Every week, another headline promises a breakthrough. A new compound doubles endurance. A peptide reverses aging. A supplement protocol outperforms everything that came before it. The claims arrive fast, the citations look legitimate, and the enthusiasm is contagious. But enthusiasm is not evidence — and knowing the difference is one of the most valuable skills you can build as someone who takes their health seriously.
Start with the study design
Before you read a single result, look at how the study was structured. The hierarchy of evidence matters enormously. At the top sit randomised controlled trials (RCTs) — studies where participants are randomly assigned to a treatment or control group, reducing the chance that other variables are driving the outcome. Below that are cohort studies, case-control studies, and at the bottom, anecdotal reports and expert opinion.
When a headline cites a study, ask immediately: was this randomised? Was there a control group? Was it blinded — meaning participants and researchers did not know who received the treatment? Each of these design features reduces bias. A study without them is not worthless, but its conclusions carry far less weight.
Check the sample size and population
A study of twelve people tells you something different from a study of twelve hundred. Small samples produce noisy results — effects that look dramatic can disappear entirely when tested in a larger, more representative group. This is not a flaw unique to bad research; it is a fundamental statistical reality.
Also ask who the participants were. A study conducted on sedentary older adults with metabolic disease does not automatically apply to a healthy, trained 32-year-old. A study on elite athletes does not automatically apply to recreational gym-goers. Population specificity matters, and extrapolating results beyond the studied group is one of the most common ways performance claims go wrong.
Understand what was actually measured
Studies measure outcomes — but not all outcomes are equal. Primary outcomes are the main thing the study was designed to test. Secondary outcomes are additional measurements collected along the way. The problem is that secondary outcomes are far more likely to produce false positives by chance, and they are also far more likely to be the ones that make it into a headline.
Look for whether the outcome measured is a surrogate marker or a direct clinical endpoint. A surrogate marker is something that correlates with a health outcome — like a blood biomarker — but is not the outcome itself. Improving a surrogate does not always translate to improving the thing you actually care about. When a study reports that a compound improved IGF-1 levels, that is a surrogate. Whether that translates to meaningful changes in body composition, recovery, or longevity requires separate evidence.
Mind the gap between statistical and practical significance
A result can be statistically significant without being practically meaningful. Statistical significance tells you that an effect is unlikely to be due to chance. It says nothing about whether the effect is large enough to matter in the real world.
A study might show a statistically significant improvement in a performance marker — but if that improvement is smaller than the normal day-to-day variation in that marker, it is not something you would notice or benefit from. Always look for effect sizes alongside p-values. A small p-value with a tiny effect size is a technically real but practically irrelevant finding.
Follow the funding and the incentives
Industry-funded research is not automatically corrupt, but it does carry a higher risk of bias. Studies funded by companies with a financial interest in a positive result are more likely to produce positive results — not necessarily through fraud, but through subtle choices in design, analysis, and reporting that favor the sponsor's product.
This does not mean you dismiss industry-funded research. It means you weight it appropriately, look for independent replication, and pay attention to whether the researchers had full control over the data and publication decisions.
The headline is not the study
Science journalism operates under different incentives than science. Journalists need clicks. Researchers need grants. Neither of those incentives aligns perfectly with communicating nuance. The result is a systematic gap between what a study actually found and what gets reported.
When a headline says a compound reverses ageing or doubles recovery speed, go to the source. Read the abstract. Look at the methods section. Check the actual numbers. Most of the time, the study is far more modest than the headline suggests — and that modesty is not a failure of the research. It is how science actually works: incremental, uncertain, and always provisional.
Build a habit, not a one-time check
Research literacy is not a single skill you acquire and then apply automatically. It is a habit of asking the same questions every time a claim lands in front of you. What was the design? Who were the subjects? What was actually measured? How large was the effect? Who funded it? What does the headline leave out?
The goal is not scepticism for its own sake. It is the ability to engage with evidence honestly — to be genuinely excited by strong findings and appropriately cautious about weak ones. That combination is what separates someone who optimizes their health from someone who just cycles through the latest trends.
The performance space is full of people selling certainty. The research rarely offers it. Learning to sit with that uncertainty — and make good decisions anyway — is the real skill. Start with the study design, check the population, understand what was measured, and never let a headline do the thinking for you.