Sweeping new rules designed to curb conflicts of interest at the US National Institutes of Health (NIH) have been met with both approval and anger.

The rules, introduced by director Elias Zerhouni on 1 February, limit employees' relationships with outside organizations. They were established in response to a series of reports in the Los Angeles Times beginning in late 2003 and a subsequent congressional hearing. These showed that numerous NIH scientists were paid consultants to drug companies that also had financial stakes in agency recommendations on research.

The new policy bars NIH employees from paid and unpaid consulting work with drug companies, universities, trade groups, health care providers or any other organization that interacts with the agency. In addition, top scientists cannot own stock in companies that do business with the NIH, and all other employees are limited to investments of $15,000.

NIH staff complained that the policy unjustly punishes all 18,000 workers for the actions of a few. Zerhouni initially favored less stringent restrictions but changed his mind as new conflict of interest cases emerged and he came under increasing pressure from Congress.

Employees were particularly annoyed at having to sell stocks at a time when many pharmaceutical share prices are falling. Many staff criticized the new rules during an in-house meeting on 2 February. Several contacted the American Civil Liberties Union to inquire whether the rules are legal.

The Federation of American Societies for Experimental Biology (FASEB), a Washington, DC–based group representing over 65,000 biomedical scientists, expressed concern that the restrictions might prevent NIH scientists from joining or working with professional societies. The group is hosting a conference in June to discuss the relationship between researchers and industry.

Opponents of the rules say they could also hinder NIH efforts to work with industry in order to translate discoveries into marketable drugs or medical devices. But Mark Rohrbaugh, director of the NIH Office of Technology Transfer, says the regulations will not bar relationships between drug companies and scientists that take place on NIH time, rather than as outside activities.

Zerhouni indicated that the consulting ban is here to stay but promised that the agency will review and possibly revise parts of the rules after a one-year trial period. Interested parties can also submit comments during a 60-day period.

Despite the complaints, some outside the agency said the stringent measures are necessary to remove any appearance of conflicts of interest. Sheldon Krimsky, who has studied research ethics at Tufts University in Medford, Massachusetts, says the rules should be expanded to cover scientists serving on NIH advisory committees as well as those at other academic institutions. “There is a whole lot more that has to be done,” he says.

The NIH made another long-awaited announcement in February about open access to research results. The agency urged NIH-funded researchers to put copies of their papers in a planned online agency archive within 12 months of publication.