7 ms·

Unexpected Honey Study Shows Woes of Nutrition Research

40 points by OrwellianChild 11y ago

arnie001 11y ago
I think the short lived research timelines is like the rest of the world is heading towards. There's immense pressure in both industry and the world of investments to yield quick results which suit the perception the funding entity is looking for. From the researchers perspective, they'd need long term funding for their research so that the results of various sweeteners could be studied over time. I'm not sure who would fund this.
- OrwellianChild 11y ago
  There is an entire section of the popular research establishment devoted to "meta-studies" that find, consolidate, and review past research of a topic, judging each for rigor, reproducibility, elimination of bias, etc. Since these studies are separately funded, there is less risk of bias or exaggeration of results. If experimental results hold up to this scrutiny, broader, more dramatic conclusions can be made. The nuances of this (and potential complications) were discussed in Carroll's Healthcare Triage video about the worms and replication [1]. These are easier to trust (and safer to report on), since the heavy lifting is done by the meta-researchers rather than the time-pressured, under-paid pop-sci writer who is trying to fill a word-count quota. [1] https://www.youtube.com/watch?v=9SCFlYlNlLQ https://www.youtube.com/watch?v=9SCFlYlNlLQ
DougN7 11y ago
Already posted at https://news.ycombinator.com/item?id=10453401 https://news.ycombinator.com/item?id=10453401
- OrwellianChild 11y ago
  HN asked me to re-post since the original didn't get much attention before it fell off the front page.
neppo 11y ago
this reminds me of this study that shows how easy it is to fudge the results if you use flawed methods: http://io9.com/i-fooled-millions-into-thinking-chocolate-helps-weight-1707251800 http://io9.com/i-fooled-millions-into-thinking-chocolate-hel...
lars 11y ago
There's a huge difference between a) not finding an effect, and b) finding that there is no effect. The former is the default result, and is what you get when your test lacks statistical power (from having a too low sample size). It sucks that the former often gets reported as the latter in the press.
- Gibbon1 11y ago
  On the other hand, honey is mostly glucose + fructose in about the same ratio as table sugar and HFCS. Seriously. If two things are chemically the same despite being from different provenance, chemistry teaches to expect identical behavior for the most part. Sometimes trace elements change things significantly. Why anyone would expect that of honey though, wishful thinking?
  - LanceH 11y ago
    But it's all natural... It's bad enough when they're talking about one sugar being healthier than another, but it really gets strange when they start comparing table salt.
- OrwellianChild 11y ago
  I look at it as a series of mis-aligned incentives. Only so much funding is available, and that defines the size and scope of the experiments that can be done. This, in turn, limits the study design (and by implication the downstream breadth of the conclusions that can be drawn). Researchers, by and large, do the research they can afford to do. Once that research is done, career and future funding potential comes from how much publicity and notoriety a study can get. Credit in many research communities goes to the authors of the papers that get the most citations, and you get citations by producing popular papers. Novel conclusions breed popularity. In the hands of a pop science journalist/communicator, traffic/audience engagement are often used as criteria for success/career advancement/job security. Thus, there isn't proper, unbiased funding for the serious evaluation and consolidation of the conclusions that can be drawn from a new research paper. This is why we get the Buzzfeed-worthy "They compared honey to corn syrup, and you'll never guess which one was healthier" headlines instead of the boring-but-accurate "No variation reported in limited time-series study of sugar intake" that might better represent the data.
vinceguidry 11y ago
We need to figure out a way to categorize research topics, fields of science, and conclusions by how firm we consider the results to be. I listen to absolutely nothing of what comes out of the field of nutrition science. I am aware that this may make me some number of shades less healthy, but the savings in time is well-worth it. I consider all nutrition studies to not be science, in the same way that I consider a kid pulling ants apart with tweezers to be not science. The information gleaned is of no practical value. What little there is that is of worth ends up getting baked into laws and regulation, so again, I don't have to worry about it. (e.g. mercury is bad for you.) I may be interested in somebody's self-experimentation results. Tim Ferriss wrote about the slow-carb diet, I gave it a shot, and lost weight with it. I similarly lost weight with the intermittent fasting diet. Do I care about the relative health effects vs. the so-called Standard American Diet? No. All I can see, all I care about is whether it makes me lose my excess fat. The human body is too complex for any nutrition study to approach a causative standard of rigor. It's impossible to control for all the variables.
- tertius 11y ago
  > What little there is that is of worth ends up getting baked into laws and regulation, so again, I don't have to worry about it. (e.g. mercury is bad for you.) I largely agree that "nutrition science" does not have rigorous standards. One of the reasons for this is that it's very hard to test hypotheses, legally, on people. And there's an over reliance on self reporting. Now, I personally find your tolerance for what would be considered unsafe way too low. High doses of mercury will kill you very quickly. Low doses of many things, legal things, will kill you slowly.
  - TeMPOraL 11y ago
    Most really harmful stuff is either banned / controlled already, or the dangers are a part of common knowledge. You don't get research articles that "touching live wires contribute to sudden unexpected heart attacks". As for "kill you slowly" things, there's slowly and there's "so slow that you may as well ignore it". Cigarettes and asbestos belong to the former group and guess what, they're controlled and people know about the dangers. Dietary stuff sits squarely in the latter - most of the "causes cancer" stuff won't have any noticeable effect on you in few decades, and if you eventually die of cancer it still won't be clearly connected to a particular lifestyle choice. It's basically paranoia coupled with inability to multiply probabilities (or knowing one should do so).
insensible 11y ago
There's a huge problem with honey labeling—often the packaged product labeled as honey is indeed mostly a cheaper sweetener.
- HillRat 11y ago
  Just to add a little more context, the international honey market is incredibly fraught; lead, antibiotics (including chloramphenicol, which can cause fatal reactions in humans) and artificial sweeteners are all commonly found in Chinese-sourced honey, much of which is transshipped through India and East Asian nations. The upshot: buy from local honey producers with known quality.

I think the short lived research timelines is like the rest of the world is heading towards. There's immense pressure in both industry and the world of investments to yield quick results which suit the perception the funding entity is looking for. From the researchers perspective, they'd need long term funding for their research so that the results of various sweeteners could be studied over time. I'm not sure who would fund this.

OrwellianChild 11y ago
There is an entire section of the popular research establishment devoted to "meta-studies" that find, consolidate, and review past research of a topic, judging each for rigor, reproducibility, elimination of bias, etc. Since these studies are separately funded, there is less risk of bias or exaggeration of results. If experimental results hold up to this scrutiny, broader, more dramatic conclusions can be made. The nuances of this (and potential complications) were discussed in Carroll's Healthcare Triage video about the worms and replication [1]. These are easier to trust (and safer to report on), since the heavy lifting is done by the meta-researchers rather than the time-pressured, under-paid pop-sci writer who is trying to fill a word-count quota. [1] https://www.youtube.com/watch?v=9SCFlYlNlLQ https://www.youtube.com/watch?v=9SCFlYlNlLQ

DougN7 11y ago

Already posted at https://news.ycombinator.com/item?id=10453401 https://news.ycombinator.com/item?id=10453401

OrwellianChild 11y ago
HN asked me to re-post since the original didn't get much attention before it fell off the front page.

neppo 11y ago

this reminds me of this study that shows how easy it is to fudge the results if you use flawed methods: http://io9.com/i-fooled-millions-into-thinking-chocolate-helps-weight-1707251800 http://io9.com/i-fooled-millions-into-thinking-chocolate-hel...

lars 11y ago

There's a huge difference between a) not finding an effect, and b) finding that there is no effect. The former is the default result, and is what you get when your test lacks statistical power (from having a too low sample size). It sucks that the former often gets reported as the latter in the press.

Gibbon1 11y ago
On the other hand, honey is mostly glucose + fructose in about the same ratio as table sugar and HFCS. Seriously. If two things are chemically the same despite being from different provenance, chemistry teaches to expect identical behavior for the most part. Sometimes trace elements change things significantly. Why anyone would expect that of honey though, wishful thinking?
- LanceH 11y ago
  But it's all natural... It's bad enough when they're talking about one sugar being healthier than another, but it really gets strange when they start comparing table salt.
OrwellianChild 11y ago
I look at it as a series of mis-aligned incentives. Only so much funding is available, and that defines the size and scope of the experiments that can be done. This, in turn, limits the study design (and by implication the downstream breadth of the conclusions that can be drawn). Researchers, by and large, do the research they can afford to do. Once that research is done, career and future funding potential comes from how much publicity and notoriety a study can get. Credit in many research communities goes to the authors of the papers that get the most citations, and you get citations by producing popular papers. Novel conclusions breed popularity. In the hands of a pop science journalist/communicator, traffic/audience engagement are often used as criteria for success/career advancement/job security. Thus, there isn't proper, unbiased funding for the serious evaluation and consolidation of the conclusions that can be drawn from a new research paper. This is why we get the Buzzfeed-worthy "They compared honey to corn syrup, and you'll never guess which one was healthier" headlines instead of the boring-but-accurate "No variation reported in limited time-series study of sugar intake" that might better represent the data.

vinceguidry 11y ago

We need to figure out a way to categorize research topics, fields of science, and conclusions by how firm we consider the results to be. I listen to absolutely nothing of what comes out of the field of nutrition science. I am aware that this may make me some number of shades less healthy, but the savings in time is well-worth it. I consider all nutrition studies to not be science, in the same way that I consider a kid pulling ants apart with tweezers to be not science. The information gleaned is of no practical value. What little there is that is of worth ends up getting baked into laws and regulation, so again, I don't have to worry about it. (e.g. mercury is bad for you.) I may be interested in somebody's self-experimentation results. Tim Ferriss wrote about the slow-carb diet, I gave it a shot, and lost weight with it. I similarly lost weight with the intermittent fasting diet. Do I care about the relative health effects vs. the so-called Standard American Diet? No. All I can see, all I care about is whether it makes me lose my excess fat. The human body is too complex for any nutrition study to approach a causative standard of rigor. It's impossible to control for all the variables.

tertius 11y ago
> What little there is that is of worth ends up getting baked into laws and regulation, so again, I don't have to worry about it. (e.g. mercury is bad for you.) I largely agree that "nutrition science" does not have rigorous standards. One of the reasons for this is that it's very hard to test hypotheses, legally, on people. And there's an over reliance on self reporting. Now, I personally find your tolerance for what would be considered unsafe way too low. High doses of mercury will kill you very quickly. Low doses of many things, legal things, will kill you slowly.
- TeMPOraL 11y ago
  Most really harmful stuff is either banned / controlled already, or the dangers are a part of common knowledge. You don't get research articles that "touching live wires contribute to sudden unexpected heart attacks". As for "kill you slowly" things, there's slowly and there's "so slow that you may as well ignore it". Cigarettes and asbestos belong to the former group and guess what, they're controlled and people know about the dangers. Dietary stuff sits squarely in the latter - most of the "causes cancer" stuff won't have any noticeable effect on you in few decades, and if you eventually die of cancer it still won't be clearly connected to a particular lifestyle choice. It's basically paranoia coupled with inability to multiply probabilities (or knowing one should do so).

insensible 11y ago

There's a huge problem with honey labeling—often the packaged product labeled as honey is indeed mostly a cheaper sweetener.

HillRat 11y ago
Just to add a little more context, the international honey market is incredibly fraught; lead, antibiotics (including chloramphenicol, which can cause fatal reactions in humans) and artificial sweeteners are all commonly found in Chinese-sourced honey, much of which is transshipped through India and East Asian nations. The upshot: buy from local honey producers with known quality.