|
We all need medication that is safe and effective. We need to know what the medication will do before we take it or give it to our children. The claim that a scientific modality, like animal testing, can be used to predict a future response is important, as society will then actually use that modality in decision-making. For example, if society believes that a drug will have the same effect in humans as it did in mice, because mice are predictive for humans, then certain decisions will be made based on how the drug affects mice. A pharmaceutical company, for example, might cancel the drug’s development thinking that it will not benefit humans.
The word ‘predict’ has a very specific meaning in science. When you say that you predicted the outcome of the football game, you are using predict in a nonscientific way. That is not to say you are using the word incorrectly, you are not. In science however, there is a difference between merely guessing correctly and predicting an outcome.
The way science uses the word prediction is different from the nonscientific use. Consider the following example: A fortuneteller says you will get married this year, have a baby next year, make $100,000 the next, discover a cure for cancer the next and retire the next year. Five years pass and in year three you did in fact make a $100,000. Did the fortuneteller predict this? In the nonscientific sense of their word, one could certainly say so. But in science the word prediction has a very specific meaning and the above does not qualify.
Another example: A university is composed of 200 faculties. This university also has a football team and hence the campus is very interested in football. At the beginning of the season, all 200 faculty wager on who will be the ultimate winner at the end of the season. Considering the finite number of serious contenders, it is highly probable one will pick the winner. Can we then say that the faculty always predicts the winner? No, for two reasons. If we are going to say the faculty per se picked the winner then the faculty would get one pick, not 200. To say otherwise is to commit the fallacy of composition, as we would be confusing the collective and the distributive forms of a word.
Second, again as in the fortuneteller case, prediction in the scientific sense does not mean merely a correlation of outcomes. If we are to say that the faculty is a good predictor, then we would need to see evidence that out of the last twenty or so football champions, the faculty as a group (getting only one pick) picked the winner a very large percent of the time. One right guess does not a prediction make; at least not in the scientific sense of the word.
Science is the best way humans have found for discovering truths about the material universe. In order for science to be conducted properly, the rules of science must be followed and those rules include what are called demarcation criteria. Demarcation criteria are how one can separate science from activities that are not science. Most agree that predictability is included in the demarcation criteria. If a discipline fails to meet the test of predictability, and yet it claims to be predictive, then in all likelihood it does not qualify to be included in the paradigm called science.
In the physical sciences, predictability is easy to define and test. Astronomers can predict with 100 percent accuracy where a planet will be next year or next century. Similarly, the second law of thermodynamics predicts a perpetual motion machine will never exist. The life sciences however are based more on probability and statistics, as we will see.
A modality, be it animal testing, astrology, or an x-ray of your chest cannot be said to be predictive simply because it exhibited a correct correlation once or even several times. If someone suggests that an animal, say a mouse, can predict human response to chemicals vis-à-vis carcinogenesis, he would need to provide data. The person making the claim should be able to provide statistical data, not anecdotes (stories). Statistical data include values like positive predictive value, sensitivity, specificity, and negative predictive value. If no such data are available or if the negative and positive predictive values are very low, then the predictive claim for the modality has been falsified.
So what does all this have to do with research and testing on animals?
Society accepts testing animals because it believes the results will predict what drugs will do in humans and how diseases will affect humans. But what if this belief were false?
Many scientific studies have been conducted that compared how drugs affected animals with how they affected humans. The scientists studied drugs that had been tested in animals and that then went to the marketplace for humans. The results are very informative. Put simply, animals cannot predict human response to drugs and disease. The research reveals that you get about as accurate results from animals, when using them to predict human response, as you would if you simply flipped a coin.
One study found that out of thirty-nine anticancer drugs tested on xenograft mice, only one mimicked the response in humans. Another study revealed that of twenty chemicals thought not to cause cancer because of studies in animals, nineteen actually did cause cancer in humans. A study published in 1990 showed that animals did not identify 67 percent of drug toxicities that occurred in humans. Many such studies exist and they all show the same thing; the negative and positive predictive values for animal testing are so low we must conclude that the modality itself is not predictive. To their credit, the pharmaceutical industry as a whole now acknowledges that animals cannot predict human response and they, along with others, are working on developing tests that will predict human response.
There are consequences to believing falsehoods; and trusting that animals can predict human response is no exception. Bad drugs have been allowed on the market while good drugs have been kept off. Furosemide, commonly called Lasix, is an example of an important medication almost lost to the public secondary to animal studies. It is a diuretic, used to treat high blood pressure and heart disease. Mice, rats and hamsters suffer liver damage from this widely used drug, but people do not. Isoniazid, an anti tuberculosis drug that has been used for decades, causes cancer in lab animals. Phenobarbital, a drug used for epilepsy would not be approved today because it causes cancer in mice and rats. The US National Cancer Institute believes society has lost cures for cancer because results from animal studies were believed. These examples can be easily multiplied.
You might have ethical issues with using animals in research and testing, or you might not. Regardless of your views on the ethics, you should be very skeptical when people tell you that animals are used in disease research and drug testing because they can predict what will happen to humans. 
Ray Greek has raised his voice against animal experimentation. He is President of Americans for Medical Advancement, Los Angeles and Europeans for Medical Advancement, London.
|
Show/hide comments
Show/hide comment form
