Tuberculosis is truly an ancient sickness. Researchers have found signs of its origin in soil 15,000 years ago. Evidence of tuberculosis has been detected in the skeletal remains of Egyptian mummies from 4,400 years ago. The Greeks called this disease that slowly ate away at the lungs phthisis, for wasting away, because its pale sufferers progressively lost weight, appearing to wilt. Persistent low fevers and coughs that frequently brought up blood were other hallmarks of the disease, which mostly attacks the lungs. Over time, it became known as consumption and the white plague.

As Europeans started living in tighter groupings in cities, particularly during the Industrial Revolution, the disease became more menacing, rising to epidemic status. And it then became forever linked with the conditions of poverty that foster its spread-overcrowded housing, poor sanitation, and malnutrition. The white plague caused one-seventh of all Europeans' deaths in the nineteenth century.

For a time, the disease came to be regarded as an affliction of romantic geniuses, and it took many brilliant artists into its clutches-Chopin, Keats, Chekhov, Poe, Shelley, Charlotte and Emily Bronte.

Speculation abounded for years about what caused this sickness-was it the soil, bad air or climate, a tumor, heredity? Then, in 1882, a German country doctor, Robert Koch, identified the tuberculin bacillus, Mycobacterium tuberculosis, that causes the disease. This discovery led to the understanding that tuberculosis is contagious-caught through prolonged exposure to droplets in the air when someone with tuberculosis coughs, sneezes, laughs, or even talks-and set off intense scientific research to find a cure.

At the same time, a number of new treatment practices were being developed, with some success: collapsing a lung or a diseased cavity, removing ribs to deflate a lung, and cutting out damaged parts of a lung. By the twentieth century, many tuberculosis patients retreated to sanatoria in Europe and the United States, where they were quarantined for up to two years. The "cure" consisted of fresh, often cold, air, healthy food, and rest. Sanatoria slowed the spread of tuberculosis mostly by keeping contagious people away from the healthy. Although many patients recovered, it's not clear whether sanatoria improved long-term patient survival.

Major medical breakthroughs beginning in the 1940s heralded a new era. One drug after the next was discovered that could destroy the bacteria. Now people were truly restored to health, no longer having to give up jobs and family lives to reside at the sanatoria.

Effective drug treatment meant that tuberculosis's days as a virtual death sentence were over. In 1907, tuberculosis was the country's reigning killer disease, taking 156,000 lives, but 40 years later this scourge ranked seventh, with a death toll of 60,000. More remarkable progress would come: From 1953 to 1984, the number of active tuberculosis cases declined 5 to 6 percent a year, down from 84,304 to 22,201-a 74 percent drop.²

As people became confident that tuberculosis was being eradicated, fear of catching it faded away. With the threat seemingly over, efforts to stay on top of the disease ended. Government funding for programs that monitor, control, and treat tuberculosis were reduced to nothing in 1972. In its place, states received block grant money to use as they chose. Treatment shifted from hospitals, which no longer maintained isolation rooms to control contagion, to outpatient clinics. And by 1989 the U.S. Department of Health and Human Services' Advisory Council for the Elimination of Tuberculosis optimistically called for tuberculosis to be eliminated in the United States-less than one case per 1 million people-by the year 2010.³

But the optimism proved to be ill-founded, and tuberculosis staged an awful comeback. It began showing up in increasing numbers of people in the late 1980s and early 1990s-and this time in a more virulent form. From 1985 to 1992, the trend was unmistakable: the annual number of cases grew from 22,201 to 26,673-a 20 percent climb.⁴ Tuberculosis was no longer receding into the past or solely the problem of developing countries with poor infection control. In New York, cases skyrocketed, and outbreaks in prisons and hospitals evoked age-old fears.

The biggest trigger for this tuberculosis resurgence was the human immune deficiency virus that causes AIDS. Ordinarily, about 90 percent of people who are infected with tuberculosis will show no signs, and the disease will be latent and noncontagious in them for the rest of their lives. But 10 percent of those who are infected will become ill with an active form of the disease, half of them within a year and the others usually when they are older and their immune system is weaker. With HIV, the immune system is so compromised that tuberculosis infection proceeds to active disease quickly, often within two months. And, in another deadly twist, tuberculosis speeds up HIV's disease progression, leading to death more rapidly. In fact, it is the foremost cause of death for people with HIV.⁵

This new epidemic had another ominous characteristic-drug resistance. Ordinarily, tuberculosis patients were treated for six to twelve months with four medications, because resistance to one drug wasn't uncommon. But now patients were showing up with bacteria that didn't respond to two or more of the most effective, least costly drugs that also cause the fewest side effects-usually to isoniazid and rifampin, the workhorses of the tuberculosis medicine cabinet.

This necessitated treatment with second-line drugs, which would take two years or longer, cost more, create other health problems, and might not cure the patient. Caring for someone with multiple-drug resistant tuberculosis can be 100 times as costly as for someone with a regular strain of the disease, reaching $250,000 for the medications and likely hospitalization; treatment for regular tuberculosis ranges from several hundred to a few thousand dollars.

And someone with tuberculosis of the multiple-drug resistant sort could spread this lethal form of the disease to others.

Ironically, the resistance develops from the nature of tuberculosis treatment. Patients with regular tuberculosis must swallow from four to 12 pills a day-a task people might give up on once they started feeling better in a few weeks. That's what was happening. But failing to complete the medication can lead to a relapse and, even worse, can allow the tuberculosis bacteria to mutate into a form that no longer responds to the medications that are taken.

The disease was preying mostly on poor and vulnerable people living at society's fringes-the homeless, people living with AIDS, migrant workers, alcohol and drug abusers, the incarcerated, and immigrants-people who are more likely to stop taking the pills prematurely. These individuals tended to be unattached to any medical system and lacked the social support to help them stick to a long course of therapy.

This combination of factors was not just a peril in the United States. Worldwide, the same trends were unfolding and in much greater numbers in developing areas such as sub-Saharan Africa and Asia, which lacked adequate health care systems, and in the jails of Russia, where the health infrastructure was in tatters. A 1991 article in the British medical journal The Lancet alerted readers that the intertwining of AIDS and tuberculosis amounted to one of the greatest public health disasters since the bubonic plague.⁶ And in 1993 the World Health Organization declared tuberculosis a world health emergency.

As this wave of fear swelled, The Robert Wood Johnson Foundation decided to address the tuberculosis crisis. The Foundation, which at the time didn't direct grant money to specific diseases, made an exception for tuberculosis because of the threat that multiple-drug resistance posed. In 1993, the Foundation funded a three-year $6.65 million program called Old Disease, New Challenge: Tuberculosis in the 1990s designed to test innovative models for reducing barriers to health care among the people most at risk for tuberculosis. "This was a little toe in the water, to see what worked," said Nancy Kaufman, a vice-president of The Robert Wood Johnson Foundation. "This was a time when a crisis was looming, and it was important to be concerned. It was an opportunity to elevate, which we do with our name and money, the seriousness of this issue, and to be a stimulus."

Five demonstration projects around the country were chosen. The sites were Atlanta, Baltimore, New York City, North Florida, and San Diego.

The projects-directed toward the hard-to-reach, the disaffected urban poor, migrant workers, and people who cross the border with Mexico-were to incorporate collaborations among local government, medical schools and hospitals, and community-based organizations. They would screen for latent and active disease. Most would use directly observed therapy, a method of insuring that patients complete their medications by having someone observe them swallow their pills each day. This approach was just beginning to gain acceptance at the time, and today is standard practice. The use of culturally sensitive outreach workers would be crucial to the success of the program. The funding also included smaller grants for related tuberculosis projects such as forecasting how the epidemic was likely to progress and examining the ethical issues that tuberculosis raised.

In an unusual approach, anthropologists were hired to do a "qualitative evaluation." They were to "live" with the projects in the field and, by using extensive participant observation and interviews, get a close-up view of the workings of each project. They were to feed insights back to the project so changes could be made. As it turned out, their reports gave textured accounts of the various players and how they carried out their goals.

To hear Danita Evans tell it, Bellevue Hospital's fight against tuberculosis was filled with glory days. A number of social service agencies that serve the poor would skin-test their clients for tuberculosis, and Evans and several other community liaison workers would whisk those who came up positive to Bellevue by subway or taxi. There she slashed through the red tape that ordinarily overwhelms the dispossessed people who were at the heart of the tuberculosis epidemic in the 1990s.

Patients no longer had to slog through the registration and appointment processes, which in the past had sometimes meant waiting a month to see a doctor. Evans got them a clinic card, a doctor's appointment, and the services they needed-all in one day. She shepherded them through getting an X-ray, a blood test, a sputum smear, and a doctor's evaluation. If left alone, she feared, these alienated, out-of-their-element patients, many of them substance abusers, homeless, mentally ill, or living with HIV, would find the hospital system too daunting and leave.

Evans worked with a passion stoked by the grim reality that New York City was the epicenter for this comeback epidemic in the United States, and that these people at society's margins were its favored prey. "It was like a rush for me," she says. "I'd get in the taxi, have three patients in back with me, many who hadn't been in one for a long time. It was exciting because I'd get the patients to Bellevue quickly. And I found that once I got hold of patients, they were mine. They'd come back for treatment."

Community liaison workers like Evans were at the heart of The Robert Wood Johnson Foundation project at Bellevue Hospital. The Shared Responsibility Program for Tuberculosis Casefinding and Treatment was conceived to try a new approach to getting the disaffected poor people most in danger of catching tuberculosis to trust an institution they had stayed away from in droves.

Bellevue, the oldest existing public hospital in the country, has a mission to serve the poor, and started the nation's first tuberculosis clinic in a public hospital 100 years ago. By the 1990s, this 1,242-bed academic facility, affiliated with New York University, was thought of as too big, imposing, and unreceptive to the very people who most needed it. Many people couldn't cope with navigating this rambling building with rules and regulations that seemed designed to keep them out.

At this time, active tuberculosis cases had climbed from the city's all-time-low of 1,307 in 1978 to 3,811 in 1992-a case rate that translates to 50.4 per 100,000 people. One-quarter of them had the drug-resistant forms of the disease. City residents were alarmed and frightened as they learned of tuberculosis outbreaks in group-living settings like prisons and hospitals and among health care workers. "In places like Harlem, the case rate was 250 per 100,000 population, which was incredibly high," said Neil Schluger, a co-principal investigator for the project who headed Bellevue's Chest Service.

With its long commitment to caring for tuberculosis patients, Bellevue was again a major player. In 1992, the hospital carried the largest active tuberculosis caseload in the state, with 511 newly diagnosed patients. Moreover, many of the patients didn't respond to more than one of the most effective drugs, and the hospital had the highest caseload of patients with both HIV and tuberculosis in the country.

To address the public health crisis and attract hard-to-reach patients, Bellevue proposed a collaboration with five well-established community-based organizations on the Lower East Side: the Lower East Side Services Center, the Educational Alliance, the BRC Human Services Corporation, Housing Works, and the Community Health Project. These agencies met many of the needs of the people at risk, and could connect them to Bellevue's tuberculosis program. Their wide range of services included sheltering the homeless, providing food, and treating substance abusers and people with HIV/AIDS.

"Our clients are a population that's well advertised but not well served," said Ronald Williams, a program director at BRC, which provides residential and outpatient treatment for clients with substance abuse problems, many of them homeless and mentally ill, and referrals to other services. "People at Bellevue had automatic attitudes about our clients, because they could tell they were using. Our clients didn't fit the nice-client mode and got pushed away. But when Danita Evans walked the same patient through, he became a real person, someone they'd actually look at and deal with."

The idea was that the community-based organizations would give skin tests and questionnaires assessing tuberculosis risk to all of their clients; those with positive tests or suspicious symptoms such as a persistent cough would go to Bellevue for chest X-rays, further testing, and evaluations. Those with active disease would be put on directly observed therapy, the supervised taking of pills. Preventive medication would be given to some with latent disease, but it would not be observed. The community liaison workers would be the bridge between Bellevue and the organizations, making sure that patients got appointments and kept them, were fully assessed, and then followed through with their treatment if one was given.

For the most part, this project clicked.

The project was led by hospital administrators who believed in it and took pains to bring it to life. Instead of telling the service agencies how the project would work, the hospital administrators invited them to help design the program. David Cohen, then Bellevue's medical director, provided strong leadership.

"We tried as much as we could to have these community-based organizations be equal to us in their input into the program," Schluger said. "We felt we had a lot to learn from them. These clients have all kinds of medical and social needs, and they're all equally important. You can't do much about one if there's a problem with the other."

The agencies let Bellevue know that many patients would feel insecure in a large institution that was new to them and, if faced with too many roadblocks on the way to the chest clinic, would simply give up. The project set out to remove barriers for these clients. The project's manager, Naomi Wolinsky, bent many hospital rules to blaze a new and truly easy way for these tuberculosis patients to get served expeditiously. She arranged for two chest clinic appointments to be set aside each day for the project's patients, insuring that they would almost always be seen.

Because of Wolinsky's efforts, the community liaison workers were able to get patients a chest clinic card quickly and escort them through the security guards and registration offices that would have discouraged them in the past. The workers were with the clients every step of the way, even giving them lunch and showing them where the restrooms were. They also provided tuberculosis education to the agencies and trained them to administer the purified protein derivative, or PPD, skin tests.

Those who were treated for tuberculosis were given an incentive to keep returning for the long course of directly observed therapy. After completing the first month, they would get 10 subway tokens each week that they kept their appointments.

The hospital allowed each agency to define how its relationship would work. In some cases, Bellevue gave funding directly to an agency. The agency's own staff member would function as the liaison worker, administering skin tests and escorting clients through the Bellevue maze. This occurred, for example, at the Educational Alliance, an agency with services for children, people with AIDS, substance abusers, and elderly people.

"You don't know how great it felt to do tuberculosis right," said Peter Cordero, the Educational Alliance's tuberculosis/AIDS coordinator who was also its community liaison worker. "If a patient was positive, we would take him to Bellevue for an X-ray and run interference for him. People feel very uncomfortable going to that place the first time." The Educational Alliance also provided tuberculosis education and screening to other local community organizations, particularly those serving Southeast Asian immigrants.

This project got some noteworthy results. Altogether, 3,828 people were evaluated, and 20 cases of active disease were found-representing a high tuberculosis rate of 522 cases per 100,000 population. Eight patients were diagnosed in each of the first two years of 1994 and 1995, four in 1996, and none in 1997, reflecting the nationwide decline in the epidemic.

Thirty-three percent of those who returned to have their skin tests read were found to be latently infected-1,163 people. From this group, 466 were further evaluated at Bellevue, and 55 were put on preventive therapy of isoniazid for six months; just 20 people completed it. Doctors viewed most of the clients as bad candidates for preventive medication because homeless people and drug addicts were unlikely to take pills for so long without directly observed therapy, which was not available to patients without active tuberculosis.

Schluger concluded that large-scale screening was worthwhile for finding active disease in a high-risk group such as this one and that treating those with tuberculosis through directly observed therapy had proven to be cost effective. At the same time, he determined that screening for latent tuberculosis as a way to prevent future cases doesn't make sense without treatment that includes directly observed therapy to insure that the drugs are taken to completion.⁷ "We found that if all you were going to do is find latent tuberculosis without treating it, you have to think long and hard about whether this is how you should spend your dollars for tuberculosis control," Schluger said in an interview. "Are you doing everything else well before you focus on the latent cases?"

Nearly everyone involved applauded this project for creating a model that linked a big-city academic hospital with community organizations serving very difficult-to-reach populations and affording access to people with tuberculosis who might have gone untreated otherwise. In 1995, the project earned an honorable mention from the U.S. Department of Health and Human Services' Models That Work competition.

"It was an eye-opening experience," said William Rom, the Chest Service's current chief. "They got to street-level people we didn't know then. We got referrals and saw patients we wouldn't have seen otherwise."

The grant had some important side benefits as well: clients often complained of other health problems while at Bellevue and received treatment. Bellevue set up primary care clinics at some of the community organizations-such as pediatric and adult primary care clinics at the Educational Alliance. This was a plus both to the agency and to the hospital, which was looking to expand its revenue sources.

When Shared Responsibility drew to a close in 1997, the community-based organizations pushed for the services to continue, and administrators tried hard to find replacement funding. The New York City Department of Health agreed to fund two community liaison workers, including Evans, primarily to help more stable low-income people with latent infection get through directly observed preventive therapy, and to provide tuberculosis education.

But the scope of the liaison workers' job is much narrower today, and the chest clinic no longer reserves time slots for the agencies' clients. The community-based organizations lament that their agencies don't have a dedicated liaison like Evans and that they can no longer sail smoothly through Bellevue. In fact, few of them go there anymore. Most tuberculosis testing and treatment take place now through the Health Department anyway, but not that many clients go for the screening.

The Educational Alliance's Peter Cordero sometimes sends clients to Bellevue for skin testing, but he and Williams say the hospital is again regarded as an impenetrable institution. The registration process has been streamlined some for all patients, but not nearly enough for their clients. "I'm sorry to say it, but Bellevue has reverted to its byzantine method for dealing with patients," the BRC's Ronald Williams said.

Nevertheless, Bellevue soldiers on. Rany Condos, the Chest Service co-director and an assistant professor, says the hospital learned better how to meet the needs of the underserved. And even if it isn't tending to many of the agencies' clients, the Chest Clinic continues to offer directly observed therapy to active tuberculosis patients, directly observed preventive therapy for those with latent infections, and to provide isolation beds for those who are hospitalized.

And the community-based organizations say the program has had one important legacy: heightened tuberculosis awareness. "The first questions nurses here at BRC used to ask were whether you were in detox before and what drugs you use," Williams said. "Now they ask what drugs do you use and have you been tuberculosis tested. "

Pedro Perez sits in a modest Tijuana medical clinic, his navy sweats hanging loosely, his eyes cast downward as he speaks softly. It was 25 years ago, in 1975, that he first learned he had tuberculosis. He was driving a cab in Tijuana then, and he coughed a lot.

"I took the pills for a while until I stopped," Perez, 51 now, says somberly. He was feeling better and, he adds sheepishly, he really wanted a drink. Told that liquor and the tuberculosis medicines could be toxic to the liver, he chose alcohol.

When the cough flared up again, he returned for medical care. He was to repeat this starting and stopping routine five times, seeking treatment in Tijuana and just over the border in San Diego County, where he visited relatives from time to time. With all this interrupted treatment, the tuberculosis bacteria altered, and by his sixth attempt it no longer responded to the seven most effective medicines. And no other drugs were available in Mexico to cure his disease.

Perez was filled with despair-he coughed up blood, had feverish night sweats, was depressed, and feared for the health of the nine people living in his family home. Then he got a break. In 1994, his Tijuana clinic was able to obtain the pricey medications he needed through its new partnership with San Diego County as a demonstration site under the Old Disease, New Challenge program. Equally important, he received the kind of special treatment and support he desperately needed to see the treatment through. In the past, Perez had either been sent home with enough medicine for a month or with a prescription he sometimes couldn't fill.

Now a nurse watched to be sure he swallowed the pills each day, either at his home or at the clinic, and gave him a supply for the weekend. He was closely monitored and tested regularly. When at first he didn't improve, part of a diseased lung was removed and the drug treatment was resumed for 18 months. Family members received medicine to keep from becoming sick, too. This time, Perez said, he took his tuberculosis seriously and was cured.

"It if wasn't for this program and the effort put in, I know he wouldn't be here," his wife, Marcela Rodriguez says. "He was given so much attention from people here. We wished others could have a program that focuses on an individual's well being like this."

Perez's personal journey is a chilling introduction to the real people behind the global tuberculosis epidemic. The incidence of the disease and its drug-resistant form are highest in developing countries such as Mexico, where there is little or no directly observed therapy to help with adherence, and there are none of the drugs patients need when the most common ones no longer work.

For San Diego County, the world tuberculosis crisis was very close to home-virtually spilling over the backyard fence. In 1992, San Diego County's new tuberculosis case rate spiked up 75 percent, even higher than California's own alarming 45 percent rise. About 44 percent of the patients were Latino, two-thirds of them from Mexico.⁸ Baja California, where Tijuana is located, had the highest active tuberculosis rate of the six Mexican border states. The county was finding-and still does find-that 20 percent of its tuberculosis patients were resistant to first-line medications. Drug resistance was even higher in Tijuana, at 30 percent, according to a 1994 study.⁹

"The need to do something was so obvious, because people cross this border more than 65 million times every year," said Alberto Colorado, who today is the county's binational coordinator. "There is constant movement and constant risk of exposure. These are people who take care of our children, our yards, serve food at McDonald's, work in the malls, movies, the racetrack, run the buses, visit family. For public health reasons, we need to understand that there are now more international interactions among people."

With a sense of urgency, San Diego County designed a program called A Model for Cooperation: U.S.-Mexico Tuberculosis Control. It would become the Foundation's first binational program, and the most problem-riddled of the Old Disease, New Challenge grants.

Ambitious goals were set: to cure patients with regular and multiple-drug resistant tuberculosis and lower the number of new infections; to ensure that patients continued to receive medications when they crossed the border; and to educate people living in this high-risk area and their medical personnel about tuberculosis. The key to achieving the objectives would be beefing up the Tijuana clinic and developing a model for directly observed therapy there.

"The challenge in Mexico, the U.S., really everywhere in the world is not so much the diagnosis and getting medication," said Kathleen Moser, county tuberculosis control chief and the grant's principal investigator. "It's once a person is diagnosed and on medication, insuring that they take the medication the right way every day for six to nine to 12 months or more. The missing piece at that time in Tijuana was completion of therapy, and that's the ballgame."

The county subcontracted for services with two health clinics frequented by people who live and work in the border region: the San Ysidro Health Center, just two miles over the border into the United States, and the Centro de Salud Urbano, the downtown Tijuana public health clinic where Perez was treated.

A more modest binational collaboration between the San Ysidro clinic and the Centro de Salud Urbano had been running since 1984. Led by Benjamin Sanchez, a physician who would later become part of the new project's team, it was funded by the federal government. Under the new program, the United States government would continue to fund medicines and testing of sputum cultures, and The Robert Wood Johnson Foundation would cover the outreach with directly observed therapy to all patients and the very expensive medications for patients with multiple-drug resistant tuberculosis.

The two clinics would conduct tuberculosis screening, diagnosis, treatment, and directly observed therapy for patients with active disease and latent infection. They would also call ahead when they knew a patient was crossing the border to help insure the continuation of treatment.

The San Ysidro-Tijuana partnership had a number of pluses. It treated 170 patients, with about 75 percent completing treatment-an improvement over Tijuana's track record, which was estimated at 50 percent. The program put 20 of the patients with multiple-drug resistant tuberculosis in Tijuana on medications. Half of them completed the long regimen, which can take more than two years.

The project also allowed experimentation with approaches to stanching this disease that hadn't been seen before in Tijuana. Each day, two nurses and a social worker drove all over the city to visit patients, watch them take their pills, and monitor their progress. Additionally, they drove to people's homes to inform them they had active tuberculosis and searched for patients' close friends, co-workers, and other regular contacts so they could skin test them. (As it turned out, this was an arduous, time-absorbing process, and the project staff concluded that the model of delivering directly observed therapy was too inefficient and expensive.) Another benefit was the training of Tijuana and San Ysidro clinic personnel in the latest standards for tuberculosis screening, treatment, and infection control-an educational opportunity staff members say continues to pay dividends. The Tijuana clinic was also given sorely needed diagnostic equipment-an X-ray machine, a film-processing unit, and a small, separately ventilated sputum induction booth where patients who had difficulty producing enough sputum for a sample could cough without spreading bacteria to others nearby. And an informal referral system developed in which project staff members would at times call ahead to one another when a patient was relocating across the border and needed follow-up attention.

The project also ran into problems that severely limited its impact. Today, a sadness lingers as participants tell of the misunderstandings and the bitterness that overtook them. They attribute the discord largely to the difficulties inherent in bringing two countries of differing resources and cultures together.

Leadership was a major shortcoming. The grant burned through two overall project managers, both based at San Ysidro, who were viewed as out of their depth. The first manager was slow to iron out basic yet complex startup issues-how to pay workers in Tijuana and buy and transport equipment, two cars, medical specimens, and medication over the border. And both managers never addressed a crucial issue-that many of the San Ysidro staff members were uncomfortable providing tuberculosis services because they feared for their own health and legal risk to the clinic if other patients caught the disease. As a result, commitment to the project was never solid. And the second project manager created serious personnel problems at the clinic that led to tension, bad morale, and disrupted service before he finally left.

Financial disagreements also plagued the project. Both clinics involved in early project planning had contracts with the county that spelled out that they would be paid for delivering certain services. As it turned out, both clinics believed they should have gotten grant money to use as they saw fit, and ill will resulted as they waited instead to be given money to cover specific expenses and were told what to do. The Tijuana staff viewed this as a rich country dictating the rules to a poor one, leaving it no better off in the end.

"The feeling at the Tijuana clinic was 'Where is the money,'" said Benjamin Sanchez, the project's coordinator. "And they had a different idea of how the money should be spent. They even stopped collaborating with us at the end because of this. They said that they weren't getting the money so they wouldn't do anything." Anger and resentment worsened when the peso was devalued and project nurses working at the Tijuana clinic, who were paid in American dollars, earned seven times as much as the clinic's own nurses. "To this day, they say it undermined the whole tuberculosis control effort in Baja California, which is hard for me to understand," said Kathleen Moser, the project's principal investigator.

Most serious of all, though, San Ysidro's understanding of its mission was at odds with San Diego County's. At monthly case conferences, the county learned that San Ysidro was taking some tuberculosis patients recommended for directly observed therapy by their doctors or the county health department off of it. The project covered this cost, and it's not clear why the clinic dropped these patients. It was speculated that the clinic wished to use its outreach worker differently. San Ysidro was also responsible for caring for close associates or contacts of its tuberculosis patients regardless of whether they had health insurance. However, San Ysidro referred uninsured contacts to the county's tuberculosis clinic, 15 miles north, increasing the chances that they might stop treatment. As a result of these disputes, the county relieved San Ysidro of its duties two years into the three-year grant, and took over caring for the patients at its own tuberculosis clinic and at another county clinic closer to the border.

Moser and other county officials don't deny that the breakdowns and turbulence stunted the project's effectiveness. But they say the conflicts and problems in getting started stemmed from the complexity of creating a binational program and the need to stay sensitive to cultural subtleties and differences. This project, they said, really needed far more time to find its way. In fact, having learned that sending outreach workers to patients' homes wasn't practical, the county asked the Foundation to extend the grant an extra year and use unexpended money to change the directly observed therapy model in Tijuana to one that was delivered by promotoras, health promoters from within the local community.

This request led to the bitterest chapter of this grant's story. "We felt our program was just beginning, and we were just starting to understand how the binational model could work better," Moser said. "It is so difficult to get something going involving international relations. Three years was way too short to start up something new and difficult like this." The Foundation said no to the extension requested. The project hadn't performed well, and it was time to cut the losses, said Marilyn Aguirre-Molina, the Foundation's program officer for the project at the time. "We had a lot of meetings on this program," Aguirre-Molina said. "It was not a decision made easily or routinely. But there was nothing there to build on. It would have been an extension to start almost from scratch."

Since the project ended, in 1996, the San Diego County tuberculosis rate has declined, as it has elsewhere in the nation. Drug-resistant tuberculosis continues at 20 percent of cases because of the influx of people from Mexico and Southeast Asia. In 1998 the World Health Organization counted Mexico among the 16 "trouble spot" countries that accounted for more than half of the people worldwide with the disease.

Despite the internal tribulations, the binational program did have some lasting benefits. The X-ray machine continues to be used in Tijuana, and expertise in tuberculosis control and management improved among health care providers. More important, relationships-and some trust-did develop among public health officials in the county and in Tijuana. Today, regular meetings on regional approaches to tuberculosis are held, and annual binational conferences that were started under the grant continue, attended by hundreds of officials from both countries.

As a result of the project, the state of California broadened the informal referral system by funding the Cure TB program out of San Diego County's binational tuberculosis coordinator's office. With an 800 number, Cure TB gives to health officials treatment information concerning about 100 active tuberculosis patients each year who move between the United States and Mexico. The project also inspired a Texas initiative called Ten Against TB, in which the 10 states of the United States and Mexico that share the border work together on tuberculosis control.

Meanwhile, when the project ended, the Centro de Salud Urbano in Tijuana was left with lesser diagnostic and treatment options. "We miss those benefits for our patients," said Concepcion Corona, the beleaguered chest clinic director there. "We got used to having resources, and suddenly they were cut off and we were left alone."

The Tijuana clinic has an active caseload of 125 tuberculosis patients, which is still high, and a disturbing 10 percent of them are children. The clinic does no drug-susceptibility tests to determine whether a patient doesn't respond to certain drugs-a dangerous omission that certainly leads many patients to be treated with the wrong drugs and allows them to spread the disease. With no real directly observed therapy occurring, patients come to the clinic once a week for their pill supplies.

At this clinic where Pedro Perez was finally cured of tuberculosis, there is little hope for patients like him whose disease can't be cured with drugs obtainable there. Corona is sure that he would have more multiple-drug resistant tuberculosis patients if the proper medications were available. "It is very sad, but when someone comes in with multiple-drug resistant tuberculosis today, I have nothing to give for medicine," Corona said. "How is it possible that after thousands of years, we finally know what is the best way to treat tuberculosis and to save people's lives, yet here, in Tijuana, we continue to give treatment with no hope?"

Brenda Luna steers a county van through the deep nighttime darkness of rural North Florida roads, past a dog barking fiercely, ramshackle houses teeming with people. Luna, a nurse practitioner, who is white, stops the van at the labor camps and houses where many of the African-American farmworkers who stoop in the cabbage and potato fields here live in poverty, most of them in virtual slavery to their crew chiefs. On her rounds through the neighborhoods, Luna walks up to a known crack house surrounded by a chain link fence-she's heard someone isn't feeling well there.

"Hey, does anyone want to see the doc tonight?" she calls out.
"He's already gone up there," a woman yells back.

On this trip, six men eventually pile into the van to be ferried to a large mobile medical bus parked several miles away. One has a toothache, another is bothered by an old hand injury from a potato grader. Someone smells of alcohol. It's a convivial group, full of light and friendly banter. This could be just another night out on the town.

When they reach the bus, parked outside a convenience store, an intake nurse wants to know why each man has come. And then she asks pointedly, "Have you been coughing? If so, is there any blood? Have you had fevers, night sweats? Any sudden weight loss lately? Are you more tired than usual?"

These are symptoms of tuberculosis, and the litany of questions and the primary health care provided on this bus are continuing signs of the good work that began here during the period covered by the Old Disease, New Challenge grant and that continues, fueled by an uncommon commitment and humanity.

This Putnam County-based project hoped to bring comprehensive testing, treatment, and tuberculosis education to 1,500 African-American male migrant workers and about 4,000 Hispanic seasonal fernery workers. They lived in four North Florida counties-Putnam, St. Johns, Flagler, and Volusia-in an isolated society with little access to health care. The project also set out to create a tracking and referral system that would insure that migrant workers continue to get medical care for tuberculosis when they follow the changing crops up the eastern seaboard to North Carolina, Maryland, and Delaware. This system wouldn't be needed for the seasonal workers, who lived locally and didn't travel to distant fields.

A companion project ran concurrently 400 miles south in Palm Beach, Glades, and Hendry counties, where about 30,000 migrant and seasonal farmworkers resided. But without the resources to reach a much larger farmworker population, it scaled back its goals and thus achieved more modest gains.

Nationally, the 4 million men and women who work the fields and nurseries are among the most at risk for catching tuberculosis. In 1992, when the return of the disease peaked in the country, the Centers for Disease Control and Prevention found farmworkers six times as likely to get tuberculosis as all working American adults. They live in crowded, substandard housing, tend to have poor nutrition and many health problems, face impediments to health care, and often don't receive continuity of care because they move frequently.

Mindful that local farmworkers were fitting into this pattern, the project set out to break down barriers for testing and treatment and established targets it came close to meeting. Directly observed therapy was provided for all active cases and most people with latent disease. The results were impressive: The number of people with either active disease or latent infection markedly declined over the four years of the project, according to an analysis by the University of Florida's Division of Biostatistics. This would suggest, the analysis said, that the project's mass case-finding program was effective.¹⁰ As a result of the project, tuberculosis came under control in this population.

"We found that by testing everyone and giving directly observed therapy both to people with active and latent disease we were preventing additional cases," said Cheryll Lesneski, county health administrator and the project's principal investigator. "This isn't something you would do in just any population-you do it only in a population with a high rate of tuberculosis, which our migrants had. In high-risk groups, it's money well spent, but it's not money that's available anymore."

A large number of people had active tuberculosis-15 migrants and a child of a seasonal worker who had recently emigrated from Mexico. None had a drug-resistant form. Treatment was completed by 93.3 percent of the migrants and 100 percent for the seasonals, with the one child adhering to the therapy. This essentially met the project goal of 95 percent completion for active tuberculosis cases.

In all, 1,052 migrant farmworkers and 1,209 seasonal workers were evaluated, some many times, with 28 percent of migrants testing positive for latent infection and 23 percent of the seasonals positive. About 151 migrants received directly observed preventive therapy with 77 percent completing it, while 205 seasonals were given this treatment with 86 percent finishing it. The goal had been for 90 percent of those with latent infection to complete directly observed preventive therapy.

But numbers just scratch the surface of what was accomplished here. The people who made up the staff for this project brought health care to the farmworkers on their own terms, rather than ask them to fit into a traditional 9-to-5 system they clearly wouldn't use.

Some key ingredients made this project work: A "migrant scout" with deep connections in this community found where the men would be working each day; the project was flexible, retooling quickly to solve problems; and employees worked crazy hours to make it as convenient as possible for the farmworkers. When obstacles arose, the project's workers repeatedly seemed to scratch their heads and say, "Let's try it a better way."

"There was a lot of caring going on," said project director Laurey Gauch. "The feeling of the folks here was that this was an emergency."

The project started out with one big advantage-Putnam County had laid the groundwork early and was able to avoid startup delays. In 1992, the county had found that the number of tuberculosis cases among farmworkers was on the rise; three of them had died. Putnam County's public health unit swiftly mounted an aggressive campaign to screen as many workers as possible. Staff members and a retired physician parked their vans and station wagons at fields and worked late into the night administering PPD skin tests. At first, the often-brutal crew leaders were distrustful, balking at the intrusion and the lost work time.

Cheryll Hampton, a migrant scout, had grown up in labor camps, and her participation was crucial in winning the farmworker community over and gaining access. She had entrÈe into the culture and persuaded crew leaders and farmworkers of the health care emergency.

"The crew leaders were ugly about us going to the camps or fields, because they really didn't care about their workers and didn't want the crews slowed down," said Hampton, whose father and grandfather had been crew chiefs, but of a kinder stripe. "They have to know and trust you to let you do it."

The county workers pulled up to the fields at times that were most convenient to the farmworkers-whether it be during work hours or after, at dawn or late at night. Sometimes a project staff member would jump onto a tractor or start pulling potatoes to replace a farmworker who needed an extensive consultation.

The screening told grim news: 70 percent of the migrants and 20 percent of the seasonal workers tested positive for tuberculosis. The numbers reflected both new infection and longstanding infection that hadn't previously been detected. They also demonstrated that the problem was greatest with the migrants, who usually lacked the family structure and support the seasonal workers had.

With limited resources, the county workers began treating the farmworkers from their vehicles, focusing most on the migrants. They soon noticed that many of them weren't taking their pills. Some had no place to store them, and others with latent infection and no symptoms lost interest because they didn't feel ill or simply didn't want to take pills for months. The county workers concluded that they needed to find funding to do more.

That's when the county applied for and secured a grant from The Robert Wood Johnson Foundation. With the infusion of funding, Putnam County planned to expand the screening and treatment and to provide directly observed therapy for those with active or latent tuberculosis. The grant would pay for tuberculosis education for workers, health care providers, and patients, and would go toward creating an interstate tracking system to ensure continuity of care for farmworkers as they traveled.

The farmworkers were tested each year unless they had already been found to be positive and had been treated. Outreach workers met patients with active tuberculosis in the fields or at home to watch them take their medication each day. Most of those with latent disease were also observed as they swallowed pills twice a week.

Hampton's inroads into the migrant community were critical to staying on top of testing follow-up and treatment. To be sure that outreach workers could find the highly mobile patients, she visited convenience stores and other locations where workers congregated to get the latest on where crews and individuals would be that day. As she became accepted, workers and crew chiefs would tip her and other scouts off when someone wasn't well or had a cough or other possible symptoms.

"I would be out in the field or stopping by a store where a bunch of them got coffee and someone would come up to me and say so-and-so has a cough, I know it's a tuberculosis cough," Hampton said. "This one's spitting up blood. Or they'd say that guy youlooking for, he moved to such-and-such labor camp."

The arrangement did have flaws. Patients who came for consultations lacked privacy on the medical bus. The portable X-ray machine was cumbersome to move around to different locations, and processing the film in the field was a challenge. And because a Catholic health organization furnished the bus and wouldn't allow family planning on board, the project couldn't educate workers about condoms. In addition, the county discovered that many farmworkers had other serious health problems that couldn't be treated on the bus, such as HIV, hypertension, and diabetes.

The county adapted by establishing night clinics twice a week, first with a community health clinic but ultimately solely at the county Public Health Unit office in Palatka. Using a van bought with grant funds, outreach workers drove farmworkers the 15 miles to the county clinic. There they experienced a new approach to their health care. They could get skin-tested for tuberculosis, have an X-ray taken, get tested for HIV, and have other health concerns addressed by a physician. The clinic proved to be popular. Local churches and restaurants donated meals, and a closet was well stocked with items the migrants, who traveled light, needed-blankets, gloves, socks, even toilet paper. They watched videos and enjoyed the evenings so much that they were often reluctant to leave.

"We worked until 11, 12, even 1 in the morning, and in the course of an evening saw 14 to 40 patients," project director Gauch said. "This holistic approach to the whole person created trust. They were so grateful to us. I've never worked with more appreciative and hard-working people."

The same determination was directed at developing a referral system to assure continuity of care when the migrants moved north for other crops. These efforts were less fruitful, although not for lack of trying. Many meetings were held with officials and providers in North Carolina, Maryland, Delaware, New York, and other states, and communication improved and important contacts were made. With the grant funding, the North Carolina Primary Health Care Association developed an extensive migrant farmworker resource manual. Still used today, it contains CDC guidelines for care, useful medical journal articles, and phone numbers for health departments, rural health centers, and other agencies along the East Coast migrant stream to notify when a sick worker relocates.

An Interstate Migrant Medical Record was designed, too, to standardize the information that would be relayed about farmworkers with tuberculosis, but it didn't gain lasting acceptance. The county had also hoped that a computerized farmworker data base would be the backbone of a migrant referral system. However, this became bogged down among migrant health care providers, state health offices, and the proj-ect principally because of concerns about patient privacy that couldn't be resolved. Putnam County also tried a system of faxing patient information to other providers, but it was scotched when Florida adopted new patient confidentiality rules. A plan for an outreach worker to travel with the migrants to help assure continuity in treatment for active and latent infections was also abandoned. There were too few farmworkers to follow to justify the cost.

Nevertheless, Putnam County made contacts in other states that proved useful. The county would call ahead to them to give a heads-up that an infected migrant was relocating, and to relay shared medical information. The contact would often provide the same courtesy when workers were headed back to Putnam County. This informal, low-tech networking continues today.

At present, in North Florida, tuberculosis is under control among farmworkers. When the project ended, outreach was scaled way back, and mass screenings were stopped. And although the county's public health department provides directly observed therapy to anyone in the community who needs it, including farmworkers, preventive therapy is rare.

But the spirit of this project lives on. As a direct result of the proj-ect's strong performance, the county had the credibility to secure funding from St. Vincents Health Services, Inc., and from the federal Health Resource Services Administration for a more comprehensive health care service for farmworkers-the mobile medical bus. Other gains that weren't lost include the contacts with other providers and the heightened tuberculosis awareness of farmworkers and county public health staff. And because of trust built during the project, farmworkers freely board Brenda Luna's van to be seen at these clinics two days a week. They are seen for many reasons-including Pap smears, high blood pressure, diabetes, injuries, the flu, and tuberculosis. And the nurses continue to ask whether they have any tuberculosis symptoms.

"These people are just beautiful," one farmworker aboard the van says. "They always treat you like you matter. Always."

Demonstration projects were mounted in two other locations: Baltimore and Atlanta.

A collaboration between Johns Hopkins University and a group of churches had hoped to test whether neighborhood health workers offering financial incentives could increase voluntary screening and treatment in a low-income East Baltimore neighborhood. But the project found that the city's own tuberculosis control program had been so effective that the infection rate was very low and the project actually wasn't needed. It was consequently trimmed way back. The project did demonstrate, however, that the outreach workers from the community, who were to test, refer, and counsel residents and people in high-risk settings like prisons, needed considerably more training and supervision.

The outreach workers were also to staff a health care referral system for prisoners with latent infection who were about to be released, but this never worked out. The prisoners did, however, produce a comic book, "2 B TB Free," to encourage inmates to complete their treatment after release.

Atlanta had the highest tuberculosis rate of American cities in 1991, and its project made some headway toward fostering coordination among the many groups that until then had worked separately on tuberculosis control. A broad coalition that was established included the county and state health departments, the public Grady Memorial Hospital, Emory University School of Medicine, Morehouse School of Medicine, Mercy Mobile Health Care, correctional facilities, the CDC, and the American Lung Association.

This project conducted a mass screening-7,246 people at high-risk locations-and concluded that it wasn't an effective way to bring latent tuberculosis under control. Just 20 percent of the patients completed their six-to-nine-month drug regimen. "Until it's possible to give directly observed preventive therapy or some other therapy, treatment for latent infection is doomed to fail," said Dr. Henry Blumberg, the project's principal investigator.

An attempt to increase care for patients discharged from the hospital got good results. Outreach workers, or liaisons, met with patients in the hospital to arrange care at a local clinic after release. About 58 percent of the patients had been lost to follow-up in 1992, but just 6 percent were lost after the grant's first year. And the median time from discharge to follow-up care dropped from 81 days in 1992 to just six in 1994.

The Atlanta project made progress in creating a tuberculosis computer network and a data base for reporting and confidential sharing of patient data. This system, expected to be up and running by the end of 2001, was to centralize information that had been dispersed at various locations. It will link all providers, improving communication among them, and also surveillance and control. The state of Georgia will maintain the system.

The grantee also published the "Georgia TB Reference Guide," a pocket-sized health care education handbook that was started under the grant and has been updated since with state funding. It offers treatment guidelines for latent infection and active disease based on CDC recommendations, and lists resources and phone numbers. "It's a little book physicians put in their white coat and fits in very nicely," Blumberg said.

The Robert Wood Johnson Foundation funded a number of smaller, related grants that supported tuberculosis control efforts.

One that attracted considerable notice used mathematical modeling to predict how much treatment it would take to eradicate tuberculosis worldwide. The modeling team, led by the mathematical biologist Sally Blower, currently at the University of California, Los Angeles, developed mathematical equations that described many aspects of how the tuberculosis epidemic behaves. The models, which were published in Science in 1996, identified the outcomes that could be expected from a range of treatment strategies for latent and active disease. The forecasts generated interest and controversy because they found that the number of patients the World Health Organization called for treating would not lead to global disease eradication.

Another grant set out to answer one of the major ethical questions of this epidemic: Just what do we do when tuberculosis patients refuse to take their medicine and endanger other people's health? Led by Bernard Lo, a team at the University of California, San Francisco, studied all 50 states' policies on criminal and civil detention and tuberculosis case reports in California counties. The study found that with sanatoria long closed and no money to pay for hospitalization, counties often had no alternative to incarcerating people. Many counties used criminal detention as a last resort, but others imposed it more aggressively. The major upshot of this grant was a collaboration with the state of California that led to civil-as opposed to criminal-confinement in two newly created state facilities for those who fail less restrictive measures. The facilities also provide some counseling, drugs, and other services.

Another grant provided support to the National Coalition to Eliminate Tuberculosis, a group of 70 organizations working to coordinate tuberculosis control. With the grant, the coalition hired a coordinator and a support staff person, improved communication among members through an online bulletin board and a newsletter, held an annual meeting and quarterly task force conference calls, and conducted tuberculosis awareness campaigns.

One aspect of the projects merits special mention. Whatever the location, staff members with an uncommon humanitarian zeal energized the projects. Workers stretched out a hand to forgotten, often ignored people, doing unseen and dedicated work. One project worker after another wouldn't accept the idea that people down on their luck and with no money were suffering and dying needlessly from a curable disease. So they rode out in the dead of night on lonely roads to find farmworkers living in ramshackle trailers. They put in long hours for relatively low pay, to be exposed to a disease that is not only highly infectious but also is sometimes resistant to drug treatment-and they did it gladly. They were an inspiring breed.

While three years is probably too short a time period for complex projects involving hard-to-reach people to plan, operate, and show definitive results, the projects demonstrated that disaffected people can be reached when local government, community groups, and health care facilities collaborate. But the partners must be compatible and committed to the work. Most important, health workers must be sensitive to the culture of their potential clients, and projects must be flexible and willing to change preconceived notions to meet actual circumstances. In addition, screening for tuberculosis should be targeted toward at-risk populations and must be paired with directly observed preventive therapy and directly observed therapy. Given the ease with which diseases can spread across borders, binational work is increasingly essential, yet the challenges posed by different cultures, financial circumstances, and legal structures are daunting.

Over the life of the grants, most of which were extended to a fourth year, the alarming rise in tuberculosis was stunningly reversed, and the disease again came under control. As a result of aggressive efforts to strengthen control, tuberculosis rates declined 34 percent from 1992 to 1999, when 17,531 people were diagnosed with tuberculosis.¹¹ The disease was at its lowest recorded level of 6.4 new cases per 100,000 people in 1999.¹² Improvement came at a high cost, however. In New York City alone, it took more than $1 billion to rebuild a public health apparatus to meet the need.¹³

There is always a danger of tuberculosis re-emerging. A report by the National Academy of Sciences' Institute of Medicine, issued in 2000, warned the United States to maintain and even step up its tuberculosis control, to be sure it doesn't repeat history. "The question now confronting the United States is whether another cycle of neglect will be allowed to begin or whether, instead, decisive action will be taken to eliminate the disease," the Institute of Medicine said. Much more headway remains to be made among foreign-born people in this country. The number of cases among those here from foreign lands has mushroomed from 24 percent of active tuberculosis patients in 1990 to 43 percent in 1999. Almost half of these patients were natives of three countries-Mexico, the Philippines, and Vietnam-and many likely came here with latent infection.¹⁴ "Until we can make tuberculosis go away in places like South Africa and Tijuana, people will still be able to bring it into the U.S., even if we keep working on pockets, " said Rany Condos, co-director of the Chest Service at New York's Bellevue Hospital. "That's why we need the infrastructure to remain. I hope we've learned this from the epidemic."