Having looked at the rampant conflicts of interest surrounding the anthrax vaccine and a bit at the logistics of the proposed pediatric vaccine trial let’s look at the trial in the context of ethical principles.

Ethical context

The need for ethical guidelines came to vivid attention during World War II, when the Nazis tortured many of their victims under the guise of conducting research. In 1946, 23 doctors were indicted; in 1947, at the Nuremberg trials, 7 of these were sentenced to death and 16 were imprisoned. The verdict included a section on “Permissible Medical Experiments,” subsequently known as the “Nuremberg Code.” This code for conducting research required that participants’ consent be voluntary and that the risks to those individuals should be understood and weighed into their decision to participate.

Over the next 20 years, more ethical lapses in studies came to light. It was only 1964 before the World Medical Association, in a somewhat tardy response to the Nuremberg horrors, formulated the Declaration of Helsinki. This code differentiated between therapeutic and nontherapeutic research. It, too, required informed consent, but it expanded this practice to allow for others to provide (surrogate) consent when the research participant was incapable of providing the consent (because of physical or mental incompetence). In 1974, federal regulations expanded consent requirements, allowing surrogate consent from a legal representative but adding protections for subjects by requiring the participant’s “assent” in the case of a consent from a surrogate. (Assent is the willingness to participate indicated by an individual who is not entirely competent in a legal definition.)

The Declaration of Helsinki and the subsequent Ethical Principles and Guidelines for the Protection of Human Subjects of Research (aka the “Belmont Report”) in 1979, form the basis for most of the FDA and ICH good clinical practice guidelines. The Belmont report concluded that all clinical research should meet three general principles: respect for persons, beneficence, and justice.

Respect for Persons

The first Belmont principle is respect for persons. This refers to an individual’s autonomy, or right to make decisions for himself or herself based on a review of available information. In other words, the individual is given enough information about the risks and benefits of a trial to make a decision about participating. This principle is expressed in the elements of the research informed consent, which requires that

Information necessary to make a decision must be presented—that is, the risks and benefits, if any, of participation.

The information must be presented at a level that can be understood by the patient or study subject.

Participation must be voluntary, and be free of coercion or pressure.


The second Belmont principle, beneficence, requires that a study should not only meet the maxim “First, do no harm” but should also provide some benefit to the subject. One of the interesting dilemmas here is whether the benefits outweigh the risks. Who decides? Does the volunteer, does the IRB, or does some well-meaning but paternalistic governmental agency?

The other half of the principle of beneficence “requires that we protect against risk of harm to subjects and also that we be concerned about the loss of substantial benefits that might be gained from research.” This means that researchers should avoid being too paternalistic (with well-intentioned safeguards), preventing patients access to therapies that some might view as too risky and thus depriving them of any possible benefits from the treatment. AIDS activists used this principle to challenge the lack of access to potential AIDS therapies in the 1980s.


The third Belmont principle is that of distributive justice, meaning that the risks and gains from research participation should be equitably distributed among different populations. Historically, this has been interpreted to mean that classes of vulnerable patients, such as the indigent, the institutionalized, the disabled, minorities, pregnant women, and children, would not be subjected to risks because they were readily available or defenseless (as in the Tuskegee experiment). At the same time, an obligation to help vulnerable subjects means that this protection must be balanced with attempts to study some of these same populations to determine whether they have unique needs. For example, it is now known that women metabolize drugs differently than men do. So while one might want to protect women (especially if pregnant) from drug toxicities, excluding women from trials might also prevent them from benefiting by having the drugs prescribed in the most efficacious manner. Brief examples of research on vulnerable populations follow.

Vulnerable Populations: Children

The proposed pediatric anthrax trials perfectly illustrate some of the ethical concerns involving research on children. On the one hand, children have unique needs and cannot simply be considered “pint-sized people.” On the other hand, children may be less capable of understanding risks and the consequences of their actions, given their educational experience, emotional maturity, and limited life experiences, than are many adults. Finally, because they are minors and in a dependent relationship with their adult caretakers, children are more susceptible to coercion. Because of these factors, they are considered to be vulnerable research subjects and historically were protected from research.

Until the late ‘90s, children’s medicines were dosed by guess and by gosh. There was no data; dosages were extrapolated based on weight. One 1994 survey examined the 10 drugs most commonly prescribed for children that lacked pediatric labeling—these drugs were prescribed more than 5 million times. Three were for asthma; albuterol inhalers alone were prescribed more than 1.6 million times. Other medications included ampicillin, antidepressants, and Ritalin.

In 1997, under the FDA Modernization Act, Congress offered pharmaceutical companies large incentives to do pediatric testing, in recognition of the unique characteristics of children. Congress gave the manufacturers an extra 6 months patent protection and exclusive marketing in exchange for testing specific drugs in children. The Pediatric Studies Rule, passed by Congress and published in 1998, required that new drugs that will be commonly used or important for treating children include specific pediatric labeling information.

The Institute of Medicine sponsored a roundtable in 1999, called Rational Therapeutics for Infants and Children, to explore the direction such research should take. However, the Pediatric Studies Rule testing requirement was subsequently challenged in court in a typical “industry versus regulatory” battle. In 2002, a U.S. District Court ruling barred the FDA from enforcing it, saying the FDA had exceeded its statutory authority.

Fortunately, increased recognition of the unique needs of this population is resulting in slow improvement. Under the Best Pharmaceuticals for Children Act of 2002 (BPCA 2002), the National Institute of Child Health and Human Development (NICHD) was given the authority and responsibility for pediatric drug development. The NICHD responded with a priority list of drugs including antibiotics (azithromycin, ethambutol, rifampin, acyclovir), diuretics (furosemide), antihypertensives (hydrochlorothiazide), antidepressants (bupropion), and anticancer agents (methotrexate, vincristine). Studies conducted under BPCA showed that prior dosing really put children at significant risk.. The studies showed significant differences in absorption, distribution, metabolism, and elimination of drugs, varying with both the age and sex of the patients. No predictable pattern was seen in the required dosing changes; some doses were ineffective, others initially toxic. Fully 20 percent of the trials showed no efficacy in children, and another 20 percent revealed new side effects that were previously unknown. For example, five studies in children showed that sumatriptan, a very effective treatment for migraines in adults, not only didn’t work in kids but led to serious side effects, such as stroke and loss of vision. Consequently, 87 percent of drugs approved for pediatric use had subsequent changes in labeling. As of 2007, post-marketing safety reporting is also required.

Pediatric trials should now be integrated with adult trials, not done later; they continue under the guidance of the FDA. Attempts have been made by the American Academy of Pediatrics and the International Conference on Harmonisation to develop ethical guidelines for studies in children. Recommendations include the following:

Ideally, the pediatric subject should have the potential to benefit from participation in the trial.

The study must take into account the unique characteristics and needs of children and their special needs as research subjects.

The study design should also take into account racial, ethnic, gender, and socioeconomic characteristics of the population being studied.

The research and procedures should be explained to the subject in age- appropriate language.

Children aged seven and older should assent in addition to full informed consent being obtained from their parent or guardian. The children should also understand that they can withdraw from the study except in life-threatening circumstances.

Compensation should be “token,” so as not to be coercive.

Pediatric anthrax vaccine

How does the proposed pediatric vaccine trial fit with the precepts outlined above?

First, the Belmont principle “respect for persons” is violated, in that this vaccine trial, using the adult anthrax vaccine (AVA) has known serious risks and other potentially unknown risks in kids, and no benefit. We won’t even know if any antibodies formed would be protective.

Belmont principle of benefits to the participant is violated since there are none. Furthermore, U.S. regulations (45 CFR 46, Subpart D) state that children may not be exposed to risk if a medical experiment does not offer them a direct benefit. Even the NBSB report notes “Currently, U.S. children are not at immediate risk from anthrax and would not benefit directly from pre-event AVA [anthrax vaccine] administration”…There is only the potential future benefit. This is further outlined by the Institute of Medicine and by Dr. Meryl Nass, an ardent critic of the anthrax vaccine.

And it appears that the Belmont principle of justice will be violated. A vulnerable population will be used without assent. (How many kids would willingly take a painful course of five injections, even without worry of other side effects?) How will the children’s continued participation be ensured? Coercion? Or bribes? Reportedly, the study is to be done in New York City, where there are 9/11 first responders who would be willing to volunteer their children. These parents, traumatized by 9/11, are unlikely to be able to review an informed consent form critically or objectively. Would there be such enthusiastic parents in rural America? I doubt it. I like the biting suggestion by Dr. Anthony Robbins, “How about the children of people who have the national security clearance required for the government to share with them all the evidence that adds up to a “credible threat?” With that information in hand, these parents would be able to make the choice-an informed decision for their children-that the rest of the public surely cannot.”

Dr. Daniel Fagbuyi, chairman of the NBSB’s Anthrax Vaccine Working Group and Medical Director, Disaster Preparedness and Emergency Management Children's National Medical Center, used fear-mongering and misrepresentation when he urged, “Or do we want to say: ‘How do we best protect our children?’ We can take care of Grandma and Grandpa, Uncle and Auntie. But right now, we have nothing for the children.” That pulls heart-strings, but made me further question his credibility.

As Dr. Fagbuyi should know, we don’t know that we can take care of Grandma and Grandpa either, as the vaccine has not been tested in elderly populations. In fact, it is well known that the elderly have a generally poor response to vaccines (“immune senescence”); influenza vaccine is given to those >65 years old in a dose that is 4x more potent.

Given the concerns I outlined above and in Parts 1 and 2 of this series, I am shocked at the National Biodefense Science Board (NBSB) recommendation to proceed with this trial and can’t really fathom their reasoning. The decision whether to proceed with this trial was recently referred to the President’s Bioethics Committee; a transcript of their meeting is available here. I hope they will ask hard questions—as who will really benefit from this proposal?


“Molecules to Medicine” banner Michele Banks

Syringe image by Debora Cartagena/CDC

U.S. Navy photo by Mass Communication Specialist Seaman Matthew Jackson via Wikimedia

Baby receiving injection by Amanda Mills/CDC

Some material in this post is adapted from my book, Conducting Clinical Research: A Practical Guide for Physicians, Nurses, Study Coordinators, and Investigators.