Recent revelations about the safety and efficacy of the use of antidepressants by children and teenagers have brought public attention to the practices of pharmaceutical companies in reporting the results of clinical trials. Although some trials sponsored by these companies failed to demonstrate the effectiveness of antidepressants in pediatric populations and indicated that children and adolescents taking antidepressants were more likely to commit suicide and to have suicidal thoughts, the companies making these drugs did not release those findings. Now that the pharmaceutical companies’ attempts to conceal the results of these trials have become public, the claim is being made that pharmaceutical companies have an obligation to release unfavorable as well as favorable results of their clinical trials. This is an intuitively plausible claim. It is further supported by the recent movement of twelve leading medical journals to require the registration of a clinical trial upon its commencement in order for an article containing its results to be included in their publications. However, many questions need to be asked about the nature and the implications of this purported obligation.

To whom? On what basis?
If pharmaceutical companies have an obligation to report both the favorable and unfavorable results of the clinical trials they sponsor, who is the subject of that obligation? That is, to whom do pharmaceutical companies have this obligation? Further, where does that obligation come from? What generates it? There are numerous possible subjects of the pharmaceutical companies’ obligation and numerous possible sources of that obligation. In the following paragraphs, I identify several different ways of conceptualizing this obligation and explain why each is, at least at first glance, plausible.

Special relationship. The obligation in question could be generated by the special relationship that exists between pharmaceutical companies and research subjects. This relationship could include the research subjects who participated in the specific clinical trial whose results should be released or the subjects who have participated in any of that pharmaceutical company’s clinical trials. One could argue that pharmaceutical companies are indebted to the human subjects who made the development of their products possible and that they therefore have a duty to protect at least those subjects from harms that those products are known to cause by making all results, especially unfavorable ones, available to them. When understood this way, companies may have an obligation to release results to their research subjects, but not necessarily to release them publicly.

This argument for an obligation of disclosure may be particularly compelling in the case of research subjects participating in non-therapeutic trials. These subjects take on the risk of participating in the trial for little or no benefit of their own. As a result, pharmaceutical companies may have stronger or more robust obligations to them than they have to participants in therapeutic trials.

Pharmaceutical companies and physicians may also have a relationship that generates the obligation to release clinical trial results. At least for prescription drugs, physicians serve as gatekeepers between pharmaceutical companies’ products and the consumers who purchase them. Because physicians play this vital role in the market, they have a special relationship with pharmaceutical companies. One could argue that physicians generally seek to promote the well being of their patients, but cannot do so effectively if they lack significant information about the treatments they prescribe. It seems plausible that pharmaceutical companies are obligated to help physicians help their patients by ensuring that they have the best available information about their products.

Implicit contract. A second way to understand this obligation is as an implicit contract between the pharmaceutical company and the research subjects who took part in the clinical trial. The releasing of trial results is not generally part of the explicit agreement made between trial sponsors and research subjects when the subjects enroll in a trial. It seems plausible, however, to view it as part of an implicit agreement. This implicit contract could exist between the pharmaceutical company and a) the subjects who participated in the trial whose results may be released, or b) the subjects who have participated in any of the company’s trials. At least some research subjects choose to participate in clinical trials with altruistic motivations. These subjects agree to enroll on the assumption that the results of those trials will be used to further the goal of making safe and effective treatments available and to promote the well being of themselves and others. Because the release of both favorable and unfavorable trial results is likely to (although in some cases may not) contribute to these goals, the requirement to release that information may be included the implicit contract made when human subjects agree to take part in clinical trials.

A second type of implicit contract that could serve as the foundation of the pharmaceutical companies’ obligation is an implicit contract with physicians. Physicians choose to prescribe drugs to their patients based upon the information provided to them by the pharmaceutical companies with the expectation that that information is adequate to make appropriate treatment decisions. They form treatment recommendations from their knowledge of the drugs’ risks and benefits, which are specified by the pharmaceutical companies. If information material to such treatment decisions were withheld, it is possible that physicians could legitimately claim that some breech of an implicit contract occurred.

A final version of an implicit contract that could create the obligation in question is a contract between pharmaceutical companies and society at large. The development of drugs is made possible not only by the company’s labs and researchers, but is built upon many years and millions of dollars of basic science research that has been supported by public funds. Although an explicit agreement to promote the public’s well being in return may not have been made when those funds were allocated or when pharmaceutical companies chose to utilize their findings, it does not seem implausible that such an implicit agreement exists. Making both favorable and unfavorable results of clinical trials available could be one way of promoting the public’s well being and therefore could be covered by such a contract.

There are several difficulties with all three of these contractual accounts of the pharmaceutical companies’ obligation. Problems with implicit contracts of this type include ambiguities about when they are made, what they include, how the participating parties know that they have been created, and how they know what is included in the contract. All of these ambiguities plague a contractual account of the obligation of pharmaceutical companies to release both favorable and unfavorable results of their trials. Nonetheless, it seems like a theoretically plausible source of that obligation.

Duty to warn. People must have access to accurate and complete information about both the chances of benefit and the risks of harm posed by a drug in order to make choices that are in their best interests. The possible obligation of pharmaceutical companies to release clinical trial results could come from a duty to warn consumers of the dangers of products they produce so that the consumers have the information they need to protect themselves. This duty has been established in other industries; consider, for example the Firestone tire recall or warnings on cigarette and alcohol packaging. It seems likely that pharmaceutical companies have an analogous obligation to inform their consumers about the risks associated with their products.

Vulnerability. A fourth possible account of the obligation that pharmaceutical companies may have to release clinical trial results is the vulnerability of their consumers. Those individuals who consume the pharmaceutical companies’ products are often vulnerable in a way that threatens their happiness, their ability to pursue their life projects, or their lives themselves. Perhaps the fact that pharmaceutical companies have a unique ability to help ameliorate these fundamental vulnerabilities means that they have a special obligation to do so. We generally hold that physicians have special obligations to act in their patients’ best interests because of the vulnerability of their patients, that is, the patient’s inability to protect or help themselves. Pharmaceutical companies deal with these same vulnerabilities. Perhaps they therefore have analogous moral responsibilities, including the obligation to disclose all available information about the risks and benefits of the drugs they sell. The subject of the obligation according to this account would be the drugs’ consumers, rather than the research subjects or physicians.

Risks and benefits. One ethical requirement for research with human subjects is a favorable risk-benefit ratio, that is, the risks of the research must be reasonable in relation to the expected benefits. The disclosure of both favorable and unfavorable information is likely to increase the benefits of a drug’s use and to decrease the harms, and therefore to improve the risk-benefit ratio. It seems plausible, then, that pharmaceutical companies have a moral obligation release their results because doing so could help ensure that the use of human subjects in their research is ethically justifiable. Although research is prospectively, rather than retrospectively justified, an obligation to promote benefit to a drug’s consumers is consistent with this requirement. According to this model, there is no specific subject of the pharmaceutical companies’ obligation. In other words, this is simply a free-floating moral obligation rather than a duty to a particular person or set of persons.

This cursory discussion suggests that there are many possible subjects and sources of this obligation. Establishing who could be the subject of the pharmaceutical companies’ obligation and determining the foundation upon which that obligation could be based may shed light upon the details of that obligation. However, the work necessary to determine which of these possibilities stand up to more rigorous analysis is beyond the scope of this article. I will therefore simply identify some of the many questions about the content of pharmaceutical companies’ possible obligation that will need to be explored once a defensible account of that obligation is found.

To do what?
What is the nature of the obligation that pharmaceutical companies may have to disclose the results of their clinical trials? That is, what, specifically, does this duty require of pharmaceutical companies? There are at least three elements of disclosure that must be considered: the information that should be released, the individuals or institutions to whom it should be released, and the method by which it should be released.

Pharmaceutical companies are generally eager to release results with favorable findings. But what does it mean to be required to release ‘unfavorable’ findings? Results demonstrating that a particular drug can cause harm clearly qualify. Sometimes, however, studies show that a drug is effective, but simply less effective than another drug. And sometimes the research fails to produce any statistically significant results – either positive or negative. Do such results count as unfavorable? Should pharmaceutical companies be obligated to release all or some of these kinds of results? Further, how many studies’ results should be released? Every study? Only the largest ones? Only those that novel results?

To whom should study results be released? Perhaps only the subjects of the clinical trial in question should be informed. Perhaps the results should be offered to any consumer of the drug. Physicians and the Food and Drug Administrations may also have a moral claim to that information. Alternatively, the pharmaceutical company may be morally required to make the results available to the public in general.

A third factor that determines the content of the pharmaceutical companies’ possible obligation is the method of disclosure. How should this information be released and distributed? How many resources must pharmaceutical companies invest to make the results known? Is issuing a press release sufficient? Publishing the findings in an academic journal? Perhaps companies are morally required to be more proactive, incorporating findings of recent clinical trials into their product packaging or informing physicians of relevant results using methods similar to those they use to market their products.

Once untangled, this web of considerations will give content to the obligation that pharmaceutical companies may have to release both favorable and unfavorable clinical trial results. These are difficult questions in need of answers. However, working through the questions raised above about the subject and source of this obligation can provide guidance. By understanding the moral foundation upon which the obligation is based, it may be possible to derive a theoretically and practically defensible set of answers to these difficult questions.