How to read a clinical study

A simple guide for thinking consumers of health information

Every week, a new headline tells us that something is good for us, or terrible, or causes cancer, or prevents cancer. One day coffee is the elixir of life. The next, it is shortening it. Eggs are out, then in, then out again. It is enough to make anyone tune out of health news entirely.

But the science itself is rarely the problem. The way it is reported, often by journalists with deadlines and editors hunting for clicks, is. Learning to read a clinical study, even at a very basic level, transforms how you make decisions about your own health. You stop being at the mercy of headlines, and you start being able to weigh what you read for yourself.

Here are seven questions to ask every time you read a health claim, however bold the headline.

1. Who funded the study?

This is the very first place to look. Most reputable journals require authors to declare their funding sources and any conflicts of interest, usually in a small section at the end of the paper. It is one of the most important sentences in the entire study, and it takes seconds to find.

Was the study funded by a government body, a university, or a charitable research foundation? Or was it funded by the company that makes the product being studied? Industry-funded studies are not automatically untrustworthy, but they do need a closer look, particularly when the result happens to support the funder's product.

Look for phrases like "This research received no external funding" or "The authors declare no conflicts of interest." These are good signs. Phrases like "Funded by [Pharmaceutical Company X]" warrant a closer read of the full study.

An example of how funding and conflict of interest sections appear in real published studies.

2. What kind of study is it?

Not all studies are created equal.

A randomised controlled trial (RCT) is generally the gold standard, where participants are randomly assigned to either the treatment or a control group.

A cohort study follows a group of people over time but does not assign treatment.

An observational study simply looks at what people are already doing.

A systematic review or meta-analysis gathers and analyses the available studies on a topic, using specific criteria (for example certain controls or excluding certain participants), to summarise the overall evidence.

A case study describes a single patient or small group. Each type of study has its place, but an RCT or a well-conducted systematic review carries far more weight than someone reporting that their cousin felt better after trying something.

An example of how the type of study is described in a real published paper.

3. How big was the study?

A study of 10,000 people tells you something quite different to a study of 12. Larger studies are statistically more reliable because they reduce the influence of chance and outliers.

Look for the letter "n" in the study, often in the abstract or methods section. "n = 1,247" means there were 1,247 participants. "n = 18" means there were eighteen.

A useful rule of thumb: be cautious of any study with fewer than 100 participants, unless the results are extreme or the study is a follow-up to a larger one. A study with thousands of participants, conducted across multiple sites and multiple years, carries far more statistical weight than a small pilot study.

4. Who were the participants?

A study done on healthy young men in their twenties may not apply to a 55-year-old woman in menopause. A study done on rats or mice may not apply to humans at all. A study done in Japan may not translate perfectly to South Africans, given differences in diet, genetics and lifestyle.

In the methods section of any good study, you will find a description of the participants: their age, sex, health status, ethnicity, and any conditions they had. This is sometimes called demographic data. If a study showed that a supplement helped postmenopausal women with osteoporosis, that does not necessarily mean it will help a healthy 30-year-old. The findings apply most directly to people like those in the study. Another important point to be aware of:

in vivo means experiments performed inside a living organism (eg humans or animals)
in vitro means experiments are performed outside a living organism in a controlled lab setting, such as a test tube or petri dish

An example of how in vitro and in vivo studies are described in a real published paper.

5. How long did the study run?

Health effects often take weeks, months or years to show up. A two-week study can tell you about very short-term changes, but very little about long-term safety or sustained benefit.

For supplements, look for studies of at least 8 to 12 weeks, ideally longer. For lifestyle interventions, multi-year studies carry the most weight. The longer the follow-up, the more confident we can be that the effects observed are real, sustained, and not just initial enthusiasm or placebo response.

6. What is the actual effect size?

This is where many health headlines stretch the truth to the point of breaking.

A study may report that "X reduces the risk of Y by 50%". That sounds dramatic. But if the baseline risk was 0.2%, a 50% reduction takes it to 0.1%. Statistically significant, yes. Clinically meaningful for an individual, probably not. The absolute risk has gone from "very small" to "slightly smaller".

There is an important difference between relative risk (a 50% reduction) and absolute risk (going from 0.2% to 0.1%). Headlines almost always report relative risk because it sounds more impressive. The full study almost always contains the absolute numbers.

Look for the actual numbers. A treatment that reduces deaths from 4 per 1,000 to 2 per 1,000 is more meaningful than one that reduces a 0.2% risk to 0.1%, even though both can be reported as "50% reductions".

7. Has it been replicated?

A single study, no matter how exciting, is just that. A single study. New findings need to be tested, retested and confirmed by other research groups before we can be confident they reflect a real phenomenon and not chance or a quirk of that particular study.

If a finding has been replicated multiple times, by different research groups, in different countries, with different populations, that is a much stronger basis for action than a single dramatic result.

This is also why systematic reviews and meta-analyses are so valuable. They pull together the results of multiple studies into one overall picture, which is far more reliable than relying on any single trial.

One last thing

The best science is actually usually quiet. It rarely makes for sensational headlines, but it accumulates slowly over years and decades, building a body of evidence that gradually shifts our understanding.

The dramatic headlines are often the least reliable. The steady, replicated, boring research is what truly moves the needle.

You do not need to become a research scientist. But asking these seven simple questions, especially the first three, will already put you ahead of most people consuming health information. It takes practice, and you will not always have time to dig into every claim. But for the things that really matter, the supplements you take, the diets you follow, the lifestyle choices you make, it is genuinely worth a few extra minutes of curious, careful reading.