How long would a person live without bacteria was recently closed as primarily opinion based. This is puzzling to me.

It can be argued that this (eliminating all microorganisms) has never been tested in humans, but much of out knowledge about human physiology has not been derived from experiments on humans; indeed the vast majority of what we know about human physiology and medical conditions has been derived from animal studies.

The dog genome has been mapped out not only for it's importance to veterinarians, but also because dog-models have been so vital (and suitable) to research of human disease. The role of mouse models has been invaluable. As stated in Mouse Models of Developmental Genetic Disease:

It is the recognition that humans and mice share the same organ systems, similar reproductive cycles, skeletons, biochemistry, physiology, and most importantly pathologies that have enabled us to progress in our understanding of the basis of human disease and this shows no sign of abating in the foreseeable future.

The use of gnotobiotic mice has recently been important to the better understanding of microbiomes and obesity, cholesterol metabolism, inflammatory bowel disease and infection-mediated diseases, type 1 diabetes, allergies, Antibiotic-Associted Diarrhea (C. difficile) and cancer, to mention only a few diseases.

In defending the use of gnotobiotics as a crucial element in studying human diseases, Rolf Freder points out

…[I]f one chooses to be rigorous in the definition of what constitutes the reproduction of a human disease, one may well state that no such model has ever been discovered.

Why, then, is a question that absolutely can be answered by studies in the murine model judged to be POB?

I would appreciate any insights, as well as what would need to be done to make this question on topic, as I believe an interest in the human (and therefore animal model) microbiomes and disease (or lack thereof) will prove an illuminating subject of study in the coming decades.

Leading the way: canine models of genomics and disease (Disease Models & Mechanisms)
Animal models of human disease: zebrafish swim into view
Mouse models of human disease
Mouse Models of Developmental Genetic Disease
From Structure to Function: the Ecology of Host-Associated Microbial Communities
Methods and Special Applications in Bacterial Ecology

  • $\begingroup$ Are you talking only about physiology? What about drug trials? Why should they be conducted in humans if animal studies are enough? I agree that most of our knowledge of physiology came initially from animals. But, based on animal studies, human studies are carried out. Are they not? $\endgroup$
    – One Face
    Commented Feb 28, 2015 at 2:03
  • 1
    $\begingroup$ Looking at the big picture, even in drug studies, far more research takes place on animals than humans; that research is then used to design human studies. That studies are then done on humans does not negate the point that animal research has a direct correlation to what we know about humans which cannot be ignored or even downplayed. From Wikipedia: "The classic model vertebrate is currently the mouse (Mus musculus)." (My references are more detailed, though.) $\endgroup$ Commented Feb 28, 2015 at 2:30
  • $\begingroup$ I think this entire conversation is irrelevant for this specific question. In your own citation it clearly gives examples (bubble boy) that showed humans can and did survive in a sterile environment. You don't need to extrapolate from mice studies since there is already a human example. $\endgroup$
    – March Ho
    Commented Feb 28, 2015 at 8:29
  • $\begingroup$ @March Ho - I agree that there was a gnotobiotic human, who would have survived quite well to old age in the bubble - physically - and did so until given a bone marrow transplant that killed him a short time later. I was hoping to convince people with science, not hit them over the head with facts. Sadly, it appears science isn't good enough for many scientists. $\endgroup$ Commented Feb 28, 2015 at 8:37
  • $\begingroup$ I don't see how the bubble boy example is a poorer answer than the experimental mice example though. Both are equally scientific (the gnotobiotic mice didn't appear to have a control group either). $\endgroup$
    – March Ho
    Commented Feb 28, 2015 at 8:38
  • $\begingroup$ please read on both the sides the questions. ncbi.nlm.nih.gov/pubmed/15060191. $\endgroup$
    – One Face
    Commented Feb 28, 2015 at 13:52
  • $\begingroup$ @MarchHo Bubble Boy experiment does not tell us anything about lifespan. It tells us only about survivability. The question also asks for lifespan. And it is good enough to prove that humans can survive without micro-organisms. Mice studies are infact inferior to prove survivability when compared to human model. $\endgroup$
    – One Face
    Commented Mar 1, 2015 at 2:01
  • $\begingroup$ @MarchHo I think you can't extrapolate the bubble boy data to general population as such because he was not normal. (He had scid).. $\endgroup$
    – One Face
    Commented Mar 1, 2015 at 3:34
  • $\begingroup$ How can you extrapolate data from an immune deficient person to a normal person? I didn't think that was possible without some correction atleast. Can we continue in chat? $\endgroup$
    – One Face
    Commented Mar 1, 2015 at 4:34
  • $\begingroup$ I've cleaned up the comments here. Keep it civil, folks. $\endgroup$
    – hairboat
    Commented Mar 2, 2015 at 17:19

3 Answers 3


It doesn't matter, at least not here on meta. We don't close questions just because we suspect that the answer isn't known. Even if the actual answer isn't known, a post explaining the current understanding and the limits of out knowledge is a valid answer.

The "primarily opinion based" close reason was created to deal with questions like "What is your favorite text editor?" or "Are tabs better than spaces?". This kind of question is problematic in cases where every answer is equally valid. It doesn't mean that any question that doesn't have a single objective answer is automatically off-topic.

The question shouldn't be closed in my opinion, it could benefit from a bit of rewriting, but it is a valid question.

  • 1
    $\begingroup$ It is sometimes difficult to screen stuff especially when there are not many reviewers. I edited the question and voted to reopen. Also, as I pointed out elsewhere, stackoverflow dynamics is totally different from here. Practically anyone can ask a bio question but to even ask a question in stackoverflow you need to know something about programming. $\endgroup$
    Commented Feb 28, 2015 at 20:05
  • $\begingroup$ @MadScientist, please answer my question about health advice too $\endgroup$
    – One Face
    Commented Mar 1, 2015 at 4:27

The post if not opinion based, is broad. And it is just closed and there is a good deal of chance to reopen it.

Whether a model is sufficient or not depends on the question that is being asked. For basic questions like DNA repair, even yeast model would suffice whereas questions related to reproductive biology of humans would require primate models (especially for questions related to female reproductive biology).

For dependence on microbiome, I think data from mice would not accurately represent human situation because the metabolism and microbiome constitution are quite different. Even primate models may not be appropriate. If I am not wrong, then microbiome constitution also depends on geographical location and dietary practices.

To conclude, the answer to the question - "Do animal experiments really mean anything when it comes to extrapolating findings to humans?" is:

Depends on what is being extrapolated.

  • $\begingroup$ If mice can live without a microbiome, do you think humans are so different that it's unlikely that a human can? Perhaps you aren't familiar with the story of the "bubble boy", born with Severe Combined Immunodeficiency Disorder, who was delivered by C-section, placed in sterile drapes, then immediately put into isolation, becoming the first germ-free human. Does that change your answer? $\endgroup$ Commented Feb 28, 2015 at 8:35
  • $\begingroup$ @anongoodnurse I know about the SCID boy. But if mice can live without microbime it doesn't really mean any other organism can too. For example if humans totally lack the biosynthetic pathway for an essential metabolite which is provided by the gut bacteria; this pathway may be just partially impaired in mice. So the loss of microbiome will not have same effects. However, it is possible that the essential metabolite is provided to the human externally (which requires the knowledge of that metabolite). $\endgroup$
    Commented Feb 28, 2015 at 8:44
  • $\begingroup$ Mice, rats, guinnea pigs (the go-to model for so much research into human disease), birds; all have survived gnotobiotically. The primary reason larger mammals have not been raised this way is size. If you knew about David Vetter, than why do you think any other human would do less well? $\endgroup$ Commented Feb 28, 2015 at 8:49
  • $\begingroup$ @anongoodnurse I am not saying they cannot. You are asking two different questions here: 1. How well can models represent human condition 2. Can humans survive without microbiota. In the above post I answered point-1. For 2: David Vetter is already a human model, you do not need animal models. How long can a human survive without a gut microbiome depends on how well can you artificially supplement what the microbiome provides. $\endgroup$
    Commented Feb 28, 2015 at 8:59
  • $\begingroup$ I also asked a third question: what can be done to make it on topic. It's not too broad for me, and it's not primarily opinion based. Vetter was kept alive and "normal" developmentally with little of the knowledge (compared to today) on the part of his caregivers of what microbiomes supply. $\endgroup$ Commented Feb 28, 2015 at 9:14
  • $\begingroup$ @anongoodnurse What can be done to make it on-topic: State assumptions clearly. And how would they die? part of the question is opinion-based. They can die of diarrhoea or perhaps even cancer (based on recent research studies on the effect of microbiota on cancer). This is just opinion based $\endgroup$
    Commented Feb 28, 2015 at 9:23
  • $\begingroup$ @anongoodnurse can we continue this discussion on chat. I am not against reopening the post but we need to clarify some misunderstandings. $\endgroup$
    Commented Feb 28, 2015 at 9:36

Animal experiments have contributed much to our understanding of mechanisms of disease, but their value in predicting the effectiveness of treatment strategies in clinical trials has remained controversial [1]–[3]. In fact, clinical trials are essential because animal studies do not predict with sufficient certainty what will happen in humans. In a review of animal studies published in seven leading scientific journals of high impact, about one-third of the studies translated at the level of human randomised trials, and one-tenth of the interventions, were subsequently approved for use in patients [1]. However, these were studies of high impact (median citation count, 889), and less frequently cited animal research probably has a lower likelihood of translation to the clinic. Depending on one's perspective, this attrition rate of 90% may be viewed as either a failure or as a success, but it serves to illustrate the magnitude of the difficulties in translation that beset even findings of high impact.

And summary from the same post:

Summary Points

The value of animal experiments for predicting the effectiveness of treatment strategies in clinical trials has remained controversial, mainly because of a recurrent failure of interventions apparently promising in animal models to translate to the clinic.

Translational failure may be explained in part by methodological flaws in animal studies, leading to systematic bias and thereby to inadequate data and incorrect conclusions about efficacy.

Failures also result because of critical disparities, usually disease specific, between the animal models and the clinical trials testing the treatment strategy. Systematic review and meta-analysis of animal studies may aid in the selection of the most promising treatment strategies for clinical trials.

Publication bias may account for one-third or more of the efficacy reported in systematic reviews of animal stroke studies, and probably also plays a substantial role in the experimental literature for other diseases.

We provide recommendations for the reporting of aspects of study quality in publications of comparisons of treatment strategies in animal models of disease.


  • $\begingroup$ Answers like this are pretty unhelpful. I guess part of that decision depends on if you think answers should be more than a block of text lifted lifted completely out of one source or not. (I think they should be more.) $\endgroup$ Commented Feb 28, 2015 at 20:10
  • $\begingroup$ And if I tell something, you will tell something else almost as if just for the sake of arguing! As far as you are concerned, I would be safer just quoting, so you can argue against the ones who wrote the paper and not the poor me! $\endgroup$
    – One Face
    Commented Mar 1, 2015 at 1:26

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .