Better Health Research for Rocky Flats: Study the People Who Were Actually There

For decades, people who lived near Rocky Flats have asked whether living near the nuclear weapons plant affected their health. It is a difficult question. It deserves rigorous research.Colorado has conducted cancer-registry analyses around Rocky Flats, and those studies provide useful information. But there is an important difference between monitoring cancers diagnosed within a geographic area and studying the long-term health of people who actually lived near Rocky Flats during historically relevant periods. Rocky Flats Downwinders believes that distinction matters.

We are not asking researchers to begin with the assumption that Rocky Flats caused the cancers or illnesses reported by members of our community. We are asking for something more fundamental.

Study the people who were actually there.

What question are we trying to answer?

When former Rocky Flats-area residents contact RFD with health concerns, the question they are usually asking is not: How many cancers were diagnosed among people living in this geographic area during a particular period? They are asking: What happened to the health of the people who lived near Rocky Flats during periods when releases, fires, waste handling and other events may have created opportunities for exposure? Those sound like similar questions. Scientifically, they are not.

That distinction is at the center of RFD's review of the Colorado Department of Public Health and Environment's Rocky Flats cancer analyses, our Truth and Transparency Project.

The problem with residence at diagnosis

Cancer registries generally record where a person lived when their cancer was diagnosed.

That makes sense for many public-health surveillance purposes. But it creates an important limitation when trying to investigate a historical environmental exposure. Imagine someone who grew up several miles from Rocky Flats during the plant's operating years. That person moves to another state at age 25 and develops cancer at age 55. The cancer registry records the diagnosis where that person lives at 55. Their decades-earlier residence near Rocky Flats is generally not reconstructed by a geographic cancer-registry analysis. Now consider the reverse.

Someone moves into a community near Rocky Flats many years after plant operations ended and later develops cancer. That diagnosis may be counted in an analysis of the current geographic population even though that person was not living there during the historical period being questioned. Neither situation means the cancer registry is inaccurate.

It means residence at diagnosis is not the same thing as residence during a potentially relevant exposure period. And that matters enormously when the disease under investigation may develop many years after an exposure.

People move. Cancer can take decades to develop.

This is one of the fundamental challenges of studying historical environmental exposures.

Populations do not remain stationary. Children grow up and leave home. Families relocate. Neighborhoods change. New subdivisions are built. People who lived in an area during one decade may be scattered across Colorado—or the country—several decades later.

Cancer also does not necessarily appear immediately following an exposure. Different cancers can have different latency periods. A study attempting to investigate a historical exposure therefore needs to consider not simply where someone was when diagnosed, but where that person lived during relevant periods and how much time passed before diagnosis. The existing geographic cancer-registry approach cannot reconstruct those individual residential histories. That limitation does not prove that an elevated cancer rate exists. But it affects the question the research is capable of answering.

Geography matters, too

Another issue identified in our methodological review is how the study population is geographically defined. Environmental releases do not necessarily follow city boundaries, ZIP codes or convenient circles drawn around a facility. Historical wind patterns, documented releases, fires, drainage, soil contamination and other potential pathways may produce very different geographic patterns.

A large study area can combine people with very different potential exposure histories.

That creates the possibility of exposure misclassification: people with potentially very different histories may be analyzed together simply because they fall within the same geographic boundary. Future research should therefore consider exposure-informed geography rather than relying only on administrative or broadly defined geographic areas.

Rare cancers and small numbers create another challenge

Cancer studies around a relatively small geographic area can encounter another statistical problem: some cancers are uncommon. When the number of cases is small, statistical estimates can become unstable and confidence intervals can become wide. That makes interpretation difficult. A result that is not statistically significant does not automatically demonstrate that no increased risk exists. It can also mean that the available data were insufficient to distinguish an increase from normal statistical variation. This is particularly important when cancers are grouped together.

Combining different cancers may produce a larger sample for analysis, but it can also obscure patterns involving particular cancer types, ages or populations. The appropriate response is not to assume an effect exists. It is to communicate the uncertainty accurately.

Cancer is not the only health question

There is another limitation to relying primarily on cancer-registry analyses to understand community health around Rocky Flats: Cancer registries study cancer. RFD has heard from community members reporting concerns involving autoimmune diseases, thyroid disorders, neurological conditions, reproductive problems, birth defects and other illnesses. Those reports do not establish that Rocky Flats caused those conditions.

But cancer-registry studies cannot answer questions about diseases they do not collect.

That is one reason RFD believes the next generation of Rocky Flats health research should be broader than another geographic cancer-count analysis alone.What would better research look like?

There is no single perfect study design, and RFD should not dictate methodology to epidemiologists. That is precisely why we want qualified, independent researchers involved.

But research intended to answer the historical Downwinder question should at least explore whether it is possible to incorporate:

• Residential history — identifying where people actually lived during historically relevant periods.

• Migration — following former residents rather than losing them from the study when they move away.

• Latency — allowing appropriate time between potential exposure and disease diagnosis.

• Exposure-informed geography — considering historical contamination, releases and potential pathways rather than geography alone.

• Disease-specific analysis — examining appropriate cancers and other health outcomes rather than relying exclusively on broad groupings.

• Age and life-stage considerations — recognizing that childhood and adult histories may need to be evaluated differently.

• Longitudinal follow-up — studying a defined population over time instead of relying only on repeated snapshots of whoever lives in an area when diagnosed.

• Transparent methods and uncertainty — clearly reporting what the research can establish, what it cannot establish and where the data remain insufficient.

  
  

The goal should be a research design developed by qualified epidemiologists and other relevant experts, with meaningful participation from the community whose health history is being studied.

Independent research matters

RFD believes future research should involve researchers with the expertise to evaluate environmental epidemiology, radiation and toxic exposures, spatial analysis, cancer and other chronic diseases. Community involvement also matters. Downwinders can help researchers understand historical residential patterns, schools, neighborhoods, community experiences and questions that may otherwise be overlooked. But the scientific conclusions must belong to the researchers and the evidence. RFD's role is not to decide what the study should find.

Our role is to help make sure the question is finally studied in a way capable of producing meaningful evidence.

We should be willing to follow the evidence wherever it leads

Better research could identify patterns that deserve further investigation. It could also find no meaningful association. It could identify effects for certain diseases, populations, locations or periods but not others. Or it could reveal that some questions cannot be answered reliably because too much historical information has been lost. Those are all scientifically legitimate possibilities. What should not happen is for a narrower surveillance question to be mistaken for a definitive answer to a broader historical health question.

That principle is at the heart of RFD's Truth & Transparency Project: Say what was studied. Say what was not. Study the people who were actually there.

The question remains worth studying

Rocky Flats Downwinders began because people who lived near the plant started finding one another and sharing their health experiences. Those stories were compelling enough to raise questions—but stories alone cannot answer them. That's why RFD helped bring community health concerns into research and why we continue advocating for independent health studies today. The organization's larger strategy has consistently distinguished community reports from epidemiological proof and identified independent health research and long-term investigation as priorities.

We do not need science that starts with the answer we want. We need research designed to ask the right question—and the willingness to accept what rigorous research finds.

Help Move the Research Forward

Rocky Flats Downwinders is seeking engagement from epidemiologists, environmental-health researchers, universities, healthcare professionals, research institutions and others interested in helping determine how the long-term health of the Rocky Flats Downwinder population could be studied more effectively.