ES · EN
Book 30 min
Methodological integrity
Whoever funds a study defines the question, not the answer. This is what you can demand of us and what we will not do, written before you hire us.
Whoever commissions an evidence report almost always has an expectation about the result. A manufacturer hopes its technology comes out well; a payer hopes the opposite; a ministry hopes the recommendation is affordable within its budget. Those expectations are legitimate. What they cannot do is determine the conclusion.
This statement exists so that you know, before hiring us, what you will get and what you will not.
The client defines the question, the scope and the deadline. The client does not define the result. The methodological team determines the direction and strength of a recommendation, and the rating of certainty of evidence, using criteria set in advance.
This is agreed in writing before we start. It is not a verbal promise.
The question, the inclusion criteria, the critical outcomes and the rules for rating certainty go into the protocol before the search begins. Changing those criteria once the results are in is the most common way of biasing a review, and the hardest to spot, which is why the protocol is the first deliverable, not the last.
If something has to change mid-project, we document the change, its date and the reason, and all of it appears in the final report.
Everyone involved in a project declares their financial and intellectual interests before starting. Not just the financial ones: positions already published on the topic bias judgement as much, or more.
We document and manage those interests. Depending on their magnitude, that may mean excluding someone from a specific judgement, from a recommendation, or from the project. The declaration goes into the deliverable.
Asking for declarations of interest means starting at home. These are the current affiliations of both directors, beyond Based Evidence. None of them shapes the method or the conclusion of a project, and any of them may be reason enough for one of us to step back from a particular commission.
If one of these affiliations touches a project, we declare it in writing in the deliverable, exactly as we require of anyone else involved. We keep this list current; if you spot something missing, tell us and we will correct it.
You can argue the method with us, and you should: if you find a study we did not include, a criterion misapplied or an extraction error, we correct it and thank you for it. That scrutiny makes the work better.
The conclusion that follows from the agreed method is not open to negotiation. If the disagreement persists, you are entitled not to publish the report. You are not entitled to have us sign something other than what we found.
If a deliverable is altered after we hand it over, we withdraw our authorship.
We document every judgement: what went in, what stayed out, with what certainty and why. That trail lets a third party reconstruct the reasoning and reach the same conclusion, or point precisely to where they disagree. It is also what carries a decision through an audit, a committee and peer review.
Last updated: August 2026. We attach this statement, signed, as an annex to every proposal.