By LAWRENCE K. ALTMAN

TLANTA, Oct. 29 — Most postal, government and media workers who were advised to take an antibiotic to prevent inhalation anthrax a year ago failed to complete the full course of therapy, a federal study has found.

No cases of anthrax developed, including among those workers who took no antibiotics at all.

Advertisement

Adverse reactions, a perceived low risk of developing anthrax and a fear of long-term side effects were the reasons workers cited for not fully complying with the government's recommendations, said the epidemiologists at the Centers for Disease Control and Prevention here who conducted the study.

The study, whose results were reported today, found that 56 percent of those advised to complete the course did not. It also found that mild and transient adverse reactions like headache, dizziness, nausea, vomiting, stomach aches and diarrhea occurred commonly among people taking the antibiotics prophylactically.

But serious effects requiring hospitalization were rare, Dr. Colin W. Shepard, the study's leader, said in reporting the findings.

This is believed to be the first large-scale study of prophylactic antibiotic therapy for inhalation anthrax.

Federal health officials had recommended a 60-day course of an antibiotic for more than 9,300 workers who were exposed to anthrax spores in four states and the District of Columbia in 2001. In making their recommendation, officials had to weigh the benefit of preventing cases of anthrax and the risks of side effects from the drugs themselves.

About two-thirds of the 9,300 exposed workers, or 6,178 people, participated in the study.

Thirteen percent of the participants, or 787 people, did not fill the original prescription for the antibiotic or did not take a single pill if they did. Because of this, health officials raised questions about whether antibiotics were taken by too many people with little, if any, risk of developing anthrax.

The study did not include those people who were told to stop taking antibiotics after laboratory tests determined that they had not been exposed to anthrax spores and so were not at risk for the disease.

Dr. Shepard said that his team was confident that antibiotic prophylaxis had prevented some cases of anthrax among those exposed to the letter sent to Senator Tom Daschle, the majority leader, because of the heavy exposure of spores in the Senate and the Brentwood postal processing center in Washington.

But Dr. Shepard said it was "impossible to determine how many cases of anthrax prophylactic antibiotic therapy prevented."

The antibiotics available to the exposed workers were Ciprofloxacin, Doxycycline and Amoxicillin, licensed drugs that the government distributed free of charge to workers. Ciprofloxacin was the drug used most often in the initial days of the anthrax attacks. It can cause convulsions as well as other adverse effects on the brain and central nervous system, and also lead to rupturing of the Achilles tendon and others.

To reduce the risk of bacteria becoming resistant to Ciprofloxicin, a powerful and key antibiotic, officials advised many people to switch to another antibiotic, doxycycline. Dr. Shepard said health officials did not know what people had done with the pills they did not take and had not taken steps to get them back.

The low rate of compliance and other findings have serious implications for future bioterrorist attacks, Dr. Shepard and other health officials said. The study, they said, illustrates the need to give clear messages at the time of an outbreak about real and perceived risks in an effort to improve compliance.

Health officials said they needed to improve assessment of an individual's risk of developing inhalation anthrax after exposure to spores as a way of determining whether they needed to take antibiotics prophylactically.

Dr. David Sencer, a former director of the disease control centers, and Dr. Ralph Henderson, a former official of the World Health Organization, told participants at a meeting today that the findings underscored the importance of communicating risks to individuals and of counseling them about adverse events.

Dr. Shepard and Dr. Henderson said the low rate of compliance in the anthrax study was comparable to that found in other studies of antibiotic use.

The anthrax study was conducted at six sites where spores were released in Connecticut, Florida, New Jersey, New York and Washington, D.C.

Of the workers who took at least one antibiotic pill, 16 percent, or 855, sought medical attention. In 5 percent, or 267 cases, their doctors stopped the antibiotic because of adverse reactions.

All four of the definite or probable severe adverse reactions resulted from symptoms suggesting a potentially fatal type of allergy known as anaphylaxis.