One of the consequences of excessive social media usage is social media addiction, which has become a topic of interest for several researchers ([1]; [9]; [18]; [19]; [23]; [32]; [48]). According to [17], addictive behavior has six components: salience, mood modification, tolerance, withdrawal, conflict, and relapse. It is further indicated that if an individual’s social media-related behaviors meet the above-mentioned components, the behavior can be defined as an addiction ([17]). For the purposes of this article, social media addiction, problematic social media use, and social media use disorder are treated as equivalent constructs and can be considered different variations of one construct.

An individual addicted to social media is described by [19] as someone: 1) Whose social media usage becomes excessive and compulsive; 2) Who spends an unreasonable amount of time thinking about social media; 3) Who experiences an uncontrollable need to use social media. Furthermore, symptoms of social media addiction can appear in the form of cognitive, physical, and emotional responses, and can result in personal, social, and psychological problems ([19]). For instance, excessive use of social media has been associated with decreased levels of job performance ([22]; [51]), poor academic performance ([31]), as well as certain mental health issues such as depression, anxiety, and loneliness ([20]; [48]).

Taking the above-mentioned into consideration, it is clear that social media addiction is a growing issue of concern. Consequently, the next critical area of interest would be identifying how social media addiction is being addressed or approached. Therefore, the purpose of this scoping review was to identify the available literature on therapeutic techniques, screening tools, or interventions for social media addiction. Conducting a scoping review will not only identify the scope of available literature on the therapeutic techniques, screening tools, or interventions for social media addiction, but it will also assist in identifying knowledge gaps on this topic ([4]; [33]; [38]). Thus, by identifying the scope of literature on the topic, this review study will provide recommendations for future research and potentially result in more research studies being conducted.

In accordance with the guidelines by [4], to reduce the risk of excluding relevant literature, researchers should keep the review question as broad as possible. Therefore, the guiding research question in this scoping review was: What literature is available on therapeutic techniques, screening tools, or interventions for social media addiction?

A bibliographic database, EDS Search, was utilized to identify and search for relevant literature. The EDS search via the North-West University Library Information System allowed access to 76 academic databases, from which 32 were accessed to identify relevant literature. Furthermore, to expand the scope of the available literature, additional searches were manually conducted on Google Scholar and by manually searching the reference lists of the initially included literature ([4]). All searches were conducted in June and completed in July 2023. The final search terms included: TITLE (“social media”) AND ALL (“addiction”) AND ALL (“therapeutic techniques” OR interventions OR “screening tools” OR questionnaires OR surveys). Database-specific field codes were used where needed. Parentheses ensured that studies focused on social media addiction and included either an intervention concept or a screening/assessment concept.

Searches were limited to full-text peer-reviewed journal articles published in English from 2010 to 2023. The initial database search yielded n = 1061 articles. After setting the limiters and removing exact duplicates, the search yielded n = 447 articles for screening and selection. An additional n = 25 articles were identified for possible inclusion through other databases, totaling n = 472 articles for title and abstract screening. The excluded literature n= 428 did not fit the aim of the current review, with n = 44 being included for full-text screening. The final pool of literature included for analysis was n= 18 (also see Figure 1).

Figure 1.Flow diagram for search and inclusion.

In accordance with [4], the researchers identified clear inclusion and exclusion criteria to eliminate any literature that did not address the research question. Inclusion and exclusion criteria such as eligible populations, study designs, intervention types, and whether platform-specific studies were included were guided by the research question and research aim. This ensured that literature that did not address the research question/aim was eliminated and that only relevant literature was included.

The research team started with an initial evaluation of the title and abstract against the inclusion criteria. Furthermore, to ensure that all relevant literature was initially included, the initial evaluation was followed by the retrieval of the full-text articles to screen and select the full-text articles against the inclusion criteria. The researchers conducting the review consisted of four members (HM, JD-S, REdT, and WdK). Three reviewers (HM, JD-S, REdT) conducted the title and abstract screening and the full-text screening. During each step of the review process, the three reviewers (HM, JD-S, REdT) worked independently throughout title and abstract screening, full-text screening, data extraction, and data analysis, and continuously held in-depth discussions throughout the review process to ensure that relevant literature was selected to address the review question and that they agreed on the findings from the analysis phase.

WdK acted as the independent reviewer, ensuring overall quality assurance during the screening and selection process and verifying that the data extraction and findings were accurate and appropriate.

Furthermore, the researchers provided a descriptive summary of the available literature without critically evaluating it ([37]). During this stage, the focus was on compiling a data extraction table to map the key data of the included literature ([4]). The data mapping was divided into two main tables, with Table 1 including literature related to social media addiction therapeutic techniques and interventions available, and Table 2 including literature on the scales available to assess social media addiction. Table 3 includes additional information on the validity scores for the scales included and discussed in this scoping review study.

Table 1.Therapy and interventions.

Table 2.Scales.

Table 3.Validity scores for scales.

Note. [2] discuss the Bergen Social Media Addiction Scale (BSMAS), which is an adaptation of the Bergen Facebook Addiction Scale (BFAS) ([3]). The adaptation was minor and involved replacing the word Facebook with “social media only” and defining “social media” as “Facebook, Twitter, Instagram, and the like” in the instructions to participants. The article by [2] therefore focuses on the BFAS’s validity to support the validity of the BSMAS.

The guidelines provided by [25] were applied during data analysis of this scoping review study. [25] identifies three phases that should be followed during content analysis: 1) Reducingthe data; 2) Groupingthe data; 3) Formingcategories that answer the review/research question. Firstly, the researchers familiarize themselves with the data by reading through the relevant literature several times ([25]). Any data that is related to the review/research question was given an open code ([25]). The researchers then compared all the open codes and grouped together similar codes to form sub-concepts ([25]). Sub-concepts were then grouped together to form concepts ([25]). Finally, concepts were grouped into main categories ([25]).

Interventions Used for Reduced Social Media Addiction and Usage

Interventions Targeting Social Media Addiction

[53] used a two-week short-term CBT-based abstinence intervention. [30] conducted weekly, 2 to 3-hour group therapy sessions for four weeks. The sessions used acceptance and commitment therapy to address psychological distress and social media addiction in college students. [29] used an eight-week facilitator-led group activities-based (GA) motivational enhancement therapy (MET) (GA-MET) program. [19] developed a two-step cognitive behavioral approach intervention program. [24] developed a 10-week life skills development program based on cognitive and behavioral therapy, as well as group or occupational therapy, where participants attended one sixty-minute session per week that focused on increasing life skills in adolescents. [13] adapted features from a mobile phone addiction intervention for a web-based intervention system to assist in the management of SNS addiction, specifically Facebook addiction, in higher education settings.

Interventions Targeting General Social Media Use Reduction

[47] conducted a one-week abstinence intervention in which participants were instructed not to use social media. [15] used a values-alignment messaging approach to increase motivation towards reducing SMU in adolescents. [46] used a one-week contingency management intervention.

Outcomes of Social Media Addiction and Usage Interventions

Interventions Targeting Social Media Addiction

[53] found that life satisfaction improved with short-term abstinence from problematic social media use. They also found that the level of addiction influences the response rate at which this intervention is effective on users. Users with higher SMA respond more quickly compared to users with lower SMA. Their intervention allowed users to form rational SMU behaviors rather than quit social media ([53]). Qualitative analysis of their findings on the abstinence intervention revealed positive influences such as active behaviors (hobbies or exercising) that also lead to affirmative feelings, being less distracted, and increased self-efficacy, efficiency, and productivity. Negative influences included passive behavior, where participants would spend time aimlessly. They would also resort to external matters to avoid accessing social media. Negative feelings of FOMO and perceived inconvenience were present during abstinence ([53]).

[30] four-week ACT program indicated no substantial changes in SMA post-intervention; however, on follow-up, participants displayed lowered addiction severity scores and decreased SMU. Psychological distress only decreased on follow-up, while psychological flexibility increased post-intervention and on follow-up. [29] found that the GA-MET program did not decrease SMA but did decrease the average duration of social media usage on weekdays and weekends. It also decreased emotional behaviors and depression. [24] found that their life skills enhancement program was statistically significant in reducing risk behaviors of social media addiction in adolescents.

[13] found that there were five web-based intervention features that can be used in web-based interventions for Facebook addiction. These five intervention features were automatic notifications, automatic limitations, automatic reward, manual limitation, and manual monitoring, and can assist in managing or intervening with Facebook addiction. [19] found that the cognitive behavioral intervention was effective in decreasing SMA, that daily reflections were helpful in reducing SMU, and that the intervention led to an improved emotional state and increased learning engagement.

Interventions Targeting General Social Media Use Reduction

During the one-week abstinence intervention, participants experienced substantially increased social pressure from their community to engage with their social media. This pressure decreased back to baseline levels postintervention ([47]). Social media abstinence led to decreased positive affect during the intervention; however, after the intervention, positive affect increased. One week of abstinence had no significant effect on negative affect. Everyday boredom levels substantially increased during the intervention phase and decreased after the intervention subsided. Craving levels considerably increased during the intervention phase ([47]).

[15] found that a values-alignment message increased adolescents’ motivation to control SMU immediately after the intervention compared to receiving traditional messages sharing the benefits of decreasing their SMU. [46] found that a one-week contingency management intervention revealed decreased smartphone and SMU in some participants during the intervention. Post-intervention changes in SMU, however, were minimal.

The Bergen Facebook Addiction Scale(BFAS)and the Bergen Social Media Addiction Scale(BSMAS)

[3] developed the BFAS based on the six core elements of addiction described by [16]. The BFAS originally consisted of 18 items, which were divided into 3 items per core element of addiction, e.g., salience, mood modification, tolerance, withdrawal, conflict, and relapse, scored on a 5-point Likert-type scale between 1—very rarely and 5—very often. Within each core element of addiction, the 6 items with the highest corrected item-total correlation were included in the final scale ([3]). The confirmatory factor analysis (CFA) indicated a one-factor analysis across loadings of the six core elements. Furthermore, a subsample of the original sample completed the scale again after three weeks, where the test-retest correlation coefficient was .82. Categorization of problematic Facebook use was not examined at the time of the development of the BFAS. However, total scores will range between 6 and 30. Additionally, the higher the score, the greater the indication of Facebook addiction. The BFAS has also been further translated or adapted to several applicable contexts (for example, see ([7]; [28]; [39]; [45])). The authors ([3]) concluded that the scale has acceptable psychometric properties and that it can be used in epidemiological and clinical contexts. They further recommend that future studies focus on determining appropriate cutoff scores relating to Facebook addiction.

The BSMAS was adapted from the BFAS during a large-scale cross-sectional study conducted by [2]. The BSMAS consists of 6 items reflecting the six core elements of addiction ([16]). The scoring ranges from 6 to 30, and with the introduction of the BSMAS, it yielded an internal consistency of alpha = .88. Upon inception and adaptation from the BFAS, the original adaptation study for the BSMAS (conducted by ([2])) did not focus on confirming validity. However, several researchers have focused on measuring the psychometric properties of the BSMAS within different contexts and/or for different language groups (for example, see (Dadiotis[8]; [26]; [32]; [44])).

Social Media Disorder Scale for Parents(SMDS-P)

The SMDS-P ([5]) was developed for parents to evaluate the social media addiction of adolescents. The SMDS-P was developed based on the SMD Scale of [50] discussed below. The SMDS-P uses 9 items on a forced-choice scale, yes or no, and the CFA indicates a one-factor model. The purpose of this scale is to give parents the chance to rate their adolescent’s social media usage and/or problematic social media usage (PMSU). The authors ([5]) indicated that they were the first to assess the psychometric properties of a parental scale focused on adolescent PSMU. They further concluded that the scale showed promising psychometric properties, and that the internal consistency (α = .85) of the scale was almost exactly the same as the original SMDS (α = .84). Other similarities highlighted between the parental scale and the original SMDS included: 1) both indicated a one-factorial model, and 2) identical cutoff scores. They concluded that the scale should be validated for the clinical setting.

Social-Media Addiction Screening Scale(S-MASS)

The S-MASS ([6]) was developed as a self-report questionnaire that can be used to screen for social media addiction. The scale consists of 16 items (three-factor model) based on the three key elements of behavioral addiction as found in the ICD-11. The items are derived from the 9 criteria for Internet Gaming Disorder (IGD). The S-MASS utilizes a four-point Likert Scale ranging between 0—definitely nottrue and 3—definitely true. The summated score ranges from 0 to 48, with a higher score indicating a greater likelihood of social media addiction. Furthermore, a CFA was conducted to test the proposed three-factor model, and it was concluded to be a good model fit. [6] concluded that the scale was a reliable and valid screening tool for social media addiction (SMA). However, they indicated that future researchers should focus on further validation studies, specifically focusing on criterion validity (e.g., predictive and concurrent validity). Additionally, it is recommended that future researchers also consider assessing whether the scale is sensitive to change after SMA interventions.

Chinese Social Media Addiction Scale

After reviewing the literature on existing measures addressing behavioral and social media addiction, the Chinese Social Media Addiction Scale ([27]) was developed considering the seven classic dimensions related to behavioral addiction. These included compulsive use, withdrawal, negative consequences, mood alteration, salience, tolerance, and relapse. Furthermore, [27] also considered two extra dimensions: preference for online social interactions and continued use. However, during the validation study, it seemed that a six-factor model was more appropriate. The final six dimensions that explained total variance to 63.83% are preference for online social interaction, mood alteration, negative consequences and continued use, compulsive use and withdrawal, salience, and relapse. The CFA also indicated that a six-factor model was a good fit in relation to the data.

Furthermore, items were compiled from other measures such as the Internet Addiction Test, Facebook Addiction Scale, and Internet Gaming Disorder Scale. Respondents score each item on a 5-point Likert scale ranging from 1—strongly disagree to 5—strongly agree. The authors ([27]) concluded that, overall, this scale was appropriate for assessing broad SMA in China and that it has adequate psychometric properties. Additionally, this scale can be used in clinical and non-clinical settings. However, [27] indicated that future research should focus on: 1) Assessing test-retest reliability; 2) Assessing whether it is applicable for other population groups such as young adolescents; 3) Whether the scale can be used when assessing for other SM platform addictions (e.g., Facebook).

Social Media Use Disorder Scale for Adolescents(SOMEDIS-A)

This self-rating screening tool ([35]) was adapted from the Gaming Disorder Scale for Adolescents (GADIS-A) to measure social media use disorder (SMUD) among adolescents. It is based on the ICD-11 criteria for Gaming Disorder (GD), and during the validation study, it was indicated that a two-factor model (cognitive-behavioral symptoms and negative consequences) seems to be the best fit. Furthermore, the scale included 10 items that respondents had to complete. Items 1 to 9 (focused on SMUD symptoms) utilized a 5-point Likert Scale ranging from 0—strongly disagree to 4—strongly agree. Item 10 (focused on frequency and duration of use) was measured based on four responses, not at all, only on single days, for longer periods, or nearly daily. Overall, the internal reliability for the scale in relation to the population included in the validation study was considered good (Cronbach’s α= .82). It was also concluded that this scale can distinguish between adolescents who struggle with SMUD and those who do not. [35] concluded that this was the first screening tool that measured SMUD in relation to the ICD-11 criteria for GD. Furthermore, it showed appropriate psychometric properties, specifically internal reliability and criterion validity. They recommend that future researchers should focus on clinical validation studies.

Social Media Addiction Scale—Short Form(SMAS-SF)

The SMAS-SF ([43]) was conceptually developed by including literature on addiction and social media use in conjunction with opinions of field experts. The scale originally consisted of a pool of 41 items that are scored on a Likert scale ranging from 1—definitely notappropriate to 5—appropriate. The final form includes 29 items with 4 sub-dimensions, e.g., virtual tolerance, virtual communication, virtual problems, and virtual information. [43] indicated that the scale showed good psychometric properties and can be used to identify SMA within students (specifically Turkish students between the ages of 12 and 22 years). Furthermore, [43] concluded that the psychometric properties of the scale should be assessed in other studies focused on different population groups and/or other age groups.

Social Media Addiction Scale(SMAS)

The SMAS was developed specifically to identify SMA within university students ([49]). During this development and validation study, it seemed that a four-factor model explained 59% of the variance. These four factors were identified as: occupation, mood modification, relapse, and conflict. The factor structure for this scale was identified by means of an Exploratory Factor Analysis (EFA); however, a CFA was not conducted to assess the model fit. It was also indicated that the content of the items that had high loadings for a specific factor was assessed, after which, based on the content of the items, each factor was named. E.g., items 13, 14, 15, 16, and 17 showed high factor loadings for the second factor, after reviewing the content of these items, it was identified to be related to mood modification. The scale includes 41 items that are scored on a Likert scale ranging from alwaysto never.Essentially, the higher one scores (maximum 205) when calculating the results, the more likely the person is to be addicted to social media.

The Social Media Disorder Scale27-item and9-item(SMDS)

The SMD Scale ([50]) was developed and validated as a 27-item and 9-item scale (related to the SMDS-P above). This scale was developed based on the DSM-5 criteria for IGD. The 27-item scale includes 3 items per criterion, e.g., preoccupation, tolerance, withdrawal, persistence, displacement, problems, deception, escape, and conflict. The 9-item scale includes 1 item per criterion. The calculation of both scales is done by summing all yes answers. The scale was developed with the purpose of including clear cutoff points that distinguish between “disordered” and “high-engaging non-disordered social media” users ([50]: p. 478). The authors ([50]) concluded that both the 27-item and 9-item scales showed good psychometric properties. They recommend that future research should focus on identifying whether the nine criteria for IGD can be used as the core features for SMD. Additionally, focus can be placed on whether the items included in the 9-item SMDS are indeed suitable for diagnosing SMD in clinical and non-clinical sample groups.