Agentic education · Learning science

Actionable companion · Feedback & scaffolding

Learning with agency: Usage of an App-Based Addiction Prevention Program for German Vocational Students: Secondary Analysis of Data From a Cluster Randomized Controlled Trial

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paperSearch paper agentCC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Secondary school. Verify the participant population and task before transferring the findings.

Original authors’ abstract, reproduced under the linked source license:

Abstract Background Digital interventions have been successfully used to address addictive behaviors in adolescents and young adults. However, maintaining consistent usage remains a challenge. There is limited evidence on the determinants of usage with multiple behavioral interventions among vocational students. Objective This paper aims to describe how vocational school students used the app-based addiction prevention program “ready4life” and to analyze student characteristics as potential determinants of intervention usage. Methods A 2-arm cluster randomized trial evaluated “ready4life” among German vocational students aged ≥16 years. After downloading the app during class, students completed an anonymous screening and received individual risks and competencies feedback. Intervention participants (n=1286) received 4 months of individual app-based coaching, with weekly chat contacts with a conversational agent. They were asked to choose 2 of 6 modules: alcohol, tobacco, cannabis, social media and gaming, stress, and social competencies that were delivered sequentially in a random order. In addition to the weekly chats, users could self-initiate chat dialogues at any time. Chats included media (videos, images, or links), quizzes, and contests. Control participants (n=1282) received a link to health behavior information and could access coaching after 12 months. Results Usage was low among intervention participants who received their assigned intervention (n=1266; females n=569, 44.9%; mean age 19.53, SD 3.57 years). On average, participants started 4.8 (SD 5.1) and completed 4.3 (SD 5.2) of 16 weekly in-app chats. Most students (n=903, 71.3%) completed no self-initiated chats, and 50.2% (n=636) stopped using the app before week 3. Unadjusted negative binomial multilevel regression models showed that females completed significantly more weekly dialogues (P<.001; incidence rate ratio [IRR] 1.55, 95% CI 1.33‐1.80), while fewer were completed by individuals with higher self-efficacy (P=.04; IRR 0.96, 95% CI 0.93‐0.998), higher social competencies (P<.001; IRR 0.97, 95% CI 0.95‐0.98), and individuals engaging in more addictive behaviors (P<.001; IRR 0.87, 95% CI 0.82‐0.93). Concerning specific educational tracks, professionals, technicians, associate professionals, and vocational grammar school students had the highest number of completed weekly dialogues. Determinants of completed self-initiated chats and usage time largely aligned with the findings for weekly dialogues. Additionally, those with higher perceived stress completed significantly more self-initiated chats (P<.001; IRR 1.19, 95% CI 1.08‐1.31). Age and year of education were not significantly associated with any of the usage parameters. Conclusions Our study supports the existing evidence that maintaining consistent usage among adolescents and young adults is a major challenge for digital interventions. An important finding was that students with higher needs for support in terms of self-efficacy, social competence, and perceived stress showed higher intervention usage. In terms of health equity, additional efforts should be made to increase intervention usage among males, those with lower levels of education, and those with higher levels of addictive behaviors.

What kind of evidence is this?

Conversational & pedagogical agents · Randomized study reported

Stage: Secondary school. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Participant ages · source passage

Results Usage was low among intervention participants who received their assigned intervention (n=1266; females n=569, 44.9%; mean age 19.53, SD 3.57 years).

Study methods · source passage

This is a secondary analysis of data from a cluster randomized controlled trial among German vocational students testing the efficacy of the app-based addiction prevention program “ready4life.” The protocol and main results have been published elsewhere [32,33]. The trial was registered in the German Clinical Trials Register: DRKS00022328.

Limitations and uncertainty · source passage

The results of this study should be considered in light of its limitations. First, as a secondary analysis, the original study was not designed to address the research questions posed in this paper. Second, although previous research has suggested that usage is often associated with efficacy [20,21], we have not yet tested this association in our study [51]. Third, usage was analyzed based on 3 parameters: weekly chats, self-initiated chats, and usage time. Other components, such as media content viewed, ask-the-expert, and number of logins, were not included in this analysis because the data were not automatically stored. Fourth, completion of weekly chats was rewarded with credits (and credits were tied to a prize draw), which may have influenced students’ intervention usage. Although we did not experimentally test this assumption, previous reviews have suggested this relationship (eg, [12]). Furthermore, in the focus group conducted at the end of our study, students cited the increased chances of winning prizes as a reason for continued use [52]. Fifth, findings may not be generalizable to other adolescent populations. Additionally, our previous analyses [36] showed that participants in this study differed from nonparticipants on several baseline variables, with students from nonprofessional backgrounds (eg, service and sales workers, or vocational preparation) and males being underrepresented among study participants. Conversely, those with higher support needs were overrepresented among this study’s participants, including students with lower social competence, lifetime cannabis use, higher problematic internet use, and higher perceived stress. Sixth, data collection during the COVID-19 pandemic may have influenced addictive behaviors or usage. Seventh, there may be other variables influencing usage that we have not analyzed. For example, previous studies of digital substance use interventions among adolescents or young adults found that nonimmigrant backgrounds and perceived benefits of smoking cessation [24,26] were associated with higher usage. Finally, reliance on self-report data carries the risk that the results may have been influenced by social desirability.

Design the system before the pilot

Define whether the agent asks questions, gives hints, explains or assesses. Test hallucinated feedback, excessive praise, misleading confidence and responses when the learner is confused. A conversational interface alone does not establish autonomous agency.

  • Give feedback on the learner’s reasoning, then fade hints. Record whether the learner can complete the next task without that scaffold.
  • Measure engagement separately from achievement. Check whether novelty-driven participation persists and whether it translates to independent performance.

A 90-day district pilot

  1. Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
  2. Days 16–30 · Design. Ask learners to attempt a problem unaided, compare an agent hint, explain a revision and verify the final reasoning against a trusted source. Make the permitted use explicit for each assignment. Name an accountable owner and set the baseline, success threshold and budget before implementation.
  3. Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
  4. Days 61–90 · Decide. Measure unaided transfer and delayed retention, calibration of confidence and the ability to identify an incorrect AI explanation. Record whether help-seeking becomes answer copying. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Diana Guertler; Anne Möhring; Dominic Bläsing; Christian Meyer; Hannah Schmidt; Florian Rehbein; Merten Neumann; Arne Dreißigacker; Anja Bischof; Gallus Bischof; Svenja Sürig; Lisa Hohls; Susanne Wurm; Stefan Borgwardt; Severin Haug; Hans-Jürgen Rumpf. Usage of an App-Based Addiction Prevention Program for German Vocational Students: Secondary Analysis of Data From a Cluster Randomized Controlled Trial. Journal of Medical Internet Research (2025-07-28). https://doi.org/10.2196/68754

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.