I hope you never have this experience: A loved one is hospitalized. Her doctors tell you her infection is resistant to antibiotics. She dies. More than 60,000 American families go through that experience each year -- and the number is almost certain to rise.

Multidrug-resistant organisms are showing up in top-flight hospitals -- like the klebsiella found this year in the National Institutes of Health's Clinical Center that may have led to the deaths of seven patients. Even infections that used to be a breeze to treat, like gonorrhea, are becoming incurable.

In much of the world, of course, bacterial disease is a routine cause of tragedy. Tuberculosis alone kills 1.4 million people a year. One reason for this staggeringly high figure is that most people in the world are too poor to pay for most medicines. But another reason is that some strains of tuberculosis bacteria have become resistant to most of the drugs we have. Even after two years of toxic treatment, drug-resistant tuberculosis has a fatality rate of about 50 percent.

What makes the rapid loss of antibiotics to drug resistance particularly alarming is that we are failing to make new ones. We are emptying our medicine chest of the most important class of medicines we ever had. And the cause can be traced, for the most part, to two profound problems.

The first is economic. Historically, the drug industry thrived on antibiotics. But if an antibiotic is useful against only one type of bacterium, relatively few people need it during its patent life. And if an antibiotic is "broad spectrum," meaning it works on many different types of bacteria, wider use shortens its commercial life because it quickens the pace at which bacteria develop resistance. Moreover, antibiotics are designed to cure an acute disease -- not to palliate a chronic one -- so people need them only for a limited time. Compared with drugs that are used for years to treat widespread conditions like high cholesterol or asthma, antibiotics pale as a corporate investment.

The second challenge stems from the nature of bacteria. Though brainless, they are brainy, enjoying a highly effective collective intelligence. Large numbers of independently mutating bacteria test adaptations to group problems, like how to survive antibiotics. What works -- like modifying the bacterial proteins to which antibiotics would otherwise bind -- wins. As bacteria become more adept at evading antibiotics, it has become much harder to find drugs that can beat them back.

Merge these two problems -- scientific and economic -- and the result is a drug-development disaster: The prospects are so discouraging that few companies bother to try anymore.

How can we confront the critical shortage of new antibiotics when both the scientific approach and the economic model are letting us down? We can change both paradigms.

Drugmakers survive by selling what people or governments buy in amounts and at prices that maximize profit. Monopoly protects the ability to set price for profit. Patents allow monopoly. Secrecy protects intellectual property until it is patented.

But what if we take a page out of the pathogen playbook? Many pathogens exchange DNA, sharing what they learn. Drugmakers can operate in the same way: They can do science "open lab"-style, working in teams with academic and government scientists and other drug companies to share what they learn and to bring fresh scientific ideas and technological tools to bear. Relaxing the traditional insistence on secrecy allows collaboration, and with it, innovation.

Did I hear you say, "It'll never happen"? It already has. GlaxoSmithKline opened its campus at Tres Cantos, Spain, to outside academic, government and biotech scientists in order to collaborate on finding antibiotics for neglected infectious diseases. The independent Tres Cantos Open Lab Foundation selects the projects and helps cover visiting researchers' expenses.

In another version of the open-lab concept, the Bill and Melinda Gates Foundation organized a TB Drug Accelerator program that brings together research teams from seven major companies (Abbott Laboratories, AstraZeneca, Bayer, Eli Lilly, GlaxoSmithKline, Merck and Sanofi) with scientists from four academic and government institutions. The companies have exchanged more than a thousand compounds and provided the academic and government scientists with access to millions.

These experiments show that even competing research teams can share knowledge, risk and reward in anti-infective drug development, test diverse approaches and avoid redundant efforts. (I'm involved in both of these projects.)

Philanthropic efforts have financed these open labs, but they can't substitute for market forces. Nor can the current economic model give antibiotic development a permanent, prominent place in drug company portfolios.

There are, however, other ways for drugmakers to profit beyond using monopoly to protect prices. As Thomas Pogge of Yale and Aidan Hollis of the University of Calgary have pointed out, an intergovernmental fund for drug discovery could reward drugmakers for products in proportion to their impact in reducing the loss of healthy years of life.

The lower the cost of a lifesaving drug, the greater the number of people who could use it; the more lives protected, then, the greater the monetary reward. An investment of $20 billion a year could encourage more open-lab collaborations to find new medicines in challenging settings like antibiotic discovery and make them accessible to all who need them.

If we don't make new antibiotics, we will lose the ability to practice modern medicine. A new collaborative model for drug discovery can help make sure this doesn't happen.

* * *

Carl F. Nathan is chairman of the department of microbiology and immunology at Weill Cornell Medical College. He wrote this article for the New York Times.