The question the pyramid answers
Does this intervention cause this outcome. For that question, randomisation is the strongest available tool and a synthesis of randomised trials is stronger still, which is why they sit at the top.
Everything in the ordering follows from how well a design controls confounding for that specific question, and it is a genuinely useful ordering when that is the question you are asking.
Questions the pyramid does not answer
Why people do not take a medication. What it is like to live with a condition. How a policy is experienced by the people it applies to. For these, a randomised trial is not merely unnecessary, it is the wrong instrument entirely.
A well-conducted qualitative study is the highest level of evidence available for a question about meaning or experience, and citing the hierarchy against it is a category error rather than an appraisal.
Level and quality are different axes
A poorly conducted randomised trial is worse evidence than a large, careful cohort study, whatever the diagram says. The level tells you what a design can establish in principle; appraisal tells you whether this study did it.
Most appraisal tools separate these deliberately, and rubrics asking you to appraise are asking for the second rather than a restatement of the first.
Where systematic reviews sit
At the top, when they are good, and a poor review is weaker than the strongest study inside it. What makes the difference is the method: a documented search, criteria fixed in advance, and appraisal of what was included.
A review that gathered whatever came to hand and summarised it is a narrative review with a misleading label, and it is worth checking which you are looking at.
Using it in writing
State the level when you cite something central, and pair it with what the study actually did. Saying this randomised trial found is more informative than either the level alone or the finding alone.
Where your strongest evidence sits low on the hierarchy, say so plainly and explain why that design suits the question. A reader is far more persuaded by that than by silence about it.
Which hierarchy to use
Your program usually names one and they differ in their details. Some separate individual studies from syntheses more finely, some include expert opinion at the base, and some are designed for questions other than effectiveness.
Use the one you were given, name it, and apply it consistently. Switching between hierarchies within a paper is visible and it looks like the level was chosen to suit the argument.
Applicability, alongside level
The question a hierarchy cannot answer is whether a finding transfers to the setting you are writing about. Population, setting, comparator and outcome all have to be close enough for the evidence to bear on your claim.
Saying so explicitly is what a strong appraisal looks like. The strongest study in the world, conducted somewhere unlike your context, supports your recommendation less than a weaker one carried out where you are.
This is also the paragraph that most often distinguishes a practice doctorate project from a coursework essay, because a project has to be implemented somewhere specific and the question of transfer stops being academic.
Questions this raises.
Is expert opinion evidence?
It sits at the base of most hierarchies and it is not nothing. Where no research exists, professional consensus is what practice runs on, and citing it honestly as consensus is legitimate. What it cannot do is outweigh research evidence, and presenting it without labelling it is what causes trouble.
Where do clinical guidelines fit?
Slightly outside the hierarchy, because a guideline is a synthesis plus a judgment about what should be done. A good one states the strength of its own recommendations and the evidence behind each. That internal grading is usually more useful to cite than the guideline's position on any pyramid.
Can qualitative research be top-level?
For questions about experience, meaning or acceptability, it is the appropriate design and therefore the best available evidence. Several frameworks now publish separate hierarchies for those questions for exactly this reason. Applying an effectiveness pyramid to a question about experience is a mistake about the question, not about the study.
Does a higher level always mean a better citation?
No, and this is where the hierarchy gets misused most often. A randomised trial conducted in a population unlike yours is weaker support for your particular claim than a cohort study carried out in exactly your setting. Applicability sits alongside level rather than underneath it, and a rubric asking for appraisal wants both of them considered and both of them mentioned.
How do I show I appraised rather than ranked?
By saying something about the study beyond its design. The sample, the setting, the outcome measure, the follow-up period, and what each of those means for whether the finding transfers to the context you are writing about. A sentence doing that demonstrates appraisal. A sentence naming the level demonstrates only that you can read a diagram, which is not what the criterion is measuring.
Dr. Sigrid Nyholm
PhD, Nursing, on the evidence-based practice side of the hall. Writes for The Footnote and coaches the students who bring this work to the desk. The rest of the specialists.
Elsewhere in the hall.
Send the assignment and the deadline.
A specialist credentialed in your own field reads what you actually have and replies within two hours with a plan and a price. The reading costs nothing, and if the honest answer is that we are not the right hall for it, you will get that instead.
