Table 2 All studies included in the review

Intervention studies

Lee et al. [3]

Socially Withdrawn Youth (N = 41), Middle/High School Students (N = 239)

16.4 ± 3.5, 16.3 ± 1.5

Korean

31, 10

Only data on clinical withdrawal characteristics were extracted

Global Assessment Functioning Scores (GAF) (Continuous scale; scored 0–100)

 Higher score = Higher level of day-to-day functioning

 Score of 41–50 = any serious impairment in school/work/social functioning

 Score of 51–60 = moderate difficulty in school/work/social functioning

Minimum duration of social withdrawal identified as 3 months

Internet usage (Mean hours ± SD): 5.20 ± 3.40

Overall GAF Scores for all participants before and after the home visitation psychotherapy intervention (Mean ± SD):

 Pre 44.6 ± 11.1 to post 53.4 ± 13.2, p < 0.001

48.8% of socially withdrawn youth showed no change in GAF scores post intervention

Malagón-Amor et al. [39]

Socially Withdrawn Adults (N = 190)

39.1 ± 18.1

Spanish

136, 54

Self-developed questionnaire assessing social network of relationships

Sociodemographic data and clinical history were extracted from the patients’ files

No statistically significant difference in social networks at the following time points:

 4 months: p = 0.200

 8 months: p = 0.947

 12 months: p = 0.991

Chan [75]

Hidden Youth (N = 502)

12–21 years

Chinese

384, 118

Self-developed questionnaire assessing participation in therapeutic activities

Empowerment Scale per Rogers et al. (28-Likert items; scored 20–80)

  Higher score = levels of empowerment

Psychological Capital Questionnaire (24-Likert items; scored 21–126)

 Higher score = higher psychological well-being

Relationships by Correlations and P-values of Hikikomori Play Therapy to:

 Empowerment = 0.59, p = 0.0000

 Psychosocial well-being = 0.60, p = 0.0000

Level 3 Hierarchical Regression:

 Empowerment = 9.56, p = 0.0000

 Play Therapy = 0.64, p = no statistical significance

No scale scores were reported

Law et al. [74]

Youth

(N = 373)

12–29 years

Chinese

171, 202

Used a proxy checklist as per Uchida and Norasakkunit to measure social withdrawal (4-item yes/no answers)

Decrease in social withdrawal by 4.61%

No scale scores reported for socially withdrawn youth

Yokoyama et al. [18]

Hikikomori

(N = 5)

16–35 years

Japanese

1, 4

Self-developed survey administered post intervention

C-BED intervention:

Decrease in anxiety, increase in willingness to participate post intervention

No measurement scales used in the study

Cross-sectional studies

Koyama et al. [62]

Hikikomori (N = 19), Community Residence (N = 1641)

20–49 years

Japanese

14, 5

Face-to-face household survey assessing sociodemographic data, hikikomori symptoms, and psychiatric history

Demographic data of males and females from Japan, age 20–49

Kondo et al. [40]

Hikikomori

(N = 337)

24.2 ± 5.4

Japanese

252, 85

Psychiatrists elicited information on sociodemographic data and psychiatric history during consultations

Education level (%):

 Junior High: 34.4%

 High School: 39.5%

 University/College: 25.5%

Demographic data of males and females from Japan, with a mean age of 24

Nagata et al. [64]

Hikikomori (N = 27), Patients with Social Anxiety Disorder

(N = 114)

27.4 ± 7.5

Japanese

12, 15

Information on sociodemographic data and psychiatric history were elicited from retrospective clinical data

Sociodemographic data on the distribution of males and females from Japan, with a mean age of 27, and with 11.7 ± 1.7 years of education; however, it was unclear whether the data on education included the preschool years

Umeda and Kawakami [63]

Hikikomori (N = 15), Community Residence (N = 693)

36.3 ± 11.1, 30.1 ± 8.6

Japanese

10, 5

Face-to-face population-based survey conducted in metropolitan, urban, and rural areas of Japan assessing sociodemographic data, hikikomori symptoms, psychological history, childhood social class, parental psychological history, and childrearing practices

Education level (%):

Junior High: 3.8%

High School: 71.4%

University/College: 24.8%

Demographic data on males and females from Japan, with a mean age 36

Krieg and Dickie [37]

Hikikomori (N = 24), University Students (N = 59)

22.84 years, 20.59 years

Japanese

14, 10

Trait Shyness Scale (16-Likert items; scored 16–80)

Maternal Attachment Scale

 Insecure/avoidant attachment subscale (8-Likert items; scored 1–4)

 Insecure/ambivalent attachment subscale (7-Likert items; scored 1–4)

Parental Rejecting Behavior Scale

 Ignoring subscale (1-Likert item; scored 1–7)

 Threaten subscale (1-Likert item; scored 1–7)

 Lock-out subscale (1-Likert; scored 1–7)

 Composite score (Sum of subscale scores)

Peer Rejection Scale (1-Likert item; scored 1–8)

Maladjustment to School Scale

 Difficulty with adjusting to peer group work (1-Likert item; scored 1–7)

All Japanese versions. Higher score = higher disturbance/intensity

Hikikomori vs University students for various scales (Mean ± SD):

Shyness score: 52.83 ± 12.27 vs 46.89 ± 9.76

Maternal attachment

 Avoidance score: 2.21 ± 0.70 vs 2.09 ± 0.71

 Ambivalence score: 2.08 ± 0.75 vs 1.51 ± 0.51

Parental rejection score

 Locking out score: 3.58 ± 2.21 vs 3.21 ± 2.37

 Threatening loss of relationship score: 3.71 ± 1.94 vs 2.05 ± 1.61

 Ignoring score: 3.00 ± 2.09 vs 2.15 ± 1.57

 Composite score: 10.29 ± 4.44 vs 7.41 ± 4.02

Peer rejection: 3.85 ± 2.31 vs 2.41 ± 2.15

Maladjustment to school: 4.50 ± 1.62 vs 3.20 ± 1.85

Hikikomori Correlation and P-values:

 Shyness: 0.285, p < 0.05

 Ambivalent maternal attachment: 0.400, p < 0.01

 Adjustment to middle school: 0.391, p < 0.01

 Parental rejection: 0.302, p < 0.01

 Threatened loss of relationship: 0.405, p < 0.01

 Ignoring: 0.219, p < 0.05

 Peer rejection: 0.287, p < 0.01

Chan and Lo [60]

Hidden Youth

(N = 588)

12–30 years

Chinese

373, 215

World Health Organization Quality of Life scale-brief (WHOQOL-BREF)

(28-Likert item; scoring 0–112)

Higher score = Higher Quality of Life

Hidden Youth QOL Correlations and P-values

 Low level of social withdrawal = 0.550, p = 0.0000

 High level of social withdrawal = -0.850, p = 0.0000

WHOQOL-BREF scores not reported

Malagón-Amor et al. [59]

Hikikomori

(N = 164)

40.0 ± 18.3

Spanish

121, 43

Information on sociodemographic data and psychiatric history were elicited from clinical data on consultations

Sociodemographic data on the distribution of males and females from Spain

Uchida and Norasakkunkit [58]

Hikikomori (N = 114), NEET (N = 86), Employed Adults (N = 7525)

20–39 years

Japanese

–

NEET Hikikomori Risk Scale

 Unclear ambitions (2-items; scoring 0–12)

Higher score = Higher level of unclear ambitions

Hikikomori vs NEET vs Employed Adults (Mean ± SD)

Unclear ambitions:

5.48 ± 1.60 vs 5.05 ± 1.41 vs 4.26 ± 1.37

Frankova [61]

Hikikomori (N = 35), Control group (N = 28)

18–40 years

Ukrainian

14, 21

Chaban Quality of Life Scale (10- item scale, scoring not reported, newly developed scale from Ukraine)

Higher score = Higher Quality of Life

Hikikomori (with and without psychiatric comorbidities) vs Control group

 11.7 ± 2.70; p = 0.001 and 13.7 ± 3.3; p = 0.001, vs 19.3 ± 3.50

Chauliac et al. [38]

Socially Withdrawn Youth

(N = 66)

23.2 ± 4.75

French

53, 13

Data on clinical withdrawal characteristics extracted from patient records (Frequency only)

Relationships maintained (%) with:

 Family: 63%

 Friends: 8%

 Family and friends: 8%

 Family and sentimental/school: 3%

 None: 19%

Going on outings (%):

 Alone: 38%

 Accompanied: 35%

 None: 27%

Poor hygiene (%):

 Yes: 33%

 No: 67%

Yuen et al. [41]

Hikikomori

(N = 104)

19.02 ± 3.62

Chinese

62, 42

Self-developed questionnaire to assess sociodemographic data, daily activities, and health history

Modified Berkman–Syme Social Network Index (7 yes/no items; scored 0–7)

 Higher score = higher connectedness

Types of Daily activities (Mean hours ± SD):

Sleeping: 7.83 ± 1.99

Computer use: 5.09 ± 4.97

Tablet/mobile use: 3.11 ± 5.03

Eating: 1.90 ± 1.03

Reading comics/animations: 0.95 ± 2.38

Watching television: 0.90 ± 1.13

Other reading: 0.65 ± 2.09

Idling/Facing the wall: 0.40 ± 0.99

Social Network Index (Mean ± SD): 2.79 ± 1.80

Yong and Nomura [43]

Hikikomori (N = 58), Non-Hikikomori (N = 3024)

15–39 years

Japanese

38, 20

Interpersonal difficulties assessed by 4 yes/no questions:

 Fear of meeting others (Q1)

 Anxiety about meeting familiar people (Q2)

 Anxiety about people’s impression of oneself (Q3)

 Cannot blend into groups (Q4)

Hikikomori vs non-hikikomori (% Interpersonal difficulties; p-value):

 Overall = 74.1 vs 36.0; < 0.001

 Q1 = 36.2 vs 8.1; < 0.001

 Q2 = 48.3 vs 7.1; < 0.001

 Q3 = 51.7 vs 28.3; < 0.001

 Q4 = 53.4 vs 14.6; < 0.001

Social class of hikikomori (%)

 Upper: 3.4%

 Middle: 77.6%

 Lower: 19.0%

Wu et al. [42]

Socially Withdrawn Adults (N = 168), Non-Socially Withdrawn Adults (N = 258)

28.82 ± 0.60

Taiwanese

78, 90

Self-developed online survey assessing sociodemographic data, social withdrawal behavior, and psychiatric history

Education:

 Bachelor Level: 90.0%

Lived in areas of (%):

 Low income: 72.6%

 Middle income: 18.5%

 High income: 3.6%

Longitudinal studies

Yuen et al. [35]

Hikikomori

(N = 104)

19.02 ± 3.62

Chinese

62, 42

Chinese Interpersonal Support Evaluation List (ISEL)- Short version (12-Likert items; scored 0–48)

 Appraisal Support subscale (4-Likert items; scored 0–12)

 Belonging Support subscale (4-Likert items; scored 0–12)

 Tangible Support subscale (4-Likert items; scored 0–12)

 Self-Esteem Support subscale (4-Likert items; scored 0–12)

Higher score = higher social support

Modified Berkman–Syme Social Network Index (7 yes/no items; scored 0–7)

Higher score = higher connectedness

ISEL (Mean ± SD):

 Time points 1–3: 24.60 ± 6.30, 24.63 ± 5.99, 24.75 ± 6.89

 Appraisal Support: 6.81 ± 2.16, 6.90 ± 1.87, 7.11 ± 2.18

 Tangible Support: 6.20 ± 1.71, 6.29 ± 1.53, 6.20 ± 1.73

 Belonging Support: 6.00 ± 2.45, 5.89 ± 2.53, 6.05 ± 2.44

 Self-Esteem Support: 5.59 ± 2.08, 5.55 ± 2.14, 5.38 ± 2.39

SNI (Mean ± SD):

 Time points 1–3: 2.79 ± 1.80, 2.93 ± 2.06, to 3.09 ± 1.87

Pilot case–control studies

Katsuki et al. [65]

Hikikomori (n = 22), Non-Hikikomori (N = 18)

33.14 ± 9.33, 37.94 ± 8.93

Japanese

12, 10

Rorschach Comprehensive System; 10-inkblot items, scored as per Takahashi et al. for the Japanese population and scored by a clinical psychologist

 Form Color; higher scores indicate a higher level of passive aggressiveness, a tendency to adjust one’s emotions to the environment and people, an inclination to suppress the expression of emotions when feeling shaken in social situations

SumT; total number of texture-related responses, higher scores indicate a need for and an openness to forming close emotional relationships

Hikikomori vs Non-hikikomori (Mean ± SD):

 Form Color: 2.50 ± 1.68 vs 1.39 ± 1.14, p = 0.037

 SumT: 0.50 ± 0.67 vs 0.11 ± 0.32, p = 0.033

Mixed-method studies

Chan and Lo [60]

Hidden Youth

(N = 363)

21.11 ± 2.93

Chinese

244, 119

Self-developed survey assessing sociodemographic data

 Relationships with Parents Scale as per Zeng and Zeng’s (8- items; scored:?)

 Relationships with Siblings Scale as per Zeng and Zeng’s (7-items; scored:?)

 Relationships with Teachers as per Deng; Xu and Ma’s (7- items; scored:?)

 Relationships with Peers as per Deng; Xu and Ma’s (9- items; scored:?)

 Rosenberg Self-Esteem Scale (10-Likert items; scored 0–30)

  Higher score = higher self-esteem; < 15 = low self-esteem, ≥ 15 = high self- esteem

Self-esteem of Hidden Youth in Correlations and P-values with a good relationship with:

 Parents = 0.73, p = 0.0000

 Siblings = 0.66, p = 0.0000

 Teachers = 0.13, p < 0.05

 Peers = 0.16, p < 0.05

No scale scores were reported

Chan [76]

Hidden Youth

(N = 502)

12–24 years

Chinese

384, 118

Self-developed questionnaire assessing sociodemographic data and uses of counselling services

World Health Organization Quality of Life scale-brief (WHOQOL-BREF) Taiwan Version (measured well-being after receiving counselling services)

Young Person’s CORE (10- Likert items; scored 0–40)

Lower score = positive results

Qualitative Semi-structured interviews assessing perceived advantages & usefulness of three forms of counseling

Total Quality of Life scores (Mean ± SD):

 Online counselling: 3.02 ± 0.43

 Offline counselling: 2.52 ± 0.27

 Integrated counselling: 3.74 ± 0.60

Young Person’s Core (Mean ± SD):

 Online counselling: 1.56 ± 1.29

 Offline counselling: 0.62 ± 0.42

 Integrated counselling: 2.59 ± 1.18

Interview results: online counselling offered platform for communication while offline counselling offered opportunity for mediation during conflicts between youth and their family

Chan [55]

Hidden Youth

(N = 357)

12–30 years

Chinese

–

Self-constructed questionnaires assessing forms of communication and friendship levels

Miller Social Intimacy Scale as per Miller & Lefcourt (17 items; scored: 17–170)

Higher scores = higher levels of social intimacy

Qualitative semi-structured interviews assessing how youth choose usage of forms for communication

Forms of communication used:

 Public text: People you only know (100%); Friends (100%); Good friends (100%); Best friends (100%)

 Public voice: People you only know (3.2%); Friends (49.0%); Good friends (100%); Best friends (100%)

 Public camera meeting: People you only know (0%); Friends (6.0%); Good friends (91.6%); Best friends (100%)

 Private text: People you only know (3.2%); Friends (43.0%); Good friends (100%); Best friends (100%)

 Private voice: People you only know (3.2%); Friends (30.0%); Good friends (100%); Best friends (100%)

 Private camera meeting: People you only know (0%); Friends (17.0%); Good friends (92.6%); Best friends (100%)

Scores for Miller Social Intimacy Scale not reported

Case studies/series

Hattori [73]

Hikikomori

(N = 35)

21.5 years

Japanese

25, 10

Clinical data extracted from clinical records and consultations

Recovery time: minimum 2 years

Patient mistrust of therapist

Spent 6–12 months testing the reliability of the therapist. Psychotherapy: 50% attrition rate

Sakamoto et al. [47]

Hikikomori

(N = 1)

24 years

Omani

1 Male

Clinical data extracted from a patient case

Exhibited the following behaviors: confined self at home, spent the majority of time in own room, did not engage in social relationships, reversed sleep/wake cycle, and refused contact with family

Demographic data of a male from Oman, age 24

Teo [49]

Hikikomori

(N = 1)

30 years

American

1 Male

Clinical data extracted from a patient case

Exhibited the following behaviors: confined self at home, spent the majority of time in own room, and did not engage in social relationships. Poor hygiene practices & urinated/defecated in jars/bottles

Demographic data of a male from the United States, age 30

Suwa et al. [51]

Hikikomori

(N = 1)

25 years

Japanese

1 Male

Clinical data extracted from a patient case

Exhibited the following behaviors: exhausted from effort to maintain relationships, inability to relate well with others, fear in entering adult society, no confidence to cope with society, felt ashamed of himself, and feared others opinions of him being unemployed

Ovejero et al. [45]

Hikikomori

(N = 1)

25 years

Spanish

1 Male

Clinical data extracted from a patient case

Exhibited the following behaviors: confined self at home, spent the majority of time in own room, and did not engage in social relationships

Demographic data of a male from Spain, age 25

Teo et al. [36]

Hikikomori

(N = 36)

18–49 years

American, Japanese, Korean, & Indian

29, 7

LSNS-6 (6-Likert item, scoring 0–30)

 Measuring social connectedness

 Family subscale (3-Likert items; scored 0–15)

 Friends subscale (3-Likert items; scored 0–15)

 Total overall score > 12 = social isolation

UCLA Loneliness Scale (20-Likert items, scoring 20–80)

Higher score = Higher level of loneliness

LSNS-6 Scale score (Mean ± SD):

 Overall score: 9.7 ± 5.7

 Family score: 5.4 ± 3.0

 Friends score: 4.3 ± 3.5

UCLA Loneliness Scale (Mean ± SD): 55.4 ± 10.5

Demographic data on the distribution of males and females, from across four countries; education level: 56.0% hold a bachelor’s degree or above

Ranieri [46]

Hikikomori

(N = 2)

13 years

Italian

2 Females

Clinical data extracted from patient cases

Exhibited the following behaviors: confined self at home, spent the majority of time in own room, did not engage in social relationships, and refused contact with family

Demographic data of two females from Italy, age 13

Psychotherapy recovery time 4 years

Kato et al. [44]

Hikikomori

(N = 1)

39 years

Japanese

1 Male

Clinical data extracted from a patient case

Exhibited the following behaviors: confined self at home, spent the majority of time in own room, did not engage in social relationships, avoided face-to-face contact with others, reversed sleep/wake cycle, and left home once a month to visit an outpatient clinic

Demographic data of a male from Japan, age 39

Matsuguma et al. [77]

Hikikomori

(N = 1)

17 years

Japanese

1 Male

Rosenberg Self-Esteem Scale (10-Likert items; scored 0–30)

 Higher score = higher self-esteem; < 15 = low self-esteem, ≥ 15 = high self-esteem

Subjective Vitality Scale (measures feelings of alertness or being energized, scoring 0–4)

Higher score = higher vitality

 Higher score = higher vitality

Kessler Psychological Distress Scale (measures anxiety and depression, scoring 10–50)

 Lower scores = lower distress levels

Strength-based coaching intervention pre-post scores:

 Rosenberg Self-Esteem Scale: 16 to 25

 Subjective Vitality Scale: 1.8 to 3.4

 Kessler Psychological Distress Scale: 17 to 6

Ranieri [46]

Hikikomori

(N = 1)

Mixed group

19 years

–

1 Male

Clinical data extracted from a patient case

Exhibited the following behaviors: social anxiety, mistrust in parents, inability to approach the opposite gender when interested. Demographic data of a male, age 19

Silic et al. [48]

Hikikomori

(N = 1)

24 years

Croatian

1 Male

Descriptive data extracted only

Exhibited the following behaviors: confined self at home, spent the majority of time in own room, did not engage in social relationships, refused contact with family, and used furniture to block entry to room to avoid contact

Demographic data of a male from Croatia, age 24

Roza et al. [72]

Hikikomori

(N = 1)

35 years

Brazilian

1 Male

Clinical data extracted from a patient case

Exhibited the following behaviors: confined self at home, spent the majority of time playing computer games, and did not engage in social relationships. Neglected self-care and hygiene. Demographic data of a male from Brazil, age 35

Qualitative studies

Ogino [56]

Hikikomori

(N = ?)

26.7 years

Japanese

–

Empirical data from 20 h site contact and qualitative interviewing with hikikomori, and staff and group leaders in support group

Life experiences: motivation to return to society but obstacles they could not overcome, lack a positive sense of identity; feelings of anxiety, unable to do anything, fear of failure, lacking in self-esteem or self-confidence; having lack of qualifications on resumé; difficulties to explaining themselves to people because lack of social identity; use of intervention called Free Space Wood

Kaneko [52]

Hikikomori

(N = 1)

Mid-30’s

Japanese

1 Male

Empirical data of field research and qualitative interviewing

Life experiences of: lack of trust in people

Wong and Ying [54]

Socially Withdrawn Youth

(N = 88)

13–24 years

Chinese

67, 21

Individual and focus group qualitative interviews

Life experiences: intimate relationships online, but no intention of meeting the other person; use of intervention involving social worker engagement, social workers have accompanied socially withdrawn youth to outings to providing sense of security

Recommendations: process of recovery a “yo-yo process” with setbacks of reversal in progress noted; recovery needs to be at pace of youth; a non-linear process; building trust is an important stage in their recovery process; and starting where client is at

Wong [6]

Socially Withdrawn Youth

(N = 252)

13–24 years

Chinese

202, 50

Empirical data from clinical data and literature

Life experiences: face-to-face interactions only with family members, mother had no face-to-face contact with child for months; use intervention of home visiting withdrawn youth, can have sudden refusal from youth to take part in social activities/face-to-face contact during social worker reengagement

Supplementary data: NEET can be considered hikikomori, but some NEET have an active social life

Recommendations: home visiting requires sensitivity and awareness, recognition when client needs privacy, sensitivity to client’s surroundings provides clues in relation to hobbies/interests which can be topics for discussion during reengage of client

Wong [70]

Socially Withdrawn Youth

(N = 30)

15–24 years

Chinese

–

Individual and focus group qualitative interviews

Life experience: sensitive name calling of “withdrawal guys or hidden youth”; self-secluding nature of hikikomori, the majority of care provided in the home; recommendations of “starts where the client is”

Tajan [11]

Recovered Hikikomori

(N = 4)

29–50 years

Japanese

3, 1

One-to-one qualitative interviews

Life experience: positive and inspiring relationships from meeting someone they trusted, receiving encouragement from someone they knew and liked, feeling that no one would care if they died and living like an animal

Rubinstein [71]

Hikikomori

(N = ?)

–

Japanese

–

Empirical data from 50 interviews on site of support groups, hikikomori and mental illness communities with parents, children, hikikomori-related program staff and field observation

Life experience: parents desired to avoid stigma and labeling of mental illness therefore chose to call their child a hikikomori, even after receiving a psychiatric diagnosis; did not claim that the hikikomori experience was pleasant

Yong and Kaneko [57]

Hikikomori

(N = 5)

Proxy respondents

(N = 3)

Online forum participants

(N = 160)

–

One-to-one qualitative interviews and empirical data extraction from internet forum from first person and second person experiences

Life experiences: cautious about establishing relationships over the Internet due to fear and inability to trust people, loss of trust in people, low self-esteem, feelings of hopelessness for future, having non-competitiveness, ineffective communication, lack a positive sense of identity and confidence, negative appraisals from others and self, negative thinking or thoughts of self, fear of social interactions, unable to secure a job, feelings of inadequacy or incompetence, complains society too demanding and unfair, felt unable to do anything, exhibits confinement at home, little social contact

Li and Wong [53]

Socially Withdrawn Youth

(N = 30)

14–29 years

Chinese

22, 8

One-to-one semi-structured qualitative interviews

Life experiences: lifestyle centered on confinement at home, with little social contact, losing touch with the outside world, lacking face-to-face contact with others, having no peer relationships, unable to get along with others, lack of trust in people, experienced a traumatic life event, use the Internet to find positive self-identity and positive and inspiring, having negative feelings or thoughts, boredom, sense of no longer stand staying at home, a loss of interest in computer games, some enjoying their seclusion and feeling freed from restraints and timelines