyle choices such as smoking, alcohol consumption, diet, exercise and so on."

One of the most telling and problematic factors for looking at cancer rates in the Washington County area is that a large percentage of the population did not reside there more than five years ago, yet the latency for most cancers is typically decades long, Bender noted.