A major government study to find out whether video display terminals cause miscarriages in female workers has been blocked on scientific grounds by the Office of Management and Budget and may never be completed, according to the official formerly in charge of the study.

The action by the budget agency, operating under its mandate to reduce government paperwork, came after BellSouth, one of the telephone companies whose workers were to be studied, hired scientists from Harvard and Brown universities to look for flaws in the project proposed by the National Institute of Occupational Safety and Health, according to Dr. Philip Landrigan, a former NIOSH official.

"In my opinion, it was a pretty overtly political move," said Landrigan, who testified in April before a House subcommittee. "The companies were just trying to get the government off their backs."

An OMB spokesman denies that rejection of NIOSH's proposed study was a result of interference by the corporations involved. The subcommittee on health and safety of the House Committee on Education and Labor will hold a hearing today on the issue.

Landrigan said OMB's denial of approval set the study back two years and may have jeopardized the project's feasibility because of a rapid changeover to VDTs of workers needed for the study's "control" group.

Since being announced in 1983, NIOSH's study has been eagerly awaited by unions, industry officials and occupational health experts in the hope that it would resolve questions about whether VDTs, which are used by an estimated 7 million American working women, are hazardous during pregnancy.

Since 1980, according to NIOSH officials, 12 reports of unusually high rates of miscarriages, birth defects or other pregnancy complications have been filed by offices equipped with VDTs. In two such clusters, the agency confirmed high miscarriage rates but could not show that they correlated with VDT use. In a third cluster, NIOSH did find a significant association between miscarriages and VDT use, but the numbers were too small to determine whether VDTs had caused the pregnancy losses.

The proposed research would compare pregnancies in two groups of female telephone workers -- long-distance operators, who currently do not use VDTs, and directory-assistance operators, who do. Landrigan said NIOSH chose these groups because they were similar in other respects, so that any difference in their miscarriage rates might be attributable to VDT exposure.

But the agency's original design for a study, monitoring pregnancies in progress, had to be abandoned when BellSouth and AT&T, the companies involved, announced plans to have long-distance operators switch to VDTs this year. The current proposal, recently revised and resubmitted to OMB, is for a retrospective study, in which long-distance operators would be questioned about pregnancies that occurred before they switched to VDTs.

"Clearly, if we don't move relatively quickly we could lose this group," said Dr. James Melius, who last year became director of the NIOSH division handling the study.

NIOSH proposed to administer questionnaires to about 1,500 telephone operators who used VDTs and 1,500 who did not. The plan was reviewed by scientists inside and outside the agency, Landrigan said. In December 1984, it was presented to the companies and unions involved.

"The companies came out and said they didn't really like it. They said they didn't see why it had to be done in their establishment and not some place else," Landrigan said.

Company representatives had pressed for a multi-industry study so that any adverse findings would not reflect badly on the telecommunications industry, according to David Le Grande, director of occupational safety and health for the Communications Workers of America. Le Grande said AT&T was willing to proceed, but BellSouth continued to oppose the project. "BellSouth has done everything they can to prevent the study from being done," he said.

Last fall, when NIOSH released its research proposal, BellSouth hired epidemiologists, Brian MacMahon of the Harvard School of Public Health and Sally Zierler of Brown University, to review it. In November, after NIOSH submitted the proposal to OMB, the company forwarded MacMahon and Zierler's report to OMB and NIOSH.

The report's language was scathing. "It is in our view inconceivable that the study would yield results that are definitive, unequivocal or credible and, even with the modifications that we will propose, such results cannot be assured . . . ," MacMahon and Zierler wrote.

Among their criticisms were that the number of women included was too small, that definitions of VDT exposure and complications of pregnancy were vague and that responses would be biased by awareness of health concerns about VDTs.

A spokeswoman for BellSouth denied that the company had tried to prevent the study from being done. She said BellSouth merely wanted to ensure that the results would be valid.

"When we originally got involved in the study, it was to be a multi-industry study," she said. "It's now narrowed down to a one-industry, one-company study, but we're still willing to do it . . . . And we are the only company in the country that does want to do this."

On Dec. 13, OMB notified the Department of Health and Human Services -- parent agency to NIOSH -- that the study had been rejected, citing criticisms of its statistical design similar to those of MacMahon and Zierler.

Melius said several of the epidemiologists' criticisms were valid, but that OMB had never before rejected a NIOSH study on scientific grounds. "We've had problems with OMB holding up studies before, but we haven't had this particular situation," he said. "It hasn't happened as openly."

He said NIOSH revised the proposal, taking into account the epidemiologists' criticisms. "Most of them were fairly readily responded to," he said. "There is some ongoing difference of opinion as to what is the best approach to this study."

Gary Bass, executive director of OMB Watch, said this was not the first time the budget agency had rejected a survey for scientific reasons but that the action was unusual. "In almost all cases that would not be the sole criterion," he said.

OMB spokesman Edwin Dale disagreed. "We're supposed to minimize the burden on the public," he said. "It's completely normal and standard for us to reject surveys on the basis of flawed statistical techniques . . . . The work by the Harvard guy, Brian MacMahon, was received, but it was not a major influence. We were already moving in the same direction . . . . "

Landrigan, now director of Environmental and Occupational Medicine at Mt. Sinai School of Medicine in New York, is planning a separate, large study on pregnancies in VDT operators. Melius said NIOSH hopes to move ahead with its project. But even if OMB approves the revised proposal, the delay may cost the government $300,000.

"Originally, two years ago, $100,000 to $150,000 was a realistic estimate" for the total cost, he said. Now, because of an increase in what contractors charge to conduct telephone surveys, the price tag would be "on the order of $400,000," he said.