Participants
Individuals (male and female) who were 18 years or older were included in the study. Participants were excluded if they presented with a possible comorbid psychotic disorder (based on the L module of the Mini International Neuropsychiatric Interview [31]). Participants were recruited from several general Facebook groups and one private group dedicated to Hikikomori: “Hikomori France. Communauté francophone des Hikikomori et reclus sociaux” (all French speaking). After asking the administrators for their permission, we posted a message explaining the aim, duration, and anonymity of the study, as well as a link to the questionnaire (with a full explanatory note containing the ethical requirements). Participants who agreed to participate in the study had to provide their informed consent before accessing the questionnaires (for minors, one of the parents also had to give consent). For the Hikikomori Facebook group, the questionnaire was distributed through its creator, known by the pseudonym Ael, himself having been Hikikomori for 13 years.
All participants received information regarding the survey and all participants provided written informed consent to participate.
The two groups (control group and Hikikomori group) were formed on the basis of the following measures: (1) their score on the 25-item Hikikomori Questionnaire (HQ-25; see Measures subsection) and (2) their answers to questions created for the study. As recommended by the authors of the HQ-25, given the high rate of false-positives [32], and because of the restrictions linked to the COVID-19 pandemic situation when recruitment took place, three additional questions were asked:
1. Outside of the current restrictions linked to COVID-19, how often do you go out alone or with friends for shopping, sports, or socializing? (once a week, several times a week, every day, once a month, less than once a month, almost never, never)
2. Outside of the current restrictions linked to the COVID-19, do you go outside only for vital needs (food or medical appointment)? (yes, no)
3. Does the current pandemic context have an impact on what you usually want to do in terms of outings, leisure, or any other activity outside your home? (yes, no, not at all)
The control group was composed of 101 participants (30 males, 70 females, 1 other; mean age ± SD = 36.2 ± 12.8 years). Most participants were either single (39.6%), or married or in a relationship (50.5%). The majority were employed (65.3%) and had less than a high school graduate education (84.2%). They either lived alone (44.0%) or with a partner (36.0%). Most had parents who lived together (45.5%).
The Hikikomori group was composed of 28 participants (13 males, 15 females, mean age ± SD = 30.1 ± 9.1 years). Most were single (67.9%), unemployed (60.7%), and had less than a high school graduate education (75.0%). They either lived alone (39.3%) or with one parent (39.3%). Most had divorced parents (42.9%).
Measures
Participants’ characteristics were evaluated, including age, gender, education, and marital and professional status.
The HQ-25 [32] was used to assess the severity of symptoms of Hikikomori over the preceding 6 months. This self-administered instrument composed of 25 items evaluates the psychological features and behavioral patterns of typical Hikikomori syndrome, such as lack of social connectedness, active social isolation or withdrawal behavior, avoidance of social contact, and a sense of alienation from society. All items of the HQ-25 were rated on a scale ranging from 0 (strongly disagree) to 4 (strongly agree). Authors of the HG-25 proposed a cutoff score of 42 (out of 100), which was associated with a sensitivity of 94% and a specificity of 61% in their clinical study. In our study, the scale showed high internal consistency, with a Cronbach’s alpha of .92.
The French version of the Big Five Inventory (BFI-Fr) [33] is a 45-item self-report questionnaire that assesses five personality domains with two facets per domain: Neuroticism (Anxiety and Emotional Volatility), Agreeableness (Compassion and Respectfulness), Conscientiousness (Organization and Responsibility), Extraversion (Anxiety and Emotional Volatility), and Openness (Aesthetic Sensitivity and Creative Imagination). All items of each dimension were rated on a scale from 1 (strongly disapprove) to 5 (strongly approve), for a total score ranging from 5 to 25 in each dimension. Each domain demonstrated high reliability and a clear factor structure. The BFI-Fr yielded adequate internal consistency in the current sample for neuroticism (Cronbach's a = .83), agreeableness (Cronbach's a = .80), conscientiousness (Cronbach's a = .85), extraversion (Cronbach's a = .88) and openness (Cronbach's a = .80).
The Brief Coping Orientation to Problems Experienced (Brief-COPE, [34]), French version [35], was used to assess various coping styles. The scale consists of 28 questions, including 14 subscales (two questions per subscale) in a Likert scale format (0 to 4 points). These subscales, or coping styles, include active coping, instrumental support, planning, acceptance, emotional support, humor, positive reframing, religion, behavioral disengagement, denial, self-distraction, self-blame, substance use, and venting. In this study, the scale showed good internal consistency, with a Cronbach’s alpha of .70. The problem-focused coping dimension included active coping, planning, and instrumental support. The emotion-focused coping dimension included emotional support, religion, positive reframing, acceptance, humor, and venting. The dysfunctional coping dimension included self-blame, denial, self-distraction, behavioral disengagement, and substance use [34].
The French version of the Hospital Anxiety and Depression Scale (HADS [35]) is a14-item self-report scale that was used to screen participants for anxiety (seven items) and depression (seven items). This tool has good psychometric properties and is quick to administer and thus suitable for field research. Cutoff scores for the depression and anxiety subscales are as follows: 7 or 8 indicates “possible presence”, 10 or 11 suggests “probable presence”, and 14 or 15 refers to “severe presence”. In this study, the two subscales showed good internal consistency, with a Cronbach’s alpha of .80 for depression and 0.74 for anxiety.
Statistical Analysis
All statistical analyses were performed with SPSS software (version 20). For sociodemographic, personality, and coping characteristics, we used t-tests or chi-squared tests for group differences, with Cohen’s d or Cramer’s V for effect size for continuous or categorical variables, respectively. We considered d > 0.5 as a medium effect size and d > 0.8 as a large effect size [36]. We also used univariate logistic regression to examine associations between social withdrawal and predictor variables (personality traits, coping, depression, and anxiety) and the presence of co-occurring social withdrawal. Odds ratios (ORs) and 95% confidence intervals were generated by using logistic regressions.