Topic

History of Clinical Trials

Clinical trials are organized attempts to compare medical interventions in human beings. Their history links bedside observation, hospital records, statistics, ethics, drug regulation, public trust, and the difficult task of deciding when a treatment has been tested well enough to guide care.

The history of clinical trials is not simply the invention of one modern method. It is a long movement from informal therapeutic comparison toward planned, documented, ethically reviewed, and statistically interpreted studies.

Scope: Therapeutic comparison, hospital records, placebo controls, randomization, masking, research ethics, drug regulation, trial reporting, and evidence-based medicine
Key links: James Lind, medical statistics, tuberculosis, medical ethics, antibiotics, cancer treatment, public health, and medical education
Search focus: History of clinical trials, randomized controlled trial history, James Lind scurvy trial, streptomycin trial, and clinical research ethics

Read the topic Back to topics

Testing Treatment

Clinical trials grew from the problem of therapeutic uncertainty

Physicians, surgeons, apothecaries, patients, hospitals, armies, and states all wanted to know whether a treatment worked. The historical difficulty was that recovery, relapse, selection, expectation, changing diagnosis, and uneven records could make ineffective treatments appear successful.

Early modern and nineteenth-century medicine was rich in observation but often weak in controlled comparison. Practitioners compared cases, remembered striking recoveries, published case series, and argued from experience. These habits mattered, yet they could not easily separate the effect of a remedy from the natural course of disease, nursing care, diet, environment, or chance.

Clinical trials became historically significant because they changed the question from "Have some patients improved after this treatment?" to "How does one group fare compared with another group observed under defined conditions?" That shift connects this subject closely to the history of medical statistics and to the larger history of medical authority.

Early Comparisons

Before modern trials, physicians tested remedies by comparison

The modern clinical trial did not appear fully formed. Its components emerged from older experiments, military medicine, hospital practice, inoculation debates, and attempts to compare outcomes in groups of patients.

James Lind compared scurvy treatments at sea

In 1747, the Royal Navy surgeon James Lind divided twelve sailors with scurvy aboard HMS Salisbury into six pairs and gave each pair a different remedy, among them vinegar, salt water, and citrus. The two sailors given orange and lemon were fit for duty within about a week. Lind published the comparison in his 1753 A Treatise on the Scurvy. The study was small and not randomized, but its deliberate pairing of treatments made it a durable landmark in therapeutic evaluation; the Royal Navy did not make citrus a standard part of the ration until 1795.

Inoculation debates made risk comparison public

Eighteenth-century arguments over smallpox inoculation asked whether the risk of deliberately inducing a mild infection was lower than the risk of natural smallpox. The issue belonged to the history of vaccination, but it also helped make preventive medicine a matter of numbers, comparison, and public trust.

Hospitals supplied cases, records, and controversy

By the nineteenth century, hospitals concentrated patients in settings where physicians could compare treatments more systematically. The history of hospitals is therefore also a history of case notes, ward records, clinical teaching, and arguments about whether groups of patients could be made comparable enough for conclusions about treatment.

Numerical Medicine

Nineteenth-century reformers challenged therapeutic custom

In Paris, Pierre Charles Alexandre Louis (1787–1872), a professor at the Faculty of Medicine, became associated with the "numerical method." In his 1836 Mémoire sur la pneumonie, he compared the outcomes of patients with pneumonia who had been bled with those who had not, and concluded that bloodletting did not reduce mortality. He did not randomize or assign treatment; his comparisons were retrospective aggregations of hospital case records. Louis did not create the modern trial, but he made a powerful case that treatment claims should be tested against aggregated outcomes rather than professional confidence alone.

Numerical medicine exposed problems that later trial design tried to solve. Patients differed in age, severity, timing of treatment, diagnosis, nursing, and social condition. Physicians could select easier cases for a favored remedy without noticing it. Records could omit failures. Outcomes could be counted differently in different hospitals.

The nineteenth century therefore supplied a crucial lesson: comparison was necessary, but comparison was fragile. Clinical trials would become more formal partly because medicine needed safeguards against its own habits of selection, optimism, and incomplete memory.

Placebos And Blinding

Trial methods developed to control expectation and observation

Placebo controls and masking did not arise from cynicism about patients. They developed because researchers recognized that expectation, observation, and the authority of treatment could shape reported symptoms, clinical judgment, and the interpretation of ambiguous outcomes.

Placebos made comparison sharper

The word "placebo" (from the Latin for "I shall please") had long described medicines given more to reassure a patient than to treat. In clinical research, placebo controls became a way to compare a proposed intervention with an inactive or standard comparator under similar conditions. During the Second World War, the British Medical Research Council's multicentre trial of patulin for the common cold used alternate allocation and masked patients and staff; its negative results were published in 1944. The 1948 MRC streptomycin trial then paired concealed random allocation with prospectively defined comparisons and became an influential model for the modern randomized controlled trial.

Masking reduced observer bias

Blinding, or masking, limited the ability of patients, clinicians, or assessors to know which treatment had been assigned. This mattered especially when outcomes involved symptoms, interpretation of images, or judgments about improvement rather than an unmistakable event.

Defined endpoints made results easier to judge

Trials increasingly specified what would count as success or failure before results were known. Mortality, relapse, sputum conversion, tumor response, pain scores, adverse events, and functional measures each carried different historical assumptions about what medicine was trying to prove.

Randomization

Random allocation became a landmark of twentieth-century trials

Randomization addressed a basic problem in clinical comparison: if patients were assigned by preference, prognosis, convenience, or social status, treatment groups might differ before therapy began. The idea came from R. A. Fisher's work on agricultural experiments, especially The Design of Experiments (1935). Random allocation did not make trials perfect, but it gave researchers a disciplined way to reduce selection bias and support statistical interpretation.

In medicine, the 1948 Medical Research Council trial of streptomycin for pulmonary tuberculosis, published in the British Medical Journal, is often cited as a landmark early randomized controlled trial. It randomly assigned 80 patients to streptomycin plus bed rest or to bed rest alone, and measured sputum culture conversion at eight weeks; the streptomycin group converted at a substantially higher rate. The trial was not the first randomized experiment in medicine, but it became a model because it joined a new drug, a defined endpoint, random allocation, and statistical interpretation in one design.

The trial also provoked methodological controversy. Fisher, who had championed randomization, criticized the MRC's open-label design and its use of significance testing, and the dispute helped push the field toward more rigorous blinding and pre-specified analysis. That study belongs in the history of tuberculosis and antibiotics as well as clinical trials, and it helped establish the randomized controlled trial as a model of reliable therapeutic evidence.

Regulation

Drug disasters and public law changed what trials were for

Clinical trials were not shaped by science alone. They were also shaped by injury, scandal, legislation, pharmaceutical markets, military research, and the growing expectation that governments should evaluate medicines before widespread use.

Safety became a regulatory question

In 1937, S. E. Massengill Company of Bristol, Tennessee, marketed a liquid preparation of sulfanilamide made with diethylene glycol, a toxic solvent. The company had not tested the preparation for toxicity, and more than 100 people, many of them children, died. The disaster led directly to the 1938 Federal Food, Drug, and Cosmetic Act, which for the first time required manufacturers to show that a new drug was safe before it could be marketed. The episode was not a clinical trial milestone in a narrow sense, but it strengthened the expectation that drug safety required public oversight.

Thalidomide made efficacy and fetal risk unavoidable

Thalidomide was marketed in West Germany from October 1957 as a sedative and for morning sickness. Reports of babies born with phocomelia, severe limb malformation, began in 1961, and the drug was withdrawn from most markets by 1962. In the United States, the FDA reviewer Frances Kelsey refused to approve thalidomide, pending further safety data. The 1962 Kefauver-Harris Amendments (the Drug Amendments of 1962) then required "substantial evidence" of effectiveness, as well as safety, and required informed consent for investigational drugs, making controlled trials central to drug approval.

Pharmaceutical research became institutionally complex

By the later twentieth century, trials increasingly involved sponsors, academic centers, contract research organizations, regulators, statisticians, ethics committees, and patient recruitment systems. Treatment evaluation became a specialized infrastructure, not only an activity of individual physicians.

Ethics

Human-subject research made consent central

Clinical trials depend on people who accept risk for uncertain benefit. That fact made research ethics central to trial history. Consent, disclosure, fair selection, independent review, and the right to withdraw became especially important after wartime abuses, coercive studies, and research conducted among vulnerable populations.

The Nuremberg Code of 1947, drafted in the wake of the Doctors' Trial at Nuremberg, made the voluntary consent of the human subject its first principle. The Declaration of Helsinki, adopted by the World Medical Association in 1964 and revised repeatedly since, extended these principles to clinical research generally. In the United States, the Belmont Report of 1979 distilled research ethics into three principles — respect for persons, beneficence, and justice — and underpinned the federal regulations governing human-subject research. Independent ethics review became standard infrastructure. These documents did not end ethical controversy, but they changed the language of legitimate research. Trial design had to account not only for scientific validity, but also for voluntary participation, proportional risk, social value, and protection from exploitation.

This history is closely tied to the history of medical ethics. A poorly designed trial is not merely inefficient; it can expose participants to risk without producing trustworthy knowledge. Ethics and method therefore became historically inseparable.

Disease Fields

Trials changed differently across medical specialties

The clinical trial did not spread evenly. Each field adapted trial methods to its own diseases, endpoints, technologies, patient populations, and institutional pressures.

Cancer trials made survival, toxicity, and staging central

In the history of cancer treatment, trials became vital because surgery, radiotherapy, and chemotherapy required careful comparison of survival, recurrence, tumor response, side effects, and quality of life. Multicenter cooperation became especially important where single hospitals lacked enough comparable cases.

Antibiotic trials joined laboratory and bedside evidence

The history of antibiotics and penicillin shows how bacteriology, culture results, dosing, resistance, toxicity, and clinical recovery had to be interpreted together. A drug could look promising in the laboratory and still require careful human comparison.

Chronic disease trials required longer horizons

Trials for diabetes, cardiovascular disease, mental health, and prevention raised different problems from short acute infections. They required attention to adherence, long-term outcomes, surrogate markers, adverse effects, and the daily realities of living with treatment.

Reporting

Reliable trials required transparent reporting

As trials became more influential, incomplete publication became a historical problem of its own. Negative or inconclusive studies could disappear. Outcomes could be changed after results were seen. Selective reporting could make a treatment appear more effective or safer than the full evidence allowed.

Trial registration, reporting standards, data monitoring, systematic reviews, and evidence-based medicine all responded to this problem. The Cochrane Collaboration, founded in 1993 in the spirit of the British epidemiologist Archie Cochrane's 1972 book Effectiveness and Efficiency, systematized the search, appraisal, and pooling of trial results. The CONSORT statement, first published in 1996 and revised in 2001, 2010, and 2025, set a minimum checklist for reporting randomized trials. Trial registration became a requirement: the U.S. ClinicalTrials.gov registry (launched in 2000 and expanded by the 2007 FDA Amendments Act) and the EU Clinical Trials Directive (2001) required sponsors to register trials and, increasingly, to report results. Together these mechanisms extended the trial beyond the moment of patient enrollment and into the public record: what was planned, what was measured, what was omitted, and how results were combined with other studies.

The history of trials therefore also belongs to the history of medical education. Modern clinicians were increasingly expected to read trial reports, judge their methods, understand uncertainty, and apply group evidence to individual patients.

Debates

Clinical trials changed medicine, but did not remove judgment

Trials gained authority because they could discipline comparison. Their history also shows why no method can answer every clinical or moral question by itself.

Internal validity and general relevance can diverge

A tightly controlled trial may answer a precise question while excluding older patients, pregnant patients, people with multiple illnesses, or those outside specialist centers. The result may be scientifically strong yet difficult to apply broadly.

Placebo use raised ethical questions

Placebo-controlled trials could be methodologically useful, but they became ethically disputed when effective treatment already existed. The successive revisions of the World Medical Association's Declaration of Helsinki addressed placebo use, permitting it where no proven intervention exists and, under limited conditions, for compelling methodological reasons. The question became especially acute in the 1990s, when placebo-controlled trials of antiretroviral drugs for HIV were proposed in low-income settings, and critics argued that withholding effective treatment from participants was unethical even where it was not standard local practice. Debates over placebo controls show how method, patient welfare, and global inequalities entered the same argument.

Evidence hierarchies could oversimplify care

Evidence-based medicine gave randomized trials special authority for treatment evaluation, but historical practice still required judgment about mechanism, diagnosis, values, access, cost, and patient preference. Trials changed clinical judgment; they did not abolish it.

Reading Path

Where to go next on Historia Medica

These related pages show how clinical trials grew from older histories of statistics, ethics, hospitals, infectious disease, cancer care, and public health.

James Lind
Lind's scurvy experiment remains a useful starting point for the history of therapeutic comparison, even though it was not a modern randomized controlled trial.
The History of Medical Statistics
Follow the statistical background that made randomization, endpoints, inference, and trial reporting intelligible.
History of Tuberculosis
Tuberculosis connects sanatorium care, bacteriology, radiology, streptomycin, and one of the landmark randomized trials.
History of Medical Ethics
Read the broader history of consent, experimentation, professional authority, patient rights, and institutional review.
History of Cancer Treatment
Cancer care shows why multicenter trials, staging, survival endpoints, toxicity, and patient follow-up became central to modern medicine.
The Birth of Evidence-Based Medicine
Follow how trial evidence, systematic review, and the language of "evidence" reshaped clinical judgment and medical authority.

Legacy

Clinical trials remade the authority of medical evidence

The legacy of clinical trials is visible in drug approval, treatment guidelines, public-health recommendations, systematic reviews, patient information sheets, ethics committees, trial registries, and the everyday language of evidence. Modern medicine often treats trial evidence as the strongest way to evaluate treatment effects in groups of patients.

That legacy is also cautionary. Trials depend on trust, honest reporting, fair recruitment, careful follow-up, and methods suited to the question being asked. When those conditions fail, the authority of the trial form can conceal weak evidence rather than strengthen it.

Historically, clinical trials matter because they changed both medical knowledge and medical responsibility. They made therapeutic claims more public, more contestable, and more dependent on institutions designed to protect patients while producing reliable evidence.