A Bitter Pill to Swallow: Nonadherence With Prophylactic Antibiotics During the Anthrax Attacks and the Role of Private Physicians

Bradley D. Stein; Terri L. Tanielian; Gery W. Ryan; Hilary J. Rhodes; Shalanda D. Young; Janice C. Blanchard


Biosecur Bioterror. 2004;2(3) 

In This Article


This study is the first to examine individual decisionmaking during an emergent public health crisis. People had to decide whether or not to follow public health recommendations in an environment characterized by uncertainty. Despite the fact that the nature of the exposure and the official responses to the event were quite different across the two sites directly affected in the Washington, DC, area, the decision-making processes of HSOB and BPF participants were similar. Participants used information and support from a variety of sources, as they weighed the risk from anthrax resulting from their level of exposure and their susceptibility against their perceptions of the advantages and disadvantages of treatment.

Challenges faced by public health officials in communicating the risk from anthrax and recommended actions with the public during the anthrax attacks have been described elsewhere.[1,8,15] One particularly challenging area of communication concerns recommended changes in protocols for screening or treating at-risk individuals; changes in protocol can greatly increase confusion or be perceived as inequitable when not communicated appropriately.

Despite these communication problems, however, fear and uncertainty about anthrax led almost all participants to begin antibiotics as recommended. But over time, information and support provided by family, friends, coworkers, and private physicians became increasingly important as participants reconsidered their decisions. Public health officials should anticipate that people will turn to other sources of information and support in a crisis, including family, friends, coworkers, private physicians, and the media. Efforts at federal, state, and local levels to provide risk communication guidelines and training to public health officials may improve their ability to communicate in an initial crisis response.[16,17] However, our findings suggest that officials should seek to give people reliable information throughout the entire period of risk, not just during the initial crisis response.

In making decisions, participants used not only what public health officials told them but also were influenced by officials' actions or inaction. The most prominent example was the nasal swab test. Despite announcements that the nasal swab test is an epidemiologic screening tool and should not be used for disease diagnosis or treatment, participants inaccurately believed that nasal swab testing could confirm whether they "had anthrax." Despite the fact that one of the postal workers who died had a negative swab test, many of those with negative results believed it meant that they had not been exposed and didn't need to take the antibiotics. In contrast, participants with positive swab results acted as if it confirmed infection, taking all antibiotics as recommended as well as the anthrax vaccination. And participants who did not receive their results were uniformly concerned or angered by that lack of information.

Diagnostic tests and treatment are familiar to most people; epidemiologic tests and prophylactic medications are not. Public health officials should be aware that even with better education and personal guidance regarding epidemiologic surveillance and prophylaxis, many people may still have difficulty grasping these concepts. Our data suggest that the nasal swab tests became an important tool for participants in judging personal exposure. Empirical studies are needed to examine whether people falsely reassured by negative epidemiologic tests are less likely to follow public health recommendations.

Even with superb communication, a public health crisis involving an emergent infectious disease will be characterized by insufficient information and conflicting opinions. Changing recommendations are likely as more is learned during outbreak investigations.[18] Our participants described how these changing recommendations were often perceived as conflicting information and advice, making it difficult to determine the appropriate course of action regarding antibiotics. The role of private physicians in individual health care decision-making must also not be underestimated; the overwhelming majority of our participants followed their physician's advice. Unfortunately, the difficulty in determining the best course of action may not be limited to at-risk individuals. Many participants reported that physicians contradicted public health recommendations regarding antibiotic prophylaxis, often providing vague or uncertain information about the risks of anthrax and the benefits of prophylactic antibiotics. As one postal worker related, "I told [the doctors] 'they put me on amoxicillin because I was allergic to the other two [antibiotics].' And the doctors told me that I was in God's hands because amoxicillin does not protect you against anthrax."

Our findings are consistent with a survey of emergency physicians conducted in the months following the anthrax attacks that found that many were uncertain about who should receive antibiotic prophylaxis and the utility of nasal swabs in diagnosing anthrax.[19] More recent findings indicate that only 24% of private physicians believe they are personally prepared to respond to bioterrorism.[20] Physicians must have accurate, up-to-date information regarding any emergent infectious disease so that they can appropriately advise patients about compliance in the face of invisible and potentially dormant health risks. During the anthrax crises, efforts were made to send fax communications to area hospitals and physician offices; however, at the time of our analyses these communications were no longer available. Future studies should examine what information public health officials provide to private health care physicians, how physicians assimilate this information during a public health crisis, how the information affects physician behavior during public health crises, and how physicians communicate this information to patients.

The challenges inherent in integrating private physicians and the public health system to respond to a public health crisis extend beyond those related to more effective communications. The primary responsibility of private physicians and other clinicians is their patients' health; the primary responsibility of public health officials will be to craft a population response that balances overall risks of illness with other factors -- for example, how best to distribute a limited supply of antibiotics or vaccines, or how to minimize the risk of infection to uninfected individuals. In some situations, there may be fundamental tensions between appropriate advice on a broad public level and the needs of at-risk individuals for expert advocates for their own health. Clinicians may also be appropriately concerned about potential liability for "poor" clinical decisions based on public health recommendations. A thorough examination of these issues and proactive steps to address some of the fundamental tensions that may arise during a public health crisis are likely to enhance the effectiveness of a public health response during a crisis.

Perceived disparities in the response between HSOB and BPF have been discussed elsewhere.[1,7] A number of our participants mentioned how these perceptions may have resulted from the racial and socioeconomic differences between HSOB and BPF personnel. We found no substantial differences in the frequency of such statements between adherent and nonadherent individuals. However, it should be noted that perceived disparities in public health actions as a result of racial and socioeconomic factors have the potential to affect many components of a public health response to a bioterrorism event, including the sources that individuals turn to for advice and whether they follow instructions from officials.[8,21] Additional studies are needed to provide a more comprehensive picture of this important issue.[22,23]

Limitations to our study include the interim period of 18 months (on average) between the initial exposure and the interviews, during which time other events and experiences may have influenced participants' recall of events and factors affecting their decisions. We have only the participants' descriptions of the discussions with private physicians; without interviewing private physicians, we are unable to verify the information that physicians actually provided to their patients. Biases about the process may have made people more likely to volunteer for the interviews, and our convenience sample is not intended to be statistically representative of a larger group of exposed people. Rather, the interviews seek to capture the range of peoples' experiences, thereby deepening our understanding of how people make decisions in an emergent public health crisis.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.