Background: Despite the popularity of concurrent use of e-cigarettes and cigarettes (dual tobacco use) among young adults, few interventions address the cessation of both tobacco products. Even fewer interventions applied human-centered design (HCD) in the development process. Objective: This study employed a HCD approach to develop smartphone-based intervention messages for dual tobacco cessation for young adults. Methods: Intervention messages were developed based on theories, cessation guidelines, existing messages, and our previous formative study. Three rounds of message testing were conducted asynchronously via an online platform with 35 young adults (18-29 years old) who currently used both e-cigarettes and cigarettes and were motivated to quit either smoking or vaping in the next 6 months. In each round, a new sample of 10–12 participants evaluated the messages individually. For the quantitative assessment, participants viewed and rated each message on a scale from 1 (“very low degree”) to 5 (“very high degree”) across four components: Comprehension (“This message is easy to understand”), Usefulness (“This message is useful for encouraging tobacco cessation”), Tone (“The language is clear and non-judgmental”), and Design (“The design is appealing”). For the qualitative assessment, participants used an app-enabled feature to place markers on specific parts of messages they liked, disliked, or found confusing, and then provided brief explanations for their feedback. Initial messages were assessed during the first 2 rounds of testing, and those with low mean scores were revised based on participants’ feedback and re-tested in the third round. Results: We found significant improvements in message ratings after refinement. The overall mean score increased from 3.6 (SD = 0.4) to 4.6 (SD = 0.2) (P < .001). Specifically, mean score of “Comprehension” improved from 4.0 (SD=0.5) to 4.9 (SD=0.2) (P < .001), mean score of “Usefulness” increased from 3.0 (SD = 0.6) to 4.4 (SD = 0.4) (P < .001), mean score of “Tone” increased from 3.8 (SD = 0.6) to 4.8 (SD = 0.2) (P < .001), and mean score of “Design” increased from 3.4 (SD = 0.48) to 4.4 (SD = 0.3) (P < .001). The qualitative assessments highlighted design elements related to message liking, such as clear layout, minimalistic imagery, italicized quotes, and highlighted keywords. Conversely, design features related to message dislike included color shades, lengthy text, and confusing wording. Conclusions: This study demonstrated the use of HCD in developing smartphone-based intervention messages to support dual tobacco cessation among young adults. Integrating remote message testing improved the feasibility of rapid prototyping while enhancing the relevance and appeal of message content and design. Future interventions targeting emerging health behaviors among young adults may benefit from incorporating a remote testing method to efficiently gather user feedback and refine intervention messages in a timely manner.

Background: Nicotine and cannabis vaping are common among young adults, potentially leading to adverse health consequences. Identifying real-time antecedents of vaping events may provide insights into intervention targets pertinent to these behaviors. Objective: This study aimed to examine real-time antecedents of nicotine and cannabis vaping and same-moment co-vaping among young adults. Methods: We collected ecological momentary assessments (EMAs) via a smartphone app among California young adults (ages 18-29) in 2023-2024. Participants completed four random prompts each day for 30 consecutive days. Outcomes were defined as whether participants reported being about to vape nicotine, cannabis, or both substances (same-occasion co-vaping) in a given EMA. We used mixed-effects logistic regression models to examine real-time antecedents of each outcome. Results: Overall, 113 participants (mean age 23.8 years, 63% female) completed 9,001 EMAs. Similar antecedents of all three vaping outcomes were craving and using alcohol. Increased cravings for a given substance were associated with a higher likelihood of vaping that substance or co-vaping. Craving for cannabis vaping was associated with lower odds of reporting nicotine vaping (Adjusted odds ratio [AOR]: 0.87, 95% CI: 0.82-0.92). Feeling happier was associated with higher odds of reporting co-vaping (AOR: 1.13, 95%CI: 1.01-1.27) while feeling more stressed was associated with lower odds of vaping nicotine (AOR: 0.95, 95%CI: 0.91-0.98) or cannabis (AOR: 0.91, 95%CI: 0.86-0.97). Seeing tobacco advertisements was associated with higher odds of vaping nicotine (AOR: 3.09, 95%CI: 1.48-6.46) and co-vaping (AOR: 4.15, 95%CI: 1.18-14.52). Cannabis vaping was more likely to occur in the afternoon (AOR: 1.52, 95%CI: 1.16-1.98) and nighttime (AOR: 1.95, 95%CI: 1.45-2.63) than in the morning. Co-vaping was also more likely to occur in the afternoon (AOR: 1.59, 95%CI: 1.14-2.22) and nighttime (AOR: 1.84, 95%CI: 1.26-2.71) than in the morning, but the association was not held for nicotine vaping. Nicotine vaping was more likely to occur in weekends compared to weekdays (AOR: 1.25, 95%CI: 1.09-1.45), but no significant associations were found for cannabis vaping and co-vaping. Conclusions: We found similar antecedents (craving, alcohol use) and unique antecedents (mood, advertising exposure, and time of day) for nicotine vaping, cannabis vaping, and same-occasion co-vaping, suggesting targets for future vaping cessation interventions.