SAY "WORKPLACE injury" and many people still think of one-time accidents. But nearly a third of the most severe workplace injuries -- the ones resulting in days lost from work -- are now thought to result from motions performed over and over again, leading to tendinitis, carpal tunnel syndrome, back trouble and other ailments. How to protect workers from this large but ill-defined class of dangers? The Occupational Safety and Health Administration, after years of trying, is set to issue a regulation on so-called "ergonomic" injuries. The House of Representatives, in a familiar countermove, this week passed a bill to block the new rule for about 18 months, until results come in from a freshly commissioned National Academy of Sciences study.

The bill to block passed the House by only eight votes, and the president has promised a veto even if the Senate follows suit, so the rule's chances of going forward are high. But while the would-be blockers make some complaints that ring hollow -- such as the notion that there is "no good science" on the fairly well-studied repetitive stress syndromes -- some troubling vagueness does lurk in at least the draft version of the OSHA rule. It doesn't help that the regulation defines these injuries sweepingly as any that result from "a mismatch between the physical demands of the task and the physical capacity of the worker." Though designed to be flexible, the regulation would cover any workplace where a single ergonomic injury had been reported; employers would then have to file a plan for training, treatment and prevention in all similar jobs -- though not much is known about the specific factors that affect one worker but not another with repetitive stress injuries on the same job.

Employers' groups have pushed for more use of "one-time" fixes and for clearer guidance on what kinds of programs OSHA inspectors would judge acceptable once the regulation was triggered. They also say most employers already are moving to address ergonomic problems because of the costs involved (one study showed more days off for a repetitive stress injury on average than for an amputation). OSHA says this progress is evidence that a wider rule will burden only those companies that are recalcitrant.

It's plain that repetitive stress injuries are a real problem for some large number of workers; speeded-up assembly lines and increased computer use make the problem worse. Those workers need the protection an OSHA regulation would bring. But how far does an employer's responsibility extend in cases such as these? Must every "mismatch" between physical capacity and the demands of a job be ironed out? What are the boundaries if the answer is no? The agency has a duty to be as clear about the limits of this rule as about its breadth. The current language is provisional only. To keep the final rule from becoming more of a battleground than it should, it ought to be carefully tightened up.