The Committee formed small teams to review the Assessment chapters in depth. The teams focused primarily on the Key Findings, but they also provide additional suggestions for chapter authors where appropriate. Detailed, line-by-line edits are compiled in Appendix A.
CLIMATE CHANGE AND HUMAN HEALTH
The stated goals of this first chapter are to: (1) provide background information on climate change in the United States and, how through environmental and other stressors, this can affect human health; (2) provide an overview of approaches and methods used in the quantitative projections. There are no key findings for this chapter.
Generally, the chapter addresses the issues related to Goal 1 very well. The text is informative and written in a way that will be easily understandable by people from a range of backgrounds and experience. The chapter does an adequate job of summarizing climate change and how this relates to human health. In addition, much of the language and vocabulary that is needed in the later chapters is introduced. The definition of climate change provided on page 26, lines 2-4 is succinct and clearly defines the relevant issues that are being considered in this report.
For Goal 2, the overview is also adequate, although it contains very little detail or reference to other chapters. Most notably, while the general model used (top of page 35) is given and the models for the expected change in exposure are very well described, models of background rates for health impacts and exposure-dose-response relationships are not; they are often treated as general knowledge. Limited references are given for the nature of a relationship and very little detail is provided on how these data were reviewed to provide the relationships that appear in the later chapters (as noted previously in the “Answers to the Statement of Task Questions”).
- Missing components: As noted in Answers to the Statement of Task Questions, Chapter 1 needs better explanations of the system(s) used to identify and select relevant literature for the Assessment and of the definition and application of criteria for likelihood and confidence. The definition of vulnerability should be moved here from Chapter 9. Further, Chapter 1 notes that the approach to quantifying uncertainty is addressed in more detail in the Technical Support Document. That document provides sources of uncertainty but does not have details on the methods used to quantify uncertainty. It is
important to provide this information in a way that can be easily understood by decision-makers and the general public.
- The Changing Climate: This section has done a good job of introducing the concepts associated with climate change (what is climate, how is it changing, how we know this, etc.). This section is a high level overview, and while the cited references tend to focus on IPCC, USGCRP, and EPA, it may also be useful to consider literature reviews and consensus reports beyond these organizations.
- Human Health: This is a very strong section and critical to the remaining document. It needs to be better referenced to clearly demonstrate those aspects that are supported by a depth of knowledge and where there are remaining uncertainties, especially concerning the evidence of health effects from current warming. Statements like “weather and climate affect the survival and movements of mosquitoes, ticks, and rodents that carry diseases like West Nile or Lyme disease” (page 27, lines 24-25) and “Some major indicators of health, such as life expectancy, are consistently improving, while others, such as obesity and diabetes, are getting worse” (page 30, lines 20-22) need to be appropriately supported by scientific references. There are numerous examples of lapses in oversight regarding literature both here and in the later chapters, including discussion of known or suspected relationships and trends with regard to health (e.g., toxic algal plumes and health, pollen and asthma) that fail to cite references. Finally, the trends in human health and the demographic shifts in the US population are nicely linked into concerns about climate and health.
- Mortality, Morbidity, Early Death—Prevalence and Incidence: The chapter shares a great deal of information with the audience regarding health, health trends, and climate change. While the concepts of incidence versus prevalence are widely used in the scientific literature to discuss these issues, the authors here have tried to use simpler language to present their case. However, rather than provide greater clarity, this leads to greater confusion and the authors are encouraged to use a more traditional language with definitions provided in this chapter. Also, mortality, morbidity, and reduced life expectancy are not carefully explained and the complexities involved in reaching sound conclusions regarding climate change and any of these endpoints are not clear to the reader.
TEMPERATURE-RELATED DEATH AND ILLNESS
General Comments and Key Findings
Chapter 2 addresses the increases in both average and extreme temperatures and the potential contributions to death and illness, as well as implications of prolonged exposure to high temperature. The authors of this chapter have generally done a thoughtful and careful job of reviewing the major literature in this important area and capturing the key findings that can be drawn from this literature. The new modeling they cite (i.e., Schwartz et al., 2014) is well done and comprehensive and consistent with the series of other studies they have reviewed. At the
same time the authors have acknowledged and described the major uncertainties, and in general they have justified their judgments of the likelihood of and confidence in each finding. However, there are several ways in which the communication of the key findings can be improved.
Key Finding 1: Future Increases in Temperature-Related Deaths
Future climate warming could lead to thousands to tens of thousands of additional deaths each year from heat in the summer, as calculated by extrapolating statistical relationships and without considering potential adaptive changes [Very Likely, High Confidence]. Climate warming will also lead to a decrease in deaths from cold in the winter [Very Likely, Medium Confidence], but this reduction in deaths is projected to be smaller than the increase in summertime heat-related deaths in most regions [Likely, Medium Confidence].
The modeling and analysis underlying this finding seem appropriate. The statement of the finding, however, is not consistent with the underlying text and the traceable accounts: the first sentence (and throughout) should refer to “additional premature” rather than “additional” deaths and should read: “Future climate warming could lead to thousands to tens of thousands of additional premature deaths each year from heat in the summer by the end of the century.”
In addition, this finding is silent on the current state of evidence and understanding of the potential effects of current warming, an issue of considerable public and media interest and debate on which Chapter 1 touches in citing the National Climate Assessment (2014) (page 26, lines 20-22): “There have been changes in some other types of extreme weather events over the last several decades. Heat waves have become more frequent and intense, especially in the West. Cold waves have become less frequent and intense across the nation.” This issue has also been brought further into the foreground by a paper just published (Fischer and Knutti, 2015) which suggests that there have already been substantial changes in extreme temperature and precipitation due to warming, including in the United States. Chapter 2 should explicitly address the state of knowledge of the effects of current warming and the degree of confidence that the authors have in that evidence.
Also, the discussion of deaths from cold in the winter is useful, and the conclusions, especially that reduction in such deaths is projected to be smaller than the increases in heat-related deaths, are appropriately given lower likelihood and confidence. However, while studies using International Classification of Disease (ICD) codes for cold-related deaths are likely understating effects (as noted on page 56, lines 3-15), the evidence that the heat related deaths from increased temperature will be larger than the cold related deaths avoided does seem relatively weak (only one study?) and thus might merit “low” rather than “medium” confidence. The authors might also consider the recent publication of a comprehensive multinational analysis of this question (Gasparrini et al., 2015) which, although international in scope, may provide insights for the U.S.-focused Assessment.
Finally, the conditioning of these conclusions by “without considering potential adaptive changes” is appropriate, and well discussed in Key Finding 3, but a significant uncertainty for the whole chapter.
Key Finding 2: Illness and Deaths Are Related to Deviations from Seasonal Average
Days that are hotter than normal in the summer or colder than normal in the winter are both associated with increased illness and death. While large health effects are observed for extreme temperature events, mortality effects are also seen for smaller deviations of even a few degrees from seasonal averages, and small deviations from average temperature occur much more frequently than extreme events. Due to climate change, more hot days and fewer cold days are expected in the future. [Very Likely, High Confidence]
Although the first part of this finding, concerning the increases in illness and death related to extreme temperature events, is well documented in the text (e.g., the Chicago 1995 example), the second part: “mortality effects are also seen for smaller deviations of even a few degrees from seasonal averages, and small deviations from average temperature occur much more frequently than extreme events” does not appear to be supported well in the accompanying text. This is also true in the Traceable Account where (at page 69, lines 7-15) the entire discussion concerns potential “mortality displacement” or “harvesting” and is silent on evidence concerning small deviations from seasonal averages. While such evidence may exist, the current text does not adequately convey it. There also does not appear to be evidence on morbidity effects (as described on page 68, lines 14-32) to justify a finding of high confidence, e.g., “Cardiovascular and respiratory illness has been most commonly examined in relation to extreme heat, but the association is more complicated for illness than for mortality.” If this evidence exists, it should be better cited in this section.
Key Finding 3: Changing Tolerance to Extreme Heat
An increase in population tolerance to extreme heat [Very Likely, High Confidence], but not extreme cold, has been observed over time. This could be related to increased use of air conditioning, improved social responses, and/or physiological acclimatization [Likely, Medium Confidence]. Including this adaptation trend in human health projections will reduce but not eliminate the increase in future deaths from heat [Likely, Low Confidence].
This is a very useful and well documented key finding and one of the better examples of explicit discussions of the potential for adaptation— and the likely effect of that adaption on overall risk. Other Chapters should strive to have similar discussion of the potential for personal, behavioral, and societal adaptations, including information about populations of concern, where available.
Key Finding 4: Some Populations at Greater Risk
Elderly persons and people working outdoors have a higher risk of dying due to increasing frequency, intensity, and duration of future heat and heat waves. Children and working age adults have increased vulnerability to heat-related illness. The socially isolated, economically disadvantaged, some communities of color, and those with chronic illnesses are also especially vulnerable to death or illness. [High Confidence]
This is a useful finding and appropriately documented for those populations cited. It is especially valuable to include occupational and socially disadvantaged populations who may not have access to adaptive behaviors such as air conditioning. However, this finding and section needs to be reviewed in light of Chapter 9 (Populations of Concern) and made consistent with the descriptions of vulnerable populations there and throughout the report. The authors should also identify any populations identified in other chapters (e.g., Chapter 8 Mental Health and Well Being) that also might exhibit related impacts.
AIR QUALITY IMPACTS
General Comments and Key Findings
This chapter briefly reviews the literature that addresses how global change will likely impact human health via air quality exposure pathways. In this case, the term “air quality” refers to both traditional air pollutants and other airborne materials, in particular aeroallergens. Human exposures may be modified by changing the contaminant level, physical/chemical characteristics of the contaminant, and/or duration of potential exposures. The authors identified two key findings: Climate change will likely impact health (1) due to increases in ozone, and (2) due to exposure to, and the potential increased reaction to, pollen-derived material. They also identified modified exposure to indoor air pollutants as an emerging issue. Recent modeling-based papers (Fann et al., 2015 and Ilacqua et al., 2015) that were directly used in the chapter were cited and provided. The authors have done a commendable job in their review of the literature, their characterization of potential changes in exposures, the additional modeling they conducted, and their careful communication of their findings. However, there are areas that require further consideration or discussion, e.g., wildland fire impacts on air quality and how our current regulatory structure is adaptive to climate change with respect to modifying the climate-air quality response.
Key Finding 1: Exacerbated Ozone Health Impacts
Changes to the climate will tend to make it harder for any given regulatory approach to reduce ground-level ozone pollution in the future as meteorological conditions become increasingly conducive to forming ozone over most of the United States. Unless offset by additional emissions reductions, these climate-driven increases in ozone will cause premature deaths, hospital visits, lost school days, and acute respiratory symptoms. [Likely, High Confidence]
The literature and modeling analysis provide strong support for a link between global change and ozone levels. In particular, increased temperatures and decreased ventilation lead to increased levels of ozone in most areas, as compared to levels likely to be observed in the absence of climate change and the same levels of anthropogenic emissions. Ozone is linked with
premature death and other adverse effects. While much of the discussion and review is based upon pre-existing literature, the authors also specifically use the results of a recent modeling study by Fann et al. (2015) (discussed later in more detail) for more quantitative analysis to support the “Likely” finding. Based upon the literature review and the modeling conducted, the report concludes that the impact of climate change on ozone and adverse health, in the absence of additional controls, is likely with high confidence. As the authors conclude, this is well supported both by the literature and the modeling conducted. While both their modeling analysis and the literature support their Key Finding, it should be recognized and well communicated that the models (climate, emissions, air quality and health) upon which this conclusion is derived are subject to uncertainties, particularly when applied to estimating future health impacts.
Key Finding 2: Worsened Allergy and Asthma Conditions
Changes in climate, specifically rising temperatures, altered precipitation patterns, and increasing atmospheric carbon dioxide, are expected to contribute to increasing levels of some airborne allergens and associated increases in asthma episodes and other allergic illnesses, compared to a future without climate change. [High Confidence]
The authors find that the literature strongly supports that climate change will impact the level, duration, and characteristics of aeroallergens, in particular pollen-derived material. Climate warming is found to increase the length of the pollen season, thus likely increasing the exposure duration and can increase levels as well. Increased atmospheric CO2 concentrations can affect the characteristics and abundance of pollen and pollen fragments, and the literature suggests that the changes will lead to an increased allergenicity of the pollen due to physical and compositional modifications to the pollen. Global change will also lead to changes in transport and loss of pollen and pollen fragment. The health outcomes of concern are respiratory diseases (e.g., asthma) in response to exposure. The authors state that the finding of potential adverse health outcomes is of “high confidence.” However, factors that may lead to reduced adverse allergic and asthmatic responses (e.g., some areas becoming drier, potential shortening of the pollen season due to plant stress) should be discussed briefly. This is particularly true as the authors did not do any quantitative modeling to assess potential future impacts. Their finding that climate change will lead to worsened allergy and asthma conditions is supported by a large body of literature on this subject, including articles that have reviewed the literature.
Climate Impacts on Particulate Matter from Wild Land Fires and Dust
The authors (appropriately) found that the literature is mixed in terms of the likely impact of global change on particulate matter (PM, in particular PM2.5), focusing primarily on PM from traditional emission sources. This conclusion is supported by a large body of literature that has likewise found that climate warming can lead to increases and decreases in PM depending on region and time period. In this chapter, the authors discuss the potential air pollution impacts from changes in wild land fires (which can be from both intentional/prescribed and wildfires, but is often just discussed in terms of wildfires), and dust, due to climate change. These two topics are also discussed in the Extreme Weather chapter in more detail, and the linkages between those two chapters should be made clearer. In both chapters, the authors note that there is a body of
literature on the potential impacts of increased emissions from fires due to severity, number, length, etc., of both wild and prescribed fires. They state (without caveat), in the chapter, that wildfires are increasing due to climate change. There is a strong and growing body of literature on the adverse impacts to human health of biomass burning-derived pollutants. Thus, there is likelihood, and a body of literature supporting a potential finding, that global change will alter wild land fires and that this will impact air quality, including both primary and secondary PM and ozone, with resulting impacts on health. Chapter 9 of the NCA (Melillo et al., 2014) includes wildfires in the “Key Messages”. The authors should consider moving this to a Key Finding, along with the choice of strength of this finding they view as appropriate. It seems as though this issue may have been overlooked, e.g., between the air quality and extreme weather chapters.
The authors were careful in the wording used to characterize the response of ozone to global change in relation to changing (dramatically reducing) anthropogenic emissions. In particular, the reduced ozone precursor emissions from major anthropogenic sources will alter the response to global change (generally dampening the response). Further, the regulatory structure of the United States is “adaptive” in that if the pollutant levels exceed the National Ambient Air Quality Standards (NAAQS), further controls are required. Thus, if current and future climate changes occur, additional controls will likely be employed to further reduce emissions to meet the air quality targets. While there will be areas affected by adverse impacts on ozone, the location and extent of the impact will be a function of a complex air quality management process that is already in place and may already have started to adapt. Some additional caution may be due in terms of how the authors report the impact that climate change has had on ozone. Leibensperger et al. (2008) state: “Such a long-term decrease in mid-latitude cyclone frequency over the 1980-2006 period may have offset by half the ozone air quality gains in the northeastern US from reductions in anthropogenic emissions.” The Leibensperger et al. (2008) discussion, like the draft chapter, notes the impact of controls. While the statement that climate change has impacted ozone is supported by the literature, the authors are advised to better explain the term “climate penalty.”
The authors note that additional research is needed on how air quality and aeroallergens will respond to climate change and on the potential for increased exposure to contaminants indoors. The literature on this issue is not extensive at this time and does not strongly suggest that there will likely be adverse health outcomes from altered exposure to indoor air pollutants. However, the potential exists and is an area for further study; the Assessment authors therefore find that this is an emerging issue of concern, a conclusion with which the Committee agrees. They cite the need to better understand wildfire response. The research needs, while broad, are appropriate. The authors should consider noting that there are areas where current scientific knowledge limits our ability to understand the formation and fate of contaminants in the present atmosphere (e.g., secondary organic aerosol formation). This will hinder our ability to understand how air quality will respond to future changes. Further consideration of potential increase in dust-borne disease should also be included.
Comments on the Modeling
Air Quality Modeling of Ambient Air Quality Response to Climate Change
Fann et al. (2015) conducted a photochemical air quality modeling and health assessment study examining the impact of ozone formation under different climate forcing levels in the near future (2025-2035). This work followed an approach that has been used by others, i.e., downscale the results of a global climate model for an historic period and a future period, and apply a regional air quality model over both periods using a similar emissions inventory to examine how climate changes impact pollutant formation. Their modeling period is suitably long (11 years) and they use a future emissions inventory (which is not always done, but is appropriate). They find an ozone “climate penalty” of 1-5 ppb. They use the air quality model results in BenMap using a projected 2030 population. They find a potentially large number of premature deaths and increased morbidity from the increased exposure due to climate change. The modeling appears to be well done and the results align with other studies.
Indoor Air Quality Modeling
In support of their identification of indoor air quality being an emerging issue, the authors conducted an indoor air quality modeling study using a traditional infiltration/emissions mass balance approach (Ilaqua et al., 2015). The study was conducted using an appropriate approach and identified potential issues of concern, though, as noted, there are many uncertainties and variabilities that are important to assessing potential future exposures. Similarly, the modeling appears to be well done and is appropriate.
General Comments and Key Findings
Chapter 4 discusses the ways in which vectorborne diseases are influenced by climate factors, including the short- and long-term effects on patterns of transmission and infection. This chapter does a nice job of balancing that for which we have empirical evidence with respect to vectorborne diseases and climate change with that which we suspect will happen, but have limited evidence. Throughout, the authors take a restrained view of the climate change and vectorborne disease modelling literature, making careful distinctions between risk from exposure to vectors and the occurrence of disease. Despite the authors’ care and clarity in presenting the climate change and vectorborne disease literature, there is room to improve this chapter. First, the Key Findings should be edited to reflect the Committee’s suggestions detailed above in the section on “Answers to the Statement of Task Questions.” Specifically, the health-related outcome(s) should be described first wherever possible. In addition, there may be opportunities to disaggregate confidence determinations in the key findings. Addressing these concerns will help achieve consistency with other chapters. Second, the “Emerging Issues” and “Research Needs” sections should be expanded to reflect more explicitly what we do not know and to more
directly address the uncertainty described in the traceable account. For example, the uncertainty around estimating occurrence of human disease is described as related to “viral evolution, changes in vector control and human behavior,” yet the Research Needs include no mention of these issues. Finally, despite the Federal Register’s call for “…special attention to research that frames risk in terms of … adaptive capacity,” there is limited discussion of this topic in the chapter.
Key Finding 1: Changing Distributions of Vectors and Vectorborne Diseases
Climate change is expected to alter the geographic and seasonal distributions of existing vectors and vectorborne diseases. [Likely, High Confidence]
This key finding is well stated and supported by the literature. Here, as throughout the chapter, the modes for attenuation through adaptive capacity are not addressed. In addition, this Key Finding and the literature to support it can be used to further delineate some of the Research Needs.
Key Finding 2: Earlier Tick Activity and Northward Range Expansion
Ticks capable of carrying the bacteria that cause Lyme disease and other pathogens will show earlier seasonal activity and a generally northward expansion in their habitat range in response to increasing temperatures associated with climate change [Likely, High Confidence]. Longer seasonal activity and expanding geographic range of these ticks may increase the risk of human exposure to ticks [Low Confidence].
Key Finding 3: Climate-Driven Mosquito-Borne Disease Dynamics
Rising temperatures, changing precipitation patterns, and a higher frequency of some extreme weather events associated with climate change will influence the distribution, abundance, and infection rate of mosquitoes that transmit West Nile virus and other pathogens by altering habitat availability and mosquito and viral reproduction rates [Extremely Likely, High Confidence]. Alterations in the distribution, abundance, and infection rate of mosquitoes may increase human exposure to bites from infected mosquitoes, which may increase risk for human disease [Low Confidence].
Key Findings 2 and 3 are similar in that the chapter takes Lyme disease (Key Finding 2) and West Nile virus (Key Finding 3) as case studies for describing the state of the science on these diseases and quantitative predictive modeling of vectorborne disease risk more generally under future climate scenarios. However, the language used for both key findings should be better coordinated with the other chapters to draw attention to the possibility of significantly higher exposure to and impacts of vectorborne diseases on human health in the near future. As implied in Key Finding 1, it seems justified (from the published evidence cited in the chapter) to state more strongly that future impacts of vectorborne diseases on human health due to climate change-induced alterations to vector populations are likely to include cases of vectorborne
disease in regions where people are unprepared to deal with them. This will, in turn, result in increased exposure to the vectors and potential increased burden of vectorborne diseases on human health.
The authors presented new modeled forecasts of climate change related extension of the Lyme disease season in support of Key Finding 2. The methods used by Monaghan et al. (2015) have been well developed and implemented and are consistent with other forecasting models of vectorborne disease risk.
Key Finding 4: Climate and Non-Climate Factors Determine Human Vulnerability
Non-climate factors that affect vulnerability to vectorborne disease (such as age, gender, socioeconomic status, geography, and occupation) also influence risk for disease occurrence. [High Confidence]
While this is an important characteristic of all climate change and human health concerns (as exemplified in Chapter 9: Populations of Concern), it is not clear that this is a novel key finding. The Committee suggests removing this statement as a Key Finding in order to lend greater room for the novel findings specific to this report. Text addressing populations of concern should remain in the document.
Barring removal, this Key Finding could be more clearly written; the difference between “vulnerability to vectorborne disease” and “risk for disease occurrence” is not clear. The lack of a likelihood statement is indicative of the paucity of data and should be highlighted in the section on “Research Needs.” The Committee also suggests the authors review the way vectorborne diseases are addressed in Chapter 9 and in the other chapters more generally for the sake of consistency.
Key Finding 5: Emergence of New Vectorborne Pathogens
Climate change will interact with other driving factors (such as travel-related exposures or evolutionary adaptation of invasive vectors and pathogens) to influence the emergence or re-emergence of vectorborne pathogens. [High Confidence]
Key Finding 5 is an important finding that is somewhat lost with the number of other Key Findings of this chapter. The confidence statement comes from the bulk of review papers all suggesting a positive association, but the lack of a likelihood statement is indicative of the limited evidence base. This Key Finding could serve as the basis for structuring the “Emerging Issues” and “Research Needs” sections of the report.
Emerging Issues and Research Needs
This chapter has evaluated two diseases for which domestic human risk is great and for which there are quantitative models on the incidence of disease (compared to other models evaluating entomologic risk, vector abundance, etc.). Many questions remain about the links
between vectorborne disease more generally and climate change, as highlighted in this chapter. As noted in the section on “Emerging Issues,” recent events like chikungunya in the Caribbean and dengue outbreaks along the southern border of the United States (including Yuma, AZ in 2014) bring travel-related vectorborne disease and importation risk to the minds of many Americans. As scientists try to quantify the probability of introduction, modeling of importation is an emerging discipline (e.g., Ruiz-Moreno et al., 2012). However, the introduction of vectors or vectorborne disease is not the only concern. Additional emerging issues with nascent scientific research include vector adaptation (e.g., Bradshaw and Holzapfel, 2001); understanding future risk in times of expanding vector risk (e.g., Ogden et al., 2014); and interactions with temperature, vectors, and insecticide use (e.g., Glunt et al., 2014). Moreover, with the burgeoning field of predictive vectorborne disease modeling, there emerges a need for guidance on interpreting the models with respect to where and when they can be applied.
The section on “Research Needs” states a need for better long-term human and vector data to feed into evidence-based models without highlighting a need for the empirical studies to provide those data. For example, Reiter et al. (2003) is a frequently cited paper for adaptive capacity with respect to dengue. However, it has yet to be replicated in other regions and with other vectorborne diseases. There is currently a significant increase in evidence-based models, all of which suffer from a paucity of empirical studies to parameterize the modeling efforts (see, for example, Ellis et al., 2011). The researchers conducting these empirical studies are finding that slight nuances in the experimental design have significant impacts on the model outcomes. For example, recent work shows that mosquitoes reared in fluctuating temperatures versus constant temperatures have considerable influence on temperature-dependent daily development rates—a critical driver behind most mosquito abundance models (e.g., Paaijmans et al., 2013). Similarly, as more models are developed, the field is seeing the implications of vector adaptation (e.g., Bradshaw and Holzapfel, 2001). Given the impact such studies will have on the capacity to adequately model risk, the literature on these research needs should be included.
In summary, Chapter 4 and its Traceable Accounts were well cited and explained most of the uncertainty surrounding vectorborne disease impacts from climate change. However, there are ways in which the state of the science regarding climate change and vectorborne diseases can be articulated more clearly and consistently with other chapters.
General Comments and Key Findings
This Chapter explores some ways in which climate and weather factors can have an impact on properties of water-related pathogens and toxins as well as impacts on human exposure pathways. In general, this chapter is well-written and does an excellent job of assembling relevant information concerning the likely and possible effects of climate change, especially rising ambient temperature, on selected water-related illnesses. While the initial sections and Table 1 make mention of a fairly broad array of agents and pathways, including illnesses related to ingestion of water containing etiologic agents or their products that can cause
human illness, the authors have clearly decided to focus the chapter on pathogens associated with consumption of fish and shellfish, including illnesses resulting from ingestion of harmful algal toxins.
Key Finding 1: Seasonal and Geographic Expansion of Waterborne Illness Risk
Increases in both coastal and inland water temperatures associated with climate change will expand the seasonal windows of growth [Very Likely, High Confidence] and the geographic range of suitable habitat [Likely, High Confidence] for naturally occurring pathogens and toxin-producing harmful algae. These changes are projected to increase the risk of exposure to waterborne pathogens and algal toxins that can cause a variety of illnesses [Medium Confidence].
Key Finding 2: Exposure Risk from Extreme Precipitation Events
Recreational waters and sources of drinking water will be compromised by increasingly frequent and intense extreme precipitation events [High Confidence]. Surface runoff and flooding associated with heavy precipitation and storm surge events increase pathogen loads originating from urban, agricultural, and wildlife sources and promote blooms of harmful algae in both fresh and marine waters. Greater pathogen or algal toxin loading in drinking and recreational water sources following an extreme weather event will increase risk of human exposure to agents of water-related illness [Medium Confidence].
Key Finding 3: Water Infrastructure Failure or Damage
Increases in some extreme weather events and storm surge will increase the risk of failure of, or damage to, water infrastructure for drinking water, wastewater, and stormwater [Medium Confidence]. Aging infrastructure is particularly susceptible to failure. A breakdown in water infrastructure would contribute to increased risk of exposure to water-related pathogens, chemicals, and algal toxins.
Given the decision to focus this chapter on toxin-producing harmful algae and non-cholera Vibrios, the three key findings presented appear to be reasonable and supported by the evidence cited. One area that requires clarification and possibly revision relates to the evidence concerning climate change, especially increases in ambient temperature, and the risk of human illness caused by non-cholera Vibrio species. The discussion of the effects of rising temperatures (and other changes) on the growth and distribution of Vibrios appears to conflate the effects on the abundance of Vibrios in a given area with effects on seasonal windows and geographical range.
A further potential problem is that the non-cholera Vibrio studies do not have a particularly mature grounding in peer-reviewed literature; the primary cited paper has not yet been published, and that paper relies on self-citations for V. vulnificus models and on a single FDA report for V. parahaemolyticus models. It also appears that the V. vulnificus results are based on a dataset that is not publicly available for analysis by other researchers who use other
modeling methods, which is troubling for a research highlight in such a high profile Assessment report. While these studies may be fully robust, there is always risk in featuring a result when the peer-reviewed foundation is limited and potentially susceptible to rapid shifts. Following on our general recommendation for all quantitative modeling results, we urge the authors to provide a discussion of model-based uncertainties and information on how to access underlying data either in this chapter or in the Technical Support Document.
Content Areas Missing
While there is no doubt that the agents and illnesses that the authors chose to focus on are important, it is not clear why the authors chose to focus on these agents and illnesses from among the much longer list of “Agents of Water-Related Illnesses” in Table 1. It also is not clear why other such agents are missing from the table altogether. Missing from the table are the Legionella species, a substantial cause of community- and hospital-acquired pneumonia in the U.S. and many other industrialized countries. Also missing from the table are the schistosoma species that cause “swimmer’s itch,” a non-fatal, but common condition resulting from contamination of recreational waters in which people swim, wade, or play. Leptospires and V. cholera are listed in Table 1, but they receive little or no attention in the text, presumably because the illnesses caused by these organisms, while serious problems internationally, are relatively infrequent causes of illness in the United States. At the same time, however, primary amoebic meningoencephalitis caused by Naegleria receives attention, despite the fact it causes only a handful of cases in the United States each year. Other agents that can be acquired from ingesting contaminated water or shellfish, such as hepatitis A and hepatitis E viruses, are also missing from the table and receive scant mention (hepatitis A) or no mention (hepatitis E) in the text.
To the uninformed reader, the process for deciding which agents and diseases to include in Table 1 and which to make the subject of detailed discussion is opaque. One might assume that V. cholera and leprospires were largely excluded from consideration because the report is intended to focus on the likely effects of climate change on diseases and health in the United States; even so, it is plausible that increasing ambient temperature and other aspects of climate change could lead to them spreading to the United States and becoming more common here. The Committee urges the authors to discuss their decision making criteria more clearly within this chapter.
The exclusion from the table and text of the Legionella species and the primary illness it causes, Legionella pneumonia (i.e. Legionnaires’ disease), should perhaps be reconsidered given the importance of the disease in the United States in terms of the morbidity and mortality that it causes; the clear association of the disease with inhalation of contaminated aerosols of fresh water; the effects of both temperature and the presence of other organisms, such as amoeba, on the growth of Legionella species; the clear link between large scale air conditioning systems (e.g., cooling towers and evaporative conditioners) and the risk of Legionella pneumonia; and the likelihood that increasing ambient temperatures are highly likely to lead to increased need for and use of air conditioning in large parts of the United States. The Committee acknowledges that this chapter cannot be encyclopedic in scope and can deal with only a limited number of water-related infectious diseases, but suggests that authors could add some rationale behind the
exclusion of Legionella and legionellosis, especially if the authors decide that there is not enough evidence in existing literature to support its inclusion.
Emerging Issues and Research Needs
In the section devoted to “Emerging Issues,” much of the focus is on Naegleria fowlera, which causes no more than a handful of cases of amoebic meningoencephalitis a year in the United States. While these cases are severe, often fatal, and tragic, it is not clear that this organism and disease warrant highlighting rather than legionellosis or perhaps the growing evidence that hepatitis E virus infection is more common in the United States than previously recognized, and its link to fecally contaminated water uncertain.
In the section on “Research Needs,” while the call for “sustained collection of public health…data” and “targeted studies…” (page 178, lines 36-37) appears reasonable and difficult to dispute, it is quite vague and general. While this report may not be the right forum for presenting a specific research agenda, the current section is not particularly helpful in providing guidance or setting priorities with regard to research in this area.
FOOD SAFETY, NUTRITION, AND DISTRIBUTION
General Comments and Key Findings
This chapter reviews the literature that addresses the impacts of climate change and extreme weather events on selected food safety issues and on the distribution and access to safe food. The chapter also reviews the impacts of rising carbon dioxide on the nutritional content of foods. The chapter is organized in three main sections: how climate change and changes in weather extremes may increase the risk of selected foodborne illness by increasing the risk of microbiological and chemical contaminants in the food chain, how rising carbon dioxide lowers the nutritional value of foods, and how climate-related extreme weather events affect food distribution and access to safe and quality foods.
Important food safety issues and related foodborne diseases caused by the contamination of fish and shellfish with Vibrios, with certain chemical contaminants, and with harmful algal marine biotoxins were covered in Chapter 5. This may confuse some readers since, according to the World Health Organization (WHO), foodborne diseases (FBD) can be defined as those conditions that are commonly transmitted through ingested food1. The decision to include those items in Chapter 5 should be better explained and key issues should be cross referenced in both chapters.
For each Key Finding, the chapter provides a description of the evidence base, the major uncertainties, and an explanation of the judgments of likelihood and confidence. The Key Findings are well described and justified in the section on Traceable Accounts, but they need to
1 FBD comprise a broad group of illnesses caused by enteric pathogens, parasites, chemical contaminants and biotoxins (WHO, 2007).
be better formulated, particularly as they are presented at the beginning of the chapter and many readers may only read these Findings.
One of the main issues to address in the Key Findings, and throughout the chapter, is the need to refer properly to:
- the likelihood of food contamination with microbiological or chemical hazards,
- the likelihood of human exposure through contaminated foods, and
- the risk of illness resulting from the consumption of contaminated foods (dietary exposure).
Key Finding 1: Increased Risk of Foodborne Illness
Although there are many practices to safeguard food in the United States, climate change, including rising temperatures and changes in weather extremes, is expected to intensify pathogen and toxin exposure [Likely, High Confidence], increasing the risk, if not the actual incidence, of foodborne illnesses [Medium Confidence].
Based on available research and evidence, it is important to note that rising temperatures and changes in weather extremes are expected to increase food contamination with pathogens and toxins (such as aflatoxins). Then, as a consequence, there may be an increased exposure to pathogens and toxins through food (depending on the risk management and risk communication strategies, such as regulatory, surveillance, and monitoring systems) and hence an associated increased risk of foodborne diseases.
Key Finding 2: Chemical Contaminants in the Food Chain
Elevated sea surface temperatures and increases in certain weather extremes associated with climate change will increase human exposure to water contaminants in food [Likely, Medium Confidence]. Climate change will also alter the incidence and distribution of pests, parasites, and microbes [Very Likely, High Confidence], which will lead to increases in the use of pesticides for crop protection, animal agriculture, and aquaculture. Increased use of pesticides may result in increased human exposure to chemical contaminants in the food chain [High Confidence].
The Committee does not disagree with this Key Finding, but suggests that the authors should be more explicit about the mechanisms. Elevated sea surface temperatures and increases in certain weather extremes associated with climate change may increase water and food contamination with chemical contaminants (such as Methyl-Hg or persistent organic pollutants [POPs]), and this may result in increased human exposure to chemical contaminants (depending on the risk management and risk communication systems in place) and hence an increased risk of associated diseases and conditions.
In addition, the increased use of pesticides may result in the increased presence of pesticide residues in foods, which may increase the chances of human exposure to pesticides and
hence result in an increased risk of associated diseases and conditions. In the case of animal production and aquaculture systems, there might be also an increased use of veterinary drugs and other chemo-therapeutants (FAO, 2008).
Key Finding 3: Rising Carbon Dioxide Lowers Nutritional Value of Food
Rising atmospheric carbon dioxide will continue to lower the nutritional value of most food crops, including wheat and rice, and can also reduce the concentration of essential minerals in a number of crop species. [Very Likely, High Confidence]
The Committee suggests that it would be helpful, especially for readers that are unfamiliar with this research, to put more emphasis on this important Key Finding and to further explain that rising atmospheric carbon dioxide lowers the nutritional value of most food crops and the concentration of essential minerals in a number of crops. This may continue as atmospheric carbon dioxide continues rising, and could have significant implications for human nutrition.
Content Areas Missing
There is a critical body of evidence on climate change and associated foodborne and waterborne illness risks, including models developed recently by the European Food Safety Authority and the European Centers for Disease Control, that should be mentioned in this chapter. Examples include: (1) The European Food Safety Authority study of the potential increase in aflatoxins in cereals in the EU as a result of climate change which includes modeling, predicting, and mapping the emergence of aflatoxins in cereals in the 27 EU countries due to climate change (Battilani et al., 2012); (2) the European Centre for Disease Prevention and Control technical report on Assessing the potential impacts of climate change on food- and waterborne diseases in Europe (ECDC, 2012); and (3) the Decision Support Tool developed by the European Centre for Disease Prevention and Control to Compare Waterborne and Foodborne Infection and/or Illness Risks Associated with Climate Change (Schijven et al., 2013).
Chemical contamination is covered in both the water and food chapters, but these chapters do not fully represent the existing body of literature. For example, the section on chemical contaminants may need more work to cover issues related to:
- Contamination with chemicals related to recurrent river floods in the United States (for reference, the EU-wide study of the impacts on food/water contamination with POPs, dioxins of recurrent floods of the Danube river; see also Umlauf et al., 2005);
- Chemicals’ concentration and consequent dispersion related to droughts followed by floods in the United States (see Rotkin-Ellman et al., 2010);
- Limited activity of pesticides in dry conditions (Muriel et al., 2001) and faster pesticide degradation related to higher temperatures and higher dose levels or more frequent applications needed to protect crops (Bailey, 2004);
- Increased use of veterinary drugs and other chemo-therapeutants associated with animal production and aquaculture systems (FAO, 2008); and
- Chemical transport systems in the Arctic (perhaps including any new findings since the previous Arctic Climate Impact Assessment [ACIA, 2004]).
Emerging Issues and Research Needs
The section on “Research Needs” would benefit from being more thorough and specific. Areas related to food chemical contamination highlighted above need to be studied in different regional contexts and geographical areas in the United States (e.g., the Arctic).
There are many research needs on climate change and food safety that have been already addressed in European countries and Canada but have not been studied in the United States yet. For example, there are no studies on the impact of climate change on food and waterborne diseases in the United States and it would be helpful to develop tools that can be used in climate change adaptation strategies for foodborne and waterborne diseases (Schijven et al., 2013). Another key research need is the development of predictive models of the emergence of aflatoxins in cereals due to climate change in the United States. Scenarios and maps could also be created to focus on potential future contamination of cereal crops. The Committee agrees that all the emerging issues highlighted in the chapter deserve further research.
The section on populations of concern needs to address the potential health risks of chemical food contamination to tribal communities in the Pacific North West and Alaska. Particular risks related to traditional diets of these tribal communities have been covered in Chapter 5, but these dietary issues are perhaps more relevant for Chapter 6.
The Chapter would benefit from including a section or table on adaptation needs and strategies for food safety including, for example, good agriculture and veterinary practices. Potential trade issues related to food contamination by aflatoxins could be mentioned in the section on food access to reflect the potential additional efforts by USDA to monitor food imports. Food prices could be mentioned in relation to the impact on food access (e.g., California). Issues related to risk benefit analysis and consequent risk communication could be included, for example, about fish contamination and consumption during pregnancy.
The section on nutrition does not refer to the “opportunities to achieve co-benefits from actions that reduce emissions and at the same time improve health by shifting consumption away from animal products, especially from ruminant sources, in high-meat consumption societies, toward less emission intensive healthy diets” (Field et al., 2014). If these issues are not addressed, perhaps the section of nutrition should be named differently, e.g., Nutritional Value of Foods.
The chapter covers two very different issues: the impacts of climate change on food safety and the impacts of rising carbon dioxide in the nutritional value of food. The title of Figure 1 should reflect this difference (i.e., rising Carbon Dioxide affects nutrient content but does not necessarily affect Food Safety). It would be useful to include a table on potential chemical contaminants related to climate change in the United States.
General Comments and Key Findings
Chapter 7 explores some of the health effects associated with extreme weather events. The authors have done a laudable job tackling a difficult task: to write about climate change extremes while avoiding some of the most obvious components of the topic—e.g., heat waves, infectious and waterborne disease outbreaks, and mental health—which are dealt with in their own devoted chapters. Given this mandate, the scope and completeness of the chapter is appropriate, and the authors have captured relevant literature in a balanced manner.
That said, the Committee feels that the chapter could be improved in three ways: more specific Key Findings, greater emphasis of the regional character of impacts, and consideration of adaptation. These points are expanded below.
Key Finding 1: Changes in Exposure Risk
Climate change may increase exposure to health hazards associated with projected increases in the frequency and/or intensity of extreme precipitation, hurricanes, coastal inundation, drought, and wildfires in some regions of the United States [Medium Confidence]. Adverse health outcomes associated with exposure to extreme events include death, injury, or illness; exacerbation of underlying medical conditions; and adverse effects on mental health.
A general suggestion of the Committee is that all Key Findings should start from health and follow with relevant physical, ecological, or social mediators. Following that model, we suggest that Key Finding 1 begin with the second sentence “Adverse health outcomes …” and include a confidence statement specific to that statement. This could be followed by a statement on how climate change affects these outcomes, similar to the current first sentence of the Key Finding.
However, we also suggest that the statement on climate change affects be phrased in more specific terms. This includes replacing the word “may” with “will,” since a confidence statement on “may” has no clear meaning. Regarding the “medium confidence” assessment, the Traceable Accounts support a more specific set of confidence statements that could be separated out by type of extreme event. As the authors note in Traceable Accounts, few quantitative studies draw the full connection from climate change to health impact. But the extent to which this connection has been made varies by type of extreme. For wildfires, for example, the authors cite studies that address climate trends in both frequency and exposure. For floods and droughts, however, the Traceable Accounts indicate that there is published evidence of climate trend but trends in exposure are less certain (though coastal floods might be an exception). For winter storms there is a lack of consensus on the climate trend itself.
These are meaningful differences that are important enough to be represented in the Key Finding. This could be done by providing different confidence statements for each type of event,
by separating events into categories of different confidence level, by dividing the Key Finding 1 statement into a sentence on climate trend and a sentence on exposure trend, or some other formulation that the authors feel is more appropriate.
Key Finding 2: Other Factors Influence Health Impacts
The character and severity of health impacts from extreme events depend not only on the frequency or intensity of the extremes themselves but also on a population’s exposure, sensitivity, and adaptive capacity. Many types of extreme events can cause loss of essential infrastructure, such as water, transportation, and power systems that are required to safeguard human health. [High Confidence]
As written, Key Finding 2 combines two concepts: impacts depend on vulnerability, and some extreme events affect health via impacts on infrastructure. The first concept would seem to fit into Key Finding 3, and the Committee suggests that it be moved to that Key Finding for clarity. This would leave Key Finding 2 to focus on infrastructure, which is certainly a critical point when linking extreme events to health. Following the formula suggested above for Key Finding 1 and for the report more broadly, we suggest that the existing Key Finding 2 statement (“Many types of extreme events …”) be followed by a statement on how climate trends are affecting this relationship. In the context of infrastructure it would also be valuable to add a sentence on how changes in infrastructure can reduce health impacts, supported by the natural disaster literature.
As with Key Finding 1, it might be useful to separate out different types of extremes in Key Finding 2 since the depth of available literature differs by event.
Key Finding 3: Certain Populations Are More Vulnerable
Key risk factors that individually and collectively shape a population’s vulnerability to health impacts from extreme events include age, health status, socioeconomic status, race/ethnicity, and occupation. [High Confidence]
This is an important Key Finding that ties closely to Chapter 9 and to the general theme of vulnerable populations throughout the report. As noted above, the Committee suggests moving the vulnerability finding from Key Finding 2 to Key Finding 3, as it seems more closely related to this point on vulnerability. More importantly, report authors should coordinate to ensure that vulnerable population findings are presented in a consistent manner across all chapters.
The Committee suggests two updates to Table 1. First, confidence statements similar to the 2014 NCA report should be added to the climate projections in column three. Second, it would be useful to insert an additional column that identifies the most affected regions for each
type of event. Including regional information in this table would also require that the phrasing in column three be formalized in some cases, which would also improve the table. Some of the statements are unclear—e.g., droughts have increased in the past “couple of decades” (which decades?); winter storms have increased and shifted northward (what is the relative confidence in intensity, frequency, and location?). Alternatively or additionally, a map could be included alongside Table 1 that draws on findings on extremes from the 2014 NCA report.
Another possible way to clarify regional impacts would be to lead off each subsection with a subheading on the “Most Affected Regions.” While most subsections cover this matter effectively in the text, a subheading that lists regions might be useful for the reader.
Currently the chapter addresses adaptive capacity only in the context of uncertainty. Given the importance of adaptation in this context, some consideration of how adaptation mediates health impacts is within the scope and important to include. While literature that specifically addresses the health impacts of adaptation to extremes under climate change might be limited, there is a wealth of relevant analyses on the health benefits of preparedness in the natural disaster literature, including studies on the siting and construction of health infrastructure (e.g., location of generators in flood-prone hospitals), the use of early warning systems, insurance incentives, and zoning in areas at risk of wildfire.
This is, admittedly, a diffuse literature, much of which is contained in analyses of infrastructure or behavior rather than of health in an explicit sense. One possible starting point on the health-specific literature is the recent book Disasters and Vulnerable Populations: Evidence-Based Practice for the Helping Professions (Baker and Cormier, 2014). Authors may also wish to consider another recent report on “Disaster Resilience: A National Imperative” (NRC, 2012). It might be just as useful, however, to build a short “adaptation” section or case study Box drawing on recent Federal Emergency Management Agency (FEMA), U.S. Army Corps of Engineers (USACE), or other analysis of a major disaster like Superstorm Sandy or Hurricane Katrina, from the numerous analyses of flood relocation programs in the United States (e.g., Kick et al., 2011), or from the literature on behavioral response to disaster warnings and evacuation orders (e.g., Tinsley et al.  and references therein).
MENTAL HEALTH AND WELL-BEING
General Comments and Key Findings
The Chapter examines a range of mental health consequences of climate change as well as specific groups of people that may be at higher risk. This is an extremely important chapter that reviews a fairly new literature and attempts to cover a wide range of topics. While the authors have done an excellent job of covering a lot of material in a limited space, and the scope and completeness of the chapter is largely appropriate, the chapter should address the following problems: alteration of the format of the Key Findings, the insufficient representation of the
literature in Key Finding 3, addressing specific gaps in the chapter, and several other literature gaps in other parts of the chapter.
Key finding 1: Mental Health Consequences of Exposure to Disasters
Many people exposed to climate-related disasters experience stress and serious mental health consequences. Depending on the type of the disaster, these serious mental health consequences include significant symptoms of post-traumatic stress disorder (PTSD), depression, and general anxiety, which often occur at the same time. The majority of affected people recover over time on their own, although a significant proportion of exposed individuals develop chronic levels of psychological dysfunction. [Very High Confidence]
A general suggestion of the Committee is that all Key Findings should start from health and follow with relevant physical, ecological, or social mediators. Following that model, we suggest that Key Finding 1 be re-ordered to be rephrased with more specificity, “Experiences from climate-related disasters cause stress and serious mental health consequences…”
Key finding 3: The Threat of Climate Change
The threat of climate change, the perceived direct experience of climate change, and changes to one’s local environment can result in substantial adverse mental health outcomes and social impacts for the American public. Virtually all Americans are exposed to the threat of climate change and to events attributed to the impacts of climate change through frequent multimedia coverage. [High Confidence]
This Key Finding claims that media portrayals of climate risks increase stress and mental health impacts of climate change. Report authors may wish to consider any available literature that demonstrates how people may temporally or spatially distance themselves from climate risks. The Key Finding as currently worded gives the overall impression that media portrayals of climate change are more detrimental than they may actually be, considering the distancing that at-risk communities tend to engage in. In addition, the second part of this Key Finding that states “virtually all Americans are exposed to the threat of climate change and to events attributed to the impacts of climate change through frequent multimedia coverage” needs to be better supported by existing literature or perhaps removed entirely from Key Finding 3.
Overall Representation of the Evidence
Possibly due to the page limitations, the chapter often simplifies very complex phenomena that might be better understood with a slightly more detailed treatment, so that they can possibly be addressed in applied settings. If this report is meant to speak to a wide audience, such as mental health professionals and others who might be in a position to deal with these issues, then more detailed discussion is important. For example, on page 298, lines 10-11, the assertion is that sea level rise affects mental health, but there is not an exploration of why.
Another example is the discussion of increased violence on page 297, line 32. Another phrase or paragraph explaining why interpersonal violence increases or for whom would help. This issue is very briefly addressed in different terms on page 300, lines 4-6, and could be moved and explained in more detail.
This general portrayal raises the larger issue of adaptation within the review of mental health and well-being. There is literature that shows that specific approaches to preparing for or responding to extreme events can mitigate the mental health impacts of extreme events. This literature should be represented in the chapter where appropriate. This is particularly true in the section on “Resilience and Recovery.” It would be ideal to expand the scope of the chapter a bit in the section on climate mitigation in order to include well-being and climate change, not just mental health impacts.
Content Areas Missing
There is a lack of discussion on access to health care and health resources necessary to deal with the mental health burden introduced by climate change. While this Assessment is not policy prescriptive and is not meant to offer policy recommendations, the lack of resources to deal with mental health impacts of extreme weather events has already played a critical role in the exacerbation of these impacts, making this issue important to raise. There is literature to refer to in this area (e.g., Blashki et al., 2009 and Shukla, 2013).
While the chapter details many specific relationships between types of weather events and illnesses that promise to exacerbate mental health impacts, there is no mention of how cumulative increases in mental health outcomes may have ramifications we have not anticipated. It is worth mentioning because most of the review is of past events, and there is an abundance of literature demonstrating that climate change will alter events in ways we have not anticipated (e.g., the past does not predict the future).
Overall, the mental health findings are very well represented in this chapter, but the findings regarding well-being specifically would benefit from additional consideration. In order to make the chapter relevant to broad audiences outside of academics, it is important to not only represent the problems presented by climate change, but also the potential solutions, which can be equally important for well-being. This might merit the addition of one specific section dedicated to this topic or the integration of such findings throughout all sections in this chapter.
CLIMATE-HEALTH RISK FACTORS AND POPULATIONS OF CONCERN
General Comments and Key Findings
The stated goal of this Chapter is to identify factors that may create or exacerbate the vulnerability of certain groups to the health impacts of climate change. The Chapter also aims to integrate information from the other report chapters to identify specific groups of people that may face greater health risks due to climate change.
Key Finding 1: Vulnerability Varies Across Individual, Time Scales, and Places
Across the United States, people and communities differ in their exposures, their inherent sensitivity, and their adaptive capacity that enables them to respond to and cope with climate change related health threats. Vulnerability to climate change varies across time and geographic areas, across communities, and among individuals within communities. [Very High Confidence]
Overall the authors present substantial evidence that there are differences in exposures, inherent sensitivity, and adaptive capacity across the United States. Extensive discussion is provided on the concept of vulnerability. Given that vulnerability is woven into the preceding chapters of the document, the discussion of the definitions and concept should be provided in the introductory chapter, as previously noted. Because much of the content on vulnerability, sensitivity, and adaptive capacity is discussed in previous chapters, the authors should note more extensive discussions elsewhere in the document.
A large issue is how to describe the vulnerability of individuals according to race and ethnicity absent socioeconomic factors. There are specific examples of vulnerability related to race and ethnicity, but they are not covered in this chapter. Neither race nor ethnicity automatically imply vulnerability, but they could point to inherent sensitivity.
Key Finding 2: Climate Factors Interact with Non-Climate Factors to Increase Health Risk
Climate change related health risks interact with some of the same non-climate factors that increase the risk of poor health generally. Non-climate factors, such as those related to demographic changes, socioeconomic factors, and pre-existing or chronic illnesses, may amplify, moderate, or otherwise influence climate related health effects, particularly when they occur simultaneously or close in time or space. [High Confidence]
This component of the chapter appears to be comprehensive, particularly the discussion of compromised mobility, cognitive function, and other mental factors. The discussion of compromised literacy could be expanded to describe the challenges that non-English-speaking populations face when they cannot comprehend warnings or health threats. The section on psychological stress has considerable overlap with Chapter 8, but does not include sufficient discussion on how fear of authority and deportation among undocumented immigrant workers could affect their susceptibility during climate events.
The statement of qualitative vulnerability assessment involving conducting surveys on the resilience of health infrastructure needs an example. It is not clear why the focus is on surveying health infrastructure if the aim is to assess pockets of vulnerability to climate change.
In general, the measurement of the extent to which non-climate factors affect vulnerability is complex and is limited by data availability, particularly at the national level. These factors are, in one sense, the essence of much of the unpredictable vulnerability that is observed after climate events.
Key Finding 3: Increased Vulnerability to Climate-Related Health Impacts is Widespread Across Ages and Stages of Life
People experience different vulnerabilities at different ages and life stages. For example, the very young and the very old are particularly sensitive to climate related health impacts. [High Confidence]
The confidence in this finding is limited due to the lack of data on the extent to which this vulnerability affects populations, particularly at the national level. The concept that youth may present increased vulnerability but also enhanced adaptability and resiliency is an important one; the authors should either cite data that is available on this question or identify in the “Research Needs” section key needs for additional evidence. Also there is redundancy in the chapter by first focusing under Key Finding 1 on children and then focusing in this Key Finding on life stages and repeating much of the previous discussion on children.
Key Finding 4: Mapping Tools and Vulnerability Indices Help to Identify Where and for Whom Climate Health Risks Are Greatest
The use of geographic data is allowing more sophisticated mapping of risk factors and social vulnerabilities, to identify and protect specific locations and groups of people. [Medium Confidence]
The authors only conclude medium confidence in this Key Finding. The figures provided in the text illustrate the capacity that is currently available to map vulnerability, but significant limitations remain. Mapping of the elderly and heat wave exposure is the most commonly cited example of identifying pockets of vulnerability. However, there are few examples of mapping other pockets of vulnerability, such as where vulnerable populations are located in coastal communities or communities threatened by wildfires. In general the first paragraph under application of vulnerability indices should include some of the more challenging examples, clearly illustrating the lack of confidence in this Key Finding.
Rather than having a Key Finding that results in only medium confidence, the authors could consider approaching it as an emerging trend (as mapping is handled in Chapter 8). If this section was relabeled “Emerging Trends,” a more inclusive discussion could address the challenges of investing in the infrastructure and resources to do mapping prospectively. It is helpful to know not only where the vulnerable pockets are, but also the challenges of investing in these pockets in order to be better prepared in emergencies. The example of mapping after Hurricane Katrina is helpful, but there is little discussion of how to obtain the resources to do this type of work prospectively in order to predict pockets of vulnerability. Ultimately the goal would be not just to develop mapping of pockets of vulnerability, but also to be able to predict (given a certain geographic site) the probability of a human health impact (e.g., if the temperature at a location exceeds 100° F, what is the probability of that exceedance lasting for 3 or more days in the current and future time periods?; or if ozone exceeds 70 ppb, what is the probability of the
exceedance continuing for the following 2-3 days?; or if a wildfire erupts in a geographic area, what is the probability that it can be controlled within a 48 hour time period?).
The section on mapping could be more inclusive. Heat is a good illustration, but it would be helpful to discuss mapping of some other vulnerability factors, such as workers’ exposure to heat and better warning systems for persons who do not have access to information or alerts in emergencies. For example, outdoor workers need some type of monitoring system, but at this time, we do not know how many workers are exposed to heat waves per year, and it is unclear which communities are stepping forward to address this challenge.
The inadequate discussion in this section is exemplified in the Traceable Accounts section on page 362. It is only here that the authors note that not all geocoded health data are available in all locations and that, in fact, vulnerable populations such as immigrant populations are more likely to be in the health databases even if they are coded. Mental health outcome data is particularly challenging to obtain and geocode, partly because the majority of cases are underdiagnosed. These uncertainties need to be addressed, not only in the Traceable Accounts section, but in the section describing mapping potential.
This page intentionally left blank.