6 ms·

Did they describe how they controlled for behavioral habits? It seems like, as a group, the type of people who would drink large amounts of sugary soda would be

Did they describe how they controlled for behavioral habits? It seems like, as a group, the type of people who would drink large amounts of sugary soda would be less likely to make healthy behavioral choices in other aspects of their life, compared to people who drink fruit juice or diet soda. This could also explain the disparity between soda and non-carbonated sugar-sweetened beverages This study doesn't seem to show that soda ages you (as the article states), but instead that sugary soda consumption indicates a tendency towards other activities/habits that may prematurely age you.

mikeyouse 12y ago
> This study doesn't seem to show that soda ages you (as the article states), but instead that sugary soda consumption indicates a tendency towards other activities/habits that may prematurely age you. No offense, but the authors are very accomplished public health researchers. If their study showed what you suggest, I think that would be their headline. What they showed with (relative) certainty is that those who drink sugary carbonated drinks have shorter telomeres than those who don't. There will be follow-up work to determine the mechanism and the causes. In either case; > Epel is co-leading a new study in which participants will be tracked for weeks in real time to look for effects of sugar-sweetened soda consumption on aspects of cellular aging. Telomere shortening has previously been associated with oxidative damage to tissue, to inflammation, and to insulin resistance.
- aaron-lebo 12y ago
  I don't see any reason to discount the authors, but a couple of things to keep in mind: 1) This is Time's popularization of their study, so Time is naturally going to put forward the most interesting and least nuanced claims of theirs. 2) They may be accomplished researchers, but if you are well-versed in an academic discipline and read material in it, it quickly becomes clear that sometimes points which seem obvious have not actually been accounted for whether by mistake or because they don't allow for interesting publishable material. Again, I'm not saying that is the case, and my points above are overly cynical, but the original poster's question is valid.
  - anigbrowl 12y ago
    The Time article is actually a very dry summary of the study's key finding, written in colloquial but not sensational language. It also has a link to the study in the very first sentence. I found the full text very easily: http://www.chc.ucsf.edu/ame_lab/pdfs/Leung%20et%20al%202014a.pdf http://www.chc.ucsf.edu/ame_lab/pdfs/Leung%20et%20al%202014a... The poster above could easily have satisfied his/her own curiosity on this point. 'Did researchers control for [extremely obvious thing]?' is becoming like 'correlation is not causation' - a smart-sounding thing to say instead of going to the effort of engaging with the material. It only took me 2 minutes to track down the paper and find that the covariate factors are discussed in detail over a full page of the paper.
    - qu4z-2 12y ago
      In fact, in this case it's more or less exactly like 'correlation is not causation'.
    - Arnor 12y ago
      1. Some of us aren't subscribers to APHA or aren't interested in spending $22 on the article. 2. Why so confrontational? If you know the answer to the question, why not summarize it? Is the answer to the question so complicated that it would take more time and energy than the two paragraphs you composed in derision?
      - mikeyouse 12y ago
        I can address #2; > If you know the answer to the question, why not summarize it? Is the answer to the question so complicated that it would take more time and energy than the two paragraphs you composed in derision? Every science article posted on HN gets the exact same treatment where well-intentioned but scientifically ignorant posters fill the comment section with drive-by posts saying "Correlation != Causation" or "Did they correct for [obvious variable]". It's exhausting and completely unproductive. If they would take 30 seconds and actually read the source information, they would answer their questions and the comment threads wouldn't devolve into this annoying meta conversation. Take the post that prompted this whole thing; > Did they describe how they controlled for behavioral habits? It seems like, as a group, the type of people who would drink large amounts of sugary soda would be less likely to make healthy behavioral choices in other aspects of their life, compared to people who drink fruit juice or diet soda. This could also explain the disparity between soda and non-carbonated sugar-sweetened beverages > This study doesn't seem to show that soda ages you (as the article states), but instead that sugary soda consumption indicates a tendency towards other activities/habits that may prematurely age you. So a poster, that later admits he didn't read the article, not only discounts the research that a half dozen PhDs in public health performed, but then provides his own summary of their research that is completely at odds with their results and experience. How is this helpful? If one were trying to troll a comment section to waste everyone's time, that's literally the best tactic since anything more obvious would be flagged or ignored.
harpastum 12y ago
I just looked through the study. They controlled for Smoking, Alcohol, Energy Intake, Alcohol Intake, Healthy Eating (via "Healthy Eating Index"), Body Mass Index, and Waist Circumference. Relevant Section: Health-related variables included smoking status (never, former, current), pack-years of smoking (0, < 30, 30---60, and > 60), physical activity assessed from questionnaire (some activity, no activity), total energy intake, alcohol intake, and Healthy Eating Index 2005 (HEI 2005) score, which is a dietary pattern de- veloped by the USDA to measure compliance with national dietary guidelines. The HEI 2005 is scored out of 100 points and comprises 12 components: total fruit, whole fruit, total vegetables, dark green and orange vegetables and legumes, total grains, whole grains, milk, meat and beans, oils, saturated fat, sodium, and calories from solid fats, alcoholic beverages, and added sugars. We collapsed HEI 2005 scores into gender-specific quartiles: for men, the cutpoints were 42.1, 45.9, and 50.5; for women, the cutpoints were 44.4, 48.6, and 53.5. We defined alcohol intake as low (0---0.5 drinks/day for men and women), moderate (0.5---2.0 drinks/day for men; 0.5--- 1.5 drinks/day for women), and heavy ( > 2 drinks/day for men, > 1.5 drinks/day for women). Adiposity measures included body mass index (BMI) and waist circumference. We calculated BMI from self-reported height (in meters) and weight (in kilograms squared), measured by trained personnel using a stadiometer and Toledo weight scale (Toledo Scale, Honolulu, HI). We defined BMI categories as underweight (BMI < 18.5 kg/m2), normal weight (BMI = 18.5---24.9 kg/m2), overweight (BMI = 25.0---29.9 kg/m2), and obese (BMI ‡ 30 kg/m2). Waist circumference (centimeters) was measured at the upper lateral border of the right ilium. We defined elevated waist circumference as 102 centimeters or greater for men and 88 centimeters or greater for women.
manicdee 12y ago
Edited from the report: Potential confounders included sociodemographics (participant’s age, gender, self-reported race/ethnicity (white, black, Hispanic, other), highest educational attainment, ratio of family income to poverty, and marital status. Health-related variables included smoking status, pack-years of smoking, physical activity assessed from questionnaire (some activity, no activity), total energy intake, alcohol intake, and Healthy Eating Index 2005.

mikeyouse 12y ago

> This study doesn't seem to show that soda ages you (as the article states), but instead that sugary soda consumption indicates a tendency towards other activities/habits that may prematurely age you. No offense, but the authors are very accomplished public health researchers. If their study showed what you suggest, I think that would be their headline. What they showed with (relative) certainty is that those who drink sugary carbonated drinks have shorter telomeres than those who don't. There will be follow-up work to determine the mechanism and the causes. In either case; > Epel is co-leading a new study in which participants will be tracked for weeks in real time to look for effects of sugar-sweetened soda consumption on aspects of cellular aging. Telomere shortening has previously been associated with oxidative damage to tissue, to inflammation, and to insulin resistance.

aaron-lebo 12y ago
I don't see any reason to discount the authors, but a couple of things to keep in mind: 1) This is Time's popularization of their study, so Time is naturally going to put forward the most interesting and least nuanced claims of theirs. 2) They may be accomplished researchers, but if you are well-versed in an academic discipline and read material in it, it quickly becomes clear that sometimes points which seem obvious have not actually been accounted for whether by mistake or because they don't allow for interesting publishable material. Again, I'm not saying that is the case, and my points above are overly cynical, but the original poster's question is valid.
- anigbrowl 12y ago
  The Time article is actually a very dry summary of the study's key finding, written in colloquial but not sensational language. It also has a link to the study in the very first sentence. I found the full text very easily: http://www.chc.ucsf.edu/ame_lab/pdfs/Leung%20et%20al%202014a.pdf http://www.chc.ucsf.edu/ame_lab/pdfs/Leung%20et%20al%202014a... The poster above could easily have satisfied his/her own curiosity on this point. 'Did researchers control for [extremely obvious thing]?' is becoming like 'correlation is not causation' - a smart-sounding thing to say instead of going to the effort of engaging with the material. It only took me 2 minutes to track down the paper and find that the covariate factors are discussed in detail over a full page of the paper.
  - qu4z-2 12y ago
    In fact, in this case it's more or less exactly like 'correlation is not causation'.
  - Arnor 12y ago
    1. Some of us aren't subscribers to APHA or aren't interested in spending $22 on the article. 2. Why so confrontational? If you know the answer to the question, why not summarize it? Is the answer to the question so complicated that it would take more time and energy than the two paragraphs you composed in derision?
    - mikeyouse 12y ago
      I can address #2; > If you know the answer to the question, why not summarize it? Is the answer to the question so complicated that it would take more time and energy than the two paragraphs you composed in derision? Every science article posted on HN gets the exact same treatment where well-intentioned but scientifically ignorant posters fill the comment section with drive-by posts saying "Correlation != Causation" or "Did they correct for [obvious variable]". It's exhausting and completely unproductive. If they would take 30 seconds and actually read the source information, they would answer their questions and the comment threads wouldn't devolve into this annoying meta conversation. Take the post that prompted this whole thing; > Did they describe how they controlled for behavioral habits? It seems like, as a group, the type of people who would drink large amounts of sugary soda would be less likely to make healthy behavioral choices in other aspects of their life, compared to people who drink fruit juice or diet soda. This could also explain the disparity between soda and non-carbonated sugar-sweetened beverages > This study doesn't seem to show that soda ages you (as the article states), but instead that sugary soda consumption indicates a tendency towards other activities/habits that may prematurely age you. So a poster, that later admits he didn't read the article, not only discounts the research that a half dozen PhDs in public health performed, but then provides his own summary of their research that is completely at odds with their results and experience. How is this helpful? If one were trying to troll a comment section to waste everyone's time, that's literally the best tactic since anything more obvious would be flagged or ignored.

harpastum 12y ago

I just looked through the study. They controlled for Smoking, Alcohol, Energy Intake, Alcohol Intake, Healthy Eating (via "Healthy Eating Index"), Body Mass Index, and Waist Circumference. Relevant Section: Health-related variables included smoking status (never, former, current), pack-years of smoking (0, < 30, 30---60, and > 60), physical activity assessed from questionnaire (some activity, no activity), total energy intake, alcohol intake, and Healthy Eating Index 2005 (HEI 2005) score, which is a dietary pattern de- veloped by the USDA to measure compliance with national dietary guidelines. The HEI 2005 is scored out of 100 points and comprises 12 components: total fruit, whole fruit, total vegetables, dark green and orange vegetables and legumes, total grains, whole grains, milk, meat and beans, oils, saturated fat, sodium, and calories from solid fats, alcoholic beverages, and added sugars. We collapsed HEI 2005 scores into gender-specific quartiles: for men, the cutpoints were 42.1, 45.9, and 50.5; for women, the cutpoints were 44.4, 48.6, and 53.5. We defined alcohol intake as low (0---0.5 drinks/day for men and women), moderate (0.5---2.0 drinks/day for men; 0.5--- 1.5 drinks/day for women), and heavy ( > 2 drinks/day for men, > 1.5 drinks/day for women). Adiposity measures included body mass index (BMI) and waist circumference. We calculated BMI from self-reported height (in meters) and weight (in kilograms squared), measured by trained personnel using a stadiometer and Toledo weight scale (Toledo Scale, Honolulu, HI). We defined BMI categories as underweight (BMI < 18.5 kg/m2), normal weight (BMI = 18.5---24.9 kg/m2), overweight (BMI = 25.0---29.9 kg/m2), and obese (BMI ‡ 30 kg/m2). Waist circumference (centimeters) was measured at the upper lateral border of the right ilium. We defined elevated waist circumference as 102 centimeters or greater for men and 88 centimeters or greater for women.

manicdee 12y ago

Edited from the report: Potential confounders included sociodemographics (participant’s age, gender, self-reported race/ethnicity (white, black, Hispanic, other), highest educational attainment, ratio of family income to poverty, and marital status. Health-related variables included smoking status, pack-years of smoking, physical activity assessed from questionnaire (some activity, no activity), total energy intake, alcohol intake, and Healthy Eating Index 2005.