Rep. Robert J. Mrazek's defense {"The Least We Owe to Animals," op-ed, June 25} of his bill to ban federal funding for all biomedical research using animals acquired from pounds and shelters contains a number of factual errors that require correction. It also miscalculates the impact his prohibitive proposal would have on critical medical research.

Although Mrazek states that the majority of random-source animals would not be affected, his bill proposes to eliminate shelter animals (and the bill's definition of shelter includes pound animals) from medical research. Most dogs used in medical research are bought from licensed, registered dealers, who often purchase them from pounds or shelters. Contrary to Mrazek's belief, a 1980 survey by the National Academy of Sciences indicated that about three times as many dogs and cats are obtained for laboratory use from pounds and dealers as are purpose-bred animals. More recently, the National Institutes of Health estimated that of 201,931 dogs used in experimentation in the United States, only some 55,000 were purpose-bred. Thus, the Mrazek bill proposes to eliminate a primary, and the least costly, source of research animals.

Cost is a primary concern in biomedical research. It makes no sense, financially or scientifically, to use a purpose-bred dog, at $275 to $600 each, rather than a pound animal, at $5 to $55 each, when the research does not require it. Despite Mrazek's statement to the contrary, the Office of Technology Assessment reported to Congress that banning the sale of pound animals would lead to higher procurement costs. The total cost of all research involving dogs could rise as much as $70 million a year if pound or shelter animals were no longer available. With no corresponding increase in research funds -- and the Mrazek bill makes no provision for an increase -- the sad result would be a serious curtailment or even elimination of research for which dogs are essential. The ultimate consequence would be fewer new treatments, cures and preventions for the diseases that now afflict countless Americans.

It is untrue that all shelter animals used for research are people's pets; most are stray or abandoned animals for which no adoptive homes can be found. As a result, 10 million to 15 million of such unwanted dogs and cats are killed in pounds each year at a cost to taxpayers of more than $500 million. In 1984, medical research needed about 200,000 of such animals, or less than 2 percent of those killed. Prohibiting use of pound or shelter animals will not save the lives of these animals; it will merely require that these animals be killed while we breed additional ones for research use at additional cost to taxpayers.

To suggest that the Mrazek bill will not adversely affect medical research is misleading. Similar legislation in Massachusetts has already had a dampening -- indeed, devastating -- effect on certain types of cardiovascular research in that state. Denied access to dogs from pounds, scientists have been forced to curtail research in order to pay for expensive purpose-bred animals or to abandon promising lines of inquiry for the relief of human suffering.

The NIH obtains some dogs and all cats used in its own research from commercial dealers, and neither specifies nor proscribes the sources of animals in intramural or extramural programs. Mrazek's statement that NIH no longer uses shelter animals for its in-house research is at odds with this NIH policy statement: "Decisions regarding the kinds and sources of animals needed are made by investigators in the context of the requirements of the research to be conducted, subject to NIH and Public Health Service policies." Thus, the fact that the NIH does not obtain animals directly from pounds or shelters does not mean that they may not use pound or shelter animals.

Will research use of a very small percentage of the millions of pound or shelter animals that would otherwise be killed undermine public confidence in animal shelters, as Mrazek speculates? Not according to past experience. Scientists have depended on such animals for generations, yet there is no valid evidence that this practice has deterred people from leaving unwanted animals in pounds and shelters. Some studies, in fact, show that people who surrender such animals have no objection to their use for research. Curiously, the willingness to use purpose-bred animals suggests that these animals have fewer "rights" not to be used in experimentation than animals abandoned by their owners.

Mrazek insists that the intent of his bill is not to prevent or slow medical research, but that will inevitably be its result. He characterizes H. R. 778 as a "compromise," an attempt to appease animal-rights groups, some of which seek to end animal research altogether. Unfortunately, it is the future of biomedical research -- and ultimately human health -- in this country that his bill will compromise. -- Michael E. DeBakey The writer is chancellor of Baylor College of Medicine, where he is chairman of the department of surgery and director of The DeBakey Heart Center.