A cross-sectional online survey of the impact of new tobacco health warnings in Colombia

Data collection method:

Data were collected via the online survey platform Qualtrics. Adult smokers (defined as ≥ 18 years of age and having smoked at least 100 cigarettes in their lifetime and currently smoking at least one cigarette per week) were recruited from the staff and students at the Universidad Nacional de Colombia and from the general public. The research team used existing mailing lists to contact members of the university across eight cities in Colombia and attended student classes to encourage students to complete the surveys. We recruited participants from the general population through social media, word of mouth and posters in public spaces. Data were collected at Time 1, prior to the introduction of the new health warnings in Colombia in 2018 (n=1985, 72% male), and at Time 2, twelve-months post-introduction (n=1572, 69% male). During data collection participants were first shown an information statement before completing a tick-box consent page. All text and questions were presented in Spanish. Participants completed screening questions and those who were ineligible were taken to the end of the study. Eligible participants completed demographic questions, responses to warnings on packs, knowledge of health risks and attitudes towards smoking. An attention check item was included in this section (Please select the strongly agree option to let us know that you read all of the survey instructions). Each of the six new warnings were presented individually on screen in a randomised order, and participants completed questions about each warning. After reading a debriefing page, participants who wished to be entered into the prize draw were directed to a second survey. Participants who participated in the first time were re-contacted at the second time to ask if they would like to participate again.

Technical details and requirements:

Data were collected using the online survey platform Qualtrics https://www.qualtrics.com Data analysis was conducted using IBM SPSS Statistics 22.

Additional information:

Each variable is titled and labelled in the SPSS datasheet. In order to examine market spread of new warnings over the implementation period we calculated percentages of participants in the two times who reported being aware of seeing each of the six new warnings. The internal consistency of the latent constructs (e.g. negative affect, cognitive elaboration) were assessed using Cronbach’s alpha. All latent constructs achieved acceptable to good internal consistency and latent variables were calculated by taking a mean of the individual items (see Supplementary Materials - Table 2).