Every health study on Doctor With Data gets two things: a plain-English verdict (such as “Promising” or “Overhyped”) and an evidence grade from 1 to 5, shown on the evidence meter. This page explains exactly how those grades are decided, so you can judge our judgements.
The 5-point scale
The evidence meter runs from Weak (1) to Strong (5). The grade reflects how much confidence the research, taken as a whole, deserves.
- 5 – Strong. Consistent findings across high-quality studies, including randomised trials. The effect is well established and unlikely to be overturned by one new study.
- 4 – Solid. Good evidence from well-designed studies, such as a large long-running cohort or a clean randomised trial. Genuinely informative, with stated limits.
- 3 – Promising. Real signal, real caveats. The study is worth your attention but the finding needs confirmation, is smaller than the headline suggests, or comes with methodological limits.
- 2 – Limited. Weak or heavily qualified evidence. Small samples, short duration, major confounding, or findings that do not survive scrutiny. Interesting, not actionable.
- 1 – Weak. Preliminary, anecdotal, or contradicted by better research. Treat as a curiosity until far better evidence arrives.
What goes into the grade
Six questions decide where a study lands on the scale:
- What kind of study is it? Randomised controlled trials and meta-analyses of trials sit at the top. Below them come prospective cohort studies, then case-control studies, then cross-sectional snapshots, then expert opinion. A big weak study does not outrank a small strong one.
- How big, and how long? Fourteen people followed for five months can produce a real signal, but it is fragile. Thousands of people followed for a decade is a different weight class.
- Is it consistent with other research? One study is a data point. A finding that lines up with the rest of the literature earns more confidence than a lone outlier.
- What was actually measured? We favour studies that measure outcomes patients care about, such as disease, fractures, or memory, over surrogate markers alone. When a headline leans on a proxy, we say so.
- Where could bias creep in? We look for randomisation, blinding, dropout rates, whether participants chose their own groups, and whether the analysis was planned before the data were seen.
- Who paid for it? Industry funding does not automatically discredit a study, but it is always disclosed and it sharpens our scepticism.
The verdict word
Alongside the number, each study gets a one-word verdict in plain English: words like Solid, Promising, or Overhyped. The word answers a different question from the grade. The grade asks “how good is the evidence?”; the verdict asks “what should a reader make of the headlines about it?” A well-run small trial can be Promising at 3 of 5; a scary headline contradicted by 106 pooled studies is Overhyped at 2 of 5.
Who grades
Every grade and verdict is assigned by Dr. Taimoor Asghar, MBBS, a physician and published community medicine researcher. No grade publishes without his review. When new research meaningfully changes the picture, grades are updated and the article notes the change.
What the grade is not
- It is not medical advice, and it is not a recommendation to start, stop, or change any treatment, supplement, or diet. Talk to a qualified clinician about decisions affecting your health.
- It is not a permanent judgement. Science moves; so do our grades.
- It does not rate researchers or institutions, only the strength of the evidence behind a specific claim.
Educational content only. Not medical advice.