Skip to content
No Vape

Why did a vaping survey move a brand answer to another device category?

Trace a typed model through the survey’s review rule and compare the raw answer with the analytical category.

No Vape · Updated

Stack of old newspapers on a table
Illustrative image · Mr Cup / Fabien Barral / Unsplash

Read how the two answers were collected

In the original BMJ Open device-type survey, people selected devices from types illustrated and defined in the questionnaire. A follow-up asked for brand and model for a type meeting the relevant reported-use condition. That creates two pieces of information about a device. A later count can depend on how authors handled disagreements between them.

Find the exact follow-up condition. Someone may never have been asked for a model if their earlier answer did not qualify. That route differs from a person who saw the brand question and left it blank. Neither situation tells you a device belonged to a different category.

BMJ Open: Types of electronic nicotine delivery systems (ENDS) used by people of different age and smoking status groups: results from a US nationally representative survey

Trace a disagreement through the published rule

The authors describe independent review by two research team members, with discussion to resolve disagreements. Their methods distinguish an answer judged to match, an answer they could not determine and several kinds of incorrect answer. When an identifiable brand or model belonged to a different type of e-cigarette, the authors describe recoding it to the other type.

For a fictional example, someone checks “category A” and types a model that the study’s specified rule identifies as “category B.” The row can retain the raw A answer and contribute to B under the recode. A second respondent leaves the field empty. Unless the methods provide another rule for that case, the empty field supplies no basis for a B reassignment.

BMJ Open: Types of electronic nicotine delivery systems (ENDS) used by people of different age and smoking status groups: results from a US nationally representative survey

Separate reassignment from exclusion

Make a short note with four columns for the original type, typed field, review result and analytical handling. A recode can move the category represented in a table. An exclusion can remove a record or a particular item from a result. Locate the table’s actual methods before assuming the same action followed every flagged response.

The same label can cover different histories. One record may have started in category B and another may have reached B after review. If the paper provides only the final distribution, you cannot reconstruct that history row by row. Keep the example fictional and avoid inferring any participant’s identity or use from a published count.

Report the final category with its provenance

A useful summary says whether the table uses self-selected or reviewed types, what could trigger reassignment and how unresolved answers were treated if the paper supplies it. This tells readers how the category was built. It cannot authenticate a consumer’s device, verify its contents or establish absorbed exposure.

Questions about this step

Does a typed model always override a checkbox?

Check the specific study rule. In this report an identifiable model from another device type could trigger recoding, while other review findings had their own classifications.

Sources and scope

Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.

  1. BMJ Open — Types of electronic nicotine delivery systems (ENDS) used by people of different age and smoking status groups: results from a US nationally representative survey

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