37/100

Emerging evidence

Trace: a gain and a problemPublished recordScienceThe Trace

Evaluating a large language model-powered system for semi-structured interview in health research

Truvace record
TRV-2026-2945raw data (JSON)
Version
v1Published
First published
Evidence type
Interviews or qualitative study
Sources
1 sourcePeer-reviewed
sha256
c70120614604…check it

Objectives: Conducting qualitative health research with marginalized populations presents persistent challenges related to trust, disclosure, and resource constraints. Traditional in-person interviews can be costly and difficult to scale, while participants may hesitate to share sensitive experiences openly. This study assessed the feasibility and acceptability of a large language model (LLM)-powered conversational interview system as an alternative to semi-structured qualitative interviews with marginalized pop…

49Gain side

How the two sides scoreClose call

48Problem side

The gain

In a feasibility study of 28 caregivers of children with asthma living in subsidized housing, participants said an AI-powered interview system was easy to use and rated it positively for clarity, comfort and accessibility.

The problem

In the same 28-caregiver asthma interview study, researchers recorded intermittent technical errors, limited contextual adaptability and uneven comfort with digital technology among participants using the AI interviewer.

A fixed formula sets both scores from the evidence and its sources. Reader votes don’t move them.

What the scores mean

G is the gain side, out of 100. It goes up when the gain is specific, when it measurably helped people and when the sources behind it are strong.

P is the problem side, out of 100. It goes up when the problem is specific, when it measurably hurt people and when the sources behind it are strong.

What this doesn’t fix

The study assessed only feasibility and acceptability with 28 caregivers, and participants met technical errors, limited contextual adaptability and varying comfort with digital technology.

Sources

1 on file
  1. JAMIA Opendoi:10.1093/jamiaopen/ooag218
    Peer-reviewed2026-11-12

Cite this record

v1

Truvace Impact Record TRV-2026-2945, v1: “Evaluating a large language model-powered system for semi-structured interview in health research.” Truvace, 2026-11-15. /record/TRV-2026-2945 (accessed on the day you copy it). sha256 c7012061460448af…

Change history

1 version

Every change made to this record since it was first published, and why. Nothing has changed since it went up.

  1. Publishedv1c70120614604…

    Certified into the record

Check this record hasn’t been altered

Check this record hasn’t been altered
How to check it yourself, without trusting this page

Download any version from /api/record/TRV-2026-2945 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.