History of clinical trials: from Lind to modern RCTs

What is considered the first documented clinical experiment?

The earliest recorded experiment resembling a clinical trial appears in the Book of Daniel in the Bible. It describes events during the reign of King Nebuchadnezzar of Babylon, who ruled for almost 60 years until 562 BC^[1]. This experiment was not conducted by a physician but by the king himself, a military leader rather than a healer^[2].

Nebuchadnezzar ordered his people to eat only meat and drink only wine, believing this diet would keep them in good physical condition^[2]. Several young men of royal blood objected and preferred a diet of vegetables. The king allowed these dissenters to follow a diet of legumes and water instead, but only for 10 days, after which their health would be assessed^[1].

When the 10 days ended, the legume-eaters appeared better nourished than the mandated meat-eaters, so the king allowed them to continue their diet. This was not a randomised, blinded or placebo-controlled study by any modern definition. It may nonetheless have been one of the first times in human history that a medical test, however rudimentary, guided a decision about public health^[1].

How did James Lind's 1747 scurvy experiment work?

British physician James Lind is widely considered the first person to conduct a parallel-arm medical experiment, and many historians call him the father of the clinical trial. In 1747, aboard the British naval ship Salisbury, Lind selected 12 sailors suffering from scurvy who all displayed similar symptoms, described in his own words as "putrid gums, the spots and lassitude, with weakness of knees"^[1].

Lind divided the 12 sailors into six pairs, isolated them from the rest of the crew, and gave all of them the same basic rations. Each pair received a different additional treatment: cider, a few drops of a weak acid (elixir vitriol), vinegar, seawater, a mixture of nutmeg and barley water, or oranges and lemons. Lind published his findings in 1753 in a paper titled A Treatise on the Scurvy^[1].

Lind's design contained most of the elements historians recognise in a controlled trial: comparable participants, isolation from confounding factors, identical baseline conditions and several parallel treatment arms^[2]. It is one of the earliest documented examples of what a modern trial calls a parallel-arm experiment, in which participants are divided into at least two groups that differ only in the treatment received^[1]. Yet the experiment still lacked randomisation and blinding. These features would only be formalised nearly two centuries later^[2].

Why did doctors rely on alternation before switching to random allocation?

Long before random allocation became standard, clinical researchers commonly used alternation: successive patients were assigned to each trial arm in strict, predictable alternating order. This practice had become the de facto method of allocation as the concept of fair clinical trials began to develop during the 19th century. Alternation is described as seemingly simple and straightforward, which may help explain its wide adoption during this period^[3].

The problem with alternation is that its predictable, systematic pattern makes it vulnerable to bias. A clinician who knows which treatment the next patient will receive can consciously or unconsciously influence who gets enrolled next. Statistician Austin Bradford Hill of the London School of Hygiene and Tropical Medicine investigated this vulnerability in the early 1930s. He identified allocation bias as a genuine risk to trial reliability^[3].

Hill's motivation for moving toward random allocation is now understood to have been pragmatic rather than theoretical: countering the risk that clinicians might bias which patients ended up in which treatment arm, whether deliberately or not. Hill recognised that an unpredictable random sequence was necessary but not sufficient on its own. The sequence also had to be concealed from the people enrolling patients, otherwise the same bias could creep back in. This two-part insight, randomise and then conceal, is what separates Hill's approach from earlier, superficially similar attempts at chance-based allocation, and it is the reason his name is attached to the birth of the modern trial rather than simply to one more variation on an old idea^[3].

What made the 1946–1948 streptomycin trial different from everything before it?

The UK Medical Research Council (MRC) trial of streptomycin for pulmonary tuberculosis, carried out in 1946 and published in 1948, is generally regarded as the study that marked the emergence of the modern randomised controlled trial (RCT)^[3]. It is described as a landmark trial, meticulous in its design and implementation, with systematic enrolment criteria and data collection compared with the more ad hoc approach of other contemporary research^[2].

The trial's method of randomly allocating patients to treatment and control arms was designed by Austin Bradford Hill, with the explicit aim of achieving a fair comparison between groups, free of biased allocation. Although sporadic use of random allocation existed before this trial, the MRC streptomycin study is the one generally credited with marking the arrival of the modern clinical trial and the decline of alternation as standard practice. Part of what made the study so influential was necessity as much as design: streptomycin was scarce and expensive in postwar Britain, which meant researchers had a strong practical reason to prove, rigorously, whether the drug's benefits justified its cost and limited supply^[3].

An earlier MRC trial, testing patulin for the common cold in 1943, is recorded as the first double-blind controlled trial, paving part of the methodological ground that the streptomycin trial would later build on. As the discipline of controlled trials grew in sophistication over the following decades, the streptomycin trial continued to be referenced as the study that set the template for rigorous clinical research^[2].

What is a randomised controlled trial, and why does randomisation matter?

A randomised controlled trial (RCT) compares two or more groups of people: one or more experimental groups who receive a new treatment, and a control group who receive the current standard treatment, no treatment, or a placebo. The decision about which treatment each participant receives is made at random, based on chance rather than a choice made by the doctor or the participant, a process called randomisation^[4].

Randomisation aims to make the two (or more) groups in a trial as similar as possible, except for the treatment they receive. This matters because it lets researchers attribute any difference in outcomes between the groups to the treatment itself, rather than to some other underlying difference between the patients in each group^[4].

If a doctor were allowed to choose which treatment a given patient received, they might, consciously or not, give the new treatment to sicker patients or to younger patients. This would make the trial's results unreliable by exaggerating or hiding the true effect of the treatment. For this reason, randomised controlled trials are described as the most reliable way to compare treatments^[4]. Seen against the earlier history of alternation, this design requirement is not an abstract statistical preference but a direct, practical answer to a problem researchers had already identified decades before it became routine^[3].

What are the phases of a modern clinical trial?

New drugs and treatments usually have to go through a series of clinical trial phases to establish whether they are safe and effective before wider use^[4]. A clinical trial itself is a systematic process intended to establish the safety and efficacy of a drug or device in treating, preventing or diagnosing a disease or medical condition^[5].

Phase	Main purpose	Who typically takes part	Approximate scale
Phase I	Test the safety of a new treatment and look for side effects, such as making people sick or raising blood pressure[4]	Often healthy volunteers[4]	Small number of people[4]
Phase II	Test whether the new treatment is safe and has some effect on the condition it targets[4]	People who usually have the condition the treatment is meant to address[4]	Usually fewer than 100 people[4]
Phase III	Assess how well the new treatment works, usually comparing it against the best available current treatment or a placebo[4]	Patients randomised to the new treatment or the comparison treatment[4]	Hundreds or sometimes thousands of people[4]
Phase IV	Gather information on the drug's effects in various populations and any side effects linked to long-term use, after regulatory approval[4]	Patients using the drug once it is on the market[4]	Varies, post-marketing surveillance[4]

There is also a Phase 0, sometimes called the micro-dosing phase, in which very low concentrations of a drug, around 1/100th of the anticipated dose, are examined for a short period of time. The aim is to study pharmacokinetics, meaning how the body absorbs, distributes and eliminates the drug, before it moves into Phase I among healthy individuals. Phase 0 and Phase II are described as exploratory trial phases, Phase I as the non-therapeutic phase, Phase III as the therapeutic confirmatory phase, and Phase IV as the post-approval or post-marketing surveillance phase. Each phase builds on the data gathered in the one before it, so a treatment can only progress if the earlier phase has produced acceptable safety and efficacy signals^[5].

Why do new treatments need to go through clinical trials at all?

No matter how promising a new drug or treatment appears during laboratory testing, it must go through clinical trials before its benefits and risks can genuinely be known. Clinical trials are described as the most reliable way of testing treatments, because a new treatment is not always better and can sometimes be worse than an existing one^[4].

A clinical trial is a research study involving patients that compares a new or different type of treatment with the best treatment currently available, if one exists. Some trials look at ways to prevent illness, for instance by testing new vaccines, rather than treating an existing condition^[4].

Clinical research more broadly is an alternative term sometimes used to describe medical research. It involves people and is generally carried out to evaluate the efficacy of a therapeutic drug, a medical or surgical procedure, or a device as part of treatment and patient management. Any research that evaluates aspects of a disease, such as symptoms, risk factors and underlying mechanisms, may also be termed clinical research, even where it does not test a specific treatment^[5].

How is a clinical trial checked for safety before it can start?

A lot of care goes into making sure a clinical trial is as safe as possible before any patient is enrolled. At the MRC Clinical Trials Unit at UCL, for example, every trial is scientifically reviewed before it can start. The plan for the trial, known as the protocol, is sent to independent scientists for comment^[4].

As a patient or caregiver, this independent review is one of the things you can ask about before agreeing to take part: who reviewed the protocol, and whether the review was independent of the team running the trial^[4].

What does this history mean if you are considering taking part in a trial?

Understanding this history helps explain two things you may be told when invited to join a trial: why you might be randomly assigned to a group, and why you may not be told in advance which treatment you will receive^[4]. This kind of concealment is intended to prevent the allocation bias that random allocation, properly concealed, is designed to guard against^[3].

If you are asked to join a trial, you can expect to be told which phase the trial is in, since this tells you what the study is trying to find out: safety in a small group (Phase I), an early signal of effect (Phase II), a direct comparison against the current standard treatment (Phase III), or monitoring after approval (Phase IV)^[4]. Knowing the phase helps you understand how much is already known about the treatment's safety and effectiveness before you decide whether to take part^[5].

If you have questions about your rights, the trial's design or the independent review of its protocol at any point before, during or after taking part, raise them directly with the research team running the trial rather than waiting. The practice of independent scientific review described above exists precisely so that these questions have documented, accountable answers^[4].

Summary

The path from Nebuchadnezzar's ten-day dietary experiment to a modern randomised controlled trial spans roughly 2,500 years, but the underlying question has stayed remarkably consistent: how do you compare two groups fairly enough to trust the result^[1]. James Lind's 1747 scurvy experiment, with its paired groups and isolated conditions, and the 1946–1948 MRC streptomycin trial, with its concealed random allocation designed by Austin Bradford Hill, mark the two clearest turning points in that story^[3].

What changed between these milestones was not the basic idea of comparing treatments, but the growing recognition that allocation itself has to be protected from bias, whether from a well-meaning clinician or a predictable alternating pattern^[3]. That same principle, applied and refined over decades, is why a patient invited into a trial today may be randomly assigned to a group, why that assignment may be concealed, and why an independent review of the protocol happens before anyone is enrolled^[4]. The phased structure of modern trials, from micro-dosing through post-marketing surveillance, is simply this same principle extended across time, breaking one large question about a new treatment into a sequence of smaller, more answerable ones^[5].

This article is for general information and does not replace individual medical advice. If you are considering taking part in a clinical trial, discuss the specific protocol, its phase and your rights as a participant with the research team or your treating clinician.

1. [1] Collier R. Legumes, lemons and streptomycin: A short history of the clinical trial. CMAJ. 2009;180(1):23-24. doi:10.1503/cmaj.081879. Available at: (accessed 2 June 2026) – https://pmc.ncbi.nlm.nih.gov/articles/PMC2612069/
2. [2] Bhatt A. Evolution of Clinical Research: A History Before and Beyond James Lind. Perspect Clin Res. 2010;1(1):6-10. PMID: 21829774. Available at: (accessed 2 June 2026) – https://pmc.ncbi.nlm.nih.gov/articles/PMC3149409/
3. [3] Matthews RAJ. The problematic history of randomised controlled trials Part 1: presumption and confusion on the road to randomisation. J R Soc Med. 2026 Feb 13. doi:10.1177/01410768261419044. Available at: (accessed 2 June 2026) – https://pmc.ncbi.nlm.nih.gov/articles/PMC12913044/
4. [4] About clinical trials (accessed 2 June 2026) – https://refine.mrcctu.ucl.ac.uk/about/about-clinical-trials/
5. [5] Kandi V, Vadakedath S. Clinical Trials and Clinical Research: A Comprehensive Review. Cureus. 2023;15(2):e35077. doi:10.7759/cureus.35077. Available at: (accessed 2 June 2026) – https://pmc.ncbi.nlm.nih.gov/articles/PMC10023071/