7
PROCESS FOR EVALUATION OF OPTIONS FOR DISTRIBUTION OF POTASSIUM IODIDE
This chapter first discusses the responsibilities for distribution of KI at different government levels and then presents an example with sample plans and multiple objectives for evaluating plans. (A decision-analysis method that could be used by local authorities to quantify the performance of plans on multiple objective is presented in detail in Appendix D.) The chapter concludes with a discussion of implementation issues.
The previous six chapters of this report have presented evidence to address the following key questions:
-
Is there a risk of a nuclear accident releasing radioactive iodine and causing ingestion/inhalation by vulnerable people?
Depending on local conditions surrounding nuclear power plants, there will be some very rare scenarios in which vulnerable members of the general public could inhale radioactive iodine. We assume that
food and milk sources will be controlled, so ingestion of radioactive iodine will be very unlikely.
-
Is KI effective in preventing the risk of thyroid cancer after a nuclear incident?
Yes, taking stable iodine (in the form of KI) within a few hours before or after exposure to radioactive iodine through inhalation or ingestion of fallout will protect the thyroid from thyroid cancer caused by such exposure in vulnerable populations. See Chapter 2 for thyroid physiology and Chapter 4 on stable iodine prophylaxis.
-
Who should take KI when an accident occurs?
The most vulnerable people are the young. It is most important for KI to be taken by children, infants and pregnant and lactating women (to protect fetuses and breast-feeding infants). See Chapter 2.
-
What is the radiation exposure intervention level at which KI or other protective measures should be taken?
We recommend that KI prophylaxis be ordered if there is likely to be an avertable cumulative radioiodine dose of 50 mGy (5 rad) to children’s thyroids. See Chapters 2 and 5 for discussion of different agencies’ recommendations on radiation dose thresholds beyond which protective actions should be taken. We also recommend that this threshold for intervention be kept under review as further information on the consequences of exposure to radiation from fallout from Chornobyl and from other radiation incidents becomes available. See Chapter 6 for radiation dose levels used in other countries.
-
Is KI safe?
Yes, in general. Relatively small numbers of people should not take KI, including those with some pre-existing thyroid conditions, iodine allergies, and some other rare medical conditions. Thyroid conditions tend to occur after the age of 40 years. See Chapter 2. It is important to avoid overdosage of the fetus and the very young child, by following the 2001a FDA guidelines. When KI is given to infants
under one month of age, FDA guidance for monitoring for hypothyroidism should be followed. See Chapter 6.
-
Does the terrorist threat add significantly to the risk beyond that with an accident?
The terrorism threat does not appear to add significantly to the risk, because of existing safety mechanisms and procedures. See Chapters 3 and 5.
In summary, KI is effective and generally safe. Nuclear power plants in the United States contain a source of radioactive iodine that, in the event of a very rare severe incident, might impose risks of exposure to radioactive iodine via inhalation and ingestion in the vulnerable population, which could lead to thyroid cancers. Given that KI is effective in protecting against thyroid cancer, KI distribution plans should be considered.
A general process for evaluating the advantages and disadvantages of different types of KI distribution plans, given the specific features of a local region, is presented in this chapter. It allows us to address the following questions:
-
Should KI be predistributed to individuals or households around NPPs? If so, within which geographical area should it be predistributed (that is, a 10-mile or a 20-mile radius or some other area)? If not, what are the other options, and on what grounds should decisions be made?
For each nuclear power plant, there is specific information on population size, ages, and location, geographical and meteorological conditions, evacuation routes, and resources available. This information can be used for quantitative analysis to aid planning in a specific region. Since the committee was not charged to carry out quantitative analysis of health risks and emergency plans at specific plants, such detailed analysis is beyond the committee’s scope. The general method presented here and in Appendix D could be used for such site-specific quantitative planning.
Responsibilities for Distribution of KI at Different Government Levels
In general, emergency preparedness requires coordination between federal agencies and state and local authorities, both during responses to an incident and in making preparedness plans. Decisions and planning actions regarding KI made at different government levels need to be coordinated.
In planning related to KI, it is important to keep in mind that its timely use reduces only the radiation exposure to the thyroid gland. That is important because it does not protect the rest of the body from radioactivity. Accordingly, KI is a supplementary measure to the primary options of evacuation and sheltering.
Federal Level
At the federal level, information and resources have been and could be provided for KI distribution in the event of a nuclear incident. Various agencies (including the FDA, the Nuclear Regulatory Commission, FEMA, and the Environmental Protection Agency) have issued guidelines regarding KI, as discussed in Chapter 6.
Potassium Iodide Distribution Before Incident
As described in Chapter 6, the Nuclear Regulatory Commission made a one-time offer to provide a supply of KI tablets to states to be made available to populations near NPPs. States accepting KI were required to develop and submit a plan within a year after receiving the KI for how it would be distributed (to people or stockpiles) in advance of an incident. The Nuclear Regulatory Commission provided only the KI, not resources for funding its distribution or for related communication or education campaigns. When the KI that was supplied reaches its expiration date, states might request resupply from the Federal Government, although no
offer to resupply KI has been made. The KI provided by the Nuclear Regulatory Commission did not come from the Strategic National Stockpile (formerly called the National Pharmaceutical Stockpile), maintained by the Centers for Disease Control and Prevention (CDC).
Potassium Iodide Availability at Time of Incident
A federal-level option is to use the potassium iodide which is included in the formulary of the CDC Strategic National Stockpile (SNS) for rapid deployment of KI to the affected part of the country in the event of an incident. The Department of Homeland Security is responsible for deploying the SNS. Day-to-day pre-emergency management of the SNS is overseen by Department of Health and Human Services, which deploys CDC to carry out the task. See the CDC Web site www.bt.cdc.gov. Two general approaches are available:
-
Stockpiling KI in Vendor-Managed Inventory. The vendor-managed inventory for the SNS has KI in it. Vendors are responsible for managing and restocking these inventories. This allows shipments of packages with just KI or any other incident-specific items that are needed within a day, or perhaps sooner. A modification of current procedures might be possible to ensure even quicker delivery of KI, in the very unlikely event that it would be needed.
-
Stockpiling KI in 12-Hour Push Packages. KI could be added to all SNS push packages, which are prepackaged with a set of supplies useful for a variety of disaster incidents. The push packages will reach the incident area first, by air to the nearest safe airport or ground transport within 12 h of the incident. KI is relatively small, so many dosages could be stored in a relatively small space.
After the September 11th World Trade Center disaster, a push package reached its New York City destination in 7 hours, and 60 loads of vendor-managed inventory packages began to arrive within 12 hours. Federal-level stockpiling provides the possibility of restocking of local stockpiles in the days after an incident if fallout exposure persists. This might also provide protection in an extremely unlikely incident away from a NPP facility or in the event of a somewhat more likely but still rare spread of a radioactive iodine plume far from a NPP if the general public is not evacuated.
State and Local Levels
State and local authorities have different specific responsibilities and authority in different locales. Two general approaches for distributing KI are stockpiling in local areas near NPPs but outside the EPZ and predistribution to people within the EPZ. As discussed in Chapter 6, some authorities have determined that KI is not needed for the public, because prompt evacuation is feasible for the entire nearby population except emergency workers, plant workers, and non-mobile special populations. Others combine evacuation plans with KI predistribution. A key decision, if KI distribution is deemed needed, is what type of distribution (stockpiling or predistribution) to use in each area surrounding each plant. Chapter 6 contains details of what decisions different states and other countries have made on predistribution areas and delivery. The next section discusses local distribution options in more detail.
Local Area Potassium Iodide Distribution Options
For optimal benefit, KI should be administered to the potentially affected populace just before, concurrently with, or within a few hours after exposure to radioiodine. (Note that we are focusing on the threat posed by inhalation, not ingestion. We assume that sufficient controls and procedures are in place to eliminate dangers
posed by ingestion. See Chapters 2 and 5.) In light of the stringent timeline, there are two primary options around fixed nuclear facilities: predistribution and stockpiling.
Predistribution
Predistribution involves providing KI to all or segments of the potentially affected population as part of the preparation for responding to an incident. The principal modes are: direct distribution to individuals or groups, making KI available on a voluntary basis to those interested in obtaining it, and a combination of the two. The main advantage of predistribution is the potential immediate availability of KI to the affected population at the time of an incident and in a time span when it can do the most good. However, predistribution raises a number of questions:
-
How will distribution occur? Some states have mailed KI to households. Others have used public officials to distribute KI door to door. Still others have instituted voluntary programs that publicize the availability of KI at such locations as county health offices, government agencies, and local pharmacies. Table 6.12 in Chapter 6 shows states’ experiences with distribution methods and success in reaching target populations. Voluntary programs have rarely exceeded 50% participation by the potentially affected population.
-
Where will it be distributed? For example, some plans predistribute to all households in the 10-mile radius EPZ around a plant. The size and location of the geographic area should be determined based on the projected averted radioiodine dose (and subsequent morbidity or mortality), see Chapter 4. The aim should be to have KI available for the vulnerable population that the state predicted would be exposed to 50 mGy (5 rad) or higher committed dose equivalent to the thyroid during an incident after other protective actions had been implemented.
-
How will those receiving KI be instructed in its purpose, proper storage, and use, including instruction that evacuation should not be impeded by attempts to locate the tablets?
-
What assurance is there that people will be able to find the KI when it is needed?
-
What resources are needed for a predistribution program?
Stockpiling
Stockpiling involves collecting and retaining KI at fixed locations, usually outside the designated EPZ, from which it is dispensed to the potentially affected population before, concurrently with, or immediately after actual or expected release of radioactive iodine into the atmosphere. Its advantages are increased certainty of the availability of KI to the affected population (those individuals who were not in the EPZ, individuals evacuated from the EPZ or individuals impacted beyond the EPZ), better control of its administration, and better recordkeeping. However, as a method of distribution to the general public, stockpiling does not work well in the relatively rare instances in which evacuation would be impractical, undesirable, or delayed. Associated issues include the location of the stockpiles and the resource requirements for a postincident distribution program, particularly in light of the short period during which KI is to be administered.
Sample Plan Options
There are tremendous regional differences in geography, population density, meteorological conditions and other characteristics related to NPPs. Thus, it is preferable for each local area to develop specific plan options and evaluate them on the basis of how well they meet their objectives in light of the area’s characteristics (such as population size, ability to evacuate the entire affected population, and risk of radioactive iodine exposure). Table 7.1 contains four sample options for KI distribution plans. The list is not comprehensive; see Chapter 6 for the large variety of actual plans.
Local areas would need to augment the plans with specifics about reaching special populations and communication and education plans. Specific details that would need to be included in a plan are discussed at the end of this chapter, under implementation issues.
Table 7.1 Four Sample KI Distribution Plansa
Objectives to Consider in Evaluating Plans
The overall goal of a KI plan is assumed to be this:
To ensure the availability of KI to the vulnerable population soon enough to be effective in incidents with a potential or actual release of radioactive iodine into the atmosphere.
Local areas should establish their own objectives to be used in evaluating their plans. Keeney (1992) provides guidelines for constructing objectives for evaluating decision
options. Objectives should specify the preferred direction for improved performance (for example, to minimize panic and anxiety). In some situations, it may be useful to characterize the objectives of different stakeholder groups (Winn and Keller, 1999, 2001). For example, Keeney, Renn, and von Winterfeldt (1987) constructed an objectives hierarchy for the former West Germany’s energy supply by combining objectives of multiple stakeholders with divergent views.
Table 7.2 contains typical objectives that may be used by state or local officials in evaluating plans. The presence or importance of each objective may vary from one region to another because of specific features of the local situation. For example, it may be important to make sure that special nonmobile populations have access to KI, but an objective of covering special populations may not be included if this population will be dealt with by a separate stand-alone plan. For a specific region, it may be that evacuation could be complete within 24 h in all scenarios, so ensuring multiple-day supplies of KI in homes would not be a planning objective. Also, objectives might explicitly include minimizing mortality and morbidity due to radioiodine exposures to thyroids.
Table 7.2 Objectives for Evaluating Plans
A. |
Minimize Radioactive Iodine Risk to Thyroid |
||
|
a.1. |
Maximize KI availability (measured by coverage of KI now, and after any planned distribution) |
|
|
|
a.1.1. |
For children and pregnant women residents (including at home and not at home) |
|
|
a.1.2. |
For other residents (including at home and not at home) |
|
|
a.1.3. |
For mobile population |
|
a.2. |
Optimize ability to take KI on time |
|
|
|
a.2.1. |
Number of people who know where KI is |
|
|
a.2.2. |
Optimal time if no evacuation (not too early; KI accessible at time of incident) |
|
|
a.2.3. |
Optimal time if evacuation (not too early; KI accessible at time of incident) |
|
|
a.2.4. |
Storage to ensure stability |
|
a.3. |
Minimize harm from inappropriate KI administration |
|
|
|
a.3.1. |
Correct dosage given (and taken) for age |
|
|
a.3.2. |
First dosage not taken too late (avoid possible increase in thyroid risk) |
|
|
a.3.3. |
Adverse side effects (non-thyroid cancer) minimized (in healthy people, thyroid patients, and those with iodine allergies) |
B. |
Minimize Harm from Other Aspects of Incident |
||
|
|
b.1. |
KI procedures don’t impede evacuation |
|
|
b.2. |
Avert mortality and morbidity from radiation or accidents (beyond thyroid risks) |
|
|
b.3. |
Minimize panic and anxiety due to KI procedures |
|
|
b.4. |
Avoid excessive resources use in KI procedures |
|
|
b.5. |
Simplify KI procedures before and during incident |
|
|
b.6. |
Educate public to respond to nuclear incident |
Once a set of alternative plan options is developed, their performance can be evaluated, and advantages and disadvantages of the plans can be identified. For example, residents may all be told to store predistributed KI tablets in a specific safe place. The advantage would be that they could help each other to remember where the tablets are stored. The disadvantage is that in the disaster people might panic and “borrow” (steal) from each other because they would know where the tablets would be.
The performance of a KI distribution-plan option may be evaluated according to the objectives by using descriptive text or a 0-10 scale, where 10 is best and 0 is worst. Appendix D contains a sample set of scales to evaluate performance on the objectives in Table 7.2.
If an option appears dominant on all objectives, it should be seriously considered for implementation by decision-makers. However, it is likely that no plan option will be dominant on all objectives. The relative importance of objectives may need to be examined in more depth by assigning importance weights. By convention, weights are normalized to sum to 100%.
The overall value of a plan can be computed by multiplying the weight of an objective by the rating of the plan’s performance on the objective, and then summing the products over all objectives. The plan with the highest overall value would be the one recommended. The evaluation template in Table 7.3 shows the format for such an evaluation. This multiple-objective evaluation approach has been used widely, for example, in evaluating Los Angeles Unified Schools desegregation plans (Edwards, 1979, 1980) and Mexico City airport siting (Keeney and Raiffa, 1976), and the merger of two professional societies in operations research and management science (Keller and Kirkwood, 1999).
Table 7.3 Plan Evaluation Template
|
Description of How Well each Plan Meets Each Objective (Rate from 0 to 10 = best) |
|||||||
|
Importance Weights (Sum = 100%) |
MM: Predistribute in Mass Mailing in KIPZ |
VP: Predistribute via Voluntary Pickup in KIPZ |
RC: Stockpile at Evacuation Reception Centers outside KIPZ |
ND: No Distribution of KI |
|||
Minimize Radiation Health Risks to Public |
|
|
|
|
|
|||
A. |
Minimize Radioactive Iodine Risk to Thyroid |
|
|
|
|
|
||
|
a.1. |
Maximize KI Availability |
|
|
|
|
|
|
|
a.1.1. |
For children and pregnant women residents |
|
|
|
|
|
|
|
a.1.2. |
For other residents |
|
|
|
|
|
|
|
a.1.3. |
For mobile population |
|
|
|
|
|
|
|
a.2. |
Optimize Ability to Take KI on Time |
|
|
|
|
|
|
|
a.2.1. |
Number of people who know where KI is |
|
|
|
|
|
|
|
a.2.2. |
Optimal time if no evacuation |
|
|
|
|
|
|
|
a.2.3. |
Optimal time if evacuation |
|
|
|
|
|
|
|
a.2.4. |
Storage to ensure stability |
|
|
|
|
|
|
|
a.3. |
Minimize Harm from Inappropriate KI Administration |
|
|
|
|
|
|
|
a.3.1. |
Correct dosage given (and taken) for age |
|
|
|
|
|
|
|
a.3.2. |
First dosage not taken too late |
|
|
|
|
|
|
|
a.3.3. |
Adverse side effects (nonthyroid cancer) minimized |
|
|
|
|
|
|
B. |
Minimize Harm from Other Aspects of Incident |
|
|
|
|
|
||
|
b.1. |
KI procedures don’t impede evacuation |
|
|
|
|
|
|
|
b.2. |
Avert mortality and morbidity from radiation or accidents |
|
|
|
|
|
|
|
b.3. |
Minimize panic and anxiety due to KI procedures |
|
|
|
|
|
|
|
b.4. |
Avoid excessive resources use in KI procedures |
|
|
|
|
|
|
|
b.5. |
Simplify KI procedures before and during incident |
|
|
|
|
|
|
|
b.6. |
Educate public to respond to nuclear incident |
|
|
|
|
|
|
OVERALL VALUE (SUMPRODUCT OF WEIGHTS TIMES RATINGS) |
|
|
|
|
|
Note that a plan may combine a number of mechanisms for ensuring redundancy in the KI distribution, such as mass mailing plus voluntary pick-up. Such a plan could be evaluated with this evaluation template, as could plan variants that considered different sizes of geographical areas for direct predistribution of KI.
When considering different geographical areas for predistribution of KI, plume dispersion models could be used to examine how far and how fast the radioactive iodine fallout might spread, as discussed in Chapter 4. When considering predistribution of KI compared with stockpiling at reception centers, as the distance from a plant increases, decisive factors will likely be 1) decreasing likelihood of potential exposures given increasing amount of time available to evacuate prior to plume passage and 2) increasing resources needed to predistribute KI to a larger region.
Appendix D contains an illustration of the process of evaluation that a local region could go through. The four generic distribution plans presented above in Table 7.1 are evaluated against the objectives above for a hypothetical plant site. The purpose of this exercise is to demonstrate a process, not to identify the best generic one-size-fits-all approach. It also demonstrates that different regions might choose different plan types, depending on regional characteristics and weights of objectives.
Implementation Issues
A number of specific details should be considered in developing a KI distribution plan.
Who Advises the Population To Take KI, and Under What Circumstances?
Disaster management officials in each state or locale will need to predetermine who will advise the public to take KI and the expected radioactive iodine dose at which vulnerable persons will be ordered to take KI if they will be remaining in an area where exposure
may occur or has occurred. The officials will also tell the public when to stop taking KI.
See Chapter 2 for discussion of the scientific evidence supporting radioactive-dose thresholds. Table 2.4 contains our recommendation to administer KI to the vulnerable population when the predicted avertable dose to the thyroid in that group due to radioiodines reaches the threshold of 50 mGy (5 rad). Table 6.2 in Chapter 6 contains different thresholds used in different countries.
What Number of KI Dosages Should Be Issued and Stockpiled for Each Type of Person (Children, Pregnant and Lactating Women, Teenagers, and Young Adults)?
See Chapter 6 for the number of tablets issued or stockpiled for different groups of people at different locations. Table 2.4 contains guidelines on the amount of KI tablets (or parts of tablets) of different sizes to take for different ages. See also Chapters 2 and 5 for related discussion on the need for simplicity in dosing instructions and confusion about old and new FDA guidelines.
Education and Communication Plans
The plan should include structuring, implementing, and sustaining a public-awareness and education campaign to support a predistribution program (if there is one) and to inform people about the use of KI and radiation risks.
For more information on communication considerations, see Improving Risk Communication (NRC, 1989), in particular Chapter 6, “Problems of Risk Communication”, and Chapter 7, “Recommendations for Risk Communication”:
“Risk communication requires its own specialized expertise and deliberate planning and evaluation. Senior managers need to devote attention and time to managing risk communication efforts per se. It is a mistake to simply consider risk communication to be an add-on activity for
either scientific or public affairs staff; both elements should be involved. There are clear dangers if risk messages are formulated ad hoc by public relations personnel in isolation from available technical expertise; neither can they be prepared by risk analysts as a casual extension of their analytic duties [NRC, 1989, p. 148].
Both the management of the process of formulating risk messages and the content of the messages should be systematically oriented to the intended audience…the best procedures for formulating risk messages have been those that elicited recipients’ perceptions and needs” [NRC, 1989, p. 148].
Of special concern is the effect that communication in a KI predistribution program might have on the public’s psychological reactions. It might be that predistribution of KI will raise fears of nuclear incidents, even though it is intended to decrease risks. Conversely, it is possible that KI might be erroneously seen as a “magic bullet” against all radiation risks, making all other risk-reduction steps wrongly perceived as unnecessary.
See Appendix E for examples of materials used in communicating with the public about KI and for informed-consent forms and response forms used for brief two-way communication. The message to be conveyed to the public, to public-health workers, and to others will need to cover many details. For example, in the event of a nuclear incident, people should take a specific amount of KI (for children, tablets can be broken into fourths or halves). Increased amount of KI should not be taken in the mistaken assumption that if a small amount is good for protecting your child or yourself, larger will be better; and taking KI will reduce your risk of thyroid disease but will not protect you from the radiation effects of other radioactive materials released with the radioactive iodine.
Plans for Special Populations
When people are in schools or other institutions (such as day-care centers, prisons, and hospitals), extra procedures may need to be planned. Institutions with the most vulnerable populations (children and pregnant and lactating women) should consider having stockpiles of KI onsite with distribution permission slips from parents of children. Professionals who provide services to children might develop specialized programs for distributing KI and information, such as pediatricians and infant services. Methods may need to be developed for tracking which institutionalized persons have taken KI. For example, if schoolchildren are given KI at school and then sent home, they might be given a sticker on their shirt or a mark on their hand so that their parents or emergency workers do not give them a second dosage. This would also give workers a quick way to make sure that everyone who should be given KI is covered. Populations that are hard to move, such as those in jails or hospitals, may need KI on site for multiple days of sheltering, especially if they include vulnerable people.
Restocking and Resupplying
Plans for restocking should be made in case the official or practical shelf-life of KI expires. It would be appropriate for the Federal Government to provide the required resources to support restocking. Plans should also include sources for resupplying—from the Strategic National Stockpile or other sources—in the days after a major disaster.
Legal and Liability Issues
Legal constraints and liability concerns may vary from state to state and should be considered in plans. Communication materials may be designed to limit liability. For example, Vermont had informed consent forms that public-health nurses instructed people to sign if they accepted predistributed KI tablets. Written parental
consent for schools to administer KI to children, in the event of an emergency, and if advised to do so by the appropriate public health authority, would appear advisable.
Resources and Opportunity Costs
Plans should determine the resources needed for KI distribution and consider the opportunity costs of forgoing other emergency or health activities when resources are devoted to KI distribution. The benefit of KI administration should be weighed against the costs.
Evaluation and Tracking System
Plans should include setting up an evaluation and tracking system to evaluate the effectiveness of predistribution programs before and after an incident. For example, the percentage of people who know where their predistributed tablets were could be tracked after 1 year, 3 years, and 5 years. The mobility of the population in the vicinity of a plant could be examined to determine how often new tablets should be predistributed.
National Registry
A national registry should be developed to be activated in the event of an incident for tracking radioactive iodine exposures (including airborne release concentrations, contamination levels in food and drink and measured thyroid radioiodine levels in individuals) and the extent of use of KI as a preventive measure (including date and time taken and dosage vs. time of release or exposure to radioiodine) and for evaluating long-term health outcomes. This would allow accumulation of greater knowledge that would be helpful in later planning.