4 ms·

Are hormones a “female problem” for animal research?

3 points by conse_lad 7y ago

gshdg 7y ago
Of course, if you exclude women and female animals from your studies because you think their hormones will interfere with your ability to draw conclusions... why would you possibly think you can extrapolate said conclusions to more than half the population?
- belorn 7y ago
  If a scientist think that hormones will interfere with their ability to draw conclusions... why should they extrapolate any conclusions? They should inject half their male test subjects with increased testosterone so it match the variation of females test subjects and see if it interfere with the data.
  - gshdg 7y ago
    I mean that if you test only on males because you think hormones will change the outcomes in females, why would you then assume your outcomes also apply to females? And yet for some reason we do.
    - belorn 7y ago
      This is a rather classic example of people talking past each other. The argument used in favor of current practice is that variables in hormones create noise and interfere when analyzing data. This conclusion is for any hormone, being anything from sex hormones like testosterone and estrogen to stress hormones like hydrocortisone and epinephrine. The arguments against current practice as brought up by the article is that this exclude female subjects, which might bias results. The problem is that both sides are right. A variation in hormones does create interference in the data, and a lack of variation in hormones create a bias in the result. Most male humans are not locked in isolation cages and share the testosterone levels of medical male animal test subjects. Most humans in general do not share the stress hormones that such animals often has. The discussing afterward tend to be that both side saying that animal testing is very inexact and almost worthless, except that it is currently better than computer simulations and it is more ethical to first test a theory on animals before doing human trials. Thus we return to the original problem: should testing be done with a variation of hormone levels or not.
      - gshdg 7y ago
        The article doesn’t just mention animal research. Tons of human research is done on men only and then incorrectly assumed to apply to women. It’s not just about animal testing or stress hormones or testosterone levels. It’s also about levels of estrogen and progesterone. It’s also about things like women’s metabolisms actually performing differently than men’s. Women respond differently than men to caloric restriction. We respond differently to antidepressants. Neither has been studied sufficiently. Dosing advice for pretty much anything is based on weight and the average male metabolism. And I’m sure there are far more examples that I’m not aware of and even more that the medical establishment is not yet aware of. This has major implications for women’s health and survival. I’m not saying don’t do the research or even not to do it on a simpler model first. I’m saying don’t extrapolate from men to women without actually doing research on women. I’m saying don’t stop after you’ve done research on men or male animals. The state of medicine for women is much poorer than for men because of decades of NOT including women in studies and then making incorrect assumptions when applying the results.
        belorn 7y ago
        The article is about animal research, not human trials. Human trials are not mentioned anywhere in the article. Metabolisms is also not mentioned. "Both testosterone and estrogen are powerful neuromodulators (12), and so if the potential for intracohort variation in ovarian hormones is concerning, then the same concerns should apply to hormone-associated variability in male subjects. But when male animals made up the majority of experimental subjects, basic scientists mostly considered hormonal variability a nonissue. No one lamented that data would be uninterpretable without careful tracking of dominant versus subordinate cage mates or testosterone assays—any variability in data was largely dismissed as natural randomness. However, when the NIH began to explore requiring its grantees to include both sexes in animal experiments, concerns regarding data variability resulting from ovarian hormones were central to the discussion" This is the core of the discussion. Something which the article miss to bring up is that part of the basic research into stress hormones came about because a famous case where a medical research into ulcers were mishandling test subjects and got uninterpretable data as a result, which promoted further studies. It also lead to stricter standard in handling of test animals in order to minimize variation of stress hormones (which is also more humane so win-win). I personally agree with the article that a variation in testosterone should be seen in the same light as a variation in estrogen. If researchers currently dismiss variation of male hormones as natural randomness then it follows that variation of female hormones of the same amplitude carries similar value. The exception is if it is possible to create studies where the variation of hormones can be decreased, in which case single-sex animal studies might be preferred. In the end however it comes down to the question of natural randomness. A mouse is a very different animal from a human and thus unless there is evidence that hormones will impact the study it might be best practice to treat hormones as natural noise and use animals from both sexes. < We respond differently to antidepressants In 1993 National Institutes of Health (NIH) Revitalization Act set a new standard which required that both women and men must be included in human trials in regard to treatment for depression. That fight is 30 years old and has already been won. No one is arguing in favor of single-sex human trials.

3 points by conse_lad 7y ago

Of course, if you exclude women and female animals from your studies because you think their hormones will interfere with your ability to draw conclusions... why would you possibly think you can extrapolate said conclusions to more than half the population?

belorn 7y ago
If a scientist think that hormones will interfere with their ability to draw conclusions... why should they extrapolate any conclusions? They should inject half their male test subjects with increased testosterone so it match the variation of females test subjects and see if it interfere with the data.
- gshdg 7y ago
  I mean that if you test only on males because you think hormones will change the outcomes in females, why would you then assume your outcomes also apply to females? And yet for some reason we do.
  - belorn 7y ago
    This is a rather classic example of people talking past each other. The argument used in favor of current practice is that variables in hormones create noise and interfere when analyzing data. This conclusion is for any hormone, being anything from sex hormones like testosterone and estrogen to stress hormones like hydrocortisone and epinephrine. The arguments against current practice as brought up by the article is that this exclude female subjects, which might bias results. The problem is that both sides are right. A variation in hormones does create interference in the data, and a lack of variation in hormones create a bias in the result. Most male humans are not locked in isolation cages and share the testosterone levels of medical male animal test subjects. Most humans in general do not share the stress hormones that such animals often has. The discussing afterward tend to be that both side saying that animal testing is very inexact and almost worthless, except that it is currently better than computer simulations and it is more ethical to first test a theory on animals before doing human trials. Thus we return to the original problem: should testing be done with a variation of hormone levels or not.
    - gshdg 7y ago
      The article doesn’t just mention animal research. Tons of human research is done on men only and then incorrectly assumed to apply to women. It’s not just about animal testing or stress hormones or testosterone levels. It’s also about levels of estrogen and progesterone. It’s also about things like women’s metabolisms actually performing differently than men’s. Women respond differently than men to caloric restriction. We respond differently to antidepressants. Neither has been studied sufficiently. Dosing advice for pretty much anything is based on weight and the average male metabolism. And I’m sure there are far more examples that I’m not aware of and even more that the medical establishment is not yet aware of. This has major implications for women’s health and survival. I’m not saying don’t do the research or even not to do it on a simpler model first. I’m saying don’t extrapolate from men to women without actually doing research on women. I’m saying don’t stop after you’ve done research on men or male animals. The state of medicine for women is much poorer than for men because of decades of NOT including women in studies and then making incorrect assumptions when applying the results.
      - belorn 7y ago
        The article is about animal research, not human trials. Human trials are not mentioned anywhere in the article. Metabolisms is also not mentioned. "Both testosterone and estrogen are powerful neuromodulators (12), and so if the potential for intracohort variation in ovarian hormones is concerning, then the same concerns should apply to hormone-associated variability in male subjects. But when male animals made up the majority of experimental subjects, basic scientists mostly considered hormonal variability a nonissue. No one lamented that data would be uninterpretable without careful tracking of dominant versus subordinate cage mates or testosterone assays—any variability in data was largely dismissed as natural randomness. However, when the NIH began to explore requiring its grantees to include both sexes in animal experiments, concerns regarding data variability resulting from ovarian hormones were central to the discussion" This is the core of the discussion. Something which the article miss to bring up is that part of the basic research into stress hormones came about because a famous case where a medical research into ulcers were mishandling test subjects and got uninterpretable data as a result, which promoted further studies. It also lead to stricter standard in handling of test animals in order to minimize variation of stress hormones (which is also more humane so win-win). I personally agree with the article that a variation in testosterone should be seen in the same light as a variation in estrogen. If researchers currently dismiss variation of male hormones as natural randomness then it follows that variation of female hormones of the same amplitude carries similar value. The exception is if it is possible to create studies where the variation of hormones can be decreased, in which case single-sex animal studies might be preferred. In the end however it comes down to the question of natural randomness. A mouse is a very different animal from a human and thus unless there is evidence that hormones will impact the study it might be best practice to treat hormones as natural noise and use animals from both sexes. < We respond differently to antidepressants In 1993 National Institutes of Health (NIH) Revitalization Act set a new standard which required that both women and men must be included in human trials in regard to treatment for depression. That fight is 30 years old and has already been won. No one is arguing in favor of single-sex human trials.