Sunday, May 3, 2009

What the CDC Can Learn from the National Research Council and the Public

In 2008, the National Research Council (NRC) -- a nonprofit institution under the auspices of the National Academy of Sciences that works to improve government decision making and public policy -- published the book "Public Participation in Environmental Assessment and Decision Making." NRC defined "public participation" as "organized processes adopted by elected officials, government agencies, or other public- or private-sector organizations to engage the public in environmental assessment, planning, decision making, management, monitoring, and evaluation." NRC's goal was to determine how public participation in environmental decision making affects the assessment of environmental problems and the development of public policy relating to these problems.

NRC concluded that:
"When done well, public participation improves the quality and legitimacy of a decision and builds the capacity of all involved to engage in the policy process. It can lead to better results in terms of environmental quality and other social objectives. It also can enhance trust and understanding among parties."
The central premise behind NRC's thesis was that, by its nature, environmental decision making is complex because it involves choices that are political, social, cultural, and economic, at least as much as technical and scientific. NRC straightforwardly acknowledged that, "There typically are multiple perspectives regarding the relative importance of issues, the best courses of action, and even the right questions to ask, with strong demands from those who may be affected by policy choices to have their voices heard."

In other words, delegating too much decision-making authority exclusively to technical experts and government officials amplifies narrow interests and understandings, while drowning out public knowledge, needs, and definitions of what constitutes an "appropriate" response to an environmental problem. According to NRC, since the 1960s, US environmental policies that left out public views have often been criticized as "bad" and "out of touch" by the communities they affected. Today public participation theorists and practitioners generally agree that good environmental policy decisions incorporate the goals and concerns of all stakeholders, including the public.

NRC's findings help elucidate a central problem with the Centers for Disease Control and Prevention's (CDC) latest move in relation to DC's lead-in-water fiasco:

On April 10, CDC issued a media statement in response to a Salon article, which alleged that CDC withheld evidence of childhood lead poisonings from DC's 2001-2004 lead-in-water contamination. Salon revealed new information about thousands of missing blood test results from the influential 2004 CDC report that found that two and a half years of astronomically high levels of lead in DC's drinking water did not cause significant harm. The CDC report misled communities around the country into believing that lead in water does not constitute a serious health risk. Salon also exposed that in 2007, CDC presented data at the annual meeting of the American Public Health Association (APHA) showing that, in fact, in 2001-2004, many DC children experienced blood lead levels (BLLs) above the CDC's "level of concern" due to the city's lead-in-water contamination. CDC chose not to publicize these findings (CDC's 2007 presentation is available on the APHA website for a fee; for additional discussion on CDC's 2004 report, see our 2/25/09, 3/22/09, 4/20/09 blog entries).

CDC's media statement was posted online hours after the publication of the Salon article to purportedly address "inaccuracies" in the Salon piece and reassure the public of CDC's commitment to lead poisoning prevention. CDC focused on the following six issues:
  • The conclusions of its 2004 report
  • The missing blood test results from its 2004 analysis
  • Allegations of scientific misconduct in relation to the 2004 report
  • CDC's communication (or lack thereof) of its 2007 findings to appropriate city and federal officials
  • CDC's decision not to publicize its 2007 findings
  • The relationship between the 2004 report and a January 2009 study, which found that hundreds if not thousands of DC infants and toddlers experienced BLLs above the CDC's "level of concern" due to the 2001-2004 lead-in-water contamination.
A close reading of CDC's media statement makes it obvious that CDC, unable or unwilling to refute the serious questions raised by Salon, chose to stay on the path of deception and obfuscation in order to try and salvage its reputation, at the expense of public health. CDC's responses to the above six issues were strikingly elusive. They either avoided the questions completely, or they made assertions that lacked critical supporting evidence. Simply put, they made little sense, leaving one confused and frustrated. To add insult to injury, they were clumsily sandwiched between opening and closing statements of self-praise about CDC's exemplary work in protecting children from the dangers of lead.

Any parent of a young child in Washington DC would want to see the CDC ably, confidently, and cogently refute the Salon findings with credible evidence. Any parent of a young child in Washington DC would also want to know that CDC respects their intelligence and educates them about public health issues with direct, honest, and accurate information that addresses their worries about their child's health. Any parent of a young child in Washington DC would hope and expect that CDC cares more about the health of their child than the advancement of its own institutional interests. Yet the CDC's response to Salon failed miserably to satisfy any of these standards.

In contrast to NRC's findings about public participation in environmental decision making, one sees that CDC, in its latest attempt to defend itself against Salon, made a strategic decision that did not take into account the public. Specifically, CDC did not regard the DC public's acquired knowledge about the science and politics of lead in drinking water, and our grave concern about the implications of the long term exposure of thousands of fetuses, infants, and young children to astronomical levels of lead at the tap.

What happened to our city's children in 2001-2004, and how they may be harmed as a result is a profoundly serious matter to us. After 5 years of disinformation about the facts of lead in water, DC deserves answers that are based on strong science and solid evidence. We can no longer accept authoritative but unsubstantiated statements blindly. We have learned to tell truth from spin. And our patience for spin has ended. In addressing our concerns, CDC must understand that unscientific, inaccurate, misleading, or imprecise information has no place in its public health messages to DC, especially when those messages are delivered purportedly to "correct" allegations backed up by science and facts of CDC wrongdoing involving data, conclusions, and decisions that had and continue to have a direct impact on our children's health.

Given its shameless evasiveness, CDC's statement of April 10 appeared as a desperate attempt to pull the wool over the public's eyes, one more time. If CDC is truly concerned about protecting children from lead in water, it might want to consider alternative answers to the six questions it chose to address. Below, in red, are some recommendations for a more responsible, appropriate, and respectful statement (CDC's original text is in black).

***
Media Statement

CDC Responds to Salon.com Article


For Immediate Release: April 10, 2009
Contact: CDC Division of Media Relations, Phone (404) 639-3286

An April 10, 2009 article appearing on Salon.com inaccurately represents the Centers for Disease Control and Prevention's (CDC) 2004 public health response to a request from the Washington D.C. Department of Health for assistance concerning lead in water in District of Columbia neighborhoods (Blood Lead Levels of Homes with Elevated Lead in Tap Water -- District of Columbia, 2004 MMWR Dispatch Vol. 53., March 30, 2004).

CDC addresses these inaccuracies and reassures the public of our commitment to preventing harmful lead exposures. For three decades, CDC has been an international leader in calling attention to the fact that childhood exposure to lead causes adverse health effects including speech, behavioral problems, difficulty learning and hyper activity. Throughout this time, CDC has provided guidance and recommendations to parents, public health agencies, doctors, and community organizations about prevention of childhood lead exposure and suggestions to seek diagnostic testing when parents are concerned about their children's potential exposure.

[The specific issue addressed by the Salon article is health effects from excessive levels of lead in drinking water. In your opening statement, please cite specifics about if, or how, CDC has ever been an international leader in calling attention to the potential health risks of lead in drinking water, and in educating parents, public health agencies, doctors, and community organizations about preventing childhood exposure to high concentrations of lead at the tap. Without this information, the above statement reads like gratuitous PR that misses the entire point of the Salon piece.]

Characterization of CDC's 2004 MMWR Conclusions
Salon correspondent Rebecca Renner mischaracterizes the 2004 Morbidity and Mortality Weekly Report (MMWR). The MMWR states that "lead in tap water contributed to a small increase in BLLs in D.C." Now, as in 2004, CDC continues to stand by its MMWR statement that, "Because no threshold for adverse health effects in young children has been demonstrated, public health interventions should focus on eliminating all lead exposures in children. Lead concentrations in drinking water should be below the EPA action level of 15 ppb. Officials in communities that are considering changes in water chemistry or that have implemented such changes recently, should assess whether these changes might result in increased lead in residential tap water."

[Here please spell out how exactly Ms. Renner mischaracterized CDC's 2004 report. What did Ms. Renner write that contradicts CDC's statements in MMWR? It is telling that you fail to quote Ms. Renner's supposedly erroneous characterizations and juxtapose them with the corresponding MMWR statements. Please allow readers to see for themselves the gap between the Salon article and the CDC report. Although your response above implies that Ms. Renner's piece got facts wrong, it provides no evidence to support the claim. Again, as it stands, this looks like one more attempt to cloud the issue rather than shed light on it.]

Missing Test Results
Ms. Renner's article contends that missing laboratory test results undermined public health. As CDC has explained to Ms. Renner many times, in interviews and in written responses, public health surveillance data uses real world health information to support public health decision-making. It is dependent on health organizations to report data accurately. In 2004, a participating commercial laboratory stopped reporting test results that fell below the CDC level of concern of 10 ug/dL. CDC believes this failure of reporting accounts for the missing data because the laboratory continued to report BLLs greater than 10 ug/dL. To the extent "missing" data would have affected overall results, it would have exaggerated the apparent problem, not masked it. As a part of the 2004 public health consultation, the CDC encouraged the D.C. Department of Health Medical Director to remind all laboratories and health care providers of their obligation to report all blood lead level tests.

[The key question is the scientific validity of your 2004 report. If you are wedded to the "negligent laboratory" explanation, try to offer as much verifiable evidence as possible. For example:
  • Can you name this laboratory and attach an official letter from them acknowledging their mistake?
  • Can you attach a complete database from the laboratory (with children's identifying information redacted, of course) showing all the blood lead data points from 2003 that they did not report to the DC Department of Health (DOH)?
  • Can you post an official letter from DC DOH explaining if, when, and how they knew that the thousands of unreported blood lead levels from 2003 were all below 10 ug/dL, and what steps they took (if any) to obtain all blood test results, as they were supposed to?
  • Can you show what steps CDC took (if any) prior to publication of the 2004 report to confirm that only test results with BLLs below 10 ug/dL were missing?
  • Can you explain why neither CDC nor DC DOH ever simply asked the laboratory to supply the missing data in order to include them in the 2004 analysis?
  • Can you debunk the allegation mentioned in Salon that high blood test results (above 10 ug/dL) were omitted as well?
Is Salon correct in its statement that the "negligent laboratory" explanation originated at DC DOH? If not, please state how you heard of this explanation. If so, please consider disclosing DC DOH's conflict of interest in the 2001-2004 lead-in-water crisis. Specifically, please acknowledge publicly that:
  • DC DOH had direct knowledge of the 2001-2004 contamination, but had taken little action to address it.
  • Some of the CDC report's co-authors worked at DC DOH and coordinated the 2004 environmental risk assessments at the homes of children with elevated BLLs, which showed in several cases that drinking water was the sole or a contributing source of lead.
  • DC DOH was the source of all the data for the 2004 CDC report and had direct responsibility for the thousands of missing blood lead test results.
  • DC DOH was sued (along with WASA) for personal injury of a DC child with severe lead poisoning, and the lawsuit alleges that the child's exposure was at least partly from high lead in drinking water.
  • Two DC DOH employees were fired for their role in the lead-in-water fiasco.
Lastly, the sentence, "CDC believes this failure of reporting accounts for the missing data..." needs rewording. "Belief" in the "negligent laboratory" explanation does not constitute scientific evidence on which the public can rely. Your readers want to know what proof CDC has that thousands of missing blood lead test results from a critical year of the study (when DC's water was highly contaminated and most DC residents were not yet aware of the problem) did not skew the results, much less merit disclosure in the 2004 report.]

Alleged Scientific Misconduct
Scientific integrity is CDC's hallmark. Scientific misconduct has a precise definition, which is "fabrication, falsification, or plagiarism in proposing, performing, or reviewing scientific activities, or in reporting scientific results." CDC's Office of Science takes any such allegation very seriously; it thoroughly investigated this complaint and found no evidence of scientific misconduct. CDC acknowledges the fact of missing data; however this fact is unrelated to fabrication, falsification, or plagiarism on CDC's part.

[The third sentence refers to "this complaint." Consider naming the source of the complaint, for clarity. Presuming that the reference pertains to the 2007 allegations of scientific misconduct by Professor of Civil and Environmental Engineering at Virginia Tech and MacArthur Fellow Marc Edwards:
  • Can you post all documents pertaining to your thorough investigation of the specific allegations about falsified and fraudulent data used in the study? The statement, "...on CDC's part," leaves open the distinct possibility that there was fraud and fabrication in the paper by the report's non-CDC affiliated co-authors. From a practical perspective, it does not matter who originated the fraud and fabrication. What the public wants to know is whether or not fraud and fabrication occurred.
  • Can you also explain why you have failed to answer Dr. Edwards' 1/2/08 inquiry about whether CDC actually investigated his 2007 allegations, and whether your purported investigation followed CDC's written guidelines for addressing scientific misconduct concerns?
Please address all of Dr. Edwards' scientific misconduct allegations (e.g., regarding the 300-parts-per-billion study, the publication procedures behind the 2004 report, etc.). As written, the paragraph above implies that Dr. Edwards' reported concerns were limited to the missing data points, which can mislead readers to the false conclusion that no other questions about the integrity of the MMWR report have been raised.]

Communication of Findings
Since 2004, the CDC Lead Poisoning Prevention Program staff and CDC Division of Media Relations have responded to numerous press inquiries about the 2004 D.C. consultation. Moreover since the 2004 investigation information has been shared regularly and rapidly with other public health authorities, local and federal agencies, including the American Public Health Association, DC Water and Sewer Authority, D.C. Lead program officials, the George Washington University School of Public Health's Water Team, and representatives from U.S. Environmental Protection Agency.

Allegations by Ms. Renner that, "Scientists from other agencies, including EPA and HUD,...were never told about the results" are untrue or misleading. For example, CDC's Lead Poisoning Prevention Branch routinely consults with EPA water experts and solicits their review and comment on a variety of topics. HUD does not have jurisdiction over issues related to lead and water; CDC works closely with HUD on a variety of other issues related to healthy housing.

[To substantiate the claim that Ms. Renner's statement about CDC's non-disclosure of the 2007 APHA presentation was "untrue" or "misleading," please provide the following:
  • The names of all agencies and individual employees CDC informed about its 2007 findings, as well as the dates and methods by which this information was delivered.
  • Copies of any and all written documents (letters and/or e-mails) disclosing the 2007 findings to agencies/individuals outside CDC.
  • Any and all public health recommendations accompanying CDC's disclosure of the 2007 findings.
As written, your statement provides no actual evidence, only more unsubstantiated assertions, that Ms. Renner's reporting was untrue or misleading.]

Alleged Failure to Share New Information

In 2007, preliminary findings of the 2004 public health consultation were presented during the annual meeting of the American Public Health Association, and the abstract was published in annual conference program guide.

It is common practice in scientific circles to present preliminary findings at scientific meetings as a way for researchers to receive comment and advice from other experts. This presentation did not present complete data analysis nor did it include all the analysis that were planned and necessary for a scientific manuscript. Since that time CDC has obtained additional statistical analyses, undertaken additional peer review, and continued to strengthen the manuscript. We believe these aspects of the scientific process are essential to conducting and communicating the best quality science. The manuscript is expected to be published later this year.

[CDC's 2004 report concluded that:
"...although lead in tap water contributed to a small increase in BLLs in DC, no children were identified with BLLs [greater than] 10 ug/dL, even in homes with the highest water lead levels."
CDC's 2007 presentation concluded that:
"The association between exposure to lead service pipe and having a BLL [of 10 ug/dL or above] remained statistically significant after controlling for potential confounders."
The 2004 conclusion indicated that no significant harm was done from lead in DC's drinking water. The 2007 conclusion indicated just the opposite: that significant harm was done from lead in DC's drinking water. If the 2007 conclusion represented the preliminary findings of the 2004 report, as you state above, please answer the following: How did the "significant harm" preliminary finding that CDC presented in 2007 get translated into the "no significant harm" conclusion in CDC's 2004 report?

If CDC deemed it necessary to publish the 2004 report, with all its horrific flaws, in less than 12 days to reassure DC residents that no significant harm was done, why didn't CDC feel an even greater responsibility to publicize its 2007 findings, and to alert the public that the agency's earlier "no significant harm" conclusion was inaccurate and misleading?

Can CDC explain its decision not to disseminate its 2007 findings of harm, despite the fact that these findings agree with a) decades of prior scientific research about the health effects of lead in water on infants and young children and b) the 2009 peer-reviewed study by independent researchers?

Lastly, in the statement above, can CDC justify continuing to defend the 2004 report when you have publicly admitted that this report:
  • is not scientific,
  • is missing thousands of blood lead test results from a critical time period,
  • failed to focus on the populations most vulnerable to the effects of lead in water,
  • makes claims about the health effects of lead-contaminated water based on blood tests of residents who didn't drink this water, and
  • is apparently being "misinterpreted" by everyone who reads it, including several of the report's own authors?
With CDC's ardent and stubborn defense of its 2004 report as background, can CDC also provide your rationale for not publicizing your 2007 presentation on the basis that it does not meet CDC's standards for "best quality science"?

Your statement above raises more troubling questions than it answers. Please consider rewriting it with clarity and honesty.]


Other Scientific Studies

Ms. Renner alleges that "a new scientific study published in January ("Elevated Blood Lead in Young Children Due to Lead-Contaminated Drinking Water: Washington D.C. 2001-2004," Journal of Environmental Science and Technology, Vol. 43, No. 5, 2009) contradicts CDC's conclusion of minimal harm. In fact this study, reaffirms CDC's conclusions that lead in water contributed to elevations in BLLs and that lead concentrations in drinking water should be below the EPA action level of 15 ppb.

[CDC's 2004 conclusion was that although a small increase in BLLs was noticed in 2001-2004, no children were identified with blood lead above the CDC's "level of concern" due to the city's lead-contaminated drinking water. The 2009 study's conclusion was that hundreds if not thousands of DC infants and children experienced blood lead concentrations above the CDC's "level of concern" due to the city's lead-contaminated drinking water. If the latter conclusion agrees with the former, please explain clearly and rationally how.]

Since 1975, CDC has been a tireless advocate for childhood lead poisoning prevention. Our agency's commitment to preventing childhood lead exposure is as strong today as it has been since we released the first statement on lead poisoning and its effects on children.

[Given CDC's complete failure to refute the main points of the Salon article with specifics, it hardly seems the time or place for self-congratulatory verbiage on the "agency's commitment to preventing childhood lead exposure." Please refrain from such statements, at least while the integrity of your influential 2004 report is being questioned.]

####

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES