Application of digital health technologies to substance use reduction among students in higher education institutions: A scoping review

Assessment of study quality

To assess the methodological quality of the included studies, we used the Joanna Briggs Institute (JBI) Critical Appraisal Tools. This tool was selected because it is well-suited for evaluating the diverse range of study designs included in scoping reviews, including quantitative, qualitative, and mixed methods research.

The quality of the studies was assessed across several domains, including methodological rigor, validity, relevance to the research question, and transparency of reporting. The results of the quality appraisal were summarized descriptively to provide an overview of the methodological strengths and weaknesses of the included studies. However, no studies were excluded based on quality to maintain the inclusive nature of the scoping review process.

Search results using PRISMA-ScR

Our search encompassed four electronic databases: PubMed,⁴¹ PsycINFO via APA,⁴² Scopus,⁴³ and Google Scholar,³² which resulted in an initial pool of 190 articles. After removing 45 duplicates, we screened the remaining 145 articles based on their titles and abstracts. This process led to the exclusion of 74 publications that did not meet our inclusion criteria. Additionally, we included one study from the reference lists of reviewed articles⁴³ and another from Google scholar.⁴⁴ Following a full-text review of the remaining 31 articles, 11 studies were selected for data extraction and analysis. The review process ensured that only those studies meeting the rigorous inclusion criteria were included in the final analysis ( Figure 1). This method of review was conducted on the basis of a predetermined protocol in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) standards for scoping reviews³⁹ (S1- PRISMA Checklist). Lastly, we searched for grey literatures, but none met the inclusion criteria.

Figure 1. PRISMA Flowchart describing the selection of studies mapping existing literature on digital health technology for substance use reduction in young adult college or university students in the LMIC.

Characteristics of the studies

The review identified 11 original, peer-reviewed studies published between 2013 and 2025 that focus on the use of digital health technology for substance use reduction in low- and middle-income countries (LMICs), meeting the inclusion criteria. Among these, five articles (45.5%) were published in the last five years (2020-2025), with 2019 having the highest number of relevant publications (n = 3, 27.3%). The years 2018 and 2025 each had two relevant publications (n = 2, 18%), while 2015 and 2020 had one publication each. Despite the growing interest in this field, the overall number of studies remains insufficient, indicating a lack of original, peer-reviewed research on this topic within LMIC contexts. The earliest study meeting the eligibility criteria was conducted in 2015. The review includes studies from various years, including three studies from 2019, two from 2018, 2023, and 2024, and one each from 2020 and 2015.

Of the 11 included primary studies, seven employed quantitative methodologies using randomized controlled trial (RCT) designs,^44–50 while one study utilized a quasi-experimental approach,⁵¹ one used an interventional study with a single group,⁴³ another employed a cross-sectional methodology,⁵² and one article adopted a qualitative approach.⁵³ The studies spanned five countries: Brazil had the highest representation,^45–48 contributing four studies, followed by Hong Kong with two,^49,53 and Iran,⁵¹ Korea,⁵⁰ and Vietnam,⁵² each contributing one study. The participants in the studies were college or university students aged 18 years or older. For instance, in Brazil,^45,46 the studies focused on college drinkers aged 18–30 who reported alcohol use in the past three months. In Hong Kong,⁴⁷ one study involved participants aged 18 or older who had consumed alcohol in the last 12 months, while another study focused on students at risk for alcohol use disorder.⁴⁹ In Tehran,⁵¹ a study included 130 students living in dormitories to assess smoking prevention behaviors. Additional studies included participants from Turkey, India, Iran and other countries ( Figure 3) (Tables 1 and 2).⁷⁴

The studies employed a range of research designs to evaluate interventions targeting substance use, showcasing diverse methodological approaches. One study⁴⁴ conducted a mixed-methods cluster randomized controlled trial, integrating both quantitative and qualitative elements. Several studies used pragmatic RCTs,⁴⁷ including a dismantling design⁴⁶ and standard approaches.⁴⁵ Other studies utilized quasi-experimental designs,⁴³ cross-sectional surveys, and semi-structured interviews to assess intervention effectiveness. The sample sizes in these studies varied widely, from a small group of 20 university students to larger cohorts, such as one study analyzing data from 5,476 participants. Other studies included 931, 4,460, and 772 participants, with one focusing on 458 participants, and another examining a cohort of 191 individuals. The studies also demonstrated a mix of large-scale and smaller, targeted research efforts, illustrating the diverse strategies employed to explore substance use in different contexts (Table 1).⁷⁴

The earliest study meeting our eligibility criteria was conducted in 2015.⁴⁸ In the review: three studies were identified in 2019,^43,45,50 two primary studies each were identified of the following years: 2018,^51,52 2023^44,53 and 2024.^47,49 One study was found in, 2020⁴⁶ and 2015⁴⁸ ( Figure 2).

Figure 2. Lome graph showing articles published per year (publication trends as of time of data collection).

Figure 3. Bar chart showing the distribution of articles per countries.

The studies in the review involved a wide range of sample sizes, further highlighting the diversity of research approaches. One study⁴⁶ analyzed data from a large cohort of 5,476 participants, while another⁴⁷ included 931 individuals, and a third study⁴⁵ involved 4,460 participants. One study⁴⁹ had 772 participants, whereas another⁵³ focused on a smaller group of 20 university students. Additionally, one study⁴⁸ involved 458 participants, while another⁴⁴ examined a sample of 191 individuals (Table 2).⁷⁴

One study⁴³ initially recruited 433 participants, completing their study with 314 individuals. Another study⁵¹ focused on a smaller group of 130 participants, while a separate study⁵⁰ included a larger cohort of 1,496 individuals. Finally, one study⁵² worked with a moderately sized cohort of 1,082 participants. Together, these studies reflect a balanced mix of large-scale and smaller, targeted research efforts, illustrating the multifaceted strategies employed to explore and address substance use across diverse populations and contexts (Table 2).⁷⁴

Type of digital health technologies

The most commonly used digital health technology in the included studies was the web-based personalized normative feedback (PNF) intervention, implemented in three studies.^45–47 One study⁵⁰ utilized a web-based intervention called the Brief Empowerment Program for Alcohol-use Monitor (BEAM), which integrated PNF and motivational interviewing (MI) components. Other technologies included web portals or Android applications,⁴⁴ a web-based program,⁴³ e-health interventions,⁵² a computer-based intervention,⁴⁸ the Telegram application,⁵¹ instant messaging apps,⁵³ and mobile chat-based instant messaging⁴⁹ (Table 1).⁷⁴

Substance use reduction

The studies in the review employ a range of methods to define and assess substance use, focusing on alcohol and tobacco, and use validated tools like AUDIT and ASSIST to measure consumption patterns and associated risks. The common thread is the evaluation of substance use behaviors to identify risky patterns and inform potential interventions.

The majority of included primary studies focused on alcohol use, with seven articles^{44–47,49,50,53} examining this substance. Three articles investigated tobacco smoking^43,51,52 and one explored broader substance use, encompassing tobacco, alcohol, marijuana, cocaine, amphetamine-stimulants, inhalants, sedatives, hallucinogens, opioids, and other substances.⁴⁸ The definition of substance use across the studies is largely consistent, revolving around alcohol consumption or smoking behavior, though the timeframes and criteria vary.

For alcohol use, several studies define it based on consumption in the previous months or year,^46,47 who focus on alcohol use in the past 3 months and 12 months, respectively, using the AUDIT-C tool to assess the frequency and quantity of consumption. Other studies in the review,^45,49 expand the definition to include hazardous drinking patterns and risk levels, which are measured using the AUDIT tool. Some studies,⁵⁰ set specific cutoffs for identifying problematic alcohol use (e.g., AUDIT-C scores of 4 for men and 3 for women).

Tobacco use is similarly defined as the act of smoking or using tobacco products. For instance studies like,^43,51 focus on smoking behaviors, with defining smoking as using at least one cigarette a day and the Health Belief Model to assess smoking behaviors and perceptions of risk. Additionally, studies like⁴⁸ examine both alcohol and drug use, including tobacco, using the ASSIST screening tool.

Measurement tools and criteria are generally based on well-established scales, including AUDIT, AUDIT-C, and ASSIST, which assess the severity of substance use and associated risks. For instance, the AUDIT-C tool is commonly used to evaluate alcohol consumption, while the ASSIST tool is employed to assess alcohol, tobacco, and other drug use. These tools help to identify hazardous drinking behaviors, smoking habits, and potential substance use disorders, often through self-reported surveys (Table 1).⁷⁴

Purpose and digital health strategies

The included studies explored various purposes related to the digital health technology: Two studies^45,46 aimed to reduce alcohol use and alcohol-related consequences. Three studies focused on reducing alcohol use specifically,^47,49,53 while one study⁵⁰ aimed to prevent problematic alcohol use and the other one on digital screening and brief intervention for alcohol misuse.⁴⁴ Two articles addressed tobacco smoking cessation,^43,52 and another focused on smoking preventive behaviors.⁵¹ Finally, one study aimed to reduce overall substance use involvement⁴⁸ (Table 1).⁷⁴

The studies in the review explored various digital interventions aimed at address substance use, particularly alcohol and smoking, in college and university settings. One study⁴⁴ developed a digital screening and brief intervention for alcohol use, delivered through an interactive digital platform tailored to help college students reduce their alcohol consumption. Another study⁴³ designed a web-based smoking cessation program that offered educational tools, resources, personalized counseling, and progress tracking to support quitting (Table 1).⁷⁴

One study created an online platform providing students with customized feedback on their alcohol consumption compared to peer norms, delivered through self-paced web sessions.⁴⁶ In a similar vein, another study⁴⁵ implemented a web-based intervention for Brazilian college students focusing on motivational strategies to reduce alcohol use. A different study⁴⁷ developed a personalized normative feedback system to address alcohol-related misperceptions among students using an online platform. Additionally, one study⁵² introduced an e-health intervention to promote smoking cessation through educational content, personalized counseling, and behavior change strategies delivered via the internet and mobile platforms (Table 1).⁷⁴

Another study⁴⁸ evaluated a computer-based intervention with three groups: one receiving computerized screening and motivational intervention, another receiving non-computerized screening with motivational intervention and a control group undergoing assessment only. One study⁵¹ applied an educational intervention based on the Health Belief Model and health literacy principles, incorporating lectures, discussions, and health education materials (Table 1).⁷⁴

One study⁵³ used instant messaging apps to deliver alcohol reduction interventions, emphasizing personalized communication and engagement. In another study⁴⁹ employed the brief interventions using either instant messaging (IM) or text messaging (SMS) to support alcohol reduction efforts. Lastly, a study⁵⁰ implemented a web-based screening and brief intervention targeting problematic alcohol use, providing tailored feedback through an online platform (Table 1).⁷⁴

These interventions highlight a wide range of strategies, including personalized feedback, counseling, motivational techniques, and behavioral support, delivered through digital platforms such as web-based systems, mobile apps, and messaging services to address substance misuse among college students (Table 1).⁷⁴

Dose and duration of the Interventions

The studies in the review implemented various interventions with differing doses and durations. One study⁴⁴ conducted a single digital session that combined screening, feedback, and motivational interviewing, followed up over a period of 3 months. Another study⁴³ provided participants access to a gradual, content-based program with weekly follow-ups or reminders over 6 months. A third study⁴⁶ implemented a single-session, web-based Personalized Normative Feedback (PNF) intervention for alcohol use with a follow-up duration of 6 months, while another study⁴⁵ also used a single web-based session featuring personalized feedback over the same duration. Additionally, one study⁴⁷ enhanced its web-based PNF intervention with booster sessions, maintaining a 6-month timeframe (Table 1).⁷⁴

A separate study⁵² focused on smoking habits, offering personalized feedback and encouraging participants to set quit dates and track their progress. Another study⁴⁸ implemented a single-session tailored intervention addressing substance involvement with feedback components, with a follow-up over 3 months. One study⁵¹ implemented an electronic educational intervention via six sessions delivered through the Telegram application over 3 months, while the control group received no intervention (Table 1).⁷⁴

Another study⁵³ explored participants’ perceptions of intervention doses, concentrating on qualitative insights into app usage without implementing a specific dose. Meanwhile, a separate study⁴⁹ featured two intervention groups: one receiving chat-based instant messaging support for alcohol reduction, and the other receiving SMS messages on general health topics, both over 3 months. Lastly, one study⁵⁰ delivered a single-session intervention with personalized feedback on alcohol use, with a follow-up period of 4 weeks (Table 1).⁷⁴

Outcomes of digital categories

The review of the included primary studies revealed diverse outcomes for the digital intervention categories. One study⁴⁶ found no evidence supporting the intervention's effectiveness, while another⁴⁵ demonstrated its success in reducing alcohol use. A third study⁴⁷ observed that the intervention was effective in reducing alcohol consumption for one month but showed no lasting impact thereafter. Another study⁵² highlighted the feasibility of integrating e-health interventions with traditional clinical or telephone-based models. One study⁴⁸: reported effectiveness in reducing alcohol use, low effectiveness for marijuana, and inconsistent results for tobacco and other drugs. An educational intervention using the Telegram application, grounded in the Health Belief Model (HBM) and Health Literacy (HL), was effective in promoting smoking prevention behaviors among university students.⁵¹ Instant messaging interventions were found to be highly acceptable.⁵³ Two studies^49,50 demonstrated the effectiveness of the mobile chat-based instant messaging and web-based interventions specifically on-BEAM, which incorporates personalized normative feedback (PNF) and motivational interviewing (MI) components. Additionally, one study⁴⁴ found that a digital screening and brief intervention for alcohol misuse was acceptable, feasible, and potentially effective among college students from low-resource settings. Lastly, another program⁴³ successfully helped students quit smoking, enhanced their self-efficacy, and facilitated the process of change toward smoking cessation (Table 1).⁷⁴

Outcomes measured

The studies included in the review assessed a range of outcomes related to substance use interventions, particularly focusing on alcohol and smoking behaviors. Several studies measured changes in alcohol consumption and related behaviors. For instance, one study⁴⁶ evaluated self-reported changes in alcohol use and associated behaviors, while a later study⁴⁷ specifically measured reductions in weekly alcohol consumption and drinking-related consequences. Similarly, one study⁴⁵ assessed alcohol use through AUDIT scores, drink counts, and consequences. One study in Hong Kong⁴⁹ also focused on alcohol use reduction, while another study⁵³ examined the feasibility, acceptability, and perceptions of using instant messaging apps for reducing alcohol consumption (Table 2).⁷⁴

One study⁴⁸ evaluated substance involvement reduction for alcohol, tobacco, and cannabis, reflecting a broader scope of intervention. Another study⁴⁴ examined changes in alcohol use behaviors and associated risks, providing insights into behavioral outcomes (Table 2).⁷⁴

Smoking-related outcomes were also a focus of several studies. One study⁴³ measured smoking cessation rates, self-efficacy, and the process of change, while another study⁵¹ evaluated the adoption of smoking prevention behaviors using the Health Belief Model combined with health literacy principles. One study⁵² assessed smoking prevalence, quit attempts, and the willingness to pay for cessation apps (Table 2).⁷⁴

Finally, one study⁵⁰ investigated reductions in AUDIT-C scores as a measure of alcohol use. Collectively, these studies highlight a variety of outcome measures, including behavioral changes, feasibility and acceptability of interventions, and the effectiveness of specific strategies to reduce substance use (Table 2).⁷⁴

In general the review targeting alcohol and smoking behaviors among university students, predominantly aged 15 and above, revealed diverse methodologies and outcomes based on context and population. Three studies in the review evaluated Personalized Normative Feedback (PNF) interventions in Brazil, finding them effective in reducing alcohol consumption, especially in motivated participants, though effectiveness diminished with low motivation or longer follow-ups.^45–47 Similarly, one study demonstrated that instant messaging interventions reduced short-term alcohol use in Hong Kong,⁴⁹ while their qualitative study highlighted user acceptance and preference for personalized, private digital interventions.⁵³ One study in the review reported on computerized screening and motivational interventions that effectively reduced substance involvement, particularly alcohol, though challenges remained in cannabis use.⁴⁸ The other study in India utilized digital tools, showing feasibility and potential effectiveness despite limited generalizability.⁴⁴ Smoking cessation by two studies in Turkey and Iran^43,51: leveraged web-based and educational interventions integrating behavioral models, improving cessation rates and preventive behaviors. Lastly, one study in the review explored e-health interventions in Vietnam, noting their feasibility but emphasizing the need for reliable information to enhance uptake⁵² (Table 2).⁷⁴

Strengths and limitations of the review

This scoping review has several notable strengths. It addresses the important topic of digital health interventions for substance use reduction among college and university students in LMICs, utilizing a transparent and reproducible approach for the search strategy, data collection, and extraction processes. The review underscores a significant gap in the literature, revealing that while many studies address substance use, only a small proportion assess the effectiveness of digital health interventions for this specific population.

However, some limitations should be noted. The review relied exclusively English-language journals, potentially omitting relevant studies published in other languages. Furthermore, its aim to provide a broad overview of the existing literature led to the inclusion of studies with diverse methodologies, which may have contributed to the heterogeneity of the reported findings.