Compulsive and excessive use of digital technologies among adolescents has emerged as a significant global public health problem, associated with sleep disruption, psychological distress, social withdrawal, and impaired academic performance.1,2 The burden is especially acute in India, home to 253 million adolescents, where studies report internet addiction in a fifth to two-thirds of school-going adolescents and smartphone addiction prevalence as high as 64.6%.3,4 National policy documents, including the Economic Survey 2025–26, the NCERT internet addiction module, and a draft Karnataka digital detox policy for schools, have called for structured, school-based prevention.5 Schools are an ideal setting for universal prevention, offering sustained access to adolescents irrespective of socioeconomic background. A systematic review and meta-analysis found school-based interventions to be effective in reducing problematic digital technology use (d = 1.47 post-intervention; d = 1.13 at follow-up), with smaller effects on screen time.6 Within schools, nurses are uniquely positioned to lead digital health promotion, combining clinical expertise, health education skills, and insight into adolescent health.7 Nurse-led interventions based on the Health Belief Model (HBM) have reduced smartphone addiction in randomized trials,8 and a nurse-led program for adolescents with problematic internet use in India demonstrated significant improvements sustained at three months.9 An Indian psychoeducation module-based intervention similarly reduced IAT scores among school adolescents.10 However, most prior interventions have been single-component, delivered outside India, or evaluated with short follow-up and without process evaluation. No trial has evaluated a comprehensive, theory-driven, nurse-led model addressing individual, peer, family, and school-environment determinants of digital addiction in Indian schools. We therefore developed the Nurse-Led Digital Health Promotion Model (NL-DHPM) and evaluated its effectiveness in reducing digital addiction, improving knowledge, self-efficacy, and screen-time behaviour, and enhancing quality of life and psychological well-being among adolescents, together with its feasibility and acceptability

Trial design We conducted a parallel-group cluster randomized controlled trial with schools as the unit of randomization and an embedded mixed-methods process evaluation. Cluster randomization was chosen because the intervention is delivered at school level and individual randomization would risk contamination between arms. Assessments were undertaken at baseline (T0), immediately post-intervention (T1, after the 8-week program), and at 3-month follow-up (T2). The trial is reported in accordance with the CONSORT extension for cluster randomized trials. Setting and participants Eight secondary schools (four government, four private) were selected from urban, semi-urban, and rural zones of the study district in India through multi-stage sampling. Eligible schools had a functional school health clinic with a designated school nurse, minimum enrolment of 100 students in grades 8–10, and ≥75% average attendance. Eligible students were adolescents aged 13–16 years in grades 8–10 with regular attendance, access to a smartphone or internet-enabled device, written parental consent, and written assent. Adolescents with diagnosed psychiatric disorders, those on cognition-affecting medication, those with disabilities precluding group participation, prior recipients of digital addiction interventions, and those planning to change schools were excluded. Intervention-delivering nurses were registered nurses employed at the schools who completed standardized training. Randomization and blinding Schools were randomly allocated 1:1 to intervention (4 schools) or control (4 schools) using a computer-generated sequence with stratification by school type and location, performed by an investigator not involved in recruitment. Owing to the nature of the intervention, participants and nurses could not be blinded; outcome assessors and the data analyst were blinded to allocation. Sample size Assuming a minimum clinically important difference of 10 points on the IAT-20 (SD 15), 80% power, two-sided α = 0.05, an intracluster correlation coefficient of 0.02 with average cluster size 50 (design effect 1.98), and 20% attrition, 176 participants per arm were required. The sample was inflated to 200 per arm (total 400; 50 per school) to preserve power for subgroup analyses. Intervention The NL-DHPM is grounded in the Health Belief Model and Pender’s Health Promotion Model and was developed through literature synthesis, expert panel review (content validity index ≥0.80 for all components), and pilot testing in a non-participating school. Trained school nurses delivered six components over 8 weeks: (1) four 45-minute psychoeducation sessions addressing the nature, consequences, and self-regulation of digital addiction, mapped to HBM constructs; (2) three 60-minute skill-building workshops on time management and goal setting, alternative activities, and peer support and social skills; (3) a peer education program in which trained peer educators conducted classroom awareness activities; (4) two parent engagement sessions on adolescent digital use and supporting healthy habits at home; (5) school environment activities including awareness campaigns, digital detox days, teacher sensitization, and a student digital wellness club; and (6) individual nurse-led counselling, informed by motivational interviewing and cognitive-behavioural principles, for adolescents with moderate-to-severe addiction. Standardized manuals, participant workbooks, parent booklets, and fidelity checklists supported delivery. Control schools continued standard school health education. Outcomes The primary outcome was internet addiction measured by Young’s IAT-20 (score 0–100; categories: normal 0–30, mild 31–49, moderate 50–79, severe 80–100).11 Secondary outcomes were smartphone addiction (SAS-SV),12 knowledge of healthy digital habits (20-item validated questionnaire, 0–20), self-efficacy in managing digital device usage (10-item adapted scale, 10–50), self-reported daily screen time (7-day log), quality of life (WHOQOL-BREF domains), psychological distress (DASS-21), and loneliness (UCLA Loneliness Scale). Process outcomes included attendance, fidelity (independent-observer checklists), participant satisfaction (10-item scale), and qualitative data from focus group discussions and interviews with adolescents, parents, nurses, teachers, and administrators. Statistical analysis Analyses followed the intention-to-treat principle. Baseline comparability was assessed with t-tests and chi-square tests. Between-group differences in continuous outcomes at each time point were examined with independent t-tests and within-group changes with paired t-tests; effect sizes are reported as Cohen’s d. Multivariate logistic regression identified baseline factors associated with internet addiction. Qualitative data were analysed thematically. Significance was set at p < 0.05 (two-tailed). Ethics The study was approved by the Institutional Ethics Committee [number to be inserted]; permissions were obtained from education authorities and schools. Written parental informed consent and adolescent assent were obtained. Control schools were offered the intervention after trial completion.

Participant flow and baseline characteristics Four hundred adolescents (200 per arm) from eight schools were enrolled and assessed at baseline. The arms were comparable across all demographic and digital-use characteristics (all p > 0.05): mean age 14.25 ± 1.05 years, 57.5% male, 55.0% urban residents, 94.0% smartphone owners, mean age at first smartphone use 11.85 ± 1.75 years, and mean daily screen time 4.85 ± 2.15 hours (Table 1). At baseline, 66.5% of participants had at least mild internet addiction on the IAT-20 and 66.0% met SAS-SV criteria for smartphone addiction. Table 1. Baseline characteristics by trial arm Characteristic Intervention (n = 200) Control (n = 200) p-value Age (years), mean ± SD 14.2 ± 1.1 14.3 ± 1.0 0.348 Male, n (%) 112 (56.0) 118 (59.0) 0.572 Urban residence, n (%) 112 (56.0) 108 (54.0) 0.671 Government school, n (%) 100 (50.0) 100 (50.0) 1.000 Nuclear family, n (%) 142 (71.0) 146 (73.0) 0.777 Smartphone ownership, n (%) 186 (93.0) 190 (95.0) 0.399 Age at first smartphone use (years), mean ± SD 11.8 ± 1.8 11.9 ± 1.7 0.569 Daily screen time (hours), mean ± SD 4.8 ± 2.1 4.9 ± 2.2 0.646 IAT-20 score, mean ± SD 45.2 ± 18.6 44.8 ± 19.2 0.833 SAS-SV score, mean ± SD 32.4 ± 10.8 32.1 ± 11.2 0.788 SD: standard deviation; IAT-20: Young’s Internet Addiction Test; SAS-SV: Smartphone Addiction Scale–Short Version. Primary outcome: internet addiction Following the 8-week intervention, mean IAT-20 scores fell in the intervention group from 45.2 ± 18.6 to 28.6 ± 14.2 (within-group change −16.6 ± 12.4, p < 0.001), compared with a minimal change in controls (44.8 ± 19.2 to 42.1 ± 18.5; −2.7 ± 10.8). The between-group difference at T1 was −13.5 (95% CI −16.8 to −10.2; p < 0.001; d = 0.82). Effects were sustained and strengthened at 3-month follow-up (26.4 ± 13.8 vs 43.6 ± 19.0; difference −17.2, 95% CI −20.6 to −13.8; p < 0.001; d = 1.05) (Table 2). Correspondingly, the proportion of intervention-arm adolescents classified as normal users rose from 34.0% at baseline to 74.0% at follow-up, while moderate addiction fell from 19.0% to 4.0% and severe from 3.0% to 1.0%; control-arm categories were essentially unchanged (Table 3). Table 2. Internet Addiction Test (IAT-20) scores across time points Time point Intervention (n = 200) Mean ± SD Control (n = 200) Mean ± SD Mean difference (95% CI) Cohen’s d p-value Baseline (T0) 45.2 ± 18.6 44.8 ± 19.2 0.4 (−3.3 to 4.1) 0.02 0.833 Post-intervention (T1) 28.6 ± 14.2 42.1 ± 18.5 −13.5 (−16.8 to −10.2) 0.82 < 0.001 Follow-up (T2) 26.4 ± 13.8 43.6 ± 19.0 −17.2 (−20.6 to −13.8) 1.05 < 0.001 Δ T0–T1 (within group) −16.6 ± 12.4 −2.7 ± 10.8 −13.9 (−16.2 to −11.6) — < 0.001 Δ T0–T2 (within group) −18.8 ± 13.2 −1.2 ± 11.4 −17.6 (−20.0 to −15.2) — < 0.001 Between-group comparisons by independent t-test; within-group changes by paired t-test. CI: confidence interval. Table 3. Change in internet addiction severity categories from baseline to follow-up Category Intervention T0, n (%) Intervention T2, n (%) Control T0, n (%) Control T2, n (%) Normal 68 (34.0) 148 (74.0) 66 (33.0) 72 (36.0) Mild 88 (44.0) 42 (21.0) 90 (45.0) 88 (44.0) Moderate 38 (19.0) 8 (4.0) 38 (19.0) 34 (17.0) Severe 6 (3.0) 2 (1.0) 6 (3.0) 6 (3.0) Secondary outcomes The intervention produced significant benefits across all secondary outcomes (Table 4). SAS-SV scores declined from 32.4 ± 10.8 to 21.2 ± 9.0 at follow-up in the intervention arm versus minimal change in controls (between-group difference −10.3, 95% CI −12.3 to −8.3). Knowledge scores nearly doubled (8.4 ± 3.2 to 16.2 ± 2.8 of 20 at T1; a 92.9% increase), and self-efficacy improved from 28.6 ± 8.4 to 43.0 ± 6.0 of 50 at T2. Daily screen time fell from 4.8 ± 2.1 to 2.6 ± 1.5 hours at follow-up (a 45.8% reduction; between-group difference −2.1 hours, 95% CI −2.5 to −1.7), approaching the recommended two-hour recreational limit. Table 4. Secondary outcomes by arm across time points Outcome / Time Intervention Mean ± SD Control Mean ± SD Mean difference (95% CI) p-value SAS-SV — T0 32.4 ± 10.8 32.1 ± 11.2 0.3 (−1.9 to 2.5) 0.788 SAS-SV — T1 22.6 ± 9.4 30.8 ± 10.6 −8.2 (−10.2 to −6.2) < 0.001 SAS-SV — T2 21.2 ± 9.0 31.5 ± 11.0 −10.3 (−12.3 to −8.3) < 0.001 Knowledge (0–20) — T0 8.4 ± 3.2 8.6 ± 3.0 −0.2 (−0.8 to 0.4) 0.521 Knowledge — T1 16.2 ± 2.8 9.2 ± 3.4 7.0 (6.4 to 7.6) < 0.001 Knowledge — T2 15.8 ± 3.0 9.0 ± 3.6 6.8 (6.2 to 7.4) < 0.001 Self-efficacy (10–50) — T0 28.6 ± 8.4 29.0 ± 8.0 −0.4 (−2.0 to 1.2) 0.626 Self-efficacy — T1 42.4 ± 6.2 30.2 ± 8.8 12.2 (10.7 to 13.7) < 0.001 Self-efficacy — T2 43.0 ± 6.0 29.8 ± 9.0 13.2 (11.7 to 14.7) < 0.001 Screen time (h/day) — T0 4.8 ± 2.1 4.9 ± 2.2 −0.1 (−0.5 to 0.3) 0.646 Screen time — T1 2.8 ± 1.6 4.6 ± 2.0 −1.8 (−2.2 to −1.4) < 0.001 Screen time — T2 2.6 ± 1.5 4.7 ± 2.1 −2.1 (−2.5 to −1.7) < 0.001 Between-group comparisons by independent t-test. SAS-SV: Smartphone Addiction Scale–Short Version; CI: confidence interval. Quality of life and psychological well-being At follow-up, the intervention arm reported significantly higher WHOQOL-BREF scores than controls in the physical health (72.4 ± 10.2 vs 64.8 ± 12.4), psychological (68.6 ± 11.8 vs 58.2 ± 13.6), social relationships (70.2 ± 12.0 vs 61.4 ± 14.2), and environment (66.8 ± 10.6 vs 60.0 ± 12.8) domains, and significantly lower DASS-21 depression (6.8 ± 4.2 vs 12.4 ± 6.8), anxiety (5.2 ± 3.8 vs 10.6 ± 6.2), and stress (7.4 ± 4.6 vs 14.8 ± 7.2) scores, as well as lower loneliness (UCLA 34.2 ± 8.6 vs 42.8 ± 10.4) (all p < 0.001). Feasibility, acceptability, and process evaluation Mean attendance across intervention sessions was 87.4% (range 78.2–94.6%), highest for psychoeducation sessions (92.3%) and lowest for parent engagement sessions (75.2%). Independent fidelity assessment indicated 92.6% of components were delivered as planned. Participant satisfaction was high: 94.7% reported overall satisfaction and 96.0% would recommend the program. Thematic analysis of focus groups and interviews identified five themes: enhanced awareness and knowledge, improved self-regulation, peer support and social connection, parental involvement, and the credibility and approachability of the nurse’s role. Adolescents described becoming conscious of their usage through self-monitoring, achieving substantial self-directed screen-time reductions through goal setting, and valuing family screen-free routines initiated after parent sessions.

This cluster randomized trial demonstrates that an 8-week, multi-component, nurse-led digital health promotion model produced large and sustained reductions in internet addiction (d = 0.82 post-intervention; d = 1.05 at 3 months), smartphone addiction, and daily screen time among school-going adolescents, alongside meaningful improvements in knowledge, self-efficacy, quality of life, psychological distress, and loneliness. The intervention was feasible within existing school health infrastructure, delivered with high fidelity, and highly acceptable to adolescents, parents, and school staff. The magnitude and durability of the primary-outcome effect compare favourably with prior evidence. A meta-analysis of school-based interventions reported pooled effects of d = 1.47 post-intervention and d = 1.13 at follow-up for problematic digital technology use, with smaller effects on screen time;6 our results fall within this range while additionally achieving a 45.8% screen-time reduction, likely attributable to explicit behavioural substitution, goal setting, and behavioural contracting components. The trajectory of IAT-20 change closely parallels an Indian psychoeducation module-based study in which scores fell from 44.75 ± 19.69 to 28.84 ± 13.98 at two months,10 and the sustained 3-month benefit mirrors a nurse-led intervention for problematic internet use in India.9 The reduction in smartphone addiction is consistent with an HBM-based nurse-led randomized trial that markedly decreased smartphone addiction prevalence,8 supporting the cross-cultural generalizability of theory-driven, nurse-delivered approaches. Several design features plausibly explain the strength of effects. First, the intervention operated across ecological levels—individual (psychoeducation, skills, counselling), interpersonal (peer education, parent engagement), and environmental (school-wide campaigns, digital detox days, policy recommendations)—consistent with evidence that non-authoritative parenting and peer norms are key determinants of adolescent digital addiction.4 Second, grounding in the Health Belief Model and Health Promotion Model ensured components systematically targeted perceived susceptibility, severity, benefits, barriers, cues to action, and self-efficacy; the near-doubling of knowledge and the large self-efficacy gains support these hypothesized mechanisms, and self-efficacy is a recognized mediator of behaviour change in this domain. Third, delivery by school nurses conferred credibility and approachability, as reflected in qualitative accounts, while embedding the program within routine school health services enhances scalability and sustainability. The improvements in quality of life, depression, anxiety, stress, and loneliness align with findings from an integrated yoga trial for internet gaming disorder in Indian schools and a nurse-led problematic internet use intervention, both of which reported psychosocial benefits.9,13 Given the strong association between psychological distress and digital addiction observed at baseline (AOR 10.82 for severe distress), addressing digital behaviour and mental well-being together appears mutually reinforcing. The reduction in loneliness is particularly notable and may reflect the peer education component fostering offline social connection. These findings carry direct policy relevance. They provide trial-level evidence for the school-based, multi-stakeholder approach advocated in the draft Karnataka digital detox policy and the Economic Survey 2025–26, and the model can complement NCERT’s internet addiction module by supplying a structured, nurse-led delivery framework.5 Because the model relies on existing school nurses, standardized manuals, and low-cost materials, it is well suited to scale-up within school health programs in India and other low- and middle-income settings. Strengths and limitations Strengths include the cluster randomized design with blinded outcome assessment and intention-to-treat analysis, theory-driven multi-component intervention with documented fidelity, validated outcome measures, a 3-month follow-up, and an embedded mixed-methods process evaluation. Limitations include reliance on self-reported outcomes, which are vulnerable to social desirability and Hawthorne effects (the small control-group improvement may partly reflect the latter); a single-district setting with limited rural representation (17.5%), constraining generalizability; the 3-month follow-up, which cannot establish long-term maintenance; and the possibility of residual contamination through out-of-school contact despite cluster randomization. Objective device-based measurement of screen time and longer follow-up are priorities for future research, as are cost-effectiveness analyses and implementation trials at scale.

A school-based, nurse-led, theory-driven digital health promotion model substantially and durably reduced digital addiction and screen time and improved psychological well-being among Indian adolescents, with high feasibility and acceptability. Integration of nurse-led digital health promotion into routine school health services offers a practical, scalable strategy for addressing the growing burden of adolescent digital addiction. Declarations Trial registration: [Registry and number to be inserted]. Ethical approval: Institutional Ethics Committee [approval number to be inserted]; written parental consent and adolescent assent obtained. Funding: [To be inserted / None]. Conflicts of interest: None declared. Acknowledgements: The authors thank the participating schools, school nurses, peer educators, students, and parents

1. Girela-Serrano BM, et al. Smartphone addiction in adolescents: a systematic review. J Behav Addict. 2024;13(2):285-303. 2. James M, Dixon C, Dragomir M, Thirlwell E, Hitcham L. Problematic smartphone use in adolescents: a scoping review. J Adolesc Health. 2023. 3. Amudhan S, et al. Technology addiction among school-going adolescents in India: epidemiological analysis from a cluster survey for strengthening adolescent health programs at district level. J Public Health (Oxf). 2022;44(2):286-295. 4. Dave D, Patel R, Sharma V, et al. Associations between smartphone addiction, parenting styles, and mental well-being among adolescents aged 15-19 years in Gujarat, India. BMC Public Health. 2024;24:2462. 5. Government of India. Economic Survey 2025-26. New Delhi: Ministry of Finance; 2026. 6. Žmavc M, Horvat J, Židan M, Selak Š. The effectiveness of school-based interventions to reduce problematic digital technology use and screen time: a systematic review and meta-analysis. J Behav Addict. 2025;14(2):571-589. 7. Pender NJ, Murdaugh CL, Parsons MA. Health Promotion in Nursing Practice. 8th ed. Pearson; 2019. 8. Effectiveness of a nurse-led health belief model-based educational intervention for reducing smartphone addiction in adolescents: a randomized controlled trial. J Pediatr Nurs. 2026;89:18-28. 9. Mathew P, Krishnan R, Bhaskar A. Effectiveness of a nurse-led intervention for adolescents with problematic internet use. J Psychosoc Nurs Ment Health Serv. 2020;58(7):16-26. 10. Narayanappa PH, Nirgude AS, Nattala P, Philip M. Does psychoeducation module-based community intervention address Internet addiction among school-going adolescents? A quasi-experimental study from Mangalore, India. J Family Med Prim Care. 2024;13(10):4237-4243. 11. Young KS. Internet addiction: the emergence of a new clinical disorder. CyberPsychol Behav. 1998;1(3):237-244. 12. Kwon M, Kim DJ, Cho H, Yang S. The smartphone addiction scale: development and validation of a short version for adolescents. PLoS One. 2013;8(12):e83558. 13. Rao NS, et al. Effect of an eight-week yoga program on adolescents with Internet Gaming Disorder in an Indian school setting: a randomized controlled trial. Front Public Health. 2026;14:1750580.