The White House blog of February 26 has a piece by Brian Deese, Senior Advisor to President Obama, entitled “7 out of 10 Doctors: Climate Change Is Already Harming Patients’ Health.” The piece claims that “Today, the American Thoracic Society (ATS) will hit the halls of Congress to educate our representatives about a new survey of more than 900 ATS members, which found the majority of doctors believe climate change is already negatively affecting the health of their patients. In fact, 77 percent of respondents reported that increases in air pollution due to climate change are worsening the severity of illnesses in their patients, and they expect these health impacts will further increase in the future. ATS members also indicated that their patients are experiencing other climate-related health problems — including injuries due to severe weather, allergic reactions, and heat-related impacts.”
This study is more red meat to the Climate Community, since it appears to bring yet more experts onto the side of the climate advocates. We should all be automatically skeptical of studies, however, since some are designed to elicit information from the surveyed group while others are structured to generate pre-ordained conclusions in support of a political agenda. The only way to tell which type of study the ATS has produced is to move beyond the executive summary and talking points and consider whether the survey adheres to rigorous methodologies. Let’s go through this step by step. I realize that this post is a bit long and tedious, but it’s really the only way to understand what we have here. (You can find the ATS study at http://www.atsjournals.org/doi/abs/10.1513/AnnalsATS.201410-460BC.)
Surveys are tricky, since the results can differ dramatically based on the sample size and composition and the way the questions are phrased. The ATS claims a membership in excess of 15,000 doctors, nurses, researchers and other health care professionals. The survey was sent to 5,420 members selected at random, of whom 915 or 17% responded. There are two issues here. The first is that the respondents account for only about 6% of the membership, which brings into question the White House claim regarding 7 out of 10 doctors. More precisely, 70% of 6% or about 4% of ATS members stated their agreement with the broad climate positions that the White House advocates.
Second, there is no way of knowing whether the sample suffers from selection bias. For example, were ATS members who are interested in climate change and supportive of the climate agenda more likely to respond? Were those who disagree with the climate agenda more likely to roll their eyes and throw the survey away without answering? There’s no way to know, but let’s assume for the sake of argument that the sample is fully representative of ATS members.
The survey starts with the question “How knowledgeable do you feel about the association between climate change and health impacts?” Seven percent of the respondents answered “Very knowledgeable”, another 31% “moderately knowledgeable” and 44% “modestly knowledgeable”. The authors suggest that the responding group is therefore authoritative since 82% of the respondents were at least “modestly knowledgeable”. It would be equally valid to say that less than 40% of the respondents were even moderately knowledgeable about climate change.
Moreover, there is a problem with self-identifying expertise. We all know people who believe they know everything, and other people who are very knowledgeable but personally modest about their abilities. Self-identification is not a very good way to determine what people actually know. It would have been better to ask the respondents a short series of questions about climate change to provide a more objective assessment. Questions such as (a) “According to the current consensus, how much have global temperatures increased in the past 150 years?”, (b) “How much have global temperatures increased in the last 15 years?” or (c) “Over the past 100 years, have the frequency and severity of hurricanes increased, decreased or remained roughly the same?”. By the way, the answers are (a) 1° C, (b) zero and (c) “remained roughly the same”.
Another interesting question here is how the respondents obtained their information. Doctors are all very busy people. How many who claimed they were knowledgeable about climate change have actually read any of the Intergovernmental Panel on Climate Change (IPCC) studies and evaluated the evidence for themselves? How many respondents, on the other hand, obtained their information from the New York Times or from TV news? In other words, do they have a detailed scientist’s knowledge of climate change issues or just a layman’s knowledge based on what they have read or heard in the news?
The first substantive question in the survey is “Do you think that climate change is happening?” As discussed often in this space, this is the wrong question to ask. The climate changes all the time. This problem is aggravated in this survey by the confusing definition offered to the respondents, which is “Climate change refers to the idea that the world’s average temperature has been increasing over the past 150 years, may be increasing more in the future, and that the world’s climate is changing as a result.” Eighty-eight percent of the respondents answered “yes”, but this statement is a tautology. Temperature is a major part of climate. If temperature rises, climate is changing by definition. This statement is as illogical as asking whether filling your bathtub would cause the water level to change.
I suspect that the study authors really meant to ask whether the temperature increase is causing changes in other climate parameters. If this is the case, they should have specified what changes they were referring to. Do they mean more frequent and severe hurricanes, more frequent and more severe droughts, more rainfall, more snow, less snow? We should also note that the actual history of these parameters, as opposed to the projection of future trends, can be determined by data, not just opinions. We have extensive data in the US on temperature, precipitation, droughts, wildfires, storms and diseases. Surveying opinion on these parameters is silly. You could, for example, ask people in Boston “How many world series championships have the Red Sox won in their history?” and you would get some insight into how knowledgeable people are about the Sox. The correct answer, however, would be eight, even if 90% of the survey respondents think the correct answer is 4.
The next survey question concerns the causes of climate change. Respondents were given six choices: (1) Caused entirely by human activities (7%), (2) caused mostly by human activities (61%), (3) caused about equally by human activities and natural changes in the environment (20%), (4) caused mostly by natural changes in the environment (8%), (5) caused entirely by changes in the natural environment (2%) and (6) none of the above because climate change isn’t happening (2%). The majority of respondents have answered in line with the “climate consensus” that more than half the observed temperature changes have been caused by human activity. Do the respondents’ answers suggest that these doctors have studied the evidence and reached a conclusion that somehow reinforces the science? Or perhaps the responses are just reflecting what the doctors have read in the newspapers or heard on TV. No way to know.
The survey then asks “How much, if at all, do you think climate change is affecting the health of your patients?” The answers are interesting. Only 10% of respondents answered “a great deal”, 34% said “a moderate amount”, 29% said “only a little” and 8% said “not at all”. The other 20% said that they didn’t know or that they didn’t see patients. This question is a prime example of a common problem in surveys – the double question.
The most notorious instance of this trick was a 1997 encounter between President Bill Clinton and Abigail Thernstrom, an outspoken opponent of racial preferences. President Clinton got in Prof. Thernstrom’s face and demanded, “Abigail, do you favor the United States Army abolishing the affirmative action program that produced Colin Powell–yes or no? Yes or no?” The question actually has two independent parts: (1) do you believe that Colin Powell’s success was a result of affirmative action by the military and (2) do you favor abolishing affirmative action in the military? By demanding a “yes or no” answer, President Clinton was trying to trap Prof. Thernstrom into saying either that she doesn’t think Colin Powell (a very popular figure at the time) should have become Chairman of the Joint Chiefs or that she doesn’t favor abolishing the affirmative action program. This is a clever debating trick, but logically fallacious. Prof. Thernstrom did not take the bait and replied that she does not “think that it is racial preferences that made Colin Powell.” In other words, she answered the double question with two answers: “no and yes”.
The ATS survey is also asking a series of double questions, although not in such a confrontational manner. The first question is whether the respondents are seeing any change in their patients’ health. The second question is whether they attribute these changes to climate. Double questions cannot be answered by “either yes or no”, since the respondent may have different answers to the two parts of the question. Let’s look at some specifics.
The first specific question is whether the respondent thinks his or her patients are currently being affected by climate change in terms of “Heat-related effects (e.g., heatstroke, heat exhaustion, cardio-respiratory illness)”. Forty-eight percent of respondents answered “yes”, but what exactly does “yes” mean? Is the respondent saying that he is seeing an increased incidence of heat-related effects, that he attributes heat-related effects to climate change or both?
According to the IPCC and other climate advocates, average global temperatures have increased by about 1° C (1.8° F) over the last 150 years. For the four hottest months of the year (June, July, August and September) the average high temperature in New York City is 80.25° F. The four-month average high in Philadelphia, about 100 miles away, is 83.25° F, a difference of 3° F. How many doctors would tell their patients that moving from New York to Philadelphia would subject them to a materially increased risk of heatstroke, heat exhaustion or cardio-respiratory illness?
This survey question as phrased is so vague and ambiguous that there is no way of knowing what the respondents had in mind. We can, however, gain a little insight by looking at the comments made by some of the responders on this topic. For example, “Frequency of chronic obstructive pulmonary disease and asthma exacerbations increased with high temperatures.” or “I had a patient with a severe chronic obstructive pulmonary disease exacerbation related to increased temperatures and several asthmatic patients who have had significant worsening of their disease in the heat, requiring increase in medications.” The thrust of these comments is that patients experience health problems when temperatures are unusually high, which is perfectly plausible, but has no bearing on temperature trends or their causation.
The next specific question asks whether the respondent thinks his or her patients are currently being affected by climate change in terms of “Vectorborne infection (e.g. Lyme, West Nile, Dengue Fever, Malaria)”. Forty percent of the respondents answered “yes” to this question. In addition to the double-question problem, the selection of diseases is odd. According to the Centers for Disease Control (CDC), malaria has been virtually eliminated in the US, except for cases where the patient contracted the disease overseas and then returned to the US. Dengue fever is a strictly tropical disease, and, according to the CDC, “Nearly all dengue cases reported in the 48 continental states were acquired elsewhere by travelers or immigrants.” Why would the study authors include these diseases in a question to American doctors? Is the implication that climate change is turning the US into a fever swamp?
Lyme disease has been increasing in the US, particularly in the Northeast, but there is significant controversy over the reason. Climate advocates like to point out that even a slight increase in temperature can cause an increase in the population of the ticks that carry the disease. Other researchers, however, note the increased populations of the deer that carry the ticks. Restrictions on hunting, changes in land use and human demographics have all contributed to the increase in the number of deer. Still other researchers attribute the increase in Lyme disease to changes the populations of other mammals, such as coyotes or red foxes, which either carry ticks or feed on animals that carry ticks. In any case, on what basis are the respondents concluding that the Lyme disease cases they see are a result of climate change? Is that even what they mean?
The last vector-borne disease in the question is West Nile virus, a mosquito-borne illness that has been increasing in the US. Evidence suggests that this disease was introduced into the US in 1999 by an infected individual returning from overseas. Once the virus is present, it can spread rapidly through birds, mosquitoes, people and other vectors. The effect of slightly increased average temperatures is simply unknown.
The issue here is whether the survey actually contains any information about the connection between increased temperature and the incidence of vector-borne diseases. It’s doubtful that many (if any) of the respondents have actually done rigorous studies of this relationship or are claiming specialized expertise. More likely, the survey is offering anecdotal evidence about the increase in such diseases coupled with the casual opinions of the doctors about the relationship to climate change.
Fifty-seven percent of the respondents thought their patients are currently being affected by climate change in terms of “Injuries due to severe storms, floods, droughts, fires.” This is an interesting result, since the data show that there has been no long-term increase in the incidence of any of these four events. The detailed comments by the respondents offer some insight. A number of the comments noted that wildfires affect patient health because of increased particulate matter in the air. On what basis, however, are the respondents attributing these fires to climate change? Wildfires have increased in recent years, but they are well below the level experienced in the 1930s. Are the respondents even offering an opinion here or are they accepting what they believe is a premise of the survey?
Several other comments noted the increased health problems associated with Hurricane Sandy. These observations are perfectly reasonable, but, again, on what basis do the respondents attribute Hurricane Sandy to climate change? There has been no increase in the frequency or severity of Atlantic hurricanes (See my post of November 13, 2012 on this topic). In fact, Sandy was a relatively weak hurricane that happened to hit a highly populated area, causing substantial damage and loss of life. I am not aware of a single climate model that predicted that increased atmospheric carbon concentrations would cause hurricanes to veer suddenly westward and strike the New York metropolitan area. The relationship between Sandy and climate change was part of the narrative presented by the Climate Community, politicians and the major media. The fact that the survey respondents are making this factual error suggests that they are simply repeating what they have heard in the media.
The same is true of the next question, whether patients are currently being affected by climate change in terms of “Air pollution related increases in severity of illness (e.g., asthma, COPD, pneumonia, cardiovascular disease)”. Seventy-seven percent of respondents answered “yes” to this question. Several of the respondents comment quite reasonably that air pollution, particularly particulate matter, causes health problems. There is, however, no connection between conventional air pollution and carbon emissions, which are totally benign at current concentration levels. A false narrative has been pushed by the Climate Community and its political and media allies by insisting on using the term “carbon pollution.” (See my post of June 14, 2014 regarding the despicable ad by the American Lung Association suggesting that carbon dioxide will poison your baby). The respondents here are reflecting a misconception probably derived from their reliance on the media for climate change information. The White House statement that “In fact, 77 percent of respondents reported that increases in air pollution due to climate change are worsening the severity of illnesses in their patients…” is simply false. Sorry, air pollution, whatever its health impacts, is not caused by climate change.
On balance, the study questions are clearly intended to document that climate change is already causing health effects but in fact the survey elicits no useful information. What we have here are scattered anecdotes about patient issues combined with a range of vague opinions about climate change and its impacts. Given the constant double questions in the survey and the often factually incorrect statements embedded in the survey questions, this study contributes nothing of any substance to the climate change debate.
Unfortunately, the poor quality of this work does not stop the authors from going to their inevitable destination – political advocacy. For example, respondents are asked to evaluate the statement “My primary place of work does an effective job minimizing its use of fossil-fuels (e.g., conserving energy/water, recycling equipment, etc.)” Seriously? Nowhere does the survey address the respondents’ views on the relationship between fossil fuels and climate change or on whether reducing fossil fuel use in hospitals or doctors’ offices would be cost-effective or have any meaningful impact on atmospheric carbon. Couldn’t the minimization of fossil fuel use, such as the elimination of diesel back-up generators or a reduction in the number of ambulances in use, actually harm patient outcomes? Would it be beneficial to replace fossil fuels with more expensive alternatives, thereby increasing health care costs?
The next statement for respondents to evaluate is “Teaching about climate change and its association with health impacts should be integrated into medical education.” Seventy-three percent of respondents either agree or strongly agree. OK, fine, but what exactly should be taught in medical schools? The standard climate narrative with all its flaws? The factually incorrect information that permeates this study?
Next comes “Physicians should have a significant advocacy role in relation to climate change and health.” Seventy-four percent either agreed or strongly agreed. What exactly are they supposed to advocate? The confused mess contained in the survey? How about asking doctors to advocate for good science instead?
I don’t fault the responders at all in this survey. They were asked vague and ambiguous questions that required combining their knowledge and their opinions on multiple topics. I suspect that doctors as a group hold opinions on climate change similar to the educated public as a whole, which reflect many of the misconceptions and deliberate distortions carefully fashioned by the Climate Community and happily spread by the media. The study authors, on the other hand, should know better. This study establishes the American Thoracic Society as yet another professional organization which has turned its back on professional standards, science and rigor and opted instead to become a political advocacy group. The result can only be a loss of professional standing and reputation over time, as people increasingly see the distortions and deceptions involved in pushing the climate agenda. The management of the ATS should (and eventually will) be embarrassed.