What the research found
This is an educational breakdown of how to interpret medical study results, not a review of new research findings. The focus is on distinguishing between two ways risk gets reported: relative risk (the proportional change) and absolute risk (the actual change in your personal probability of an outcome). A treatment that "cuts risk in half" sounds dramatic but may shift your individual risk from 2% to 1%—a real benefit, but a smaller one than the relative framing suggests.
The episode walks through several interpretation tools: understanding what a p-value of 0.05 actually signals about reproducibility, recognizing that statistical significance and clinical significance are separate questions, and using confidence intervals to gauge how precise (or uncertain) a result really is. The number needed to treat (NNT)—how many people must receive a treatment for one person to experience the benefit—emerges as a practical metric for weighing efficacy against cost and side effects.
Why it matters for you
As a MyKine user, you're likely evaluating whether a supplement, peptide protocol, or training intervention is worth your time and money. Manufacturers and wellness media routinely quote relative improvements: "increases strength by 40%" or "boosts NAD+ by 50%." Understanding the absolute change—and how many people like you would actually see benefit—lets you decide whether that intervention fits your priorities.
This also applies to biomarker chasing. If a protocol shifts your testosterone by 15% but the confidence interval is wide, you're seeing noise, not a real effect. Similarly, if a study shows statistical significance (p < 0.05) but the actual difference between groups is 2%, it may not be worth protocol changes. The NNT concept helps here too: if you're considering a supplement for recovery, knowing whether 1 in 5 or 1 in 50 people see meaningful improvement changes the decision calculus.
Caveats
- This is a tutorial on research literacy, not a new experimental finding
- No original data, studies, or effect sizes are presented—it's a framework for reading others' work
- The tools described (relative vs absolute risk, NNT, confidence intervals) apply inconsistently across study types; observational work and RCTs require different interpretation
- The episode doesn't address how headlines often omit or misreport these numbers in the first place