Prevention Needs and Target Behavior Preferences in an App-Based Addiction Prevention Program for German Vocational School Students: Cluster Randomized Controlled Trial

Abstract

Background: Vocational school students exhibit a high prevalence of addictive behaviors. Mobile phone-based prevention programs targeting multiple addictive behaviors and promoting life skills are promising. Tailoring intervention content to participants' preferences, such as allowing them to choose behavior modules, may increase engagement and efficacy. There is limited understanding of how personal characteristics relate to module choices.

Objective: This study examined the prevention needs of German vocational school students as well as their prevention preferences through self-determined module choice in the multibehavior app-based addiction prevention program ready4life.

Methods: A 2-arm cluster randomized controlled trial recruited German vocational school students aged ≥16 years. Among 376 classes from 35 schools, ready4life was introduced during a school lesson. Students were invited to download the ready4life app and completed an anonymous screening with individualized risk and competence feedback in the form of a traffic light system. Informed consent was provided by 2568 students. Intervention classes received individual app-based coaching with weekly chat contacts with a conversational agent over 4 months. They could choose 2 of 6 modules: alcohol, tobacco, cannabis, social media and gaming, stress, and social competencies. Control group classes received a link to health behavior information and could access coaching after 12 months.

Results: Prevention need was high. For 86.2% (2213/2568), ≥2 risks were reported based on yellow or red traffic light feedback. Within the intervention group, stress (818/1236, 66.2%) and social media and gaming (625/1236, 50.6%) were the most chosen topics, followed by alcohol (360/1236, 29.1%), social competencies (306/1236, 24.8%), tobacco (232/1236, 18.8%), and cannabis (131/1236, 10.6%). Module choices closely aligned with received traffic light feedback among those with 1 or 2 risks. Multilevel regression models showed that women were significantly more likely to choose the stress module (odds ratio [OR] 2.38, 95% CI 1.69-3.33; P<.001); men preferred social media and gaming (OR 0.52, 95% CI 0.40-0.69; P<.001), alcohol (OR 0.50, 95% CI 0.37-0.67; P<.001), and cannabis (OR 0.37, 95% CI 0.21-0.63; P<.001) when holding age, educational track, and prevention need for the corresponding behavior constant. Younger students were significantly more likely to choose the cannabis module (OR 0.81, 95% CI 0.74-0.90; P<.001). Educational track also influenced module choice (eg, those with a lower educational level were more likely to choose alcohol and cannabis, suggesting a positive equity impact). Students' prevention needs significantly influenced choice of the module (eg, higher alcohol consumption increased the likelihood of choosing the alcohol module; OR 1.31, 95% CI 1.20-1.43; P<.001).

Conclusions: Our study confirms vocational school students' high prevention needs regarding addictive behaviors. Students' module choices were highly congruent to their demonstrated needs, with most students being interested in the stress module. Module choice also differed by age, gender, and educational track.

Trial registration: German Clinical Trials Register DRKS00022328; https://drks.de/search/en/trial/DRKS00022328.

International registered report identifier (irrid): RR2-10.1024/0939-5911/a000811.

Keywords: computer tailoring; eHealth; module choice; multiple addictive behaviors; vocational school students.

Figure 1

Individual risk and competence feedback provided by the app (English translation).

Figure 2

Flowchart of clusters and participants. *n=11 students were identified who had registered for the study twice; these were already subtracted from the download count. **The app included multiple versions tailored to different countries or languages. Due to a technical issue, between September 2, 2021, and December 10, 2021, a total of 307 students were mistakenly directed to a different version of the app instead of the German version. As a result, these students used a version of the app that was not part of this study, and informed consent could not be obtained from them. ***App users were able to initiate the deletion of all data collected at any time point regardless of whether they had just started the screening or provided informed consent and were using the intervention already. ****Group allocation was conducted at the class level based on class-specific passwords; due to mistyping of passwords, 1.6% (20/1286) of the students from the intervention group did not receive access to the intervention, and 0.2% (3/1282) of the students from the control group received access to the intervention.

Figure 3

Percentage of students who received red, yellow, or green traffic light feedback by topic (N=2568). Traffic light feedback was based on self-reported behaviors and gave advice on potentially useful modules.