Informed consent in human experimentation before the Nuremberg codeBMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7070.1445 (Published 07 December 1996) Cite this as: BMJ 1996;313:1445
- Correspondence to: Professor Vollmann.
- Accepted 28 October 1996
The issue of ethics with respect to medical experimentation in Germany during the 1930s and 1940s was crucial at the Nuremberg trials and related trials of doctors and public health officials. Those involved in horrible crimes attempted to excuse themselves by arguing that there were no explicit rules governing medical research on human beings in Germany during the period and that research practices in Germany were not different from those in allied countries. In this context the Nuremberg code of 1947 is generally regarded as the first document to set out ethical regulations in human experimentation based on informed consent. New research, however, indicates that ethical issues of informed consent in guidelines for human experimentation were recognised as early as the nineteenth century. These guidelines shed light on the still contentious issue of when the concepts of autonomy, informed consent, and therapeutic and non-therapeutic research first emerged. This issue assumes renewed importance in the context of current attempts to assess liability and responsibility for the abuse of people in various experiments conducted since the second world war in the United States, Canada, Russia, and other nations.
First Prussian directive on informed consent
The introduction of scientific and experimental methodology into clinical medicine in the nineteenth century brought with it an increased demand for experimentation on human subjects, particularly in bacteriology, immunology, and physiology. This research was done mainly on patients in hospital, often without their consent, under an “ethos of science and medical progress.” As a result of injury to some patients subjected to non-therapeutic research, however, controversy and public debate ensued about the ethics of human experimentation.1 2 3 4
In 1891 the Prussian minister of the interior issued a directive to all prisons that tuberculin for the treatment of tuberculosis “must in no case be used against the patient's will.”5 But the first detailed regulations about non-therapeutic research in Western medicine came from the Prussian minister for religious, educational, and medical affairs in 1900. They were issued after critical public discussion and political debate on the Neisser case in the Prussian parliament and set forth the legal basis of disclosure and unmistakable consent.1 2 Of particular interest is the debate within the medical profession and the political circumstances.
The Neisser case
In 1898 Albert Neisser, discoverer of the gonococcus and professor of dermatology and venereology at the University of Breslau, published clinical trials on serum therapy in patients with syphilis. In order to find a method of syphilis prevention he injected cell free serum from patients with syphilis into patients who were admitted for other medical conditions. Most of these patients were prostitutes, who were neither informed about the experiment nor asked for their consent. When some of them contracted syphilis Neisser concluded that the “vaccination” did not work. However, he argued that the women did not contract syphilis as a result of his serum injections but contracted the disease because they worked as prostitutes. Liberal newspapers published these and other cases, triggering public debate.
Most academic physicians at the time supported Neisser. An exception was Albert Moll,6 a psychiatrist in private practice in Berlin, who collected in his Physicians' Ethics 600 cases of unethical non-therapeutic research on humans and emphasised the need for informed consent. Moll also developed a legally based, positivistic contract theory of the patient-doctor relationship, which is widely ignored in current bioethics publications.7
In 1898 the public prosecutor investigated the case, and Neisser was fined by the Royal Disciplinary Court. The court ruled that, though Neisser as a well known medical authority may have been convinced that the trials were harmless, he should have sought the patients' consent. Not questionable science but lack of patients' consent was the main principle for the legal judgment.
The Prussian parliament also discussed the case several times and in 1899 asked the government to act. As a result the minister for religious, educational, and medical affairs commissioned a detailed report from the Scientific Medical Office of Health, which was composed of leading German physicians such as Rudolf Virchow. The commission directed its attention to beneficence and autonomy. It concluded that a physician who recognised that an injected serum might cause infection had no right to inject such a serum. In any case, both informing the subject and obtaining the subject's consent were preconditions to experimentation. In a handwritten report Emil von Behring argued that, particularly with reference to the Neisser case, self experimentation should always precede experiments on patients. He personally held that purely scientific experimentation on human subjects was unethical even if they gave voluntary consent.1 4
The minister also sought legal advice on the Neisser case. Lawyers stated that conducting non-therapeutic research on a subject without consent fulfilled the criteria for causing physical injury in criminal law. The scientific validity of the experiment did not serve as mitigation. Informed consent was a mandatory precondition for any non-therapeutic research. Problems of coercion, persuasion, and the unequal authority between doctor and patient were discussed in detail, and the lawyers concluded that respect for rights and morality had the same importance for the good of mankind as medical and scientific progress. Written documentation and clear responsibility of the medical director for all human experimentation became legal doctrine.
Finally, in 1900 the minister for religious, educational, and medical affairs issued a directive to all hospitals and clinics. Medical directors were advised that all medical interventions other than for diagnosis, healing, and immunisation were excluded under all circumstances if “the human subject was a minor or not competent for other reasons” or if the subject had not given his or her “unambiguous consent” after a “proper explanation of the possible negative consequences” of the intervention. All research interventions could be performed only by the medical director or with his or her authorisation. In all cases fulfilment of these requirements as well as all further circumstances of the case had to be “documented in the medical history.”1 Despite all this, however, the directive was not legally binding and little is known of its impact on human experimentation.
Circular of the Reich minister of the interior: guidelines for new therapy and human experimentation, 1931
Because of criticism of unethical human experimentation in the political press and in parliament as well as in the context of a political reform of criminal law in Germany, in 1931 the Reich government issued detailed “guidelines for new therapy and human experimentation.” The guidelines clearly distinguished between therapeutic (“new therapy”) and nontherapeutic research (“human experimentation”) and set out strict precautions.
Besides the principles of beneficence and nonmaleficence, the regulations were based on patient autonomy and a legal doctrine of informed consent. “New therapy may be applied only if consent or proxy consent has been given in a clear and undebatable manner following appropriate information. New therapy may be introduced without consent only if it is urgently required and cannot be postponed because of the need to save life or prevent severe damage to health….” In those cases a written report must clearly outline the preconditions. But non-therapeutic research was “under no circumstances permissible without consent.”8 18 Written documentation and a clear structure of responsibility for each clinical trial were required. Though an early model of institutional review boards was discussed, the official guideline adopted the hierarchical model from the directive of 1900, in which the medical director was responsible for all clinical research in the institution.
As later formulated in the Nuremberg code, a careful cost-benefit calculation and a detailed research plan with animal experimentation beforehand were already required to minimise risk to human subjects. Some regulations were even stricter and more detailed than those contained in the Nuremberg code and the much later Declaration of Helsinki. Human experimentation on dying patients was absolutely prohibited. Publication of the results of new therapy must respect the patient's dignity and the mandate of humanity. In academic teaching every opportunity should be taken to emphasise the special responsibilities of a physician undertaking clinical trials. Even further, any exploitation of social or economic need in testing new therapies was rejected.
This paper shows that explicit directives concerned with the welfare of people subjected to medical experimentation in Germany were in place long before the Nuremberg code was devised in 1947.9 10 11 12 Critical press reports and debate in parliament forced the Prussian government to issue the first directive concerned with medical experimentation in humans in 1900. This directive was based on medical and legal scientific reports. A clear distinction was made between therapeutic and non-therapeutic research, but regulations were issued only for non-therapeutic research. The regulations were based on the principle of autonomy and represented an early model of informed consent. A “proper explanation of the possible negative consequences” of the intervention and “unambiguous consent” became the mandatory standard. In addition, legal reports carefully discussed aspects of coercion, persuasion, and imbalance of authority between patient and doctor just as in contemporary work.13 Minors and incompetent subjects were generally excluded from non-therapeutic research, as they could not give valid informed consent.
We conclude that at the turn of the century informed consent was already a legal doctrine in medical experimentation in Germany, being based on “unambiguous consent” of the subject after “proper” information had been given by the doctor, including negative consequences and side effects. Interestingly, the regulations were not initiated by doctors or research institutions but were issued by government authorities. However, it remains an open question how informed consent was applied by doctors in research and clinical practice and how it shaped the individual doctor-patient relationship.14 15 16
The guidelines issued by the Reich government in 1931 regulated therapeutic and non-therapeutic research in human subjects. Whereas without exception non-therapeutic research could be performed only with the subject's informed consent, therapeutic research could be performed without explicit consent but only in a medical emergency and if it was deemed to be in the patient's best interest.
The second part of the Prussian directive of 1900 defined a structure of responsibility in medical institutions. Because of the hierarchical structure in German hospitals only the medical director and physicians authorised by the medical director were allowed to conduct research on human subjects. However, in no case of injury to a patient by experimentation was the issue of responsibility controversial, as all medical directors and professors declared their personal responsibility. This hierarchical model of responsibility, also found in the Reich government's guidelines of 1931, differs from the modern concept of responsibility in clinical research. Under current concepts the individual researcher is personally responsible for his or her actions and ethical issues are assessed by peers on institutional review boards.
For the first time in history informed consent, the research process, and explicit clarification of personal responsibility for the experiment were required to be included in the medical record. In addition, issues of written research plans with a risk-benefit assessment, the need for previous animal experimentation, and medical self experimentation were raised. Though a system of public health insurance existed in Germany in 1931 and provided good health care for all citizens, issues of social justice and the protection of poor people in medical research were regulated. We question whether the healthcare system in the United States would meet these regulations, many patients without health insurance having no access to regular medical treatment. In order to obtain medical help these patient must rely on free experimental treatment in research institutions without having a choice whether to give free and autonomous informed consent.
Though present conceptions of informed consent differ from those in the Prussian directive of 1900 and the Reich government's guidelines of 1931, some basic elements can be identified in postwar regulations17 18 together with many ethical issues of human experimentation.19 20 21 22 23 Our primary objective was to show that the basic concept of informed consent was developed long before the second world war and before Nazi crimes in Germany, not on the initiative of the medical profession or research community but as a legal doctrine by government authorities. The guidelines of 1931 were not annulled in Nazi Germany, when unethical experiments were performed by German doctors in concentration camps. Though no other nation seems to have had such ethically and legally advanced regulations at the time, these did not prevent crimes against humanity by part of the German medical profession.9 10 11 12
New research has uncovered ethical issues of informed consent in human experimentation as early as the nineteenth century
Regulations were not initiated by the medical profession but were issued after critical public dis- cussion and political debate
Basic elements of the modern legal concept of informed consent can be found in these early regu- lations
These early regulations were not binding in the legal sense and little is known about their actual impact on clinical research
We thank Mrs M A Shiffman for help with the English translation and Dr Thomas Lennert (department of paediatrics, Free University of Berlin) for the photograph of the medical experiment at the Kaiserin Auguste Victoria Haus.
Funding Deutsche Forschungsgemeinschaft (German Research Council; grant Vo 625/1-3).
Conflict of interest None.