Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
8 Communication of Study Results There are many factors to evaluate in the communication planning and practice related to the release of the HTDS Draft Final Report. Some aspects of the communication strategies were well carried out, particularly the ones that kept the public informed about the study protocols, design, and progress over the years. However, in the fall of 199S, problems-some controllable and some not arose that had substantial and unfortunate effects on the communication efforts that finally were made in late January 1999 for the release of the HTDS findings. The release of the Draft Final Report led to unhappiness and dismay among some citizens in the Hanford area, not only because of the main message, but also because of how the message was delivered. BACKGROUND The communication of risk information to the public is an important issue that has been addressed by many individuals and groups. Numerous journal articles, manuals (Swanson and others, June 1991), and reports including one by the National Research Council (1989) have discussed how to convey information about health and environmental risks to the public. But there is no sure prescription to follow for providing such information effectively. One piece of advice that appears in most nsk-communication publications is to ensure that the information ~4
Communication of Study Results ~5 source is credible and trustworthy in the minds of the public to which it is communicating. Risk communication as a field has evolved over the last 15-20 years from a simple linear top-down communication mode} (experts translated technical information and dispensed it to the public) to a more sophisticated series of models that view risk communication as a "tangled web" of interactions that move in many directions and involve many players local, state, and federal governments; special-interest groups; citizen groups, industries; unions; and so on (Krimsky and Plough, 1988~. The communication that occurred about the HTDS is as tangled a set of risk-communication webs as one comes across. It includes not only a federal agency and a private contractor, but also health agencies in three state governments and representatives of nine American Indian nations, numerous citizen groups in the region, national and regional journalists, a class-action lawsuit involving many litigants, various consultants and potential expert witnesses, and a number of private individuals in the region who have suffered or whose family members have suffered from some type of thyroid disease. The public and private messages traded back and forth by these groups and individuals over the years have all shaped the Hanford and HTDS communication process. For example, it is important to remember that, according to citizen comments reported in the mass media and elsewhere, many of the citizens in the area had developed a distrust of government sources, particularly the Department of Energy (DOE) and its precursor agencies. When CDC began to evaluate the thyroid- disease situation at Hanford, the distrust was already in place, although not applied directly to CDC at the time, it became an important factor when the HTDS Draft Final Report was released. From citizen comments, it appears that the Fred Hutchinson Cancer Research Center in Seattle enjoyed greater public trust than the federal agencies while carrying out the HTDS.
116 Review of the HTDS Draft Final Report ORIGINAL HTDS COMMUNICATION PLAN The HTDS was conducted over a 9-year period, and it is not possible to evaluate all the written and oral communication with the public that occurred about it, but apparently a number of potentially effective communication efforts were made. As described in section X of the HTDS Draft Final Report, public meetings on the project began in 1990 and continued throughout the study, written brochures and fact sheets were developed, newsletters were sent to more than 9,000 people and organizations, and a toll-free 800 line and a Web site were available. In March 1991, the first meeting of the federally appointed public HTDS Advisory Committee was held, and this committee and several other groups approved the study protocol. The advisory committee continued to meet throughout the life of the study. Special arrangements were made to keep study participants advised of the results of their clinical evaluations of thyroid disease. The open . . communication seems to have continued almost up to the end of the study, and no one who provided information to the present subcommittee during its public meeting in Spokane or otherwise criticized those communication efforts. Given the earlier history of less than openness with the public in the Hanford region on the part of DOE and its precursor agencies, this plan for open communication was enlightened and promising. So were the decision to establish a citizen advisory group for the study and the apparent cooperation offered to various other citizen groups in the region, including the Hanford Health Information Network and the Hanford Health Effects Subcommittee (HHES). All those efforts seemed to help to build trust and credibility for the study and its investigators and for CDC. Section X of the HTDS Draft Final Report outlined a good communication plan to deliver the final information about the report that might have worked if it had been put into operation with its release in March 1999. Especially admirable was the concern shown for translating the technical information in the Draft Final
Communication of Study Results ~7 Report into an understandable booklet for the public. The use of focus groups to look at various presentation strategies for the booklet was a fine approach and resulted in a readable and understandable public summary. (More will be said about the public summary later.) Considering the many audiences that had to be informed about the study findings, the communication plan described in section X was fairly complex and involved several types of briefings, with a suggestion that some of these be conducted via satellite connections to remote broadcast sites throughout the region. Although expensive, such a plan would have probably worked far better than the telephone briefings that were eventually used when the Draft Final Report was released. The original communication plan was approved by the HTDS Advisory Committee and widely disseminated in the HTDS newsletter. According to Scott Davis, one of the principal HTDS investigators, there was never any intent to release to the public the draft report that was delivered to CDC at the end of September 1998. That draft report was supposed to go through internal CDC and National Research Council closed peer review; comments were to go to the HTDS researchers, who would then make needed changes and release a peer-reviewed final report in March 1999. However, three major factors interacted to bring about the early release the Draft Final Report, which contnbuted to the extensive communication problems encountered: public pressure to get the document out to the public and concern that CDC's internal review would alter the original findings of the HTDS investigators, the National Research Council's desire for open peer review of the draft report, and a subpoena from one party in a lawsuit that sought immediate release of the Draft Final Report. Concerning the first factor, in early October 199S, the director of CDC's National Center for Environmental Health received numerous written requests for immediate release of the Draft Final Report. According to CDC, because the agency had received the Draft Final Report only on September 30, the requests
118 Review of the HTDS Draft Final Report indicated that the correspondents expected the report to be made available to the public unchanged. As to the second factor, a communication in late October to CDC from the National Research Council said that this organization felt that the credibility of its review would be compromised if the HTDS report were not publicly available when the review process began. Given the many audiences, including study participants who needed to be informed about the study findings, given the extensive communication plan already developed, and wanting to preserve the credibility of the review process, CDC officials decided to release the Draft Final Report before the Research Council's open review. On November 12, they met with the HTDS investigators and decided to release the report on January 28, 1999. The third factor added resolve to that decision. According to CDC, during the week of November 16, 199S, the HTDS investigators received a subpoena that called for delivery of the Draft Final Report to the plaintiffs' lawyers within 30 days. The delivery date of the report had already been moved up, and the plaintiffs' attorneys, with consent of the court, indicated that they could wait until the January release. One of the problems preventing an even earlier release was the need to develop documents about the report that the public could understand (Davis, 1999b). Both public demands to CDC to release the report and not change it were exemplified by the minutes of an HHES meeting held on December 10-ll, 199S, in which a CDC official explained that the agency had heard from many people "who expressed a real interest in seeing if we could move the date of the release of this report up to make it public" (HHES, 1998~. Concerning fears that changes would be made in the draft by CDC's internal reviewers before the report was released, one person said, "if there are changes made between what Fred Hutchinson delivers and what comes out the door at CDC, I'm hoping that you have heard from this subcommittee clearly that we
Communication of Study Results ~9 want to know what those changes were and the rationale for those changes" (Hanford Health Effects Subcommittee, 1999, p.39-40~. Concern also was expressed at the meeting about plans to release the report, in particular, how HHES and other citizen groups would be briefed and the fact that the HTDS investigators were submitting an article on the study results to a scientific journal before citizens knew the results. COMMUNICATION ISSUES IN THE WRITTEN MATERIALS Besides the Draft Final Report itself, a number of written pieces were developed for the public, including the "Summary Final Report of the Hanford Thyroid Disease Study" and several HTDS newsletters. One featured a summary of the study results, another presented information on thyroid disease and how it was diagnosed, and a third included questions and answers about radiation and thyroid disease. There were also at least seven fact sheets from CDC on various subjects related to the HTDS. Most of these were included in the briefing kits for the media and public when the report was released. To evaluate the written materials provided by the HTDS investigators and CDC, a number of different factors must be considered. Among these are accuracy, appropriateness of material, and readability or ease of understanding. Accuracy From a scientific perspective, the results section of the Draft Final Report provided details on linear dose-response analyses conducted on 13 types of thyroid disease or abnormality and on a number of thyroid-related laboratory tests. Other analyses were conducted, when possible, with alternative definitions of the disease being analyzed and alternative dose-response functions. None of the numerous tests showed a statistically significant increase with dose. f
120 Review of the HTDS Draft Final Report The resulting database on the absence of dose-response relationships formed the basis for statements made in the report that no evidence was found of a statistically significant increase in effects associated with increased dose of its. Within the reported framework of the data used and the analyses conducted, that is an accurate representation of the results given in the Draft Final Report. Appropriateness Although the above statement is scientifically accurate, given the state of a draft not reviewed by outside scientists and a number of uncertainties that were already apparent to the HTDS researchers, some overstatements were made in the public summary, HTDS newsletters, the news release, and the executive summary of the Draft Final Report. In addition, there was little or no mention in any of these documents of the uncertainty issues involved in the study. Such uncertainties, already described in the present report and others, included the statistical power of the analyses, possible errors in the dosimet~y, and the reliability of some information in the computer-assisted telephone interviews related to possibly faulty recall about milk sources and amounts. At the time of the Draft Final Report's release, the HTDS investigators were trying to run an uncertainty analysis but were not succeeding in its execution; this was not mentioned in the written materials. Despite those problems, the results of the HTDS were presented with unqualified certainty, and at the time of and after the release of the report some statements attributed to the HTDS investigators appear to have overstated the certainty of the results in the Draft Final Report (emphasis added): . Thus, given that the HTDS had adequate statistical power to detect reasonably small effects, and the rigor of the study design, these results provide rather strong evidence that exposures at these levels to ]3~] do not increase
Communication of Study Results the risk of thyroid hyperparathyroidism. These disease or results should consequently provide a substantial degree of reassurance to the population exposed to Hanford radiation that the exposures are not likely to have affected their thyroid or parathyroid health Davis & Kopecky, 1998, p. Ace. Findings of the Hanford Thyroid Disease Study are clear and unequivocal fDavis and others, 19991. This was a very powerful study because it included a large number of people estimated to have a wide range of exposures to 13~{ tDavis, 1999a]. The study had sufficient statisticalpower to detect increases in thyroid disease risk that were predicted based on studies in other populations tCDC, 1999a]. The design and successful completion of the study ensured a very high probability of detecting relationships - between Hanford radiation dose and diseases under study if such relationships exist. The study was very powerful because . . . kHTDS Newsletter, 19991. 121 The subcommittee believes that such statements were ill-advised at a time when the Draft Final Report had not yet been subject to external peer review. Given the many questions raised
122 Review of the HTDS Draft Final Report about dosimetry and other issues and the problems in running the uncertainty analysis, the subcommittee feels that statements indicating such certainty should have been modified to take uncertainties into account and the uncertainties should have been listed and explained. Several paragraphs and perhaps a list of uncertainty issues should have been in the executive summary and all public documents related to the HTDS. The subcommittee recognizes that including such uncertainty information would probably have diluted the strength of the investigators' message; however, such caveats are critical to increasing public acceptance of the results of the report. Omitting them left the investigators open to the charge that they had emphasized negative results. On a related matter, the printed HTDS public summary was titled "Summary Final Report of the Hanford Thyroid Disease Study" (FHCRC 1999b), which was misleading, inasmuch as it was still a draft. From the cover of the printed summary, a reader would not have been able to tell that the report was a draft. Nor did the text of the public summary make it clear that the report was still under review and that some findings might be changed. The text explained that there had been public and scientific review of the study, but it implied that review was finished (FHCRC 1999b, p. 8~. Those items gave too much certainty to the results and the study itself. In contrast, the January 1999 HTDS newsletter included information in two places about the final stages of peer review by both the Research Council subcommittee and a journal, and it called the report the "Draft Final Report to the CDC". Later versions of the public summary given out by CDC were also stamped "Draft". The subcommittee recommends that all uncertainty issues be clearly noted and explained in the final report and all public documents related to it.
Communication of Study Results Contextual Information 123 Both the main public documents and the executive summary of the Draft Final Report lack comparisons with findings in other reports on similar subjects, such as the Institute of Medicine's report (NAS/IOM, 1999) on the nationwide fallout study by the National Cancer Institute (NCT, 1997) and the NTS report (Kerber and others, 1993~. It should be noted, however, that a CDC fact sheet, "What We Know from Other Studies of Environment Radiation Exposure and Thyroid Disease", was available. How widely it was disseminated is unknown. The subcommittee believes that not having at least a summary of this contextual information in the public summary and the Draft Final Report presented a problem because readers did not have any background information with which to compare the HTDS results if they did not have the CDC fact sheet. In risk issues, many readers and journalists need to know more than just the event at hand in this instance, the findings of only this one report. They need to see the long-term issue in context so that they can understand the variety of findings and judge for themselves the validity of the current study against the others. The subcommittee recommends that contextual information be included in both the executive summary and the body of the final report and the public summary. Readability The public summary, fact sheets, newsletters, and Web- site information about the HTDS Draft Final Report were readable and relatively easily understood. However, the executive summary of the Draft Final Report was far from that. The language was technical and often highly complex, and it did not need to be so in most places. Scientists will not be the only people reading the executive summary, and it should be understandable for a number of educated groups (such as nonscientist government officials, lawyers, journalists, and social scientists), even though it need not be as simplified as the public summary.
124 Review of the HTDS Draft Final Report There were a number of readability problems in the executive summary. For example, in the first part (pages 3-7), it was difficult to keep track of the many numbers related to how many people were in various parts of the sample. On page 4, in the fourth paragraph, the way the numbers were presented was confusing: 3,865 "potential participants . . . agreed on either the first or second attempt. The remaining 3,565 . . . agreed to participate. . . ." Those sentences should be rewritten. Throughout this section, charts would help to clarify which numbers apply where and help the reader to follow the discussion. The material on the computer-assisted telephone interviews and on how people provided names of relatives to be interviewed (pages 4-5) also needs to be explained better. Throughout the executive summary, abbreviations and uncommon words were unexplained or were explained after they had already been used. For example, finding "CIDER", "Exp.-IPl", and "realizations" on page 5 of the executive summary would probably confuse a reader. Those and other terms need to be briefly explained in the text or in footnotes. The subcommittee recommends that an effort be made to remove excess technical language and to use consistent terms, particularly for types of thyroid diseases, in the executive summary of the final report. More charts should be used in the executive summary to provide visual aids to help understand the information. in addition, the entire final report should be edited and should include a glossary of abbreviations and technical terms used. Written Materials from CDC, May 1999 After the adverse response to the release of the Draft Final Report by members of interested citizen groups and some other citizens in the region, as shown in letters to the editors in several newpapers and to CDC officials, CDC prepared some new materials that were used at two public meetings, in Spokane and Seattle, to discuss the report in May 1999. in the "Summary of the
Communication of Study Results 125 Preliminary Results", CDC (1999b) was more cautious about interpreting the findings: These preliminary results do not mean that people living in the Hanford area during the 1940s and 1950s were not exposed to IT and other radionuclides, or that these exposures had no effect on the health of people living in the Hanford area. Although no link between estimated IT and amount of thyroid disease was identified by the HTDS in the study population, the study results do not prove that a link between IT and thyroid disease does not exist. There may be individuals in the overall population who were exposed to Hanford radiation and did develop thyroid disease because of their exposure. However, in backing away from the certainty that was a main theme of the January release, CDC might have gone too far. In using "preliminary results", the agency seemed unwilling to acknowledge that the study had reached the final-report stage. Later in the document, CDC referred to the 'initial study results' provided in the Draft Final Report. Although the peer-review process had not been completed, it is clear that the Draft Final Report of this 9-year, $~S million study had progressed well beyond the point of preliminary analyses. Such terms are normally reserved for periodic progress reports. Both undennterpreting and overinterpreting the results ofthis major study are problematic. Updating the Communication Section of the HTDS Draft Final Report The subcommittee recommends that the communication section of the HTDS report be updated to reflect the development
126 Review of the HTDS Draft Final Report of a new plan to release the final version of the report and to provide some history of the communication efforts made for the Draft Final Report. The new plan should include such issues as whether the whole study or only the changes (if any) will be released and how uncertainty will be discussed. If the changes made to the Draf Final Report are minor, only minor planning is needed. However, given the public dismay at the release of the Draft Final Report, the subcommittee believes that major communication planning will be required to ensure that the integrity of the StUdY and the investigators is maintained in the , do, ~ . .. .. . ~ . ~ . ~ ~ . _ . . eyes ot both the public and the media. Some plans will also be needed for the eventual publication of the article submitted to a scientific journal, if it is accepted, because its publication could attract more media attention. Additional requirements for a new communication plan will be discussed below. RELEASE OF THE DRAFT FINAL REPORT With hindsight, one can often see why a reasonably well-planned risk-communication plan was not successful. However, during the planning of a risk-communication effort, it is often difficult to evaluate how members of the public will respond to messages or whether they will even pay attention to them. There was little worry that the messages in the HTDS Draft Final Report would reach an interested population, including national and regional journalists. However, as described earlier, three major unplanned factors led to an early release of the report. In addition, at the time of the release, another unplanned factor- a leak to the New York Times about the report put more stress on the release situation. To evaluate the outcomes of the various briefings that occurred with the release of the report, it is important to look at some different aspects, if only briefly: the planning process, the need for an information blackout, the briefings themselves, the leak to the Times, and the selection and effect of the main messages presented at the briefings.
Communication of Study Results 127 Planning Process To release a report of great interest and magnitude, such as the HTDS report, is not a simple task, and neither the investigators nor the CDC appeared to take it lightly. Plans described in section X of the Draft Final Report were ambitious and generally well designed. With the early release of the report, efforts had to be speeded up, and this possibly led to some problems because many audiences had to be dealt with: HTDS participants; the various advisory and citizen groups in the region; state public-health officials and state, county, and local government leaders in three states; tribal officials; Washington, DC, officials, including those at the Department of Health and Human Services (DHHS) and congressional delegations from the three states involved; and, of course, the mass media. In addition, the printed materials for the public had to be prepared as did material that would appear on the HTDS Web site with the release of the report. Mailings with study results had to be sent to hundreds of people, including the study participants. That is a great deal to accomplish in the 3 months between the decision to release the report early and its official release date. Nevertheless, CDC had a well-planned "rollout" schedule, listing all the tasks, who would do them, and the necessary deadlines. Complicating the planning and matenal- development process was the need to procure many clearances in CDC and DHHS that took about 2 weeks, a "fast-track" timeframe. Some concern has been expressed that as some of the draft public materials went through the clearance process, some of the qualifiers on the findings were dropped and the message became more positive. That is often the case when many messages have to be cleared through channels, and it might have occurred here, but no evidence to that effect was seen. Need for an Information Blackout Part of the planning appeared to include an information blackout of the HTDS results until official release of the report.
128 Review of the HTDS Draft Final Report Supposedly, the blackout would let the study participants and the public know the study results at about the same time. That was somewhat unrealistic, given the number of groups to be briefed. The more briefings scheduled, particularly if some are a day or so in advance of others, the greater chance that infonnation will be leaked as it was in this case. in addition, the blackout had an unhappy effect on the citizen groups, including the HTDS Advisory Committee, which had been kept informed about the study's progress over the years. Members had come to expect to be updated about what was happening in the study and became upset about not being told the main findings earlier than the day of the public release. [n retrospect, there were reasons to keep the findings of the study confidential, but perhaps the most important need to communicate first and foremost with the public-was downplayed. Trying to brief so many official parties before the public created a substantial opportunity for information leaks. This subcommittee believes that trying to maintain an information blackout, given the number of briefings needed, was problematic and unrealistic. Briefings According to the schedule provided by CDC, briefings about the report were to start in Washington, DC on January 22 and 26 with officials of CDC, DHHS, DOE, the National institute for Occupational Safety and Health, and the Agency for Toxic Substances and Disease Registry. A briefing for congressional delegations was scheduled for January 27. On the main release ~lay, January 2S, the briefings moved to the state of Washington. Two morning conference-call briefings were scheduled for state health officers and the Northwest Tribal Nations and Indian Health Service. On the 2Sth, at ~ pm, four citizen groups, including the HTDS Advisory Committee, were to be briefed by conference call. The media briefing was to occur at 3 pm, and the public meeting on the report was scheduled for 7 pm in Richiand, WA. Material concerning the study was to be posted on the Web site at 3 pm.
Communication of Study Results 129 It was an ambitious schedule, and it represented a compromise from the earlier discussion in the Draft Final Report of using satellite connections for remote meetings rather than conference calls. One participant in the calls felt that they did not work well, because they were far too impersonal and unwieldy. Participants did not know who was on the other end; they could not show graphs or other illustrations; they could only deliver an abbreviated version of information that would be presented at the media briefing and take questions. There also was frustration for the people on the receiving end of the conference calls, according to information given to this subcommittee. They briefly heard a message that they had not expected, had few details about the study, had nothing in writing, and could only ask general questions. In addition, because they had no written materials, they could not respond to questions being directed at them about the report by journalists on January 28. Another problem in retrospect was that the briefings even those for the media or the public were not taped, so they could not be transcribed later for interested parties. Several people in the region concluded that CDC did not take the briefings seriously enough to record them for review. Given that there were no transcripts, it is difficult to evaluate how well the briefings were done. One reporter at both the media beefing and the public meeting was surprised by "how absolutely confident the Hutch people were." She pointed out that subtleties and uncertainties were not discussed, nor were any problems with statistical power. She noted that scientists usually are not that positive about their studies and open make "conditional statements particularly when a study is still a draft and hasn't undergone peer review" (Steele, 19991. However, another reporter who attended the media briefing said that even if the uncertainties in the study had been stressed, the media probably would not have emphasized them. She noted that the press "wouldn't have dwelt on the uncertainties", because the media, particularly the broadcast media, do not get into all the technical
130 Review of the HTDS Draft Final Report details, and she said that they would report only that "the bottom line is this. That's the way the media operate" (Cary, 1999~. Effect of the Leak to the New York Times After the congressional briefing on January 27, the HTDS results were leaked to a Times reporter, who quickly wrote an article about them. The story was picked up by the Associated Press and was on the wire to be seen by reporters in the Hanford area about ~ pm PST that night. It was early enough, said one reporter, that she was able to add some local reaction about the findings to a story that she had been writing about the release of the report the next day. Her newspaper also ran the New York Times story on the morning of January 2S, before both the media briefing and the public meeting (Cary, 19991. Again, an uncontrolled situation had changed a carefully planned risk-communication strategy. The leaked storer sent reporters and HTDS and CDC officials scrambling. CDC media officials starting faxing materials about the study results to reporters at 6 am EST on January 2S, not waiting for the 3 pm media briefing. They also began putting all the planned information about the report on the Web site at 3 am EST. It is hard to evaluate the impact of the article with the leaked information on the planned risk-communication process, but several members of citizen groups said that they were upset because reporters were calling them for comment on the morning of January 28 and they had not yet been briefed. Even after they were briefed, they still had no written data and had not read anything official about the study to which to respond. They felt that they were put into an awkward position. There also has been some supposition that reading the results of the study in that day's newspapers made the people who attended the public meeting in the evening more angry than they would have been otherwise. It is hard to know whether that is valid. However, as one reporter said, the Times story set the tone for most of the media coverage that followed, and it was headlined "No Radiation Effect Found at
Communication of Stucly Results 131 Northwest Nuclear Site" (Wald, 19991. Many who presented information to this subcommittee during its public meeting in June in Spokane complained about the media coverage of the study and its implied dismissal of thyroid problems at Hanford. That point is exemplified by a headline found in the Salt Lake Tribune, topping an AP story on the study: "Study Disputes Hanford Poisoned People" (Associated Press, 19991. For a number of people in the Hanford region, such a conclusion was unacceptable, and they blamed the media position on the overpositive and strong message provided in the Draft Final Report and the various briefings. The Message and Its Effect in evaluating the risk-communication process and the activities that occurred around the HTDS Draft Final Report, it is hard not to question whether the public dismay with the release of the report would still have come about if the message had been different. The main message no link between radiation exposure at Hanford and prevalence of thyroid disease was not expected by concerned members of the public in the region. Given the findings of the NTS, Chernobyl, and other studies and the documented radiation releases from Hanford, a positive association was expected, according to interviews with local journalists and some concerned citizens. Delivering unpopular risk messages is itself risky. in many instances, it has to be done delicately, with great thought about how it will affect an audience expecting an opposite result. Varied audience responses have to be forecasted and planned for. Sensitivity to audience health concerns and fears needs to be shown. In this particular situation, with an audience very concerned about perceived high rates of thyroid disease in the population an audience that had been reported to have little trust in government agencies great care should have been taken to deliver the results of the HTDS sensitively and tactfully. Implications for individuals and families that have suffered from
132 Review of the HTDS Draft Final Report thyroid disease should have been not only explained, but also highlighted. Perhaps knowing that they faced a difficult task, the HTDS investigators felt that they had to deliver as strong and positive a message as they could about their findings, indicating to people in the region that they should fee! relieved that no link had been found. But that was not a sensitive way to proceed, given the audience. One important error was to emphasize the statistical group effect and not the outcome for individuals. It was only in response to questions at the briefings that they acknowledged that their findings did not negate the suffering of people in the region from thyroid disease. Later, they explained that they had left the uncertainties in the study undiscussed during the briefings and in the written public material because the focus groups that they had worked with on the public summary during the fall of 1998 had told them that anything technical was not appropriate for the public materials. CDC's role in the message selection and delivery is more complicated. People providing information to this subcommittee in Spokane questioned why CDC had not intervened to counter the overpositive message about the study given by the HTDS investigators in the briefings and the written materials. They said that the investigators were contractors and that CDC was ultimately responsible for what was said about the study. They charged that CDC had done a disservice to the people of the region. Clearly, this is an important and complex issue. It involves agency-contractor responsibilities and relationships, academic freedom, and responsibilities to the public. It is even more complicated if one remembers that in December 199S, HHES members and other citizens urged CDC not to alter the report as it came from the investigators and to release it as it was. Those concerns helped to put CDC officials between a rock and a hard place. if they asked the HTDS group to soften the tone of the findings, they could be accused of altering the investigators' report. It is apparent that CDC officials would have been criticized
Communication of Study Results 133 by various groups no matter what actions they took, because of the numerous audience expectations about the study results. CDC officials said that the main message was decided by collaboration between them and the HTDS investigators on the basis of numerous discussions. Several CDC officials noted that they had concerns that some of the messages were too strong, such as "this was a powerful study." Despite their concerns, after struggling with some of the language in the report, they decided to leave it as drafted by the HTDS investigators because of public pressure not to alter the report. The subcommittee recognizes and supports CDC's sensitivity to citizens' concerns and the needs of academic freedom for investigators, but it believes that there was a middle ground: both the HTDS investigators and CDC officials should have expressed their own views and interpretations about the Draft Final Report at the briefings and in the public documents. Although consensus might have been preferable, the differing interpretations should have been presented to the media and the public. That is preferable to having one point of view dominate the other regardless of which side dominates and then backtracking to change or soften a message. Such advice does not agree with some generalized risk-communication guidelines, but such guidelines must be adapted to specific situations. In this case, many members of the public and the journalists in the Hanford region were actively engaged in the issues and educated about them. They were not going to accept a simplified approach and message, particularly if it disagreed with their own experiences and points of view. Rather than presenting a black and white picture of the results with a positive spin, the HTDS and CDC personnel should have emphasized the shacles of gray. Despite its sensitivity to problems with the language in the report and concerns over what to do about it, CDC itself showed insensitivity to people and families with thyroid disease in the region when it announced at the public meeting on January 28 that it would recommend a change in plans for medical monitoring.
134 Review of the HTDS Draft Final Report No matter what reasons officials gave including a report by the Institute of Medicine questioning the value of medical monitoring-the public linked this action to the announced results of the HTDS. And although CDC took great pains to point out to the public and the media that the HTDS report was a draft and would be subject to peer review and public review, the agency appeared to be basing policy decisions on it. Even if the decision regarding medical monitoring was correct, announcing it at the same time that the HTDS Draft Final Report was released was a mistake and hurt CDC's credibility. Recommendations about Releasing the Final Report The subcommittee recommends that CDC continue its open-communication policy on the HTDS and improve on it for release of the final report. it applauds the development of materials for the public such as the newsletters, the background fact sheets, and the Web site-and recommends that it be continued. It is important to remember that those efforts were well received in the community, and they should not be overshadowed by the problems encountered with the release of the Draft Final Report. In writing and releasing the final report and its public summary, steps must be taken to explain alternative interpretations of the data and to ensure that findings are presented in an even- handed method that does not overemphasize one point of view. Efforts must be made both in the report and its accompanying public documents to explain the implications of the findings for individuals and families sensitively, indicating, for example, that a statistical study does not necessarily negate the existence of thyroid disease in this population and explaining why that is so. The messages in the final report must be developed with sensitivity to audience health experiences, concerns, and fears. Any substantial changes made from the Draft Final Report should be clearly outlined and explained, including why they were made. Remaining uncertainties must be acknowledged and explained, along with their implications and effect on the final
Communication of Study Results ~l35 conclusions. CDC and HTDS personnel should work together on the wording and presentation of any public messages, presenting alternative interpretations of data and conclusions as needed. Given the controversy that already exists over this report, presenting only one viewpoint will lead to more public distrust. A new communication plan should be developed for the final report. It must take into account and acknowledge the problems encountered with the release of the Draft Final Report and include dealing with possible lowered public trust in the HTDS investigators and CDC in the Hanford region. Because of serious problems encountered in trying to maintain an information blackout, such efforts should be minimized. For the final report, multiple briefings should be abandoned, and there should only be an early briefing for CDC and DHHS officials followed quickly, if possible given political realities, by one large briefing for all other parties, using satellite transmission or other~advanced technology to link groups in various locations. In particular, citizen groups that have participated in the study process over the years should not be kept out of the information flow until the last minute. All media and public briefings should be videotaped to provide a record of the proceedings. Journalists should receive copies of the final report several days in advance of its official release after agreeing not to write stories about it until the release. That practice, known as embargo, is widely used and, particularly on complicated subjects, allows journalists time to study a report and develop thoughtful articles. Given the investment of time and effort by the people who participated in the HTDS, they should be randomly surveyed as to their satisfaction with the communication of the draft study findings and their own dose results, so that communication to this group can be improved for the final report. In light of the importance of the HTDS and future CDC reports to the public on radiation, the subcommittee suggests that the agency hold a workshop of selected risk-communication experts, scientists, journalists, and members of citizen groups to
136 Review of the HTDS Draft Final Report discuss some of the important communication issues that have been raised in this case and the complex topic of the advisability of releasing unreviewed draft reports to the public. In particular, such a workshop would help to focus the growing body of social- science risk-communication research on questions about audience response to such reports as the HTDS report and simultaneously produce new research questions for systematic study. Such a workshop could also investigate how the government relates to and discusses with the public the levels of uncertainties involved in various scientific studies, as well as alternative ways of addressing public concerns about issues like Hanford. Much still needs to be known about how members of the public use and respond to risk- related messages, including the complicated roles of trust and credibility in how audiences accept and process risk messages.
