**1. Introduction**

The Internet can serve as a double-edged sword. On the one hand, it provides many benefits and increases convenience. On the other, its use can be highly alluring, which may result in excess consumption, which manifests in behavioral addiction-like symptoms [1]. Since Goldberg [2] proposed the concept of "Internet Addiction Disorder", many researchers have used similar concepts, such as Pathological Internet Use, Problematic Internet Use, Excessive Internet Use, and Internet Addiction [3].

Although these concepts are slightly different, they all have two common characteristics: (1) addiction-like symptoms: compulsive use of Internet, withdrawal reactions, and tolerance, combined with inability of users to effectively control themselves; (2) as a result, their daily functions such as academic and social activities, as well as sleep and food intake, are impaired [4]. We believe that

the concept of Internet Addiction (IA) can best reflect the above characteristics and is widely used. While the term is controversial [5], for convenience reasons, we will use in this study the term Internet Addiction to describe this phenomenon, while acknowledging the need of future research to fine tune its definition and boundaries. IA has been recognized as a public health concern [6], which can correlate with poor health outcomes such as higher likelihood of being overweight [7], impaired sleep [8], increased suicidal ideation and attempt [9], and increased depressive symptoms [10].

College students can be highly susceptible to IA [11]. Once freed from the strict control of high school and family, they tend to have free time at their disposal, which when coupled with easy access to the Internet can lead to IA [12]. In addition, these students are experiencing changes in the environment with which they must learn to cope. For example, they face pressures on handling new interpersonal relationships and balancing learning and social activities [13]. Consequently, students may seek to relieve these pressures via Internet use, which can result in IA [14]. Among Chinese college students, IA is prevalent. Between 6.56% and 13.5% of Chinese college students may be classified as addicted to the Internet [15]. Therefore, paying close attention to issues related to college students' use of the Internet, and intervening if necessary, is critical for preventing IA and its adverse impacts [16].

Various schools of psychotherapy have been applied to alleviate IA. Among them, solution-focused brief therapy (SFBT), as a relatively new paradigm of post-modern constructivism, has also been applied to IA cases. Specifically, Yang [17] provided psychotherapy with an emphasis on SFBT and comprehensive family intervention to teenagers with IA, and reported positive results showing that SFBT can be an effective therapy for treating IA. SFBT evolved out of the clinical practice of Steve de Shazer, Insoo Kim Berg, and colleagues at the Brief Family Therapy Center in Milwaukee, Wisconsin, in the early 1980s [18]. SFBT change processes were originally grounded in the constructivist approaches to communication and social interactional theories [19]. Over time, SFBT also became associated with social constructionism and the philosophical, post-structural views of language such as in Wittgenstein's language games [20,21]. The main components of SFBT include looking for previous solutions, acknowledging problems but identifying that exceptions to the problem are key to the solution, focusing on present and future, as opposed to past-orientated questions, using validation, and utilizing techniques such as miracle questions, scaling questions, and coping questions [22].

Due to the high cost of psychotherapy, people strive for a faster and more effective treatment, which has led to the widespread application of SFBT, such as for treating students' emotional behavior problems [23], substance abuse [24], adolescent crisis events, suicide and self-harm [25]. SFBT is not only applied to psychotherapy and counseling, but also applied in school class management, enterprise management and other fields [26]. SFBT emphasizes using a positive perspective to comprehend people, expressing their affirmation, and fully trusting that people will understand the methods that suit them. It is complementary to traditional Confucian culture [27] and compatible with it [28]. As such, it is highly suitable for application in China, as demonstrated by large effects in Chinese populations [29].

Comparing SFBT with Cognitive Behavioral Therapy (CBT), Jordan, Froerar, & Brovelas [30] argue that professionals using CBT assume the expert role, aiming to identify problems in people's thinking and behavior. But professionals using SFBT regard their patients as already having all the resources they need, defining their own role as supportive in eliciting the individuals' strengths, and empowering people to articulate these into achievable goals for the future. Therefore, SFBT enables patients to look for possible behaviors to change, and puts patients in control as regulators of their behavioral changes. It is not only beneficial for patients to maintain behavioral change, but also helps them to improve self-efficacy, thus making it more suitable for college students' self-exploration and continuous transformation. In addition, some studies have shown that SFBT used more of the client's exact words, and used more positive language, than other therapies such as CBT and Motivational Interviewing (MI) [30,31]. Therefore, SFBT may be more beneficial to relieve patients' anxiety and their concerns towards the issue, thereby making it easier for college students who are sensitive and care about others' evaluation and social pressures to accept this method and reduce their resistance. In addition, group counseling is an effective intervention for IA [32]. Compared to solution-focused individual counseling, applying SFBT to group counseling makes the process lively, encourages group members to share their experiences and promotes psychological growth with mutual support and encouragement from the group. Nevertheless, the efficacy of SFBT in group counseling to reduce IA in Chinese populations is not fully established. We aim at ameliorating this gap in this pilot study.

Specifically, we focused on applying SFBT in group counseling and explored the intervention effects of solution-focused group counseling on IA in college students. By examining immediate and long-term post-intervention effects, we aim at clarifying if solution-focused group counseling improves IA in college students. Based on the above mentioned findings of prior intervention studies applying SFBT in different contexts, we hypothesized that participants in the experimental group (i.e., those who will accept five-week solution-focused group counseling aimed at alleviating Internet Addiction core symptoms, such as compulsive use, withdrawal and tolerance, and Internet Addiction related problems such as interpersonal, health-related, and time management problems) will show significant changes in these aspects, beyond the placebo effects observed in a control group.

#### **2. Materials and Methods**

#### *2.1. Study Design and Participants*

Participants were screened from a sample of 27 college students who presented high IA symptomology, recruited through flyers on campus. Based on an initial interview and the Revised Chinese Internet Addition Scale (CIAS-R), we selected 18 subjects with scores ≥53. The other 9 had score <53 and were excluded. The final 18 participants were randomly assigned to experimental or control groups with inclusion criteria CIAS-R score ≥53, which represents moderate to high levels of IA. There were 9 in the experimental group, (male = 3, female = 6; age = 20.11 ± 1.45) and 9 in the control group (male = 3, female = 6; age = 20 ± 1.56). Non-parametric Mann-Whitney U tests showed that there was no significant difference in CIAS-R score between the two groups [Z = −0.71, *p* > 0.05]. We also used a self-reported Symptom Checklist-90 (SCL-90) questionnaire as an auxiliary selection tool, and the scores of each factor of the students in the two groups were less than 5, showing no symptoms of serious mental disorders. The study was given ethical approval by the Institutional Review Board of the Southwest University (No. 20180030).

#### *2.2. Procedures*

Closed semi-structured group counseling was conducted on the experimental group once a week for five weeks. Every counseling session lasted about two and a half hours. There were one leader and two assistants in the group. The leader is a certified psychotherapist, and the assistants are psychology graduate students trained for group counseling. Their roles followed basic procedures that were designed in advance. Participants were given free interaction space. The leader regulated and navigated the group process according to the needs of the members. The control group did not receive any intervention. Both groups completed surveys at t1 and t2 (t1 + 5 weeks). The intervention group also completed a t1 + 6 months survey. Instruments included in the survey are outlined in the next section.

#### *2.3. Instruments*

#### 2.3.1. Revised Chinese Internet Addiction Scale (CIAS-R)

The scale [33,34] consists of 19 items, on a four-point Likert scale. Two subscales are Internet Addiction Core Symptoms (IA-Sym) and Internet Addiction Related Problems (IA-RP). Core symptoms include four dimensions: Compulsive and Withdrawal Symptoms of Internet addiction (Sym–C&Sym-W), Tolerance Symptoms of Internet Addiction (Sym-T), Interpersonal and Health-Related Problems of Internet Addiction (RP-IH), and Time Management Problems (RP-TM).

#### *Int. J. Environ. Res. Public Health* **2020**, *17*, 2519

The experimental group and the control group completed the scale before and after group counseling, and the experimental group also carried out follow-up measurement six months after group counseling. The higher the score, the greater the risk of IA is. CIAS-R is widely used in the measurement of IA. Several studies show that CIAS-R has good reliability, with Cronbach's alpha ranging from 0.85 to 0.90 [34–36]. In this study, Cronbach's alpha was 0.86, 0.89 and 0.83 in the three rounds of administration.

#### 2.3.2. Symptom Checklist-90 (SCL-90)

The Symptom Checklist-90 (SCL-90) has a total of 90 items rated on a five-point Likert scale. It includes 9 factors, such as depression and anxiety. Studies have found that IA may be comorbid with depression and anxiety disorders [37,38]. If comorbidity exists, it is very important to evaluate and treat related mental disorders. However, our group counseling mainly focused on Internet addiction and had no strong pertinence to other mental disorders. Therefore, SCL-90 is used to screen whether the subjects had major comorbid mental disorders.
