Why Rocky Flats Needs a Multidisciplinary Public Health Policy Review

The question is no longer simply, “What did the cancer studies find?” We also need to ask whether the studies were capable of answering the questions the public believes they answered—and whether those limitations have been communicated clearly.

For decades, people who lived near Rocky Flats have asked a basic question: Did living near the former nuclear weapons facility affect our health? Colorado has repeatedly pointed to cancer-registry studies conducted by the Colorado Department of Public Health and Environment (CDPHE). These studies are important sources of information. But they also have limitations.

That creates a problem that cannot be adequately addressed by epidemiology alone.

Rocky Flats Downwinders believes the time has come for an independent, multidisciplinary Public Health Policy Review bringing together experts in epidemiology, biostatistics, environmental health, research methodology, health policy, and public-health communication.

The goal would not be to prove that Rocky Flats caused cancer. Nor would it be to prove that CDPHE is wrong.

The goal is simpler and more important:

Determine what the existing Rocky Flats studies can actually establish, what they cannot establish, and whether the CDPHE has communicated that distinction accurately to the public.

One Question—Several Different Disciplines

A complicated public-health issue cannot always be answered by one type of scientist.

An epidemiologist can examine whether the populations, cancer outcomes, time periods, and study design were appropriate.

A biostatistician or research-methodology expert can evaluate statistical power, confidence intervals, comparison populations, geographic boundaries, small numbers, and whether the statistical evidence supports the conclusions being drawn.

An environmental-health or spatial epidemiology expert can consider whether study geography appropriately relates to the historical environmental question being investigated.

And a health-policy researcher can ask an entirely different—and essential—question:

What responsibility does a state health department have to explain the limitations of its research when those findings are being used to reassure communities and inform public policy?

We need all of these perspectives.

The Difference Between Cancer Surveillance and Following Downwinders

One issue deserves particular scrutiny. CDPHE has described cancer incidence near Rocky Flats as having been “monitored” since the 1980s. But what does the word monitoring mean to an ordinary member of the public? Someone hearing that statement might reasonably believe Colorado identified people who lived near Rocky Flats during historically relevant periods and then followed their health over subsequent decades.

That is not the same thing as conducting periodic analyses of cancer-registry data.

Cancer registries perform an enormously valuable public-health function. But registry surveillance generally records cancer diagnoses and geographic information associated with those diagnoses. It does not, by itself, reconstruct every patient's complete residential history or identify everyone who lived in a particular community decades before diagnosis. That distinction matters when investigating diseases with potentially long latency periods.

A person might have lived near Rocky Flats during an historically relevant period, moved elsewhere, and developed cancer many years later. Conversely, someone diagnosed while living near Rocky Flats may have moved into the community relatively recently.

That does not make cancer-registry data invalid.

It means we need to be precise about what questions that data can answer.

The Difference Between “Not Detected” and “Does Not Exist”

This is another reason independent biostatistical expertise is important. A study that does not detect a statistically significant increase does not automatically establish that an increased risk does not exist. Researchers need to consider the number of cases, confidence intervals, statistical power, geographic aggregation, cancer groupings, comparison populations, latency, migration and other methodological issues before deciding how strongly a result can be interpreted. The appropriate conclusion might ultimately be reassuring. But reassurance should come from the strength of the evidence, not from language that is broader than the evidence supports.

This Is Where Health Policy Becomes Essential

Until now, much of the Rocky Flats debate has centered on epidemiology. We believe there is another important question: How should government communicate uncertain or incomplete scientific evidence?

That is a health-policy question. When CDPHE publishes information about Rocky Flats, those statements do not remain inside scientific journals. They can be read by residents, physicians, journalists, elected officials, local governments, developers and policymakers. Government health information can influence public perceptions and public decisions. That creates a responsibility to communicate not only what a study found, but also what it was capable of finding.

An independent health-policy review could examine whether important methodological limitations should accompany public statements about Rocky Flats and whether Colorado should establish stronger disclosure standards for environmental-health studies generally.

Six Questions an Independent Review Should Answer

A multidisciplinary research team should be asked to examine:

1. Were CDPHE's Rocky Flats cancer studies appropriately designed for the questions they are being used to answer?

2. Do their geographic boundaries, comparison populations, statistical methods and cancer groupings support the conclusions being communicated?

3. How do cancer latency, population movement and residence at diagnosis affect interpretation of the findings?

4. Does describing decades of registry analysis as cancer “monitoring” adequately distinguish cancer surveillance from longitudinal monitoring of a historically defined population?

5. Does CDPHE's public communication accurately reflect the evidentiary strength and limitations of its research?

6. What disclosure standards or public-health policies should Colorado adopt so residents and policymakers understand both what is known and what remains uncertain?

   
   

Independence Is Critical

Rocky Flats Downwinders should not determine the answers to these questions. Neither should the review be designed to produce a predetermined conclusion. Independent researchers should establish the methodology, review the evidence, disclose relevant conflicts of interest and publish their findings regardless of whether those findings support the concerns raised by Rocky Flats Downwinders. If independent researchers determine that CDPHE's conclusions and public communications are fully supported, the public deserves to know that. If they determine that the underlying research is useful but public statements need greater qualification, the public deserves to know that too. And if they identify important unanswered scientific questions, Colorado should acknowledge those questions and determine how they can be answered.

From Scientific Disagreement to Better Public Policy

This is why a multidisciplinary project could change the Rocky Flats conversation. Instead of asking the community to choose between “Rocky Flats caused cancer” and “the studies show there is no problem,” independent researchers could examine what the evidence actually allows us to say.

The result could include an independent scientific review, a health-policy analysis, an evidence-to-communication comparison of CDPHE's public statements, recommended disclosure standards, and recommendations for future research. It could also help answer a larger question: Do we need a different kind of health study to understand the experiences of people who actually lived near Rocky Flats during historically relevant periods?

That might eventually lead to retrospective cohort research, historical residential reconstruction, longitudinal follow-up, exposure-informed analysis, or research that considers health outcomes beyond cancer.

Truth and Transparency Should Not Predetermine the Truth

That principle is important to Rocky Flats Downwinders. A Truth and Transparency initiative should not begin by telling researchers what they are supposed to find. It should begin by asking qualified, independent researchers to find out. Our request is not that Colorado declare Rocky Flats responsible for illnesses that have not been scientifically attributed to it. Our request is that the strength of Colorado's public statements match the strength of Colorado's scientific evidence.

Where evidence is strong, say so.

Where evidence is limited, say so.

Where uncertainty remains, say so.

And where important questions have never actually been answered, have the courage to ask them again—with research designed to answer them.