Evidence-based medicine and clinical trials: how treatments are tested

Doctors in modern medicine, sometimes called allopathy, are trained to base decisions on tested evidence. This article explains the idea of evidence-based medicine and then looks at clinical trials, the studies that supply much of that evidence. It draws on a US National Library of Medicine textbook chapter, the US National Institute on Aging, the US Food and Drug Administration (FDA) and the US National Institutes of Health profile of India's trial rules.

1. Introduction

Evidence-based medicine is an approach to decisions that combines three things: the best current research evidence, the clinician's expertise and the patient's own values and preferences ^[1]. The idea was spread in clinical teaching in the early 1990s ^[1]. It means that a treatment is judged by how well it has been shown to work and how safe it is, and not only by tradition.

2. How evidence-based medicine works

The textbook chapter describes a repeating cycle ^[1]:

1. Form a clear question about a patient's problem.
2. Search for the best evidence.
3. Judge the quality of that evidence.
4. Apply it to the individual patient.
5. Check the result afterwards and adjust.

Not all evidence is equal

Evidence is often pictured as a pyramid. Animal studies, single case reports and expert opinion sit near the bottom. Observational studies, which watch people without testing a treatment, sit in the middle. Randomised controlled trials, systematic reviews and meta-analyses sit at the top ^[1]. A systematic review gathers all studies on a question, and a meta-analysis combines their numbers.

Why your values matter

The same chapter stresses that strong evidence does not automatically show the best choice for every person. A study may involve people who differ in age, other illnesses or setting, so the result may not apply to you ^[1]. Your own goals, worries and the burdens of treatment are part of the decision ^[1]. This is why your doctor discusses options with you.

Limits of the system

The chapter lists some weaknesses ^[1]:

• Studies with positive results are more likely to be published, which can make benefits look bigger than they are.
• Some trials are impractical for rare diseases, for ethical reasons or because of cost.

3. What a clinical trial is

The US National Institute on Aging explains that clinical trials are research studies that test whether a medical, surgical or behavioural approach is safe and effective in people. Often a new option is compared with an existing treatment ^[2]. Before a trial in people, researchers do laboratory tests and often animal studies ^[2].

Phases of a drug trial

The National Institute on Aging describes four phases for drugs and devices, with FDA approval possible after phase 3 ^[2]:

• Phase 1: about 20 to 80 people, to study safety, side effects and dosage.
• Phase 2: about 100 to 300 people, to gather early data on whether it works for a specific condition and keep examining safety.
• Phase 3: several hundred to a few thousand people, to gather more information and compare the product with other treatments.
• Phase 4: after approval, to monitor effectiveness and safety in large and varied groups over time.

The FDA's own description gives similar group sizes and notes that phase 3 studies provide most of the safety data, and their larger size makes rare or long-term side effects more likely to show up ^[4].

Randomisation, blinding and placebos

In a randomised controlled trial, chance decides who is assigned to the group receiving the tested treatment and who is assigned to the comparison (control) group ^{[2, 3]}. Some comparison groups receive a placebo, which is an inactive substance designed to have no real effect, although placebos are not used if an effective treatment is already available ^[3]. Trials may be single-blind or double-blind, which means that people do not know who is in which group, and the assignment can be revealed if it is medically necessary ^[3].

4. How participants are protected

The National Institute on Aging describes several safeguards in the United States ^[3]:

• Review by an institutional review board, a group of doctors, scientists and members of the public that checks that risks do not outweigh potential benefits.
• Informed consent, which means the team explains the purpose, length, procedures, risks, possible benefits and contacts before you decide, and keeps explaining during the study.
• Data and safety monitoring boards, made up of experts who review results as the study goes on and can recommend stopping a trial early if a treatment is not working or is harming people.

Signing a consent form is not a contract. You may leave a study at any time for any reason ^[3]. Benefits can include contributing to knowledge and, for some, early access to a possible treatment. Risks include side effects, extra appointments, a treatment that does not work and privacy concerns ^[3].

5. Clinical trials in India

The US National Institutes of Health profile of India explains that the Central Drugs Standard Control Organization (CDSCO) is the main regulator for approving and inspecting trials, under the New Drugs and Clinical Trials Rules, 2019 ^[5]. Both the head of CDSCO (the Drugs Controller General of India) and an ethics committee registered for the purpose must approve most trials before they start, and each trial site needs ethics committee approval ^[5]. The Indian Council of Medical Research issues ethical guidelines for research ^[5]. Ethics committees must include people from different backgrounds, including a lay person and a legal expert, and at least half of the members should be unconnected with the institution ^[5]. CDSCO can inspect sites and can suspend permissions if rules are broken ^[5].

6. When to see a doctor

If you are thinking of joining a trial, talk to your own doctor, and keep seeing your regular doctor during the study ^[2]. Ask the research team about the purpose, procedures, risks, possible benefits, how long the study lasts and who can see your information ^[3]. Tell your doctor about any new symptoms during a study.

Do not stop or change a prescribed medicine because of something you read about a study. Speak to your doctor first. In an emergency, call your local emergency number (112 in India, 911 in the United States).

7. Conclusion

Evidence-based medicine asks doctors to combine good research, their own experience and your values. Clinical trials, run in phases and under ethical and regulatory oversight, provide much of that research. No study fits everyone, so the best use of evidence is a conversation with your doctor about what it means for you.