An undercover investigation into the system that regulates human experimentation in the United States has revealed flaws that expose it to 'unethical manipulation', the Government Accountability Office reported last week.

The federal inquiry was launched in January 2008 to probe the network of institutional review boards (IRBs) that oversee research using human patients, such as clinical trials. The boards are often run by universities and hospitals, but, with researchers clamouring for their proposals to be reviewed more quickly, a burgeoning industry of independent, for-profit IRBs has recently emerged.

In a hearing before the House Committee on Energy and Commerce on 26 March, government investigators reported that they had registered fictitious IRBs with the Office for Human Research Protections — including one called Phaké Medical Devices, supposedly based in 'Paynesville, South Carolina'.

At the hearing, protections office director Jerry Menikoff noted that the registration of IRBs is a simple listing process that does not involve background checks. That system was recommended following a previous inspection of the programme, he said.

Investigators also advertised a bogus IRB pledging "fast approvals guaranteed", and naming the fictitious board's president after a three-legged dog called Trooper. Six companies responded to the advert, and one attempted to hire the IRB.

In a separate arm of the inquiry, an intentionally vague research protocol was concocted and submitted to three real IRBs. The proposal called for a litre of a fake gel to be poured into the abdominal cavity of women to ease recovery after surgery. Two of the IRBs rejected the proposal outright, with one board member calling it "the riskiest thing I've ever seen", investigators reported. But one IRB approved the protocol unanimously.

"Our investigation showed the current system is highly vulnerable to unethical or incompetent actors," says Gregory Kutz, managing director of forensic audits and special investigations at the Government Accountability Office.

The report falls short of a comprehensive review of independent IRBs but is still valuable, says Trudo Lemmens, a bioethicist at the University of Toronto in Canada. "It's more or less anecdotal," he says, "but it confirms that there is a problem in how the system is constructed."