There’s no better way to start the New Year than getting a jump on it. I don’t know where I found this, but “Five Easy Lies” is a wonderful example of how to sow doubt and uncertainty about any set of data. I have written before about how sowing doubt and uncertainty with environmental data is a time-honored tradition of opponents of toxic substances regulation, climate change denialists or anyone desiring to use “too much uncertainty to make a decision” as a strategy for deferring any type of economically or politically painful decision. Stated more eloquently on SKAPP’s web site:
“Doubt is our product,” a cigarette executive once observed, “since it is the best means of competing with the ‘body of fact’ that exists in the minds of the general public. It is also the means of establishing a controversy.”
The lesson about how to talk about data that isn’t telling your story works better with the visuals (visit the site if you’re interested), but here’s a summary of the principles:
Select your cutoffs – focus on just the piece of the trend or the piece of the data set that’s most favorable to your side of the argument.
Talk about the trend of the trend – very useful for engaging people who are bad at math. If there is still an increasing (or decreasing) trend in the data that isn’t helpful to your side of the story, be sure to talk about the rate of the increase or decrease. If this fails, as the post notes, “[k]eep on differentiating until you find a curve that matches your needs.” If that fails, transform the data until is resembles something that’s helpful.
Talk about the different phases – focus on the changes in the data trends – any data set will have it’s moments when it will support your side of the story, even if the weight of evidence is against you. Make sure that noone looks closely at the magnitude of the different data trends.
Focus on outliers – there’s always a case that is not readily explainable based on the preponderance of the data, especially if it’s a noisy data set. This is more easily done if you ignore error bars or other measures of data uncertainty.
Sow confusion – combine any or all of the above to increase doubts about the data set.
I found this to be such a resonating statement:
Evidence is your friend. More evidence means more cutoffs to choose from, more trends to analyze, more phases to count, more outliers to discover, and more confusion to sow. Be careful to disguise the fact that you and not the data are the source of the confusion.
We’ll talk another time about how to criticize the methods use to collect environmental data, as a technique for sowing doubt.
Happy New Year.
Well, maybe a few. Such as the quote from the West Virginia woman. . .
How can we get digital cable and Internet in our homes, but not clean water?
. . . after having to treat her kid for skin lesions because he’s bathed in water contaminated with nickel, or get crowns on the teeth of another kid of hers after metals-contaminated drinking water has eroded the enamel off the kid’s teeth.
How about, “because we care more about digital cable and Internet than clean water”. How about, “because we’ve learned to not think of clean water as a right that we have to continue to fight for”. How about “because the political will which was around 40 years ago to give us the Clean Water Act and the Safe Drinking Water Act is has disappeared”. How about, “because we abrogate this responsibility for our health, leaving it in the hands of politicians and bureaucrats”.
The tone of the New York Times article implies that the government should be doing more, such as more stringent enforcement, or beefing up the resources of regulatory agencies. Helpfully, the article provides a database that allows readers to identify violations of Clean Water Act or Safe Drinking Water Act regulations in their area. There are similar products available from EPA such as EnviroFacts and the TRI database. However these all reflect operational and management metrics – how many permit violations have occurred or how many pounds of chemicals are emitted – which don’t really say anything about environmental conditions or the risks encountered by humans or the ecosystem. These still reflect the “command and control” style of environmental management which some in EPA through the Unfinished Business and Reducing Risk reports were considering to be obsolete as long as 20 years ago. Not addressed in the New York Times article is the idea that the existing command and control regulatory structure developed in the 1970s and 1980s NPDES, RCRA, Clean Air Act permitting and enforcement – is broken.
There is some recognition that a new framework for environmental protection is needed, which acknowledges that health and ecological risks are related to where a community is located, that multiple and overlapping sources of contaminant releases might affect members in that community, and that the best measure of risk is based on what types and levels of contaminants those individuals are exposed to. EPA has developed frameworks for community-based risk assessment and cumulative risk assessment, which acknowledge:
In many cases, human health often is directly related to where one lives. Certain communities, groups, or individuals within a community may be more at risk than others from multiple exposures to chemicals based on the location of a town; the individual’s location within a town; activities, such as commuting to work or school or exercising; dietary patterns of residents; or socioeconomic status. Focusing on the community provides a rational starting point for developing, evaluating, and applying cumulative risk tools to determine the risk of chemical mixtures.
Of course, characterizing risks with this framework involves more monitoring, particularly from the locations where the people or affected wildlife are located, possibly including biomonitoring using biomarkers and genomic markers. The current tools such as EnviroFacts and the TRI, which don’t really tell you anything about what you’re being exposed to aren’t what’s needed for community-based risk assessment. Also, the existing regulatory and legal framework, which is source- and industry-based, becomes an impediment to this very sensible risk-based approach. Who “owns” the liability and responsibility for what pollution? How do you prove that my releases (says the local chemical company) are producing your body burden? That regulatory framework is also going to become an impediment to implementing sustainable chemical production techniques (“green chemistry”), an initiative that also will revolutionize environmental protection.
Without a new paradigm in environmental protection that’s community-based, or oppressive enforcement of the current command and control regulatory framework, we’ll continue to have problems such as kids getting rashes and having their teeth fall from contaminated drinking water.
Enter data judo (to be continued).
Science-Based Medicine, a physicians’ group blog takes to task a recent documentary by the Canadian Broadcasting Corporation, “The Disappearing Male”, which mixes endocrine disruption science with overheated rhetoric to raise the question of the decline of the human species.
The post represents informed advocacy, and can scarcely be considered a full characterization of potential endocrine disruptor risks. While I have some minor disagreements with a few of its sources, overall it needs to be acknowledged as a welcome counterpoint in the endocrine disruption debate.
Science-Based Medicine: “The Disappearing Male – a Pinch of Science, a Pound of Speculation”.
This is something that sounds so cool as an exposure monitoring technology that I hope it pans out experimentally and can be deployed.
Lung cancer cells may exude volatile organic compounds different than normal cells, principally as the byproducts of oxidative stress and byproducts of reactive oxygen species (ROS)-induced processes. The differences may be detectable in breath samples. A monitoring tool is being investigated as a non-invasive way to identify non-small-cell lung cancer. The objective for this tool would be to increase the odds of starting treatment while the disease is in its early stages and still localized. The analytical method involves an array of gold nanoparticle sensors in combination with pattern recognition methods; this level of description is what has been found in the press coverage, and having written it I realize I know as much now as I did before hearing about this technique, which is zip. I’m reading the paper trying out the methodology with headspace samples of tumor cell lines, published in the journal Small and realizing I have a lot of catching up to do on analytical methods. . . .
The sensor can discriminate the breath of normal individuals from lung cancer patients, overcoming the problem of high humidity in the breath samples (a problem with the prior method using carbon nanotubules) and without requiring preconcentration of the breath samples (which would require more complex laboratory techniques). Hossam Haick, the lead investigator estimates this method could become available as a diagnostic tool in about three to five years.
I lurk over at Scienceblogs, where some of the bloggers routinely express their outrage at particular leaders of the “autism community”, when those leaders speak out about vaccinations as a cause for the cases of autism observed in this country. That outrage stems from the unscientific nature of the anti-vaccination arguments, the public health risk created from certain infectious diseases if vaccination rates begin to fall, and the that fact that it appear to work – despite the wrongheadedness of it, the proponents of anti-vaccination are experiencing some success in getting their messages across in the mass media.
Ok, so maybe everyone knows this stuff already, and I’m just demonstrating a firm grasp of the obvious, but there may be a conceptual model which provides some understanding about why the anti-vaccination spokespeople are resonating, and the science bloggers. . . aren’t. Consider: if it’s selling, if you yell it loud and long enough, and if enough people start believing it, whatever “it” is becomes the truth regardless of what the facts are. Also, it helps if the spokespeople are appealing on camera and speak from their Gut (which if you allow it, passes for “common sense”).
I can’t take credit for that analysis, but have absorbed it from a recent reading of Charles Pierce’s new book. I’ll be optimistic that Pierce’s message is something the Sciencebloggers can absorb and use, because it sure seems that being rational, sensible and evidence-based just isn’t cutting it.
In solidarity with Orac over at Scienceblogs, who is annoyed with overwrought science reporting, I offer the following submittal.
I don’t follow the Peak Oil debate closely enough to know who Michael Lynch is, beyond the fact that he was invited to write an op-ed in the New York Times about what a farce Peak Oil was, and that we’re not really running out of oil.
Reading that piece prompted a remembrance from about thirteen years ago. At the time, I was working on the RCRA corrective action for an oil refinery along the Gulf Coast. I had traveled to attend a meeting near the facility. During a break near the end of the day, I was chatting with one of the client’s technologists. He excused himself to go out and have a smoke. I went with him, bummed one and we continued talking. It was fall, and getting on towards sunset. Sunset was beautiful, all reds and oranges filtered through the crappy air quality and highlighting the cracking towers, flares and processing equipment in the background. Call it perverse but there was a sense of order at that moment, standing at the locus of ravening energy consumption and toxic hot spots, inhaling carcinogens and fine particulate matter.
We drifted on to the topic of oil depletion and the future of the energy business, and he expressed the opinion that we’re unlikely to ever run out of oil, which was understandable coming from a representative of the energy bidness. I replied yeah, we’ll probably pump 2% carbon dioxide into the air before we run out of oil. Both of us chuckled, then we stubbed out our cigs and went back into the meeting.
That was satire. Currently, a typical carbon dioxide concentration in the atmosphere is 384 parts-per-million (ppm), or 0.000384% in air. The IPCC predictions are that, with emissions unchecked, concentrations could rise to over 800 ppm, or 0.0008% by 2100; I linked that value from an article that’s skeptical of anthropogenic global warming just to show I’m not biased, and to present an example of really, really, really bad science journalism. Anyhow, 2% carbon dioxide, my untutored speculation, would be 20,000 ppm in air, which if we achieved that, would probably have sedentary unfit people and asthmatics breathing a little harder, render glaciers a distant memory and make Little Rock, Arkansas beachfront property. I’m not terribly interested in commenting on Mr. Lynch’s analysis, though I’m sure the Oil Drum is on it. However, there is a description of subsurface investigation pertinent to this matter which has stuck with me, though I can’t recall the source any longer, and which goes like this: trying to understand the Earth by boring holes in the ground is like trying to understand the stars by looking at them through a soda straw. . . . Readers who don’t know anything about geological uncertainty, in other words most of them, might be fooled into thinking he has a point, when he’s really just blowing smoke. The Peak Oil mavens might be as wrong as he says, though we don’t really know that, and that being the case, how do we want to roll the dice with regard to our oil policy? Actually, we’ve already decided that, by making our oil policy the same as our defense policy, and committing to fighting a combined resource and holy war in the Middle East. Unanswered in Mr. Lynch’s editorial is the why are we fighting and dying for oil, if it’s so plentiful?
This was a waste of time, but it’s off my chest, and I can now close the link to Mr. Lynch’s editorial and the New York Times op-ed page and move on. Besides, I needed a break from writing about endocrine disruptors, carcinogens and breast cancer.
[Note: this a series of posts stimulated by this recently-published research on breast cancer risks from multiple environmental contaminants. A previous post is here.]
I have been aware of initiatives to address endocrine disrupting chemicals, early-life exposure to environmental contaminants, and cumulative risk assessment but over the past few years hadn’t paid that much attention to them. While all of these topics had public health importance, and were beginning to turn into risk assessment guidance, regulatory agencies just haven’t been requiring them to be used for making decisions about the kinds of problems affecting my clients. So, they were more of an intellectual curiosity.
That might be changing. In 2005, EPA published guidelines for assessing susceptibility from early life exposure to carcinogens. In 2008, EPA updated the screening levels used for evaluating contaminant data at Superfund sites and incorporated the early life exposure guidelines for selected carcinogenic chemicals. Guidance on how to conduct cumulative risk assessments is steadily becoming more specific, and all of this represents a sea-change in how to perform risk assessments which the National Academy of Sciences says is overdue.
A recent review article argues there is substantial evidence that hormonal perturbations early in life (either in utero or during early development) are associated with increased disease susceptibility later in life, with two examples being prostate and breast cancer. The Endocrine Society has recently issued a scientific policy statement (news items here and here; link to the report here) identifying endocrine disruptors as a significant public health concern. The thrust of these stories is that professional societies are becoming involved not in just generating the science but in encouraging that it be used in policy making.
BPA, which was mentioned in the previous post, gets its own chapter in the Endocrine Society’s report. Low-dose exposure in rat fetuses to has resulted in alterations in mammary tissue. Higher dose prenatal exposures (i.e. where the pregnant females are dosed with BPA) increase the numbers of precancerous lesions in next-generation rats later in life. BPA increased mammary tumor incidence in animals when administered along with rodent carcinogens such as nitrosomethylurea and dimethylbenzanthracene. The Endocrine Society’s summary statement is:
These results indicate that perinatal exposure to environmentally relevant doses of BPA results in persistent alterations in mammary gland morphogenesis, development of precancerous lesions, and carcinoma in situ.
Or, exposure to levels of BPA, which you might normally encounter through your daily routine, might, if you’re pregnant, predispose your female child to an increased breast cancer risk. The Endocrine Society speculates that the increased incidence of breast cancer observed over the last 50 years might have been caused in part by exposure of women to endocrine-mimicking chemicals.
Of course, it’s risky to let yourself get tunnel-vision and focus on only one answer. In 2002, reports that post-menopausal hormone therapy posed an increased breast cancer risk resulted in a rapid decline in this kind of hormone use in women. The decline in hormone use is suspected to be a contributor to the subsequent decline in breast cancer rates. Would reducing BPA exposure in a systematic manner, also result in a decline in breast cancer rates? It would be hard to say – an epidemiological investigation of post-menopausal hormone therapy is a good deal simpler than investigating BPA; 93% of Americans have detectable levels of BPA in their bodies – where do you find a control population?
Not done yet. . . .
Bisphenol-a, used to manufacture polycarbonate including plastics for food and beverage containers, has been found to leach from those containers, is consumed by us and can be detected at trace levels in nearly everyone’s blood and urine. Bisphenol-a is hormonally active (otherwise known as an endocrine disruptor), and produces reproductive and developmental abnormalities in laboratory animals including changes in mammary glands. The kinds of changes observed in mammary tissue leads some to be concerned that bisphenol-a might pose some level of risk for breast cancer.
By itself, maybe the breast cancer risk from bisphenol-a (or BPA) by itself isn’t something a woman ordinarily needs to be concerned about. There isn’t any certainty about it, and the effects observed in lab animals are pretty subtle. But we’re not exposed to BPA by itself, but as mixtures of contaminants. Other chemicals that we’re commonly exposed to that are possible human carcinogens are polycyclic aromatic hydrocarbons (PAHs). PAHs are found in tobacco smoke, air pollutants, motor vehicle exhaust, particularly diesel, and some fried or smoked foods. One PAH, dimethylbenzanthracene (DMBA) reliably causes mammary cancer in a selected strain of rat, so that DMBA-rat system is used as an animal model for breast cancer research.
A study published a few months ago in Environmental Health Perspectives explored the hypothesis that exposure to BPA early in life would produce changes in mammary tissue, creating a predisposition for breast cancer. This study investigated the interaction between BPA and breast cancer risks by exposing newborn rats to BPA through lactation, then giving the young female rats oral doses of DMBA.
Those tending to be skeptical about environmental contaminant-disease trend relationships might be inclined to note that breast cancer incidence and mortality rates have declined over the past several years. When you browse the SEER statistics directly (breast, Figure 4.2), mortality hovers a little over 50 per 100,000, while incidence hovers around 300-350 per 100,000. While a decreasing trend is good news, a worthwhile question then is how many cases are too many? Note: yes, I know zero would be nice, but let’s stay in the real world for now. . . . The CDC reports that in 2005, about 186,000 women were diagnosed with breast cancer and 41,000 died from it. Aside from the financial and human costs, there is the issue that a substantial fraction of those cases may still be avoidable.
To be continued. . . .
From a certain point of view (as Obi-Wan Kenobi might say), John Cloud’s article in TIME Magazine, “Why Exercise Won’t Make You Thin” is understandable. The man is a working writer who’s success is based on being dependable – being able to turn out a product of predictable quality and being able to stick to a schedule. That, from a certain point of view, his article resembles little more than a loosely fact-checked blog post, is more the responsibility of TIME’s editors. However, it would be too easy to chalk this article up to poor-quality writing and editing – far from it. Instead, I would argue that “Why Exercise Won’t Make You Thin” was precisely what TIME was looking for because it contained a message that resonates with a substantial potion of its readership; that message being “it’s not your fault that you’re overweight – exercise and self-control don’t work, and science has proved that”.
Cloud shows some surprise that exercise, at least the way he appears to define it, doesn’t fully offset the calories you take in, particularly if you’re trying to lose weight,
You might think half a muffin over an entire day wouldn’t matter much, particularly if you exercise regularly. After all, doesn’t exercise turn fat to muscle, and doesn’t muscle process excess calories more efficiently than fat does?
That last statement, in Cloud’s hands, becomes a strawman to be knocked down and so there is no exploration of how diet, exercise, muscle and fat are interrelated. No, exercise doesn’t turn fat into muscle. However, aerobic exercise in combination with a mild calorie deficit will burn fat. This combination by itself will also burn some muscle, so some strength training should accompany it to help you hold on to as much of your muscle mass as possible, while the fat burning is under way.
We now return to TIME:
Yes, although the muscle-fat relationship is often misunderstood. According to calculations published in the journal Obesity Research by a Columbia University team in 2001, a pound of muscle burns approximately six calories a day in a resting body, compared with the two calories that a pound of fat burns. Which means that after you work out hard enough to convert, say, 10 lb. of fat to muscle — a major achievement — you would be able to eat only an extra 40 calories per day, about the amount in a teaspoon of butter, before beginning to gain weight. Good luck with that.
Checking out this article, which is a review of the various methods for calculating resting energy expenditure (REE), does reveal as part of a conceptual model for REE that internal organs such as heart, kidneys, liver and brain have higher resting metabolic rates compared with skeletal muscle (13 kcal/kg-day) and body fat (4.5 kcal/kg-day). Checking the math indeed shows this to correspond to 6 kcal/day and 2 kcal/day (kcal = the “calorie” we hear about). However, for purposes of fat loss (hence weight loss), what we’re interested in is total energy expenditure which in the introduction to this article is described as the sum of five components: REE, physical activity-induced energy expenditure, thermic effect of food, facultative thermogenesis (energy expenditure in response to hot or cold conditions – sweating or shivering) and anabolism/growth. It’s the physical activity and the anabolism (muscle growth induced by strength training) that’s going to promote weight loss.
In the end, I can only say that John Cloud’s experience with weight loss seems to be completely different from mine. When I’m working on weight and fat loss (which isn’t continually), the strategy has been mild calorie reduction, food selections based on calculated macronutrient ratios (say 35%/35%/30% protein/carbs/fats) with a bias towards lean protein, vegetables, fruit, low glycemic index carbs, healthy fats such as fish and olive oil, combined with exercise intensity and duration along the American College of Sports Medicine recommendations (at least 150 minutes per week), mixing different modes of cardio with strength training including weight lifting. I can attest that this strategy has worked for me, allowing me to lose 50 lbs, keep it off for over eight years, and put me in striking distance of seeing my abs again. There are time sacrifices involved (you have to do some shopping and cooking), and I need to accommodate this within the schedule associated with being a success-addled corporate professional. However, I don’t regret one bit making fitness a goal in my life. I’m even starting to enjoy working out. One other thing I can recommend to John, beyond doing some weight training and paying closer attention to diet is to find a workout partner. Maybe his next article on exercise, fitness and diet might actually be informing and inspiring.