Do device photos give a vaping survey more usable data?
Read the VAPER study by separating completed photo uploads, identifiable product details and agreement on cases where both methods have an answer.
No Vape · Updated

Identify what the picture would have to answer
A product photo can be clear enough for someone to identify the device model while leaving the label on a liquid container unreadable. Start with one variable: device model, flavor or another characteristic named in the table. Then distinguish the number of participants asked, the number of uploads and the number of photos that supplied a usable answer for that field.
A self-report has its own possible gaps. A participant can be asked and say they do not know. They can also be routed past a question because it does not apply to their product. Keep those situations separate when reading a methods comparison, since each can affect the number of usable answers differently.
Look at the usability column as well as the agreement result
Crespi and colleagues compared reported characteristics of the most-used device and liquid with characteristics coded from submitted photographs in a US adult cohort of regular users. Their paper distinguishes the agreement assessment from exclusions and usable data, and discusses routing separately. The direction of the usability comparison differed by variable.
The results about device model and nicotine formulation do not extend to all liquid information. The authors found more usable self-report information for most other variables when considering the items actually asked. The study also found weaker agreement for nicotine concentration than for the other reported comparisons. A label visible in a photo is not a laboratory assay of contents or an exposure measure.
Use a paired example with an explicit base
In an invented set of ten eligible participants, eight give a usable typed device model and five upload an image from which it can be coded. Suppose four people have both usable answers and those four match. The agreement observed in those four pairs does not tell you that the photographs covered all ten participants. It also does not tell you that either method was correct for the six without a usable pair.
Try the same worksheet for a different variable. Record the eligible base, usable reports, usable photo codes and number of paired results. A field might have more usable photos in one column and fewer in another. The paired percentage is intelligible only after you know which records had two observations to compare.
Limit the claim to the products and people studied
The study considered its own recruited cohort, collection procedure and coding rules. Do not treat the findings as a guarantee that images work the same way for all devices or newer packaging. If the methods do not show what happened to a field you care about, describe the question as unresolved instead of converting a photo submission count into a success rate.
A useful reading note can say which method produced more usable answers for which field, followed separately by what paired agreement showed. This is a question about interpreting published evidence. It does not require you to collect another person’s images or infer the nicotine they absorbed.
Sources and scope
Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.
Written with AI assistance and checked against the cited sources. No clinical review is claimed. This guide covers practical planning, not individual medical treatment. How we research and write