Conceptual framework links generative-AI health safeguards to five decision
A one-month support program was linked to changes in how junior high school students in Malang Regency, Indonesia, reported using the Konco SREGEP app, which promotes adolescent health. The study also found that how useful students thought the app was remained an important part of the model used to understand whether they accepted it. The researchers studied 92 first-grade junior high school students from three Population Alert Schools in East Java, Indonesia. They used a statistical method called partial least squares structural equation modeling to examine the results. In the post-test model, students who thought the app was useful were also more likely to think it was easy to use, have a positive view of using it and intend to use it.
Students who intended to use the app were more likely to report that they had used it, while problems in their surroundings were linked to lower use. However, the study had no control group and did not collect direct records of app activity. Therefore, the results cannot prove that the support program caused students to keep using the app over time or show that one factor was the single strongest reason for accepting it. The researchers carried out the study from February to June 2022, using a pre-experimental design in which one group was assessed before and after the program.
To take part, students had to be in the first grade of junior high school, have access to a cellphone and the internet, have received no earlier information about Konco SREGEP and attend a school that agreed to join the study. The researchers used simple random sampling to select students from three Population Alert Schools, known in Indonesian as Sekolah Siaga Kependudukan, in Malang Regency. They recruited 94 students, and 92 completed the study and were included in the final analysis.
The researchers also referred to the 10-times rule for partial least squares structural equation modeling. This rule says that a study should include at least 10 times as many participants as the largest number of paths leading to any one hidden concept in the statistical model. The researchers measured the model separately before and after the program. Konco SREGEP is an Android app developed starting in 2020 by the Faculty of Public Health at Universitas Airlangga, the East Java provincial government and the Malang Regency government.
The app was designed to meet adolescents' needs, characteristics and abilities. It combines health education, self-checks, counseling and ways for young people to report local problems. Its Sinau Yuk feature offers organized learning materials on 13 topics: adolescent development, sexuality and gender, adolescent activities, body image and body shaming, nutrition and fertility, teenage pregnancy, stress and ways to cope, life skills, the GenRe program, time management, love management and assertiveness, and preparation for marriage. Each learning module includes tests before and after the lessons, along with videos. Users who complete the material can download a certificate with a barcode.
The Awakmu SEHAT? feature lets users check their own health and automatically receive scores, categories and educational information. CurCOLin Aja offers online counseling and lets users contact an administrator through WhatsApp. A separate reporting feature allows adolescents to report problems or needs involving friends, family, their local area or school.
The app also includes referral codes, a points system, updates about the status of reported problems and options for inviting other users. Its design aimed to do more than simply provide information. It combined education with self-assessment, access to support and opportunities for young people to take part. At the time of the study, Malang Regency had 30 Population Alert Schools at the junior high school level. The local government wanted students to use Konco SREGEP before they graduated.
The study questionnaire measured students' knowledge and skills, curiosity, confidence in their ability to use computers, perceived control, how important they considered the behavior, environmental constraints, habits, perceived usefulness, perceived ease of use, attitudes toward using the app, intention to use it and actual system use. All participants received training from peer educators, installed Konco SREGEP and took part in one month of support activities. These activities included presentation slides, videos showing how to use the app, testimonials from young people who had completed its features, weekly activities to encourage engagement and daily coordination through a WhatsApp group.
The activities were designed to help students understand the app's benefits, learn how to use its features and see why promoting adolescent health matters. However, the researchers did not collect app records, login frequency, session length, total time spent using the app or other direct measures of each student's activity. As a result, the study cannot show that every student used the app regularly or spent a specific number of hours using it.
Four measures changed between the assessments before and after the program: knowledge and skills, perceived control, how important students considered the behavior and reported actual system use. Attitudes and intentions to use the app did not change significantly during the same period. The statistical model after the program explained more of the differences in the main measures of app acceptance than the model before the program.
The R-squared value for intention to use the app rose from 0.650 to 0.868. This means that the factors included in the model accounted for 86.8% of the differences in reported intention after the support period. The model's ability to explain actual system use rose from 0.204 to 0.398, or 39.8%. These numbers show how much of the variation the statistical models could explain. They do not measure the cause-and-effect impact of the support program.
The finding about actual use was also based on the study's measure of reported use, rather than independently checked records of app activity. The study did not show whether students continued using the app after the assessment. Perceived usefulness was an important part of the post-test model, but the researchers did not prove that it was the single strongest factor affecting acceptance. Students who thought the app was useful were more likely to say it was easy to use, feel more positively about using it and report a stronger intention to use it.
Knowledge and skills were also linked to higher perceived usefulness and perceived ease of use, while curiosity was linked to higher perceived usefulness. In the post-test model, knowledge and skills, curiosity and perceived usefulness together explained 22.8% of the differences in perceived usefulness, compared with 29.9% before the program. Perceived usefulness and knowledge and skills together explained 64.1% of the differences in perceived ease of use. Perceived usefulness alone explained 82.6% of the differences in attitudes toward using the app.
These findings show links between factors, but they do not prove that changing students' views would necessarily make them use the app more. Other factors related to behavior and the surrounding environment also played different roles in the model. Students who considered using Konco SREGEP important were more likely to intend to use it, and stronger intentions were linked to more reported use.
Environmental constraints were linked to lower actual use. This means that problems in students' surroundings were associated with less engagement, even when students intended to use the app. In the post-test model, how important students considered the behavior and how useful they thought the app was together explained 86.8% of the differences in intention to use it. Environmental constraints and intention to use the app together explained 39.8% of the differences in actual system use. The model therefore suggested that accepting the app involved more than finding its interface easy to use. How valuable students thought it was, how important they considered using it, their intentions and the conditions around them all helped explain differences in adoption.
The researchers viewed the findings as partial support for combining the Technology Acceptance Model with the Integrated Behavior Model to study mobile health apps for adolescents. The Technology Acceptance Model focuses on how useful and easy to use people think a technology is. The Integrated Behavior Model also considers factors such as intention, how important a behavior seems, perceived control, environmental constraints and habits. The clearest links in the post-test model involved perceived usefulness, intention to use the app and actual use. However, the study found no significant changes in attitudes or intentions after one month of support.
The authors also suggested that students may already have had positive attitudes toward smartphone-based tools, leaving little room for a measurable change. This is one possible way to interpret the results, not an explanation tested by the study. The findings also apply to a specific setting: first-grade students at three schools in one district. They may not apply to all Indonesian adolescents, older students or similar programs in other places.
Several limitations affect what the study can tell us. Because the researchers studied one group before and after the program and had no separate control group, they could not confidently separate the changes from other possible influences. These included normal development as students grew older, the effect of taking the same measurements twice, school activities and other events during the study.
The study also relied on students' answers to questionnaires and did not directly track how they used the app. It included only 92 students who had access to a cellphone and the internet, which may limit how well the findings apply to adolescents with different levels of access, older students or other educational settings. The researchers assessed outcomes after just one month of support, so they did not measure whether students continued using the app or remembered what they learned over the long term. The limited number of schools and the focus on one area further restrict how widely the results can be applied. Despite these limits, the study gives a detailed picture of how adolescents accepted a locally developed health app: perceived usefulness was strongly linked to later measures of acceptance, intention was linked to reported use, and practical problems in the surrounding environment were linked to engagement.
The findings suggest that the usefulness of a digital health tool, students' access to it and continued support may be important when evaluating such tools. However, the study does not settle whether the mentoring program was effective or whether any effects would last over time.
Subject of Research: Acceptance of the Konco SREGEP adolescent health-promotion app among junior high school students in Malang Regency
Article Title: Acceptance model of adolescent health promotion application (Konco SREGEP) through modifications of technology acceptance model and integrated behavior model
Article References: Muthmainnah, Nurmala, I., Lin, C.-Y., Salim, L. A., Devi, Y. P., Salsabila, A. C., Amalia, A., Bilad, M. R., & Restuti, D. Y. (2026). Acceptance model of adolescent health promotion application (Konco SREGEP) through modifications of technology acceptance model and integrated behavior model. Discover Artificial Intelligence, 6(1), Article 1570. https://doi.org/10.1007/s44163-026-02590-5
Image Credits: AI Generated
DOI: 10.1007/s44163-026-02590-5
Keywords: Konco SREGEP, adolescent health, digital health, technology acceptance, mobile health, PLS-SEM, Malang Regency, student engagement, health promotion apps, Acceptance, model, adolescent