A natural medical experiment?
A topic in 'Science & Ethics'
Science is about acquiring a better understanding of the natural world. Scientific knowledge can be considered a 'good' – something to be valued and sought.
But the acquiring of this good has costs – time, funding, intense activity: resources spent in one activity rather than on some other alternative goal. So, even if we accepted that acquiring knowledge is of itself always inherently good, this does not mean that all activity to acquire knowledge is ethical when we consider the costs involved. (So, for example, much knowledge of human anatomy and physiology was acquired by experiments that tortured animals – there is a judgement to be made bout the ethical status of such experiments.)
The research discussed here, the Tuskagee 'bad blood' study, reflects one of the most shameful episodes in medical science. It is unlikely any one with normal human values would, in retrospect, argue this was an ethical study. Yet, the medical scientists involved clearly decided it was the 'right' thing for them to do at the time. This raises the issue of how we can explain their behaviour.
When should the investigators have realised this was an unethical study? How did they manage to convince themselves this was a good thing to do?
The text below just offers a brief overview of s complex study and an interested reader should refer to more detailed and nuanced accounts (e.g., Gray, 1998; Jones, 1993).
A longitudinal study
In 1972 it become widely reported that a longitudinal study had been running for 40 years that explored the progression of the disease syphilis. Moreover, it seemed that the subjects of the study had not been fully informed of the nature of the study; had not been been given a proper medical diagnosis and information about the nature of their disease; and had not been offered appropriate medical treatment (in order that the researchers could see how the disease progressed when unchecked).
This work not was carried out in some poor, developing nation, but in the United States. The work had not been carried out by some obscure research organisation, but by the Public Health Service. The public service meant to provide healthcare appeared to be treating {sic, not in a medical sense} sick people as no more than scientific specimens to be observed and measured.
Syphilis
Syphilis is an infectious disease, caused by infection with a bacterium (Spirochaeta pallida). Like other bacterial diseases, it can today be effectively treated with antibiotics. If it is untreated, it can lead to a variety of symptoms, some very serious. Some symptoms appear soon after infection, but others only arise much later after an extended latency period. Syphilis is still a major health issue in some parts of the world today, but only because sufferers go unidentified and so untreated (World Health Organisation). With proper medical care, this is a treatable disease.
Syphilis has been known for some centuries, and many famous people are reputed to have been infected, and indeed to have ultimately died from the infection. (The first effective antibiotic treatments only became widely available in the mid-1940s.)
Techniques to definitely diagnose syphilis are relatively modern-the 'Wasserman test' was developed in 1907 (Fleck, 1979)-and before this, like many diseases, the precise range of symptoms considered characteristic of the disease had shifted over time. As initial symptoms fade, and the later effects only become clear after a considerable time lapse, it had not previously been possible to be definitive that these issues were all caused by the same disease (and this might have seemed unlikely before the germ theory of disease was accepted in medical circles).
Also, there was considerable stigma attached to syphilis as it is a venereal disease, with infection usually due to sexual contact (or acquired in the womb by a child carried by an infected mother) and sufferers would not generally have publicly shared a diagnosis. Therefore, it is not possible to be definite about the diagnoses we might give to historical cases.
Before modern antibiotics, the common treatment for diagnosed syphilis involved the use of mercury compounds. While this did seem to have some useful effect, mercury is a toxin – so the only widely available treatment involved using a poison that it was hoped would do more damage to the disease than to the patient. (Of course many medicines have side effects, but a treatment as toxic as mercury has to be seen as something of an extreme measure.) It has been suggested (bearing in mind the serious caveats raised above) that writer Friedrich Nietzsche might have died either from the effects of syphilis or the mercury treatments, and that the composer Franz Schubert died from mercury poisoning from treatment taken for the disease.
In 1910 a new 'wonder drug' was proclaimed, Salvarsan (arsphenamine). But this was based on another toxin: arsenic, and needed dozens of doses to be administered to a single patient over an extended period.
It is fair to say that in the early twentieth century:
- syphilis was a major public health issue
- the nature and progression of the disease was not very well understood
- no effective, safe treatment was available
A 'racial' element
One of the reasons why the Tuskagee study has become so notorious is because of racist features of the project. Tuskagee is located in one of the Southern US states, Alabama, which had historically supported slave ownership before the US Civil War (that is, Alabama was part of the Confederacy). The research started in the 1930s, long before the civil rights movement of the 1960s.
Although slavery had long been abolished by the time of the study, many of the descendants of the slave populations still lived in economically challenging conditions (e.g., such as subsistence crop-sharers-in effect working for farmers who owned the land, somewhat like Medieval serfs).
Studies of epigenetics suggest that such traumatic experiences as slavery and poverty may have objective biological consequences for generations (Meloni, 2019) – operating over and above any culturally transmitted effects. Moreover, at the time there was a form of apartheid operating with people not only labelled on 'racial' lines (such as 'black' or 'white'), but then separated for purposes such as education. The level of resources available to the different groups, and so outcomes such as literacy levels and heath standards, were far from equitable.
While racism continues to the present day, at the time of the initiation of the Tuskagee study it was institutionalised to the extent of being established and maintained by the laws of the land.
But races are a fiction?
"Race serves to introduce a spurious variable – biology, or nature – into a discussion of economics, politics and morality."
Jonathan Marks, 2017
Today we know that there are no distinct races of human beings. As the saying goes, 'there is only one race-the human race'. Members of populations that become relatively isolated over extended periods tend to become genetically more similar to each other than to people from geographically distant people. But there are no absolute distinctions, and all humans derive ultimately from the same ancestral populations that originally lived in Africa. (One of those women was the so-called mitochondrial Eve who lived in Africa 100 000-200 000 years ago, and is a direct ancestor of every person alive today. *)
Yet the notion of humans falling into distinct races was once a wifely accepted idea – even among scientists. This idea is known as racialism, and was not of itself inherently racist (though no doubt racism played a part in the acceptance of racialism among scientists; and certainly racialism provided a foundational position that facilitated racism).*** Race was believed to be a meaningful biological distinction, to the extent that scientists would not only examine and report on differences in anatomic 'norms' (ignoring that some of these would have simply reflected factors like diet) but also many would ascribe inherent differences in personality patterns and moral worth between groups. It was also widely assumed (but not in the nuanced way that ethnicity is treated as a potential factor in health today) that different racial groups could respond to disease and treatments in different ways. Although a large-scale study of the progression of syphilis had been reported in Northern Europe, it was thought that this could not be assumed to reflect what would occur among American 'Blacks'.
To some extent this reflected a long-standing set of ideas that had dominated European scholarly thinking, and that saw the creation as a kind of ladder of forms, which represented levels of perfection. Humans were above the apes, but below the Angels! At one time it was assumed that the original creation had offered every variation of graduation possible – so intellectuals set out deciding which order the 'races' took on the ladder towards perfection. Strangely, the race of the European scholars who engaged in this fantasy was often identified as the highest form of human.
See 'great chain of being (scala naturae)': Historical examples of alternative conceptions
This notion should have become untenable with Charles Darwin's work, which showed there are no entirely distinct and stable groups such as species or races****: yet being ideologically based it survived. Although categories such as 'White' and 'Black' are not scientifically sound, they survive today in how many people identify themselves (and others) – and in 1930's Alabama the labels, and the distinctions they represented, had legal standing and very real consequences. Although I am uneasy about using such labels as if they are objective categories, in the context of the work being reported here people were commonly seen as, for example, White or Black.
A pilot study to treat syphilis in Black patients
The infamous Tuskagee study developed from a series of small scale pilot studies to identify and treat syphilis among poor Black people where it was believed the disease was circulating widely. (Conditions such as diet, housing and education might explain any higher incidence of diseases among the poor – but unfounded racial stereotypes about the different moral character of different 'races' influenced the thinking of many medics at the time.)
These studies could be seen as a test of concept: there were real doubts that the population worked with would be open to the unpleasant and repeated (and toxic) treatments for a disease they were unlikely to understand very well. However, it was found that cooperation was forthcoming, and such treatment programmes could be effective.
Might these small-scale treatment studies be seen as producing worthwhile knowledge, and if so might that justify them as 'ethical', even if aspects of their their motivation was questionable?
How might the public health service have built upon these studies to apply the knowledge gained?
The longitudinal programme
Unfortunately, funding to follow-up what had been learned, in order to start a large-scale and long-term treatment programme was not forthcoming. What did follow (i.e., was feasible with the limited resources available) was the beginning of a study in Tuskagee that identified Black men infected with syphilis, and then took yearly data from them. (This included painful collection of fluid by lumbar puncture.) The men were given some food supplements and a hot meal when they participated, but were not offered any specific treatment. The study was managed by White public health officials, but a Black nurse was enrolled as a key point of contact for participants. (That can be seen as a sensible, or a cynical, choice.)
Bad blood
It is impossible to now know exactly what was said between the participants and the medical researchers. There was no formal ethics procedure in place (but that was quite normal for the times) so no protocol whereby voluntary, informed, consent was obtained. The participants were willing volunteers – but not on the basis of knowing what the study was really about. Some participants considered they suffered from 'bad blood' – a vague and unscientific term – and some of the dialogue between participants and researchers likely revolved around that idea.
What clearly did not happen
There was no explicit diagnosis of syphilis, and what this meant (for example, that it made it likely that husbands would infect their wives, who might then pass on to disease to children in the womb) shared with the infected participants; and no explanation that, as this was an observation study, the doctors would be making no attempts to offer substantive treatment. One can imagine how someone might have responded to such information if it had been provided:
"You have a very serious infectious disease which, unless you control certain behaviours, you might pass on to others; and which is likely to have very serious, even fatal, effects. The treatments are unpleasant and need to be repeated a lot, and you could not afford them. Sadly I do not have funding to offer you these treatments. Do you mind if we bring you in once a year to ask you some questions and do some measurement and invasive tests and see how the disease is progressing in you? We won't ask you to moderate your behaviour as we'd like to see how quickly your wife is infected."
What asking for informed consent might have sounded like
Today, any suggestion of carrying out research on people without their informed consent would raise 'red flags'. This would normally be seen as treating people as experimental subjects-like lab rats-rather than genuine participants in research. There may be circumstances where this could be justified (see the discussion of 'Observing in the rest room:
When researchers toil in the toilets'), but certainly not in a medical context such as this. But that is to look at the research from a modern perspective which has developed from the discussion and critique of past practice.
The participants were considered to be largely illiterate with poor levels of biological education. Is it reasonable to assume the infected men could not have understood the nature of their disease? What should researchers do when they suspect participants may have a limited understanding of a study?
A natural experiment
The study may be considered as a natural experiment, where a naturally occurring situation (here a population where there were both infected and non-infected men from the same location and socioeconomic setting) is investigated in order to explore an effect (here, the progression of disease in infected men). The researchers do not bring about the conditions of the experiment (as in a 'genuine' experiment) but simply observe the 'natural' situation.
Medical practitioners promise to do no harm. There is no suggestion that the Public Health Service caused the infections in the infected men. Rather, the researchers simply observed the natural progression of disease. Is this any defence against the charge that the research was unethical?
Do the researchers have any responsibility for infection being spread by study participants who had not been informed of their condition?
Indeed the notion that the researcher did not intervene in the course of affairs is not entirely true, as active measures were taken to avoid infected participant being treated outside the research programme (Office for Science and Society, McGill University, 2019)-that would have undermined the study.
A long-term study
As mentioned above, the study lasted for four decades. This is impressive, as very few studies can be maintained so long: it requires continued resources, continued institutional approval, and usually continuation despite changes in staffing. Such long term studies are known in research as longitudinal studies, and present particular challenges to researchers.
Outcomes from the study were reported in research journals in the scientific literature. So, the editors and expert peer reviewers of the journals concerned judged that the study had led to valuable information worth sharing in the scientific/medical communities. [Today scientific journals ask researchers to report on the ethical review procedures employed in their studies – but this was not always so.]
When is gaining knowledge worth the costs?
That work from the study was published suggests that the researcher's belief that such a study could have some valuable output was justified. Learning more about a serious infectious disease, and recognised public health problem, seems a worthwhile objective. But there are always 'costs' to research, so simply producing useful knowledge cannot in itself justify research – we cannot assume that 'the ends justify the means'. We need to consider research in the round.
We should consider all those impacted by research. Better understanding of a disease might indirectly lead to better treatments and so valuable health outcomes. Research outcomes inform other researchers in their work, which might in turn have useful outcomes. But we should also consider the effect on participants. They were denied treatment – although that was never an option at the initiation of this study as the resources were not available. But they were also denied information which could have led to them changing behaviour and/or potentially seeking treatment elsewhere. But the researchers seem to have assume that would not be likely. They had selected a sample of poor men (and in a national context, the United States, which even today does not provide free medical treatment on a basis of need).
Could this be justified?
Is it possible to consider the Tustakee study as justified at its outset? Considering the context at the time, and the knowledge available to the researchers – is it possible to see the study as well-meaning and potentially ethical (e.g., had proper consent been obtained) within that context?
The antibiotic age
Penicillin was discovered by Alexander Fleming in 1928, although he initially was mainly interesting in its possible use as an antiseptic rather than as a drug taken internally It was not developed into a useful drug till work by Howard Florey and Ernst Boris Chain in the early 1940s. The advent of the second world war led to an effort to mass produce the new wonder drug, penicillin, as it had long been recognised that many soldiers hurt on the battlefield had survivable injuries, but died because their wounds became infected with bacteria. It is often suggested that in many conflicts the germs killed more than the bombs and bullets.
By the end of the 1940s there was then for the first the time the wide availability of an effective and relatively inexpensive treatment for bacterial infections – such as syphilis. Unlike mercury and arsenic compounds, antibiotics were usually tolerated by most patients without any serious side effects.
This changes the balance of considerations regarding the Tuskagee study in two ways. Firstly, there can be no arguments that treating infected participants would be problematic because of the toxicity of the treatment, or because of the difficulty of administration of, or the need for extended periods of, treatment.
Secondly, whilst there might still be some academic value in knowledge of how the disease progresses in untreated men – this became information of much less utility in medicine, as there was no need for there to be any untreated people once antibiotics could be used.
Modifying the study
How should the researchers, and/or their bosses, have responded to the availability of an effective treatment? How should the study have been modified (or perhaps simply stopped) once syphilis was a readily treatable disease?
Despite the medical developments, the Tuskagee study continued as before. There was no attempt to offer antibiotics, or to inform the infected participants that such a treatment was available and would be indicated. Instead, the research team continued to collect data about how the disease was damaging the men's bodies.
Ending the study
At no point did the research team decide for themselves that their study was no longer justified, and should be closed down. Rather news of the study was reported in the press, leading to widespread public condemnation. It was external publicity and pressure which led to the researchers ending their study.
The study had enrolled 399 poor, largely illiterate, Black Americans. By the time the study was stopped, it was reported that 28 of the 'participants' had died form syphilis, and another 100 had died form medical conditions related to being infected with syphilis. Many more were presumably in much poorer health than they would have been with proper treatment.
Moreover, 40 of the wives of participants had become infected with syphilis, as well as 19 children born with the disease because their mothers was infected.
Explaining the study
It is very hard to explain how medical professionals, public servants employed to improve public health, could have ever though it was acceptable to treat the study participants, fellow humans beings, in the way they did. But this is a historical case – it happened.
Clearly the institutionalised racism of the United States in the 1930's was a major factor – the norms of the time, and the beliefs of those involved. For example, there were institutionalised norms that segregated the different 'races' – this not only maintained inequity but also limited social mixing between groups. The participants were of a different social class to the investigators, who would likely have only engaged with them in a professional (sic) context.
We might describe the actions of the research team with a word like evil, but that would just be a label. And it does not seem likely that people who are medical professionals working in the public services can be simply dismissed in this way (even if they did some things that seem evil).
A particular mystifying issue in this case is the way in which the study continued year after year as if in isolation from the wider social context. The US civil rights movement of the 1950-1960s, and the changes it brought about did not seem to impact on the thinking of the researchers. Even more inexplicable, the antibiotic age did not lead to a reappraisal of a project which was producing scientific data, albeit increasingly irrelevant data. It seems as if once the project was under way it acquired its own momentum and become part of local custom and practice that was not held to review and reappraisal.
In conclusion:
- the Tuskagee study was seen as an observation study rather than an intervention study: but in a context where the appropriate thing for a doctor to do is treat disease we might see deliberately withholding treatment as an intervention (in the normal cause of medical treatment) which did do serious harm;
- the knowledge being sought could at one time have informed better medical practice, but this became overtaken by events when antibiotic treatment became available;
- the knowledge being obtained from the study was not being shared with those participating, who were misled about key features of the study directly relevant to them;
- the research was managed by people from one group in society that held much greater relative power than the group from which the participants were enrolled (and in a divisive social context llikely to encourage seeing those from other groups in depersonalised terms);
- the purpose was 'to observe the natural progression of the disease if left untreated', but because of the prior treatment study (see above) it is likely some of the 'untreated' men had actually received some treatment [that is, there seems to have been imperfect experimental 'control' of variables];
- the longitudinal nature of any long-term study requires periodic review of the design in the light of what has been learnt, and any other pertinent factors: but the Tuskagee study seemed to acquire inertia that kept it on track regardless;
- academic freedom means that researchers should have the right to decide what areas of research to follow based on their interests and professional judgement; but this cannot be isolation, and publicly funded research should demonstrate that it meets the needs and priorities of the wider society;
- while appropriate treatment was withheld, the participants were offered enticements to contribute (e.g., a warm meal) that were likely seen as substantive given their poverty;
- not only were participants (and their families) harmed by the combination of refusal of treatment and lack of clear information about their medical condition, but the Public Health Service, and indeed the wider community of medical researchers, suffered reputational damage (from the very public condemnation of the research) likely effecting both trust in medical studies and the medical community, and willingness to participate in future research.
Today, there are procedures and guidelines in place which SHOULD stop anything similar ever happening again. Learning about such shameful cases is important as a warning of what can happen when researchers fail to properly consider the ethics of their research designs.
* Although she is called Eve, after the scriptural Eve, there is no suggestion she was the first woman. People alive today have many (**) different ancestors from the time of mitochondrial Eve, but whereas the nuclear genes get mixed up by sexual reproduction, the mitochondria (with their own genes) are only passed from the mother to the child. We all have our mother's mitochondrial genes, which are her mother's mitochondrial genes, which are her mother's mitochondrial genes, which are…back to Eve.
** But perhaps not as many as most people assume! When considering recent generations, most people have 2 parents, 4 grandparents, 8 great grandparents, 16 great great grandparents, etc. But this pattern does not continue indefinitely – and indeed many of your ancestors take up multiple places on several different levels (generations) of your personal chart of descent / 'family tree' (as if your aunt was also your sister and your great grandma!) (Read about 'Inter-generational couplings in the family'.)
*** The distinction between scientific racialism (wrong, but not inherently wicked) and racism has been explained:
"It thus became useful to distinguish between the two related fallacies we named earlier: racialism, the empirically false scientific idea that the human species can be naturally partitioned into a reasonably small number of reasonably distinct groups; and racism, the morally corrupt political idea that natural human groups are differently endowed, are rankable and differently entitled on such a basis, and that consequently individual people ought to be judged on the basis of their membership in such group rather than on the basis of their own properties, abilities, achievements, or rights."
Jonathan Marks, 2017
While racialism was not itself evidence of racism, it is certainly true that some of the scientific studies that detailed supposed differences between racial types were based on very poor sampling and led to invalid conclusions accordingly. Not noticing (or perhaps even subconsciously selecting) inadequate methodology is more likely when flawed findings support an existing ideological belief.
**** At one time it was common to teach that separate species could not successfully mate and produce fertile offspring. This might be a useful rule-of-thumb but it is not an absolute. Many (if not all) people alive today carry genes from other human species (such as Neanderthals) that successfully bred with their Homo sapiens ancestors. That is, we have non-Homo sapiens ancestors that bred with our Homo sapiens ancestors, and gave rise indirectly to us! (Read 'Can ancestors be illegitimate?')
Work cited
- Fleck, L. (1979). Genesis and Development of a Scientific Fact (F. Bradley & T. J. Trenn, Trans.; T. J. Trenn & R. K. Merton, Eds.). The University of Chicago Press. (Original publication in German, 1935)
- Gray, F. D. (1998). The Tuskagee Syphilis Study. The real story and beyond. NewSouth Books.
- Jones, J. H. (1993). Bad Blood: The Tuskegee Syphilis Experiment (New and expanded ed.). The Free Press.
- Marks, Jonathan (2017) Is Science Racist? Polity Press
- Meloni, M. (2019). Impressionable Biologies: From the archaeology of plasticity to the sociology of epigenetics. Routledge.
- Office for Science and Society, McGill University (2019) 40 Years of Human Experimentation in America: The Tuskegee Study, https://www.mcgill.ca/oss/article/history/40-years-human-experimentation-america-tuskegee-study.
- Taber, K. S. (2013). Classroom-based Research and Evidence-based Practice: An introduction (2nd ed.). Sage.
