RICHMOND, NOV. 6 -- Two physicians whose advice underpins A.H. Robins Co.'s proposal to compensate Dalkon Shield victims praised the intrauterine contraceptive device today, testifying that it is as safe and effective as other IUDs.

Dr. Brian L. Strom of the University of Pennsylvania School of Medicine and Dr. Herbert F. Sandmire, an obstetrician-gynecologist in Green Bay, Wis., testified at a court hearing intended to resolve disputes over the number of people filing claims against the IUD that should be compensated, and over the size of the total payout.

Strom, an assistant professor whose specialties include epidemiology, was asked on cross examination whether he would prescribe the shield were it to be made tomorrow.

"Sure," he said, adding that his guidelines would be the same as for other IUDs. He based his testimony on a review of medical literature.

Sandmire, who began inserting IUDs in 1964, and who kept registers in which he recorded how his patients fared, said his studies -- and the literature -- show that sexually transmitted diseases are to blame for nearly all cases of pelvic inflammatory disease (PID) attributed to the shield. PID can be life-threatening and can cause infertility.

Much of the physicians' testimony about the safety and efficacy of the shield clashed with evidence that has been long accepted by the Food and Drug Administration, many respected scientists and numerous juries that awarded large sums to plaintiffs in shield product-liability cases.

For purposes of deciding what constitutes "full" payment to victims, U.S. District Judge Robert R. Merhige Jr. has said he assumes the shield caused the injuries.

At times, the two physicians disagreed with each other. For example, Sandmire testified that with a "very rare" exception, the shield did not cause PID, while Strom said it did, being no "better or worse than other IUDs." Sandmire said the shield "did not" cause tubal prenancies; Strom said it did.

Strom also said that in evaluating Sandmire's studies, he had to allow for the doctor's loss of his registers for three years of shield insertions.

Robins, which filed for bankruptcy protection in August 1985 proposes to pay up to $1.15 billion to claimants, including averages of $33,318 each to 30,000 victims with "potentially valid compensable claims," $300 each to 112,005 women with "minimal support" for their claims and $2,200 each to 33,234 women who have "limited" support for their claims.

The day's final witness testified for the committee of outside stockholders, which estimates a total payment of $1.02 billion to 50,000 victims. The proposal rests on an analysis drawn from a common data base by the witness, Dr. Dennis Hack, president of Statiscal Consultants Inc. in Lexington, Ky.

Under cost-examination by Marc Ellenberg, a lawyer for the Dalkon Shield Claimants' Committee, Hack testified that he had diluted the value of the claims of 98,051 women because full medical records for them were unavailable.

The dilution would apply to women who had been seriously injured, had medical records showing that an IUD -- of some kind -- had been removed from their bodies, and had sworn at the end of a 50-page questionnaire that they had used shields.

Sandmire's testimony about the shield being erroneously blamed for sexually transmitted diseases depended in part on research he attributed to Dr. Nancy Lee, a scientist at the Centers For Disease Control in Atlanta.

In 1983, Lee concluded from major studies that while all IUDs increased the risks of PID to some extent, the shield increased the risks several times more than rival devices.

Today, Sandmire testified that Lee has adjusted her data to distinguish shield users in stable, monogamous relationships from users who had multiple sex partners. He testified that he found the incidents of PID in the former group to be no higher than in women who used no contraceptives or used other IUDs. Her new work was not introduced into evidence, and reporters were unable to obtain a copy from lawyers in the courtroom.

Cross examination brought out some apparent inconsistencies in Sandmire's testimony. For example, he said that 98 percent of so-called "silent" PID, which can go undetected for long periods, was attributable not to IUD use, but to certain infections, called chlamydia, that can be sexually transmitted. A few minutes later, he put the rate at "80 percent to 90 percent." Later, he told the judge the range was 60 percent to 80 percent.