Global psychiatrists' opinions about hikikomori from biopsychosocial perspectives: International case vignette survey

Abstract

Aim

The aim of this study was to investigate whether psychiatrists around the world believe they might encounter cases of hikikomori (prolonged social isolation), and how they formulate and treat such cases.

Methods

A hikikomori case vignette was sent to psychiatrists of 34 countries around the world. Participants rated for the vignette: frequency of similar cases in one's practicing country; and aspects of formulation, diagnosis, suicide risk, and treatment plan.

Results

In total, 344 complete responses from 34 countries were returned. Eight countries/areas had 10 or more respondents: Japan (61), South Korea (54), Nepal (48), Iran (40), Thailand (32), India (23), Hong Kong (12), and UK (10); the remainder were placed in the “others” group (64). Respondents from all countries except Thailand felt that similar cases were seen. Diverse patterns of response were obtained regarding formulation and treatment. Japan, South Korea, and “others” favored psychosocial aspects in the formulation, while Iran, Nepal, and India favored biological factors. Most respondents felt the case could be treated by an outpatient visit, while others preferred hospitalization. Psychotherapy was rated highly as an intervention; Iran, South Korea, and “others” also rated pharmacotherapy highly.

Conclusion

Despite its limitations as an exploratory study, we found evidence that hikikomori‐like cases might exist around the world. However, opinions on how such cases should be formulated and treated vary significantly among countries. We believe this reflects how the experience of hikikomori is dependent on the related sociocultural context. Further comparative work, preferably with standardized assessment tools, will help to clarify how society might influence the individual experiences of practitioner and hikikomori patients.

INTRODUCTION

Hikikomori is a Japanese word that refers to severe and prolonged social isolation. Hikikomori individuals spend long periods of time withdrawn from society, typically in their own room. ¹ Efforts have been made to propose formal diagnostic criteria: (a) marked social isolation in one's home; (b) duration of continuous social isolation of at least 6 months; and (c) significant functional impairment or distress associated with the social isolation. ² , ³ The condition was first described in 1998 in Japan by psychiatrist Dr Tamaki Saito, ⁴ but in the 2010s became increasingly reported in other countries around the world, such as Spain, ⁵ France, ⁶ USA, ⁷ Hong Kong, ⁸ Taiwan, ⁹ Ukraine, ¹⁰ Oman, ¹¹ Israel, ¹² and Italy. ¹³ However, the lack of consensus around definitions has resulted in these studies being of varying quality. ¹⁴ Efforts have been made to standardize this with hikikomori diagnostic criteria ² and the standardized 25‐item Hikikomori Questionnaire (HQ‐25), ¹⁵ which has now been translated into multiple languages. ¹⁶ , ¹⁷ , ¹⁸

There are complex biopsychosocial factors that are thought to drive the hikikomori experience. Some biological links to inflammation, uric acid levels, and various blood metabolites have been found, ¹⁹ , ²⁰ , ²¹ but a concrete biological explanation has yet to be confirmed. Much more numerous are the variety of cultural factors that have been suggested. Many of these were thought to be culturally specific to Japan (e.g., haji, ijime, futoukou), but such terms do have approximations in English too (“shame,” “bullying,” and “school refusal,” respectively). Much research has also suggested a possible relation to amae, a term reflecting dependence upon the goodwill of someone in a higher position. “Sullenness” or “sullen withdrawal” was hypothesized as one transformation of amae, hence its possible association with hikikomori behavior. ²² Amae does not have a direct translation in English, but has also been found to exist in Western populations. ²³ , ²⁴ This, coupled with the mounting evidence of hikikomori‐like cases outside Japan, has seriously challenged previous research that suggested it to be a culture‐bound syndrome. Sakamoto et al. ¹¹ suggested hikikomori might instead be “culture‐reactive,” in that social isolation could be a culturally mediated reaction to stress and illness. This is similar to how the DSM‐5 TR recognizes it as a “cultural concept of distress.” ²⁵ Furthermore, hikikomori has gained traction in some Asian countries, such as Hong Kong (“hidden youth”) ²⁶ and Korea (“socially withdrawn youth”). ²⁷ However, how exactly such presentations are linked to the local culture, and how these and other such manifestations around the world may differ from Japanese hikikomori, remain unclear. On the other hand, hikikomori has also been called a “modern society‐bound syndrome.” ²⁸ , ²⁹ This views social isolation as simply an unavoidable aspect of a post‐COVID society, facilitated further with the convenience from increasingly accessible Internet technologies that enable one to live entirely from one's own room. While the diagnostic criteria for hikikomori do now specify that the condition must come with “associated impairment or distress,” ² how exactly this distress is quantified, and how this might differ around the world, remain uncertain.

Accordingly, the aim of this study was to investigate attitudes of psychiatrists around the world regarding the concept of prolonged social withdrawal: specifically, if they think such cases exist in their own countries and, if so, how they might formulate and treat such cases. While the study was initially carried out in 2012, we decided the rapidly growing body of research on hikikomori meant it would be instructive to consider the extent to which attitudes by psychiatrists around the world had been influenced by this.

METHODS

Data collection

The participants of the study were psychiatric trainees (residents) and psychiatrists of varying ages in different countries. They were asked to consider a case vignette of hikikomori, and to fill in an anonymous questionnaire regarding the formulation, diagnosis, and other issues related to their understanding. The case vignette and questionnaire were both distributed together with other case vignettes as part of the Research on East Asian Psychotropic Prescription Pattern Study (REAP), which was recently published by Nakagami et al. ³⁰ The questionnaire in Japanese or English was distributed both in online and paper formats using the psychiatrists’ networks, such as the early career psychiatrists’ section of the World Psychiatry Association and the Japan Young Psychiatrists Organization. Using these organizations, we were able to recruit a large number of interested psychiatrists who agreed to help distribute the questionnaire to colleagues in their home countries. The questionnaire was distributed to a convenience sample of psychiatrists, though we specifically requested recruitment of psychiatrists of differing ages and years of experience. All participants were requested to complete the questionnaire within the survey period from 1 June 2018 to 31 October 2019.

Case vignette and questionnaire

The hikikomori case vignette and questionnaire were initially developed and distributed in August 2010. ²⁴ The case vignette (included in Supplementary Information) was developed based on a review of the literature and expert opinion. Hikikomori experts in Japan suggest a focus on prolonged cases with problematic behavior. In addition, the hikikomori state is diagnostically complex and can mimic other psychiatric disorders, such as depression or the early stages of a psychotic illness. Despite this, no firm diagnostic test for hikikomori has been created, and the utility of typical psychiatric screening questionnaires is often debated. ³¹ These features have all been incorporated into the case vignette. While the initial survey included another case vignette featuring a younger hikikomori patient, it was not included in this survey, as evidence suggests an increase in the age of the hikikomori. ³² A complete mental status examination and follow‐up data were not described for the case in order to stimulate the imagination of participants. The questionnaire was self‐administered, and participants evaluated the following on 5‐point Likert scales after reading the case vignette: frequency of similar cases in one's country; and various aspects of formulation, diagnosis, suicide risk, and treatment plan. Demographic information about the participants, including their experience and length of training in psychiatry, was also recorded. The case vignette and questionnaire were first developed in Japanese, then translated into English using forward‐backwards translation.

Statistical analysis

Data from 34 countries were returned. Eight countries each had 10 or more respondents; these were Japan, South Korea, Nepal, Iran, Thailand, India, Hong Kong, and the UK. The remainder of the participants were grouped together in a separate “others” group. Differences in scores among these nine groups were analyzed by a Kruskal–Wallis one‐way analysis of variance (ANOVA). Significance values were adjusted by the Bonferroni correction for multiple tests. SPSS Version 29.0.1 was used to perform the statistical analysis. Where data were nominal (for example, when asked if the case could be applied to be diagnostic manuals), a chi‐square test was performed. Significance was set at a level of p < 0.05.

Approval of the study

This study was performed in accordance with the Declaration of Helsinki and was approved by Kyushu University (IRB approval number: 30‐251). All subjects were informed that participation was completely voluntary and that the return of the anonymous questionnaire implied consent.

RESULTS

Participants (Table 1)

Table 1.

Demographic information of participants.

Country of current clinical practice	Total	Japan	South Korea	Nepal	Iran	Thailand	India	Hong Kong	UK	Others
Sample size (n)	344	61	54	48	40	32	23	12	10	64
Age (years: mean ± SD)	36.7 ± 8.7	37.5 ± 5.9	34.3 ± 5.9	32.2 ± 4.9	37.7 ± 6.2	32.3 ± 3.2	33.9 ± 4.8	49.2 ± 10.6	40.0 ± 11.3	41.1 ± 10.6
Male (n, %)	207 (60.1)	50 (82.0)	44 (8.15)	31 (64.6)	12 (30.0)	6 (18.8)	18 (78.3)	6 (50.0)	7 (70.0)	33 (53.2)
Were trained in multiple countries (n, %)	35 (10.2)	12 (19.7)	0	7 (14.6)	1 (2.5)	0	2 (8.7)	3 (25.0)	0	10 (16.1)
Experience in psychiatry (years: mean ± SD)	8.4 ± 7.8	8.9 ± 9.0	6.3 ± 5.7	4.8 ± 4.0	8.0 ± 6.0	5.8 ± 2.7	6.9 ± 4.1	23.5 ± 10.6	11.3 ± 9.8	11.3 ± 8.8

Abbreviation: SD, standard deviation.

There were initially 459 responses returned. Responses with less than half of the hikikomori questionnaire filled out, or in which less than half of the responses were interpretable, were discarded. This left a remainder of 344 respondents (response rate 74.9%). In total, 34 countries were represented in this sample (Table 1). The six countries with the most participants were Japan, South Korea, Nepal, Iran, Thailand, and India (n = 61, 54, 48, 40, 32, and 23, respectively). Hong Kong (12) and the United Kingdom (10) also had 10 or more participants. “Other” countries included Sri Lanka (9), Indonesia (7), Philippines (6), Taiwan (6), Malaysia (3), Nigeria (2), Australia (2), Switzerland (2), Myanmar, Maldives, Tunisia, Turkey, Mexico, Pakistan, Croatia, Italy, Spain, Singapore, Ukraine, Poland, Portugal, Austria, Germany, Ireland, USA, and New Zealand (1 each). Forty psychiatrists (11.8%) were trained in multiple countries; this included short courses delivered in other countries. Of these, only six psychiatrists were working in a different country from their nationality. The average age of all participants was 36.7 years (SD = 8.6 years). A total of 207 participants (60.1%) were male.

Frequency of similar cases (Table 2)

Table 2.

Frequency, setting, and formulation of case.

	Total	Japan	South Korea	Nepal	Iran	Thailand	India	Hong Kong	UK	Others	Significance (ANOVA)
Do you think this kind of case is common or rare in your country? (urban or rural area)?
Urban area	Mean	3.07	3.34	3.04	3.13	2.83	2.56	3.39	2.67	2.80	3.19	JP, IN > TH
	SD	0.931	0.911	0.846	0.981	0.931	0.948	0.988	0.888	1.033	0.807
Rural area	Mean	2.49	2.93	2.42	2.34	2.33	2.00	2.43	2.25	2.30	2.69	JP > TH
	SD	1.006	0.962	0.887	0.915	1.060	0.950	1.080	0.866	1.059	1.049
Do you see this kind of case in your clinical practice?
	Mean	2.73	2.89	2.49	2.96	2.63	1.97	2.96	2.42	2.70	3.03	JP, NP, IN, OT > TH
	SD	1.071	0.877	0.933	1.110	1.170	0.897	1.224	1.165	0.949	1.086
To what extent do you think he is likely to be at risk of committing suicide?
	Mean	2.86	2.88	3.02	2.89	2.49	2.81	3.00	2.91	2.20	3.00	NS
	SD	0.983	0.804	0.858	1.088	0.970	0.965	1.243	0.831	0.789	1.071
To what extent do you think he has been affected by:
Mind (psychological factors)	Mean	3.61	3.90	3.67	3.08	3.10	3.78	3.52	3.50	3.90	3.93	JP, OT > NP
	SD	1.059	0.961	1.009	1.108	1.194	0.941	1.275	0.674	0.568	0.910	JP, OT > IR
Brain (biological factors)	Mean	3.25	3.15	3.37	2.98	3.49	3.16	3.70	3.08	3.00	3.28	NS
	SD	1.079	0.872	0.951	1.246	1.121	1.019	1.302	0.900	0.667	1.241
Social factors	Mean	3.65	4.15	3.65	3.23	2.90	3.97	3.35	3.58	4.10	3.87	TH, OT, JP > IR
	SD	1.049	0.872	0.850	1.096	1.150	0.861	1.071	0.996	0.738	1.016	JP > NP, IN
Cultural factors	Mean	3.26	3.90	3.44	2.85	2.43	3.22	2.74	2.75	3.90	3.55	JP, KR, UK, OT > IR
	SD	1.134	0.870	0.984	1.072	1.130	1.128	1.054	0.866	0.876	1.126	JP > IN, HK, NP
Family relationship (mother)	Mean	2.91	3.39	3.19	2.46	2.50	3.06	2.09	2.75	2.70	3.08	JP, KR, OT > IN
	SD	1.106	0.918	0.992	1.148	0.987	1.216	0.949	1.138	1.059	1.076	KR, JP > NP
												JP > IR
Family relationship (father)	Mean	2.82	3.39	3.11	2.33	2.30	2.97	2.00	2.92	2.70	2.97	JP, KR, OT > IN
	SD	1.080	0.881	0.945	1.078	1.043	1.150	0.905	1.165	1.059	0.975	JP, KR > IR, JP, KR > NP
School environment	Mean	3.21	3.18	3.43	3.23	2.55	3.41	2.87	3.08	3.90	3.42	KR. UK, OT > IR
	SD	1.082	0.975	0.838	1.153	1.176	1.012	1.217	0.900	0.568	1.153
Economic environment	Mean	2.71	3.21	2.69	2.52	2.23	2.47	2.35	3.00	3.40	2.79	JP > IR, NP
	SD	1.086	1.051	0.948	1.110	1.074	0.879	1.027	0.853	1.174	1.147
Tendency of psychosis	Mean	3.24	2.84	3.08	3.32	3.35	3.28	4.13	3.33	3.20	3.27	IN > JP, KR, OT
	SD	1.057	0.986	1.045	1.086	1.099	0.813	1.180	0.888	0.789	1.043
Tendency of mood disorders	Mean	3.05	2.62	3.52	2.60	2.62	3.59	2.48	3.50	3.50	3.44	KR, TH > IN
												OT > JP, NP, IR
	SD	1.069	0.986	0.746	0.948	1.091	0.946	1.163	1.000	0.527	1.081	KR, TH > IR
												TH > NP
Tendency of personality disorders	Mean	3.41	3.20	3.48	3.62	3.08	3.50	3.35	3.42	3.10	3.63	NS
	SD	1.116	0.891	0.966	1.134	1.222	1.218	1.229	1.311	0.994	1.204

Note: Comparisons deemed significant if Bonferroni‐adjusted p < 0.05.

Abbreviations: IN, India; IR, Iran; JP, Japan; KR, South Korea; NP, Nepal; NS, not significant; OT, “other” countries; SD, standard deviation; TH, Thailand.

Participants were first asked their opinion on how common similar cases were in urban or rural areas in their respective countries, as well as if they personally saw such cases in clinical practice. Overall, most participants felt that similar cases were seen in their practicing country, and that they were more frequently seen in urban areas, particularly in Japan (Table 2). Participants from India also reported such cases to be more common in urban areas. However, participants from Thailand felt such cases to be comparatively rare, both in everyday clinical practice, as well as in urban and rural settings.

Suicide risk (Table 2)

All participants felt there was a moderate likelihood for the case to be at risk of suicide. There were no significant differences in perceived suicide risk of the case among countries.

Causes and formulation (Table 2)

Participants were asked to what extent they felt that the hikikomori case was affected by a range of factors. Psychological factors were felt to be important in this regard, and typically more important than biological factors in most countries. Social and cultural factors were similarly felt to be important, particularly in Japan and the “others” group. Thai participants ranked social factors as more important, while South Korean participants emphasized cultural factors. Family relationships with both the mother and father were felt to be important in Japan, South Korea, and the “others” group. Japanese participants also felt the economic environment to be more important, but the school environment less so; while participants from the UK and the “others” group expressed the opposite opinion. This was in contrast to participants from Iran and Nepal, who disagreed that the case was less affected by psychological factors, or indeed any of the listed sociocultural factors. Participants from India expressed similar views, although the higher standard deviations for the relative contribution to the school or economic environment might reflect disagreements locally on these points. Participants from Japan, South Korea, and the “others” group placed less emphasis on tendency of psychosis, though Japanese participants disagreed that tendency of mood disorders was important. This was in agreement with Indian, Nepali, and Iranian participants on this point. Nepali and Iranian participants did not feel that tendency of psychosis was important. Interestingly, all participants felt both biological factors and tendency towards personality disorders affected the case to some extent, with no significant differences in scores between countries on this point.

Diagnosis and applicability to ICD‐10 or DSM‐5 criteria (Table 3)

Table 3.

Diagtic applicability of ICD‐10 and DSM‐5.

	Total	Japan	South Korea	Nepal	Iran	Thailand	India	Hong Kong	UK	Others
Do you think this case can be applied to ICD‐10 criteria or not?
Yes	n (%)	188 (55.0)	22 (36.1)	23 (42.6)	38 (79.2)	16 (40.0)	13 (40.6)	21 (91.3)	8 (66.7)	8 (80)	39 (62.9)
No	n (%)	112 (32.7)	38 (62.3)	24 (44.4)	5 (10.4)	6 (15.0)	17 (53.1)	1 (4.3)	3 (25.0)	2 (20)	16 (25.8)
No response	n (%)	42 (12.3)	1 (1.6)	7 (13.0)	5 (10.4)	18 (45.0)	2 (6.3)	1 (4.3)	1 (8.3)	0 (0.0)	7 (11.3)
Chi‐square test		19.13**	4.27*	0.02	25.3**	4.54*	0.53	18.18**	2.27	3.60	9.62**
* if adjusted p < 0.05
** if adjusted p < 0.01
Overall result		Yes	No	NS	Yes	Yes	NS	Yes	NS	NS	Yes
Do you think this case can be applied to DSM‐5 criteria or not?
Yes	n (%)	147 (43.0)	15 (24.6)	20 (37.0)	16 (33.3)	18 (45.0)	16 (50.0)	14 (60.9)	6 (50.0)	7 (70.0)	35 (56.5)
No	n (%)	137 (40.1)	42 (68.9)	25 (46.3)	8 (16.7)	12 (30.0)	16 (50.0)	5 (21.7)	5 (41.7)	3 (30.0)	21 (33.9)
No response	n (%)	58 (17.0)	4 (6.6)	9 (16.7)	24 (50.0)	10 (25.0)	0 (0)	4 (17.4)	1 (8.3)	0 (0)	6 (9.7)
Chi‐square test		0.35 (0.55)	12.79**	0.56	2.67	1.20	0.00	4.26**	0.09	1.60	3.50**
* if adjusted p < 0.05
** if adjusted p < 0.01
Overall Result		NS	No	NS	NS	NS	NS	Yes	NS	NS

Note: p values adjusted by the Bonferroni correction for multiple tests.

Abbreviation: NS, not significant.

There was considerable disagreement on whether diagnostic manuals were applicable to the case. Only a small overall majority of participants felt ICD‐10 criteria were applicable (55% yes; 32.7% no). The findings regarding applicability to DSM‐5 criteria were even more equivocal (43% yes; 40.1% no). Both cases also included a large number of blank responses. Interestingly, participants from Japan felt neither ICD‐10 nor DSM‐5 were applicable; whereas participants from India and the “others” group thought they both were.

When asked about details of their diagnoses with respect to the diagnostic manuals, schizophrenia spectrum disorders were mentioned most frequently (87 participants). These included, for example, simple/latent/prodromal schizophrenia, schizotypal disorder, schizoaffective disorder, or non‐specified psychotic disorder. Personality disorders were next most common (57 participants), including schizoid personality disorder. Twenty‐five participants listed mood disorders, including depressive disorder, dysthymia, or cyclothymia. Eight participants listed anxiety disorders, most commonly social anxiety. Eight participants listed neurodevelopmental disorders, most commonly autism spectrum disorder, although learning difficulties and intellectual disabilities were also suggested.

When offered the chance to describe the diagnosis they would make for the patient, 11 Japanese participants made reference to hikikomori (18.03%), as compared to nine participants not from Japan (3.18%). Other diagnoses mentioned included gaming disorder or internet addiction, as well as Hwa‐byung, a Korean anger syndrome considered to be culture‐related. ³³

Treatment setting (Figure 1)

Figure 1.

Percentage of responses to: “Where should the case be treated?”.

Almost all participants felt the case needed treatment. The most common response on treatment setting in all countries (except Hong Kong) was “outpatient visit.” The next most common response was hospitalization (Figure 1). The popularity for this varied remarkably between countries (from 5% in Japan to 36.4% in India). Some countries also had a substantial proportion of “other” responses. These included day hospitals, home visitation programs, and residential placements that did not have a specific medical focus (e.g., rehabilitation centers). One respondent each from Japan (1.64%) and South Korea (1.85%) selected, “no treatment indicated.” Several responses, particularly from Japan, also highlighted the uncertainty of the situation, and how this made it difficult to decide on a treatment plan.

Interventions (Table 4)

Table 4.

Interventions in case.

		Total	Japan	South Korea	Nepal	Iran	Thailand	India	Hong Kong	UK	Others	Significance (ANOVA)
Regardless of whether you do or not, to what extent do you think the following interventions will be effective?
Psychotherapy	Mean	3.72	3.95	3.75	4.17	3.10	3.41	3.36	3.17	2.90	4.05	NS
	SD	1.196	1.048	1.017	0.975	1.619	1.188	1.293	1.030	1.101	1.047
Pharmacotherapy	Mean	3.16	2.53	3.55	2.98	3.54	3.06	3.45	2.67	2.40	3.52	KR, IR, OT > JP
	SD	1.216	1.081	0.889	1.263	1.315	1.105	1.371	1.231	1.075	1.184
Non‐pharmacologic biological treatment	Mean	1.99	1.87	2.40	1.68	1.77	2.09	2.05	1.42	1.70	2.19	KR > NP, IR
	SD	1.127	1.016	1.098	1.181	1.202	1.058	1.090	0.669	0.823	1.212
Environmental intervention	Mean	3.89	4.13	3.72	3.65	3.53	4.09	3.45	3.83	3.60	4.32	JP, OT > IR OT > NP, KR, IN
	SD	0.971	0.873	0.769	1.041	0.922	0.928	1.224	1.337	0.699	0.864
Alternative therapies	Mean	2.26	2.47	2.09	2.74	1.56	2.03	2.14	2.50	2.30	2.39	OT, JP, NP > IR
	SD	1.130	1.033	1.005	1.188	0.754	1.150	1.037	1.168	0.823	1.310
Self‐help	Mean	3.47	3.95	3.45	3.48	3.26	3.22	2.91	3.08	3.00	3.65	JP > IN
	SD	1.176	0.999	0.867	1.130	1.141	1.362	1.342	1.165	0.816	1.356
How likely are you to choose the following interventions for him in your daily clinical setting:
Psychotherapy	Mean	3.72	3.90	3.79	4.13	3.18	3.41	3.05	3.17	3.50	4.05	NS
	SD	1.284	1.145	1.063	1.055	1.604	1.292	1.558	1.337	1.269	1.193
Pharmacotherapy	Mean	3.44	2.62	3.98	3.27	3.69	3.28	3.91	2.83	3.30	3.81	IR, KR, IN, OT > JP
	SD	1.274	1.180	0.747	1.300	1.341	1.170	1.377	1.403	1.252	1.226
Non‐pharmacologic biological treatment	Mean	1.85	1.62	2.15	1.81	1.62	1.91	1.77	1.42	1.80	2.08	NS
	SD	1.113	0.865	1.026	1.283	1.091	1.058	1.020	0.793	0.919	1.334
Environmental intervention	Mean	3.72	4.00	3.26	3.38	3.47	4.00	3.36	3.58	3.90	4.21	JP, OT > KR OT > IR, NP
	SD	1.106	0.921	1.095	1.074	1.224	0.984	1.364	1.240	0.568	0.960
Alternative therapies	Mean	2.17	2.07	2.00	2.64	1.64	1.94	2.09	2.42	2.50	2.42	NP > IR
	SD	1.201	1.006	0.981	1.276	0.903	1.190	1.192	1.505	1.354	1.421
Self‐help	Mean	3.52	4.03	3.32	3.47	3.44	3.22	3.05	3.17	3.80	3.63	JP > IN, KR
	SD	1.180	1.008	0.976	1.177	1.209	1.338	1.327	1.403	0.632	1.231

Note: Comparisons deemed significant if Bonferroni‐adjusted p < 0.05.

Abbreviations: IN, India; IR, Iran; JP, Japan; KR, South Korea; NP, Nepal; NS, not significant; OT, “other” countries; SD, standard deviation.

Participants were first asked what modalities of intervention were felt to be most effective for the case in principle. All countries ranked psychotherapy as being highly effective. Many countries also ranked environmental intervention highly. However, participants from South Korea, Iran, and the “other” group had a preference for pharmacotherapy and disfavored environmental interventions. Participants from Nepal and the “other” group expressed a preference for alternative therapies; while Iranian participants ranked alternative therapies lower.

Next, participants were asked which interventions they would choose for the case in their clinical practice. The aim of this question was to get participants to consider whether there might be interventions they would think effective in principle, but that would be unable to implement in reality (e.g., due to unavailability of services). Psychotherapy and environmental interventions again scored highly across all countries. When compared with their answers for the first part of the question, mean psychotherapy scores were relatively similar, while mean scores for environmental intervention were slightly reduced for all countries except in the UK. For biological treatments, participants from India now joined Iran, South Korea and the “other” group to express a preference for pharmacotherapy, while non‐pharmacologic biological treatment was ranked relatively lower in all countries. Alternative therapies were also ranked lower in all countries. Participants in Nepal expressed a relative preference for this, but even so it still ranked second lowest of all modalities (after non‐pharmacologic biological treatment).

DISCUSSION

This international study explores the opinions of psychiatrists around the world about the hikikomori phenomenon in Japan and other countries. It utilizes a case vignette previously developed in 2012. We were able to obtain data from many other countries on repeating the survey. Importantly, several countries that had not previously published case descriptions about hikikomori were also included: namely Nepal, India, and the UK. Thus, the current study and dataset were felt to potentially hold even further insights into how psychiatrists from different backgrounds formulate hikikomori differently. Furthermore, to our knowledge, there have not been other efforts to survey the global landscape of psychiatrists’ attitudes towards hikikomori. It is thus unclear whether there have been notable changes in this over the years. This is despite hikikomori research having advanced significantly since. It is hoped that this study could serve as a starting point in clarifying whether global attitudes towards hikikomori could be changing as a result.

Participants from all countries except Thailand agreed with the suggestion that hikikomori was seen in their clinical practice. One possible explanation may be that certain countries or cultures are less susceptible to the development of hikikomori. Alternatively, hikikomori‐like behavior may in such settings be seen as non‐pathological, or even a variant of normal behavior, and thus not warrant clinical attention.

Our survey found that participants from most countries appeared to favor psychological, social, cultural, and environmental factors in the formulation of hikikomori. However, participants from Iran, Nepal, and India favored biological factors. This was also reflected in their rating of the case's tendency towards psychosis as higher than mood or personality disorders. Research on hikikomori has found evidence to support both of these formulations to be valid. For example, sociological narratives of hikikomori conceptualize it as a life strategy ³⁴ that may be influenced more broadly by family and society. ³⁵ Biological correlates of hikikomori exist, ²⁰ but the field is less well‐researched and is typically framed in the context of depressive disorder, as opposed to the broad differential diagnosis that hikikomori cases include. It is likely that all such narratives play a part in the development of hikikomori; thus, a comprehensive biopsychosocial formulation can be of value in clinical assessment. ²² It is noteworthy how each country appeared to express a preference for specific formulations, as this may perhaps reveal individual clinicians' own implicit bias for specific sections of the biopsychosocial model—something that can occur even with those trying to adhere to it. ³⁶ , ³⁷

There was also little consensus among participants about the diagnosis of the case. Psychotic disorder and personality disorders were the most popular diagnoses offered; however, a range of other alternatives were also offered, including mood, anxiety, and neurodevelopmental disorders. While the Japanese Ministry of Health, Labour, and Welfare has acknowledged schizophrenia as one possible differential diagnosis for hikikomori, ¹⁴ the full range of other diagnoses should also be considered in a full assessment (see Kato et al. ²² for a full review). Indeed there is evidence to suggest that many hikikomori do also fulfil diagnostic criteria for other mental health disorders ¹⁴ ; however, it remains unclear whether this represents hikikomori as being a discrete disorder that has high comorbidity with known mental health disorders, or simply a late diagnosis of a known disorder. Also worth noting is the existence of a small but significant proportion of hikikomori that manifests independently of known psychiatric disorders. ⁵ , ²² Interestingly, patients experiencing a known psychiatric disorder appear to experience more severe illness and comorbidity if there is concurrent hikikomori: there is evidence for this for autism spectrum disorder, ³⁸ as well as depressive disorder in both adolescents ⁶ and adults. ³⁹ This suggests that even if hikikomori does have roots in known psychiatric disorders, there are aspects of its symptom experience that extend beyond commonalities with existing diagnoses—perhaps a syndromic manifestation of pathology, worth investigating in its own right. ³¹

Of note, 18.03% of Japanese participants identified the case as hikikomori, as compared to 3.18% of non‐Japanese psychiatrists. This could reflect how hikikomori‐like phenomena have been present in Japan for a long time, and hence more commonly known to professionals. ²² , ²⁴ Interestingly, a study of tweets (social media posts) about hikikomori found that less than 1% of Japanese tweets were related to research, epidemiology, or therapeutic aspects of hikikomori, as opposed to 42.2% of Western language tweets. ⁴⁰ This suggests how hikikomori discourse in Western society remains mainly a topic of academic interest and limited to those who know the term, whereas in Japan the concept has captured the attention of the lay public. Much of the Japanese discourse (both academic and non‐academic) also alludes to how it could be societally driven, which could represent an interesting parallel to how Japanese participants favored such accounts when formulating our case vignette. For example, accounts of hikikomori stories that have been published in Japanese newspapers often make reference to significant past events in the family and education. ⁴¹ In addition, many Japanese cultural concepts, such as amae (a form of emotional dependence on others), ijime (severe bullying) and ibasho (a place that gives one a sense of belonging) have also been linked to hikikomori in Japanese discourse; however, recent research suggests that they may not be uniquely Japanese. ¹⁴ , ⁴⁰ These concepts could represent an interesting parallel to hikikomori that was initially thought of as unique to Japan, but has become more universal than initially expected due to globalization and the influences of a modern society. ²⁸ And yet, the under‐recognition of the hikikomori concept may also contribute to an underdiagnosis of hikikomori in the West. It is worth highlighting how there are no systematic studies examining the Western hikikomori's subjective personal experience, and few studies of Japanese hikikomori (for an example, see Kaneko ⁴¹ ). A comparative analysis of cultural influences of social withdrawal between the two groups would be interesting and instructive.

Our survey found many participants were uncertain on how applicable the case was to DSM‐5 or ICD‐10 criteria. Hikikomori does not appear in the ICD‐10, or indeed in the ICD‐11. In the DSM‐5 TR ²⁵ and proposed diagnostic criteria, ²² hikikomori could occur in the context of an established psychiatric diagnosis (“secondary”) or manifest independently (“primary”). This reflects how studies of hikikomori presentations often find some that do not fulfil criteria for existing psychiatric diagnoses, though typically only less than 1% of studied cases fall into this category. ⁵ , ⁴²

Diagnosing hikikomori is also made difficult by the changing sociocultural landscape, and the possibility that this may introduce novel psychopathology and syndromes. ²⁸ As a current example of this, advances in Internet technology have reduced the need for people to go out and interact directly, and these changes could potentially have been further embedded into society following lockdown measures implemented to control the COVID‐19 pandemic. ⁴³ , ⁴⁴ Some speculate this could increase hikikomori‐like behavior across society more generally. ⁴⁵ , ⁴⁶ , ⁴⁷ , ⁴⁸ There is evidence that hikikomori symptoms are associated with Internet and smartphone addiction in Japanese students, ⁴⁹ Internet gaming disorder symptoms in US and Australian young adult players of massively multiplayer online games, ⁵⁰ and gaming disorder tendency in Japanese nonworking adults. ⁵¹ Internet gaming disorder is a new addition to the DSM‐5 TR in the section recommending conditions for further research, ²⁵ in contrast to the ICD‐11, which introduced gaming disorder as a new diagnosis. However, the concept of pathological gaming requires more research, for example into the exact boundaries between pathological and non‐pathological gaming, and whether game genre might affect the diagnosis. ⁵² , ⁵³ Similar questions could be raised regarding the issue of hikikomori and technology use—for example, would certain game genres have a stronger association with hikikomori symptoms? What level of technology use should be considered “pathological” and hence a hikikomori risk that warrants intervention? It will be important for health professionals to be aware of not just the broad range of differential diagnoses in hikikomori presentations, but also how these might be influenced by changes in the wider environment.

Indeed, the diagnostic process in mental health is inherently subjective, and practitioners must be aware of how this could be affected by their own unconscious biases. Under‐recognition by clinicians of atypical symptoms of depressive illness, such as irritable mood, increased hostility, and agitation, can make it less likely to be diagnosed than more typical presentations. It is significant that this occurs more in minority ethnic groups, such as African Americans. ⁵⁴ , ⁵⁵ There could be a parallel with hikikomori, which has been conceptualized as a cultural expression of distress, ²⁵ with extreme social isolation being a manifestation of typical depressive symptoms, such as anhedonia, anergia, and lack of motivation to engage in social activity. Hwa‐byung, a Korean‐culture‐related anger disorder that was suggested as a potential differential diagnosis by one South Korean participant, is another such example. Hwa‐byung involves naturally isolated groups of society that face stigma associated with help‐seeking for mental illness or psychological distress. ⁵⁶ Stigma has also been a barrier to help‐seeking for minority ethnic groups in the UK, ⁵⁷ and could be similar to the haji (shame) experienced in Japanese culture that is associated with the hikikomori state. ²² Clinicians thus need to be robust and thorough in the assessment of hikikomori presentations. More broadly, comparative studies between hikikomori and other culturally related manifestations of distress could clarify whether or not they are diagnostically similar entities. Standardized interviews and self‐administered assessment tools for diagnosis and evaluation ³ , ¹⁵ , ⁵⁸ would be helpful in both these instances.

We suggest that treatment decisions could be influenced by two factors. First, the cultural subjectivity that influenced differences in formulation could have led to different treatments. In our survey, countries who favored biological factors appeared also to favor their corresponding treatments, such as pharmacotherapy. Of concern is how underdiagnoses or misdiagnoses of hikikomori may then represent the initiation of an inappropriate treatment, which could lead to poorer patient outcomes. One example could be if a socially withdrawn patient with a depressive illness is misdiagnosed as having latent schizophrenia and started on antipsychotics as a first‐line treatment, unnecessarily exposing them to antipsychotic‐related side‐effects. This highlights again the need for robust processes in assessment and diagnosis of hikikomori. Second, treatment decisions could also reflect differences in treatment availability. For example, the Korean National Youth Commission has supported efforts to increase home visitation programs for people with hikikomori symptoms referred from community mental health centers and who received individualized psychotherapeutic support. ²⁷ Other countries, such as the UK, emphasize the social prescribing approach—a healthcare model that relies on partnerships with local communities. ⁵⁹ The different treatments available in each country could reflect the different manifestations of social isolation, and the importance of an approach that is tailored to the needs of individual cultures and the health systems of each country. The current evidence base for hikikomori treatments suggests a family‐oriented approach may be effective ¹¹ , ⁶⁰ ; however, these are limited to case reports and small‐sample studies. Thus, there remains no firm evidence base to recommend specific treatment modalities. ⁶¹

Limitations

This survey was intended as an exploratory study and has several limitations. First, respondents in each country were recruited by convenience sampling, and comprised mostly of young psychiatrists, hence they may not be a representative sample. Data were also collected from a limited number of sites in each country; thus, responses might not reflect the full range of diagnostic and therapeutic approaches that different institutions in the same country might provide. Detailed characteristics of the participants' experience and theoretical orientations were not obtained, thus making this difficult to comment on through this study. Second, respondents included only psychiatrists. Opinions of the wider multidisciplinary team, such as psychologists and social workers, which in some countries play a large role in managing social isolation, were thus not captured. Third, the study was unable to capture variations and nuances in response patterns beyond the current statistical analyses, as individual variations in response pattern and style were not analyzed. Thus, only general trends in response patterns between countries could be observed, and individual variations could not be taken into account. Fourth, terms used in the survey, such as “urban/rural,” “psychotherapy,” “inpatient,” were not defined specifically. While this did allow for cultural variations between countries in how each term is interpreted (e.g., the urban/rural contrast would differ between countries), this could have led to inconsistent use of the term across countries (e.g., participants might not have similar “urban” settings in mind). Future surveys could consider making the vignette more detailed. Allowing for different options (e.g., different modalities in psychotherapy, or different classes of pharmacotherapy) may also allow for more granularity in the results.

CONCLUSION

This study suggests that psychiatrists from a large range of countries identify with aspects of the hikikomori syndrome, adding to a growing body of evidence that it is no longer “culture‐bound.” However, opinions on how hikikomori should be both formulated and treated varied significantly between countries. We believe these differences reflect its nature as a socially and culturally mediated condition—sensitive to the context in which it occurs. It is crucial for practitioners to be aware of how such cultural influences might affect their diagnostic process and clinical practice, at worst resulting in missed diagnoses and initiation of inappropriate treatment. Comparative studies to understand individuals' experiences across cultures are also needed; preferably aided by standardized assessment tools that have been appropriately adapted for use in each cultural context. This will help to clarify the existence of hikikomori globally and lead to a deeper understanding of the interactions between mental illness, sociocultural factors, and social isolation.

AUTHOR CONTRIBUTIONS

Takahiro A. Kato: conceptualization of study; design of case vignette and REAP‐CV survey questionnaire; data interpretation. Marcus P. J. Tan: data analysis and interpretation, writing of manuscript. All authors are members of the REAP‐CV survey committee who assisted with finalization of REAP‐CV survey materials, and helped with data collection in their respective countries.

CONFLICT OF INTEREST STATEMENT

Shigenobu Kanba is the Editor‐in‐Chief Emeritus of Psychiatry and Clinical Neurosciences and a co‐author of this article. Toshiya Murai and Tomohiro Nakao are editorial board members of Psychiatry and Clinical Neurosciences and co‐authors of this article. Pichet Udomratn and Norman Sartorius are International Advisory board members of Psychiatry and Clinical Neurosciences and co‐authors of this article. To minimize bias, they were excluded from all editorial decision‐making related to the acceptance of this article for publication. The remaining authors declare no conflicts of interest.

ETHICS APPROVAL STATEMENT

This study was performed in accordance with the Declaration of Helsinki and was approved by Kyushu University (IRB approval number: 30‐251).

PATIENT CONSENT STATEMENT

All subjects were informed that participation was completely voluntary and that return of the anonymous questionnaire implied consent.

CLINICAL TRIAL REGISTRATION

N/A.

Supporting information

Supplementary Information.

Tan MPJ, Hayakawa K, Nakagami Y, Pereira‐Sanchez V, Park SC, Park YC, et al. Global psychiatrists' opinions about hikikomori from biopsychosocial perspectives: International case vignette survey. Psychiatry Clin Neurosci Rep. 2025;4:e70120. 10.1002/pcn5.70120

DATA AVAILABILITY STATEMENT

Data are available from the authors on request.