Food-safety oversight in the United States has been in disarray for many years. Responsibility for it is split, on historical grounds, between 15 different agencies in the federal government, operating under at least 30 different statutes. It is past time for Congress to legislate to modernize the entire system.

Late last month, the Government Accountability Office (GAO) added food safety to its list of 'high-risk' federal policies and programmes most in need of reform. The non-partisan GAO is recommending that Congress ask the National Academies to examine new ways of organizing the federal food-safety system. As part of that project, the academies would certainly examine the state of federally funded food-safety research.

“The current fragmented federal system has caused inconsistent oversight, ineffective coordination, and inefficient use of resources,” the GAO said in a statement accompanying its updated list, adding that Congress should “consider a fundamental re-examination of the system”.

The alarm sounded by the GAO is likely to resonate with the public, after a three-month stretch last autumn in which outbreaks of food-borne disease killed three people and made more than 500 others sick. The culprits ranged from spinach contaminated with the bacterium Escherichia coli, to salmonella-bearing tomatoes, to lettuce that probably infected scores of people with E. coli after they ate at Taco Bell and Taco John's restaurants.

These well-publicized incidents stand out against a far larger, latent problem. According to the Centers for Disease Control and Prevention in Atlanta, 76 million Americans contract food-borne illnesses each year, and 5,000 of them die.

A science-based system is needed. What has evolved instead is an irrational and expensive arrangement.

A science-backed regulatory system is needed to address an issue on this scale. What has evolved instead over the past century is an irrational and expensive arrangement, whereby officials examine every carcass at every slaughterhouse in the United States every day, but a major food processing plant may escape inspection for a decade.

At the same time, the government research programmes supporting food-safety regulation are neither comprehensive nor coordinated. Research is scattered between often-obscure subdivisions of several departments, from the Environmental Protection Agency to the US Department of Agriculture (USDA). The result is a patchwork of research that, according to a 1998 Institute of Medicine report that reads uncannily like the GAO's latest assessment, “raises serious concerns about duplication of effort and about the linkage of science to attempt to solve food safety problems of the highest priority”.

For historical reasons, the USDA is responsible for the safety of meat, poultry and processed egg products, whereas most of the rest — 80% of the food supply — rests with the Food and Drug Administration (FDA).

But the FDA's Center for Food Safety and Applied Nutrition in College Park, Maryland, which bears the primary responsibility for the research that supports food safety at the agency, has seen its budget and staff cut over the past decade. This year, its operating budget will fall to $25 million — a little over half of what it was in 2003. The centre has far less funding for its own science, and next to none for extramural research in areas it needs to learn more about, from microbial ecology to detection methodologies for pathogens in food.

The recent outbreaks caused by bacteria in fresh fruit and vegetables illustrate the paucity of research. Nine years ago, after a string of comparable outbreaks, the FDA issued a set of general recommendations on managing manure, irrigation water and farmworker hygiene to minimize contamination of fruit and vegetables. But it has not done the research to get the data needed to convert these general guidelines into firm, quantified regulations to be implemented on farms, and during food processing and transportation.

In the meantime, Democrats in Congress have repeatedly introduced legislation to establish a single US food-safety agency. It is tempting to believe that this approach would produce the necessary coherence in food safety. But past experience of amalgamating parts of the US federal government, from the Department of Energy to the Department of Homeland Security, does not give cause for optimism that such a consolidation would be either efficient or effective.

The best approach may instead involve three more modest steps: an inter-agency panel to properly coordinate food safety; a comprehensive revision of the antiquated and fragmented legislation now governing it, to better reflect today's risks and today's science; and a properly supported, coordinated research programme to inform food safety policy and practice.