WEASEL WORDS: It's good news that the Semiconductor Industry Association is moving toward a study of the effects of its production technologies on the health of its workers, but the wording of the trade group's announcement left me wondering whether it's really ready to confront the problem -- or just stalling for time.

As I noted here last week, there's never been a comprehensive scientific study of the question, even though there's suggestive -- and scary -- evidence that some workers have suffered unusually high rates of cancer, birth defects and miscarriages.

Environmentalists and occupational-health experts -- not to mention current and former workers -- have long pressed the industry to authorize a broad epidemiological study. That would involve reviewing the health histories of a large sample of workers to determine which if any subgroups among them have actually experienced any of these misfortunes at significantly higher-than- average rates.

That, I'm told, is standard public-health practice when, as in this case, anecdotal evidence and more-limited studies suggest the possibility of a serious problem. But for a decade the SIA and its members, which include all major chip manufacturers, have blocked efforts to begin such a study.

In 1999, however, the SIA did agree to convene a scientific advisory committee (SAC) to determine whether a study of cancer among its workers was warranted. Last week the San Jose group put out a press release saying it "will implement the key recommendations" made by that body.

It wasn't until a few days later, though, that the SIA released an executive summary of the SAC's report. Comparing that document to the SIA's press release turns out to be an interesting exercise.

The former says straight out that "The SAC recommends that the SIA commission an epidemiologic cohort study of wafer fabrication workers."

Here, by contrast, is what the SIA press release says on that score: "The SIA will conduct a preliminary review to determine if it is possible to conduct and go forward with a meaningful retrospective epidemiological study." In other words, the industry didn't actually accept its advisory group's recommendation to commission a study; it agreed only to a preliminary review of the possibility.

Granted, the SAC itself opened the door to that kind of shilly-shallying: After five paragraphs about the questions the study should seek to answer and how it should be structured, and after noting that the necessary data appear to be available, it added one sentence suggesting that the study have two phases, "a scoping phase and, depending on determination of feasibility, a full epidemiologic study."

Seizing on that sentence, the SIA can claim that it has indeed accepted the scientists' advice.

But Dr. Joseph LaDou, director of the International Center for Occupational Medicine at the University of California School of Medicine in San Francisco, told me there's absolutely no doubt in the scientific world that an epidemiological study is feasible -- such analyses are old hat in other industries, and much of the medical evidence is already collected in resources such as the state-run California Cancer Registry.

So the question is why the SIA press release fudged the issue. Perhaps it was just a rhetorical concession to recalcitrant members of the group's board, and the industry is actually prepared to proceed not just with the "scoping phase" of a study, but with the real thing. In that case, I guess the weasel wording won't matter in the long run.

But if it was really an attempt to delay the investigation further, or to fool the industry's workers and the public by pretending to face the issue without actually doing so, then shame on everyone involved.