TRIP Database

27/05/2026

Much of evidence production is shaped by the research agenda: what gets funded, studied, reviewed and turned into guidance.

But clinical practice creates its own agenda too.

It appears in the questions clinicians repeatedly ask: about safety, applicability, multimorbidity, conflicting guidance and messy real-world care.

Perhaps EBM needs to ask not only “what evidence exists?” but also “what are clinicians trying to find out?”

Evidence-based medicine has spent decades improving the supply of evidence: trials, systematic reviews, guidelines, summaries and search tools. But there is another side we understand much less wel…

17/04/2026

Positive feedback is encouraging. Negative feedback is often more useful.

A recent critique of an AskTrip answer highlighted a deeper issue: not just whether AI cites evidence, but whether it can recognise when that evidence may itself be unreliable.

That feedback has already pushed us to think about adding RIAT-related signals and Expressions of Concern into how we assess papers.

We get regular feedback on AskTrip’s answers and yesterday we got two in quick succession. The first was a question about the diagnosis of bladder cancer and the person who asked it left this…

10/04/2026

Why strong evidence doesn’t always mean a good answer

Frameworks like the GRADE Working Group help us judge how much we can trust evidence.

But they don’t tell us whether an answer is good.

In practice:
• Strong evidence can be poorly used
• Weak evidence can be well explained

At AskTrip, we separate these into:
Evidence Strength (GRADE-inspired, with Trip-based downgrades)
Answer Quality (does it actually answer the question, faithfully and appropriately?)

We’re currently testing this approach, with a planned release in early May.

A small shift—but an important one:
not just “how good is the evidence?”
but “how good is the answer?”

Full blog:
https://blog.tripdatabase.com/2026/04/09/from-grade-to-asktrip-evaluating-evidence-and-evaluating-answers/