Protecting the Crown Jewels of Medicine
A strategic plan to preserve the effectiveness of antibiotics

This report was written by Patricia B. Lieberman, Ph.D., and Margo G. Wootan, D.Sc. The authors thank Dr. Richard Novick of New York University Medical School; Dr. Louis Rice of the VA Medical Center in Cleveland, Ohio; Dr. Sidney Wolfe, executive director of Public Citizen Health Research Group in Washington, D.C.; and Dr. Karim Ahmed, Deputy Director of Health, Environment, and Development at the World Resources Institute in Washington, D.C. for reviewing this report.

Copyright © 1998 by the Center for Science in the Public Interest
Referances available by request

Executive Summary

    The recent reports of three Americans infected by Staphylococcus bacteria that were resistant to the antibiotic vancomycin triggered only fleeting news coverage. The cases should have sent shudders through the medical community and the public, because vancomycin is the last line of defense in treating deadly hospital-acquired staph infections. Despite vancomycin’s value, a recent study showed that 63 percent of vancomycin prescriptions were inappropriate. The cases of vancomycin-resistant staph, taken together with widespread resistance to penicillin, tetracycline, and other antibiotics, demonstrate the urgent need to prevent antibiotics from losing their effectiveness against diseases that are now curable.

    Despite antibiotics’ extraordinary value, the overuse of those miracle drugs in medicine and agriculture endangers their continued effectiveness. The more antibiotics are used, the more likely it is that bacteria will develop mechanisms to evade them.

    Until recently, each time an antibiotic lost its effectiveness there was another magic bullet on the pharmacist’s shelf. But now that shelf is almost empty. The development of new antibiotics has not kept up with the development of antibiotic resistance. The time has come when public and private institutions, as well as the general public, must change their policies and practices to prevent further increases in antibiotic resistance. Rather than believing that new drugs continually can be developed to treat antibiotic-resistant infections, public-health prevention measures should be adopted. Reforms should include:


    A hundred years ago, infectious diseases were the leading cause of death in the United States and abroad. But advances in public health and the discovery of antibiotics brought many of those diseases under control in the U.S. For example, in 1900, pneumonia and tuberculosis caused almost one-quarter of all deaths in the United States. By 1990, those two illnesses caused less than four percent of all deaths.

    Antibiotics are the wonder drugs of this century. They have played a major role in decreasing rates of tuberculosis, curing children of meningitis, ensuring the recovery of burn victims who often fell prey to infections while healing, and reshaping the treatment of syphilis and gonorrhea.

What is antibiotic resistance?

    Many infectious diseases, including bacterial pneumonia, tuberculosis, and gonorrhea, are caused by bacteria. Antibiotics work by a variety of mechanisms to kill bacteria and have proven vital in curing infectious diseases. However, bacteria have the capacity to evolve defense mechanisms against antibiotics and can become resistant to their effects. When such resistance develops, bacteria are no longer killed by the antibiotic and, thus, the antibiotic is no longer capable of treating or curing the disease. The more an antibiotic is used, the more likely that bacteria will "learn" to evade it.

    Natural selection plays a key role in the development of antibiotic resistance. Most bacteria die when exposed to antibiotics to which they are sensitive. That leaves more space and available nutrients for surviving bacteria (i.e., for antibiotic-resistant bacteria). As a result, the resistant bacteria can reproduce and multiply freely and pass on the antibiotic-resistant genes to the next generation.

    Not only can resistant bacteria proliferate after other bacteria are killed off by an antibiotic, but they also can transfer that resistance to other bacteria that have never been exposed to the antibiotic. Bacterial cells can join briefly and exchange loops of DNA (called plasmids) that contain genes that confer antibiotic resistance. For example, if one bacterial species becomes resistant to a broad-spectrum antibiotic, it could transfer its resistance genes to other bacteria that have never encountered those antibiotics.

From 1975 to 1991, the incidence of methicillin-resistant Staphylococcus aureus (MRSA) isolates in U.S. hospitals increased from 2.4 percent to 29 percent.

    The genes that cause antibiotic resistance function in a number of ways. Some do not permit the antibiotic to get into the bacteria; others actively pump it out of the bacteria; some produce enzymes that inactivate the antibiotic; and others modify the antibiotic’s target site in the bacteria. To make matters worse, many bacteria have become resistant to multiple antibiotics. That property can result from the cumulative effect of treating stubborn infections with multiple types of antibiotics (or acquiring a plasmid with numerous resistance genes).

    Antibiotic resistance has increased rapidly in the U.S. and abroad in recent decades. More than 90 percent of isolates of Staphylococcus aureus, one of the most common disease-causing organisms, are resistant to penicillin. Staph bacteria, which cause skin, heart valve, blood, and bone infections that can lead to septic shock and death, also are showing resistance to medicine’s next line of defense, the antibiotics related to methicillin, at an alarmingly increasing rate. From 1975 to 1991, the incidence of methicillin-resistant Staphylococcus aureus (MRSA) isolates in U.S. hospitals increased from 2.4 percent to 29 percent. In some hospitals, more than 60 percent of strains are methicillin resistant. As MRSA spreads, physicians are left with only one approved antibiotic, vancomycin, to treat that deadly infection. Although a new antibiotic, Synercid (guinupristin-dalfopristin), may be approved shortly, overuse of that antibiotic could lead to antibiotic resistance as well.

    Since vancomycin is currently the last line of defense against MRSA infections, the first U.S. cases of vancomycin-resistant Staphylococcus aureus, which have been reported in the past year, are cause for great concern. The first known case of Staphylococcus aureus that showed the beginnings of resistance, or intermediate resistance, to vancomycin occurred in 1996 in a Japanese child. The next two cases occurred in the United States., One patient had suffered repeated bouts of MRSA, for which he had received multiple treatments with vancomycin. Although the first three patients were eventually treated with a combination of other antibiotics, those other antibiotics may not be able to treat all vancomycin-resistant staph infections. Indeed, a New York man in his 70s infected with vancomycin-resistant Staphylococcus aureus died in March 1998. About five percent of hospital patients acquire bacterial infections while hospitalized. Since Staphylococcus aureus is one of the most common causes of both hospital- and community-acquired infections worldwide, the possibility that those infections may become untreatable with antibiotics is alarming.

    Streptococcus pneumoniae, a bacterium that can cause ear infections, pneumonia, blood infections, and meningitis, is becoming increasingly resistant to antibiotic treatment. In 1987, antibiotic-resistant pneumococci were unknown. By 1997, as many as 40 percent of pneumococcus isolates were resistant to penicillin and other commonly used antibiotics.

In 1987, antibiotic-resistant pneumococci had not been encountered. By 1997, as many as 40 percent of pneumococcus strains were resistant to penicillin and other commonly used antibiotics.

    One example of how overuse of antibiotics leads to increased levels of antibiotic resistance was demonstrated in a study of children in Memphis, Tennessee. Children of affluent parents with good access to health care and prescription-drug coverage were more likely to be carriers of antibiotic-resistant strains of Streptococcus pneumoniae than children with less access to medical care, because the affluent children were more likely to have been treated repeatedly with antibiotics for ear infections.

    Appendix 1 lists ten common drug-resistant microbes, the diseases caused by those bacteria, and typical antibiotics that can no longer be used to treat those diseases.

The impact of antibiotic resistance on health-care costs and mortality

    The cost of antibiotic resistance in the United States is substantial, but it is difficult to estimate with precision. One investigator estimated that the cost is between $100 million and $30 billion annually (the wide range arises primarily from different values assigned to the value of a human life). The National Foundation for Infectious Diseases recently estimated the cost to be $4 billion annually. Costs will rise as resistance becomes more prevalent.

    The U.S. Office of Technology Assessment (OTA) calculated the direct hospital costs from five classes of hospital-acquired (nosocomial) infections associated with only six strains of antibiotic-resistant bacteria. It concluded that the minimum hospital costs of those infections were $1.3 billion (1992 dollars).

The cost of treating a patient with tuberculosis increases from $12,000 for a patient with a drug-susceptible strain to $180,000 for a patient with a multidrug-resistant strain.

    Antibiotic resistance also adds significantly to the financial burden of illnesses contracted outside of the hospital. Contributing to the problem are diseases that were once thought to have been eradicated but that now are reemerging with antibiotic-resistant strains. Those resistant infections are difficult and costly to treat. For example, the cost of treating a patient with tuberculosis increases from $12,000 for a patient with a drug-susceptible strain to $180,000 for a patient with a multidrug-resistant strain. Multidrug-resistant tuberculosis (MDR-TB) is a serious public-health threat in the U.S. and in the rest of the world. In the United States, MDR-TB was unheard of in the 1980s. However, with the erosion of public-health programs in many cities, in 1992 approximately ten percent of cases of TB were resistant to the antibiotic isoniazid and four percent of patients were infected with strains of TB that were resistant to isoniazid and rifampin (MDR-TB).

    In addition to the increased cost, antibiotic resistance increases the seriousness of diseases. Treating a patient who has TB caused by an antibiotic-sensitive strain is highly effective. In contrast, fatality rates for patients with MDR-TB range from 40 percent to 60 percent in patients with normal immunity to 80 percent in immunocompromised patients.

How antibiotics are used and misused

Medical uses of antibiotics

    In the 1940s, penicillin was hailed as a cure-all for staph infections and gonorrhea. But before the end of that decade, strains had emerged that could evade penicillin. At that time, drug companies were discovering many new kinds of antibiotics, and the problem of antibiotic resistance was postponed temporarily, as new innovations kept pace with the development of resistance.

    In the 1970s, strains of the bacterium that causes ear infections and meningitis in children (Haemophilus influenzae) and the bacterium that causes gonorrhea (Neisseria gonorrhoeae) developed almost complete resistance to penicillin. One resistant gonorrhea strain was traced to a brothel in Asia where penicillin was given as a preventive measure to keep women free of disease. That strain spread around the world.

    Development of new antibiotics has not kept pace with the emergence of antibiotic resistance. In the past few decades, pharmaceutical companies have shifted their development efforts to drugs for chronic diseases, such as those used to treat heart disease and high blood pressure. In addition, with the new infectious disease challenge posed by Human Immunodeficiency Virus (HIV), many researchers have focused on anti-virals instead of antibiotics. The government has been complacent about antibiotic research, devoting only limited resources to research on antibiotic-resistant bacteria and little to the development of novel antibiotics. As a result, the FDA has approved no new classes of antibiotics in the last decade.

    In doctors’ offices, antibiotics are often seen as a quick-fix solution for patients. Doctors currently prescribe antibiotics for outpatients approximately 150 million times a year. However, according to the CDC, more than 50 million of those prescriptions (one-third) are prescribed inappropriately -- when the illness is not caused by bacteria or the disease-causing bacteria are not susceptible to the antibiotic prescribed. In their desire to treat patients quickly and in their acquiescence to some patients’ demands for antibiotics, physicians -- who may be influenced by drug advertising -- often prescribe antibiotics without testing to determine (a) if bacteria are the cause of the illness (culturing) or (b) which antibiotic would be effective against the bacteria (susceptibility testing). Cost also may be a deterrent to culturing. If a physician performed a culture and susceptibility test and waited for the results before prescribing an antibiotic, he or she might find that the patient did not have a bacterial infection and, thus, would not be helped by antibiotics, or that the bacteria present were not susceptible to the antibiotic that he or she had planned to prescribe. For example, the patient might have a viral respiratory infection, and viruses are not susceptible to antibiotic treatment.

    Increased culturing of bacteria and susceptibility testing could decrease the number of inappropriate antibiotic prescriptions for many illnesses. However, for some illnesses culturing is difficult. For example, the common ear infection, a diagnosis made 25 million times each year, is difficult to culture. Because aspirating the eardrum and culturing the fluid that collects inside the child’s ear is an invasive procedure, the physician often must rely on his or her own judgement in order to treat patients. Many physicians routinely prescribe antibiotics for children with symptoms of ear infections. However, there is much scientific debate as to the merit of treating all types of ear infections with antibiotics.,, Many ear infections are not caused by bacteria, and most ear infections go away without antibiotic treatment.

In a recent large-scale study of hospital records by the Health Care Financing Administration, 63 percent of orders for vancomycin, the last line of defense against severe hospital-acquired infections, were inconsistent with CDC guidelines.

    Overuse of antibiotics for inpatients also contributes to antibiotic resistance. Increasingly, today’s hospitalized patients are seriously ill. At any given time, 25 to 40 percent of hospital patients are receiving intravenous antibiotic treatment. However, many of those prescriptions are inappropriate. In a recent large-scale study of hospital records by the Health Care Financing Administration (HCFA), 63 percent of orders for vancomycin, the last line of defense against severe hospital-acquired infections, were inconsistent with CDC guidelines.

    Another trend that is contributing to the problem of antibiotic resistance is the shift in the types of antibiotics that are prescribed. Increasingly, stronger, newer, and more broad-spectrum antibiotics are being prescribed in place of narrower-spectrum antibiotics -- which is like using a cannon instead of a rifle. That shift is worrisome because the more broad-spectrum antibiotics are used, the more likely a wider spectrum of bacteria will develop resistance to them. Broad-spectrum antibiotics are appropriate for treating patients at high risk of succumbing to infection quickly. For example, they are appropriate for a severely ill infant or an elderly, very sick patient with a high fever and other symptoms of infection, who cannot wait 24 hours for the results of a culture and susceptibility test. However, it is important not to squander those powerful antibiotics in circumstances when narrow-spectrum antibiotics could be effective. Better surveillance data, which would provide doctors with information about the types of antibiotic-resistant bacteria prevalent in their region, also could aid physicians in selecting more narrow-spectrum antibiotics.

    Antibiotics also are used to prevent disease in humans. That use is sometimes appropriate and necessary. Prophylactic treatment with antibiotics includes giving children susceptible to ear infections low doses of antibiotics for long periods of time in an attempt to control ear-infection-causing bacteria. Most surgical patients receive systemic antibiotics to prevent infection. Even teenagers with acne are given low doses of tetracycline to decrease levels of acne-causing bacteria. Giving antibiotics preventively also contributes to the development of antibiotic resistance. Some studies have shown that patients given prophylactic doses of antibiotics were more likely to develop infections caused by resistant organisms because the antibiotic treatment allowed resistant organisms to grow in the absence of competition from sensitive bacteria., The preventive use of antibiotics should be reserved for those conditions where their value is proven and outweighs potential risks.

    In developing countries, antibiotics also are often used inappropriately. In some countries, patients with bacterial infections cannot afford to complete a full regimen of antibiotics., That allows some of the bacteria -- especially ones moderately resistant to antibiotics -- to survive and multiply. In some countries, antibiotics are available over-the-counter or can be purchased on the black market, which also contributes to inappropriate use.

Agricultural uses of antibiotics

    While medical use of antibiotics is probably the major contributor to the emergence of antibiotic resistance, agricultural uses also pose a problem. Agricultural uses account for more than 40 percent of all antibiotics manufactured in the United States. Agricultural uses include (1) treatment of diseases in animals, (2) prevention of disease in animals, (3) growth promotion and improved feed efficiency, and (4) spraying of fruit trees and vegetables to prevent blemishes due to fire blight.

    The subtherapeutic use of penicillin, tetracyclines, and other antibiotics related to those used in human medicine poses a significant hazard to human health. That use involves routine and prolonged subtherapeutic dosing of animals with antibiotics for growth promotion and improved feed conversion. Subtherapeutic use of antibiotics, including penicillin and tetracyclines, accounts for as much as 80 percent of the antibiotics used in agriculture. Poultry, cattle, and swine routinely are given low doses of those antibiotics to increase their growth rates and reduce the amount of feed required to raise an animal to slaughter size. However, it is possible to raise animals economically without growth-promoting antibiotics. For example, in Sweden antibiotics are not allowed for growth promotion and are used only sparingly for therapeutic purposes in farm animals. Swedish officials say reductions in antibiotic use can be done cost effectively. Also, according to spokespersons for Tyson Foods, Inc. and Perdue Farms, Inc., neither company uses subtherapeutic doses of human-use antibiotics for growth promotion.

Faced with vigorous opposition in Congress by agribusiness and farm-state legislators, the FDA never implemented its proposed limits on antibiotic use in agriculture.

    Soon after it became routine to add antibiotics to animal feed in the 1950s, health officials in the U.S. and abroad expressed concern that long-term treatment of livestock with low doses of antibiotics could be harmful to human health. In the 1970s, the FDA unsuccessfully attempted to ban certain agricultural uses of antibiotics. Following a task force’s recommendations, the FDA proposed rules to revoke the then-permitted uses of penicillin and tetracyclines in animal feeds for disease prevention and growth promotion.,,, The FDA also proposed restricting the use in animals of all antibiotics used in human medicine to short-term therapeutic use prescribed by a veterinarian, unless the drug’s sponsor submitted data that demonstrated that agricultural use would not jeopardize human health.

    Faced with vigorous opposition in Congress by agribusiness and farm-state legislators, the FDA never implemented its proposed limits on antibiotic use in agriculture. Instead, Congress requested that additional studies be performed to determine the risk to human health of subtherapeutic uses of antibiotics in animals. In 1979, the Office of Technology Assessment issued a report that stated that the risk to human health from the use of antibiotics in animals could not be quantified, but that it must be regarded as a threat to the therapeutic value of antibacterials in both human and animal disease.

    In 1980, a National Research Council (NRC) committee concluded that the existing data did not prove that subtherapeutic use of antibiotics in animal feed caused a significant health risk. However, the committee acknowledged that a majority of studies stated that restrictions on the use of antibiotics in animal feed should be imposed. (When considering the findings of the NRC committee, it also should be noted that the chair of the committee, Raoul Stallones of the University of Texas School of Public Health, served as a paid consultant to several animal-drug companies. Another member of the committee was a consultant to American Cyanamid, one of the largest producers of medicated animal feed.)

    In 1984, Congressman John Dingell (D-MI) introduced the Antibiotic Preservation Act. The act called for restricting subtherapeutic uses of penicillin and tetracyclines in animal feed. Although hearings were held on the issue, the bill was never reported out of subcommittee.

    The mid-1980s was a time of new discovery and new activity concerning antibiotic use in animal feed. Researcher Thomas O’Brien and colleagues from Harvard Medical School published a study showing that antibiotic-resistance genes found in bacteria infecting humans were identical to some of those found in bacteria infecting animals. Holmberg and colleagues at the CDC linked an outbreak of antibiotic-resistant Salmonella in humans to beef cattle that had been fed subtherapeutic doses of chlortetracycline for growth promotion. The Holmberg results are noteworthy because that type of study is difficult to perform. By the time an outbreak is discovered, the contaminated food is usually long gone. Although none of the contaminated meat was available for sampling, all of the victims had consumed hamburger from the same source, strongly suggesting that the infections came from cattle from the South Dakota farm cited by the CDC.

    Lester Crawford, then director of the FDA’s Center for Veterinary Medicine, said that the Holmberg and O’Brien studies supported the FDA’s contentions about the human health risk of subtherapeutic use of antibiotics in agriculture and that the Holmberg study "was as good as they were going to get." Even professor (and industry consultant) Raoul Stallones said that if studies such as the one by Holmberg had been available at the time of the NRC review he might not have been as blasť about antibiotics as he was. "The conclusions that Holmberg drew are most reasonable," said Stallones. Referring to the link between antibiotics in animal feed and human illness, he added, "I think this is as close as one can get to a direct link."

    In 1984, the non-profit Natural Resources Defense Council (NRDC) petitioned the FDA to ban the subtherapeutic use of penicillin and tetracyclines in animal feed. Based largely on the new studies by O’Brien and Holmberg, the petition claimed that there was an "imminent hazard" posed by the use of those antibiotics in animal feed. The Department of Health and Human Services denied the petition on the basis that the NRDC failed to establish that the continued use of subtherapeutic levels of penicillin and tetracyclines in animal feed presented an imminent hazard to the public’s health that warranted immediate suspension of their approval. The "imminent hazard" standard places an unreasonably high burden of proof on the petitioner and almost guarantees that, without direct evidence of imminent and quantitative harm to human health, such a petition will be denied.

    An Institute of Medicine report that was published in 1989 concluded that there was considerable indirect evidence that feeding penicillin and tetracyclines in subtherapeutic doses to animals increases resistance to those antibiotics. Results of its risk assessment indicated that subtherapeutic uses of penicillin and/or tetracyclines result in about 40 human deaths per year.

    To date, the FDA has failed to take any action whatsoever on penicillin and the tetracyclines. The United States (and Canada) still permit their use. The European Union, Japan, Australia, and New Zealand have banned subtherapeutic use of penicillin and tetracyclines.

    The routine use of antibiotics such as penicillin and tetracyclines to promote growth has rendered those antibiotics less effective in treating animal diseases. For example, in Denmark, where the subtherapeutic use of tetracyclines and penicillin is banned, tetracyclines are effective in treating calf scours. In the U.S. tetracyclines are not effective against calf scours. As therapeutic options become less effective, drug companies and veterinarians have urged the approval of additional human-use antibiotics, such as fluoroquinolones, to treat animal diseases, jeopardizing the usefulness of a greater number of antibiotics for treating human diseases.

    Another likely contributor to the inappropriate use of antibiotics in livestock is that many antibiotics commonly used in human medicine (penicillin, tetracyclines, and tylosin, which is related to erythromycin) are available to farmers over the counter for the treatment of animal diseases. In contrast, those drugs are available to human patients only through a prescription from a physician. As a result, there may be no veterinary supervision in the treatment of animal disease. There are no safeguards to prevent a farmer who is concerned about the health of his or her livestock from giving antibiotics that may not be medically warranted. Additionally, because it is often difficult to treat individual animals in a flock or herd, the entire group may be dosed with antibiotics when only a few animals are sick. Dosing the entire herd or flock greatly increases the number of animals exposed to the antibiotic and, thus, increases the chances that antibiotic-resistant bacteria will emerge.

An initial report from the Minnesota State Department of Health shows that only two years after fluoroquinolones were approved for use in poultry, there has been an increase in fluoroquinolone resistance in Campylobacter bacteria isolated from chickens, turkeys, and humans.

    The lack of appropriate controls over agricultural uses of antibiotics continues to jeopardize the usefulness of antibiotics for treating human diseases. In 1995, despite vigorous opposition by the CDC and the Infectious Disease Society of America (IDSA), the FDA approved the use of fluoroquinolones to prevent E. coli infection in poultry., The CDC and IDSA opposed the use of fluoroquinolones in poultry feed or drinking water, an approach that doses many birds when only a few may have an infection. Fluoroquinolones are an important line of defense used in humans to treat urinary tract infections, sexually transmitted diseases, invasive Campylobacter and Salmonella infections, respiratory infections, infections in patients with cystic fibrosis, and many other diseases.

    To address some of the critics’ concerns, the FDA limited possible uses of fluoroquinolones in poultry to short-term use in limited situations. However, it did allow dosing of entire flocks through their drinking water. The FDA also mandated surveillance to determine if fluoroquinolone use in poultry contributes to the development of resistant bacteria. An initial report from the Minnesota State Department of Health shows that only two years after fluoroquinolones were approved for use in poultry, there has been an increase in fluoroquinolone resistance in Campylobacter bacteria isolated from chickens, turkeys, and humans. That finding, which is consistent with the CDC finding that 13 percent of human Campylobacter isolates are fluoroquinolone resistant, suggests that the use of fluoroquinolones in the drinking water of poultry poses a risk to human health.

    Despite the disturbing finding of fluoroquinolone-resistant Campylobacter, officials at the FDA may ignore that early warning sign and wait for evidence of fluoroquinolone resistance in Salmonella before reversing their approval of fluoroquinolones in poultry. (Fluoroquinolones are the preferred treatment for invasive Salmonella infections in adults, whereas other treatment options are available for invasive Campylobacter.) However, waiting for such evidence poses a significant risk to human health.

    The use of fluoroquinolones in livestock has already produced devastating consequences in the United Kingdom. A multiple-antibiotic-resistant strain of Salmonella typhimurium DT104 emerged that is resistant to ampicillin, chloramphenicol, streptomycin, sulphonamides, and tetracyclines -- antibiotics that are or were used commonly in animals. Before fluoroquinolones were approved for use in animals, DT104 was not resistant to those antibiotics. Only two years after approval of fluoroquinolones for agriculture, fluoroquinolone resistance was found in 16 percent of the Salmonella DT104 on farms. In 1996, fluoroquinolone-resistant Salmonella DT104 caused an outbreak among people who ate turkey at a restaurant. Thirteen people were sickened by the tainted birds and one person died. Judging from that experience in the U.K., it is only a matter of time before fluoroquinolone-resistant Salmonella are found first on poultry farms in the U.S. and then in hospital emergency rooms.

    Currently, 32 percent of reported Salmonella typhimurium cases in the U.S. (approximately 3,000 culture-confirmed infections) are the penta-resistant DT104 strain, up from seven percent in 1990. If fluoroquinolones are used more frequently in agriculture, and if those strains become resistant to fluoroquinolones, we could be faced with a situation where therapeutic measures are limited to treat those resistant Salmonella infections. Treatment measures would be more costly, because they might require hospitalization for administering intravenous antibiotics and could result in more side effects.

    Unless agricultural practices are changed, even the occasional new antibiotic will become obsolete. For instance, the new antibiotic Synercid is one of the last hopes against deadly antibiotic-resistant bloodstream infections. Although it has not yet been approved for use in humans, Synercid’s value already has been compromised because resistance to one antibiotic can cause resistance to others. Thus, researchers at Wayne State University have found Synercid-resistant bacteria in turkeys that had been fed another antibiotic, virginiamycin, to promote growth. If people consume meat that is contaminated with those resistant bacteria and become ill, Synercid would be of no use.

    Another use of antibiotics in agriculture is the spraying of crops with antibiotics to prevent and treat diseases. The species of bacteria that cause diseases in plants are different from the bacteria that cause diseases in humans, though they may belong to the same families. The main bacterial pathogen that harms plants is Erwinia, which attacks fruit trees causing fire blight, which damages the fruit. Erwinia is related to bacteria that include Escherichia coli, Salmonella, and Shigella, common foodborne pathogens that infect humans.

    Streptomycin is the antibiotic typically used to combat Erwinia, although in some regions, resistance has developed, necessitating the use of oxytetracycline. Although streptomycin is rarely used in humans, streptomycin resistance is still important because often it is found in conjunction with other antibiotic-resistance genes. Using streptomycin in agriculture may select for bacteria that are resistant to commonly used antibiotics. If those resistance genes are transferred to bacteria that infect humans, those resulting infections would be more difficult to treat.

Action plan to protect the usefulness of antibiotics

    A number of health organizations in the United States and around the world have recognized the problem of antibiotic resistance and have proposed recommendations to combat it. Some of those recommendations are outlined in Appendix 2. To date, few of the recommendations have been implemented in the U.S. The Center for Science in the Public Interest proposes the following action plan to protect the life-saving usefulness of antibiotics.

Increased funding is needed to combat antibiotic resistance

Congress should provide funding for public and professional education, national surveillance (of antibiotic use, the prevalence of antibiotic-resistant bacteria, and the diseases they cause), research and development, and immunization, and should support international efforts to reduce antibiotic resistance.

Actions regarding medical uses of antibiotics

The DHHS should include in Healthy People 2010 national targets for decreasing inappropriate antibiotic use and reducing antibiotic resistance. Setting nationwide goals to stem antibiotic resistance would:

The FDA should alter its policies on antibiotic advertising to help reduce inappropriate uses of antibiotics. Restrictions on ads for antibiotics are as -- or more -- important than for other drugs, because overuse of antibiotics affects not just the patient, but the entire community.

Congress should require Medicare and Medicaid, Veterans Administration hospitals, and military hospitals -- and state governments should require health maintenance organizations and private insurance companies -- to cover the costs of bacterial cultures and susceptibility testing for sore throat, cough, and certain other infections, to help ensure that antibiotic prescriptions are necessary and appropriate.

For bacterial infections for which cultures generally cannot be obtained (otitis media, sinusitis, bronchitis, etc.) and for viral infections (colds and flu), the DHHS should develop and distribute practice guidelines to decrease unnecessary antibiotic prescriptions. The DHHS also should develop and distribute to physicians who oversee nursing-home patients practice guidelines for treatment of infections in the elderly.

The government should use its own health-care facilities -- such as VA, defense department, and prison medical clinics and hospitals -- to serve as showcases for the most prudent antibiotic-use practices.

The DHHS should exempt effective rapid strep tests from the Clinical Laboratory Improvement Act of 1988 (CLIA) to enable physicians to perform strep tests in their offices.

Hospitals and nursing homes that receive Medicare or Medicaid reimbursement should be required to offer vaccinations against Streptococcus pneumoniae to patients at risk of pneumococcal infection. The protective efficacy of the vaccine against invasive pneumococcal infections is estimated to be 60 to 70 percent.

Actions regarding agricultural uses of antibiotics

The FDA should ban all subtherapeutic uses of antimicrobial agents that (a) are used in human medicine or (b) might select for cross resistance to antimicrobials used in human medicine. For example, approvals for subtherapeutic use of penicillin and tetracyclines should be revoked. New antibiotics should not be approved for subtherapeutic use, and new formulations of already approved antibiotics should not be approved for subtherapeutic use.

The EPA should ban the use of antibiotics as pesticides.

The DHHS should fund research on the impact of agricultural antibiotic use on human health, as recommended in 1995 by the Office of Technology Assessment. The research should include identifying associations between antibiotics used in animal feed and antibiotic-resistant bacteria in humans.

The USDA should fund research on alternatives to antibiotics for growth promotion and disease prevention in livestock.

The USDA should publish practice guidelines to educate producers about alternatives to antibiotics for growth promotion. That information could be disseminated to producers through cooperative extension services and other outreach efforts.

The FDA should develop a symptom-based formulary for veterinarians that describes appropriate treatment for common livestock infections. The treatment guidelines should be based on current scientific data and susceptibility patterns.

The FDA should repeal approval of fluoroquinolones in poultry and should only allow additional approvals of fluoroquinolones if the company can show that those uses would not reduce their effectiveness in human medicine.

Congress should investigate why the FDA’s Center for Veterinary Medicine ignored the CDC’s recommendation regarding the approval of the use of fluoroquinolones in poultry.

The FDA should require that antibiotics for therapeutic use in livestock be available only through prescription by a veterinarian.

Appendix 1: Ten common antibiotic-resistant bacteria


Diseases caused

Drugs resisted

Staphylococcus aureus bacteremia (blood infection), pneumonia, surgical-wound infections Chloramphenicol, Rifampin, Methicillin, Ciprofloxacin, Clindamycin, Erythromycin, Beta-lactams, Tetracycline, Trimethoprim
Streptococcus pneumoniae meningitis, pneumonia, otitis media (ear infection) Aminoglycosides, Penicillin, Chloramphenicol, Erythromycin, Trimethoprim- Sulfamethoxazole
Mycobacterium tuberculosis tuberculosis Aminoglycosides, Ethambutol, Isoniazid, Pyrazinamide, Rifampin
Haemophilus influenzae epiglottitis, meningitis, otitis media, pneumonia, sinusitis Beta-lactams, Chloramphenicol, Tetracycline, Trimethoprim
Enterobacteriaceae (e.g. Klebsiella pneumonia, Escherichia coli, Salmonella) bacteremia, pneumonia, urinary-tract or surgical- wound infections, diarrhea Aminoglycosides, Beta-lactams, Chloramphenicol, Trimethoprim
Enterococcus bacteremia, urinary-tract or surgical-wound infections Aminoglycosides, Beta-lactams, Erythromycin, Vancomycin
Neisseria gonorrhoeae gonorrhea Beta-lactams, Penicillin, Spectinomycin, Tetracycline
Pseudomonas aeruginosa bacteremia, pneumonia, urinary-tract infections Aminoglycosides, Beta-lactams, Ciprofloxacin, Tetracycline, Sulfonamides
Bacteroides septicemia, anaerobic infections Penicillin, Clindamycin
Shigella dysenteriae severe diarrhea Ampicillin, Trimethoprim-Sulfamethoxazole, Tetracycline, Chloramphenicol

Appendix 2: Previous recommendations by expert bodies

The Institute of Medicine’s 1998 report on antimicrobial resistance identified issues and options for (1) improving national and international surveillance, (2) understanding the use of antibiotics in food production, (3) prolonging antibiotic effectiveness, (4) developing new products, and (5) legal and regulatory approaches. The report recommended:

The World Health Organization (WHO), in October 1997, concluded that excessive use of antimicrobials, especially as growth promotants in livestock, presents a growing risk to human health. The WHO recommended that:

The U.S. Office of Technology Assessment (OTA) in 1995 made 14 recommendations to prolong the effectiveness of antibiotics or to develop new antibiotics. Although that report was requested by members of Congress, no hearings were held to discuss its findings and no legislation was introduced or passed to implement its recommendations. Recommendations included:

The American Society for Microbiology in 1994 recommended that:

The Institute of Medicine’s 1992 report on emerging infections expressed concerns about agricultural conditions and practices contributing to foodborne illnesses due to antibiotic-resistant bacteria. The report recommended: