Background
Although the majority of adolescent populations reside in low and middle-income countries (LMICs), little research has focused on the identification, prevention and treatment of serious and socially disabling mental health problems in these countries. A recent priority-setting exercise for global child and adolescent mental health research [
1] highlighted the dearth of evidence on early intervention for psychosis in LMICs, with only one identified trial from China and few cross-cultural validations of screening tools. Psychosis tends to first occur during adolescence and is a leading worldwide cause of disability; with social disability often observed before, during and after the first psychotic episode [
2‐
5]. The first episode of psychosis – and the preceding ‘prodromal’ period – represent key opportunities for early intervention [
6‐
10]. The provision of evidence-based early intervention services globally is very variable, however, and standard care for psychosis rarely meets the minimum standards suggested by the World Health Organisation’s (WHO) Mental Health Gap Action Programme [
11]. Access to care in LMICs typically lags far behind the first onset of symptoms [
12,
13], which increases risk for poor long-term prognosis [
14]. Thus, identifying and intervening early for young people who are at risk for serious, socially disabling mental health problems – and especially transition to psychosis – remain critical yet largely neglected challenges in LMICs.
In Malaysia, access to mental health care is particularly limited. Malaysia is a Southeast Asian country of 32 million people. The majority ethnic group is Bumiputra (68%), comprising a majority of Malays and a minority of other indigenous people [
15]. The other major ethnic groups are Chinese (23%) and Indian (7%) [
15]. Malaysia is a Muslim-majority country but many people identify as Christian, Buddhist, Hindu, Taoist, Sikh and other minority religions [
16]. Epidemiological estimates suggest that mental health problems in Malaysia have more than doubled over the last 20 years and now affect at least 30 to 40% of the adult population [
17,
18]. Young people aged 16 to 24 years are particularly at risk of developing mental health problems [
18]; the estimated prevalence of youth mental health problems in Malaysia exceeds the average worldwide prevalence [
19] and may be increasing [
20]. Amongst young Malaysians, people from low-income and/or indigenous backgrounds show increased vulnerability to mental health problems [
18,
20]. The mean average Duration of Untreated Psychosis (DUP) in Malaysia is over 3 years, which has significant negative implications for prognosis [
13].
Where available, Malaysian mental health services are largely based on Western models of psychiatry and clinical psychology [
21]. There is evidence that Western models may have broad application, with positive impacts evidenced in South East Asia and Malaysia specifically [
21‐
23]. Nevertheless, the universality of Western approaches remains largely untested in the local context [
21,
24]. The relative importance of communality and collectivism in the Southeast Asian cultures [
24] may also complicate the ‘fit’ of Western approaches which foreground individual support and self-enhancement [
25,
26]. Moreover, the cultural validity of Western approaches in serious mental health problems is further complicated by differences in understandings of unusual experiences or psychosis [
24,
27] and significant heterogeneity of health belief systems amongst different ethnic groups in Malaysia [
27]. Thus, whilst Western approaches may benefit the development of psychological interventions in Malaysia, exploring the cultural validity of such approaches prior to and during implementation is essential. For example, this may reveal potential clashes of culturally determined values with imported therapeutic models and practices and could suggest scope for adaptation or optimisation through integrating Western approaches and Eastern philosophies [
28,
29] — or else highlight a need for ‘bottom-up’ approaches grounded in the local setting [
21].
A social recovery approach may have particular utility in Malaysia and other global mental health settings, where the social dimensions of serious mental health problems may be particularly poorly served. People accessing community-based rehabilitation services report extremely limited social support [
30] and have highlighted their needs for interventions focused on increasing self-agency, social connections, social support and around increasing contact with and acceptance from the broader community [
31]. Moreover, vocational support is rarely available in this part of the world [
32]. There is preliminary evidence from Hong Kong that ‘case managers’ can provide social support and help facilitate socio-occupational functioning in schizophrenia [
33], yet most practitioners in Malaysia lack adequate training and experience in working with individuals with complex mental health and social needs [
30].
Social Recovery Therapy (SRT; 5) may be a particularly promising intervention for the Malaysian—and broader LMIC—setting due to its focus on social recovery through personally meaningful and valued structured activity including employment, community, leisure and social activities. SRT is guided by personalised goals and values and gives specific attention to the individual’s wider context, and particularly their social networks [
8,
34]. The intervention is informed by psychosocial constructions of mental health and recovery rather than a Western bio-medical model of mental ‘illness.’ As such, SRT is not primarily focused on diagnosis and symptom reduction; rather symptoms are attended to only insofar as they form barriers to social recovery (in addition to other personal and systemic barriers of relevance). In the UK, SRT has been found to be an effective treatment for young people experiencing social disability following psychosis [
34,
35] and is currently being tested for young people with complex emerging mental health problems including at-risk mental states for psychosis [
7]. Moreover, SRT provides practitioners with an explicit theoretical framework, manualised intervention procedures, and a set of therapeutic and assessment tools to facilitate patients’ social recovery. A clear framework and structured materials have been highlighted as important practice facilitators in previous research involving non-specialist mental health workers in high-income countries [
36] and LMICs [
37]. In addition, SRT recognises the contextual and cultural dependence of recovery and supports patients to formulate personally meaningful goals which are in line with their values [
6].
The ‘fit’ of Western assessment tools needs to be explored in order to provide a foundation for applying a social recovery approach across diverse contexts. Qualitative accounts of using such tools are available from participants in a UK randomised controlled trial of SRT for 16–25 year olds with persistent social disability and complex emerging mental health problems [
38,
39]. Participants identified positive aspects of disclosure and talking about difficult experiences during screening and outcome assessments [
38,
39]. Participants also spoke of the benefits of exercises conducted within SRT in helping them to understand and manage barriers to structured activity [
38]. Whilst these qualitative accounts support the acceptability of the social recovery approach in the UK, its suitability in other contexts is unknown. There is evidence regarding the semantic equivalence, validity, and reliability of some social recovery assessments with relevant populations, for example measures of at-risk mental states for psychosis in Chinese populations [
40]; however, the majority of relevant tools are untested outside the UK.
Our aim was to extend our prior qualitative work in the UK [
38,
39] by piloting key social recovery assessment tools with young people in Malaysia, focusing on feasibility (i.e. whether the social recovery tools were easily, conveniently and successfully administered to participants; 31) and acceptability (i.e. whether the tools were favourably received by participants; 31). Feasibility was operationalised as time taken to complete assessment measures and rates of participant completion. Acceptability was examined with respect to the qualitative experiences of participants, using a deductive coding framework derived from our prior work in the UK [
38,
39].
Methods
Design
We performed a cross-sectional pilot study to assess young people’s experiences of undertaking a multi-faceted structured assessment of their mental health and social disability.. The focus was on the experiential process of completing existing standardised social recovery measures that would be completed as part of a clinical research assessment, i.e. the assessment of time use, unusual psychological experiences (e.g. hearing voices), emotional problems, and positive and negative self-beliefs [
7]. We focused additionally on the completion of assessments typically used within the therapeutic assessment and formulation process conducted within SRT therapy; i.e. a values assessment and social identity mapping exercise. We also incorporated a more generic youth mental health screening and outcome measure as potentially more viable to capture emotional problems as part of a social recovery approach in Malaysia compared to more technical and resource-intensive diagnostic assessments used in the UK [
7].
Participants
Following ethical approval from the University of Sussex (Reference: CB/321/8) and relevant local approvals, a sample of participants were recruited from a non-government educational and mental health social enterprise and partner organisations in Peninsular Malaysia. Inclusion criteria required participants to be between 16 and 30 years old, able to provide informed consent, and be vulnerable young people under the institutional care of a Non-Government Organisation (NGO) in a full residential setting. The NGO and partner organisations serve low-income populations (defined as earning 40% less than the national average) in crime-affected localities in greater Kuala Lumpur. The low socio-economic status of the vulnerable target population also manifests as a lack of access to basic services such as housing and formal education. The organisations included orphanages which serve young people who are unable to remain in the family home due to extreme poverty, neglect and/or trauma. Participants did not need to report experiencing previous or current mental health problems to participate.
Potential participants were first approached by NGO staff members. Consent from the parent or caregiver with parental responsibility was sought before approaching potential participants aged under 18 years old. Interested young people were provided with information about the study. After obtaining verbal agreement for contact from the study team, each participant was invited to meet for an interview with a researcher and an interpreter. Participants were sampled using convenience sampling approach that maximised ethno-cultural diversity across Indigenous (Orang Asli), Malay, Chinese and Indian participants and the three primary languages of Malay, Mandarin and Tamil. The final sample (N = 9) comprised 5 males and 4 females, aged 16 to 23 years (M = 19.78 years; SD = 2.86). No participants reported a diagnosis of physical or mental health problems. Four participants were referred by the social enterprise, 2 from a partner educational organisation and 3 from a looked after children’s home or orphanage. All invited participants had at least 8 years of formal education. No approached participants declined. One additional orphanage was approached but did not refer any potential participants, with reasons unknown. An additional young person who was referred was not invited to consent due to having a serious learning disability which precluded capacity to provide informed consent.
Experiential process measures: social recovery clinical research assessments
Time use survey (TUS)
The TUS is a validated semi-structured interview measure of time use in clinical and non-clinical populations [
9], derived from an Office for National Statistics (UK) survey [
41]. Respondents recall time spent in structured activities over the past month (paid and voluntary employment, education, housework, childcare, sports, and leisure) which is then averaged into weekly hours.
Prodromal questionnaire (PQ-16)
The PQ-16 [
42] is a 16-item true/false self-report questionnaire. A score of 6 or more indicates elevated risk of psychosis. There is evidence of good validity and reliability in a Chinese population [
40].
Comprehensive assessment of at risk mental states (CAARMS)
The CAARMS [
43] is a semi-structured interview capturing intensity, frequency and duration of subthreshold psychotic symptoms. Scores across unusual thought content, non-bizarre ideas, perceptual abnormalities, and disorganised speech subscales, plus Global Assessment of Functioning (GAF) scores were used to determine At Risk Mental States (ARMS) status. There is evidence of good reliability and validity in a Japanese population [
44].
Brief Core Schema scales (BCSS)
The BCSS [
45] is a 24-item self-report measure in which participants rate agreement with 6 positive and 6 negative beliefs about themselves and other people from 0 (No) to 4 (Believe totally). The BCSS has been used successfully in Japan and Indonesia [
46,
47].
Strengths and difficulties questionnaire (SDQ) adolescent version
The SDQ [
48] is a 25-item brief behavioural screening questionnaire designed to identify emotional and behavioural problems. Participants rate item agreement as Not true, Somewhat true, or Certainly true. Many translated versions of the SDQ exist - including a Malay parental informant version; however there is limited information about linguistic or semantic equivalence [
49]. Nevertheless, completion by Malay parents of either the Malay or English version of the questionnaire has been found to have negligible impact on the scores [
49].
Experiential process measures: SRT therapeutic assessment tools
Social identity map (SIM)
The SIM tool [
50] produces a visual representation of participants’ social groups. After identifying all their social groups and rating each group’s importance from 1 (not at all important) to 5 (very important)), participants rate number of days actually spent with the three most important groups in the past month (0 to 30), number of days that they would have liked to have spent with these groups (0 to 30), and inter-group compatibility (easy, moderately easy, and hard).
Values assessment
The Values Assessment is an adaptation of the Valued Living Questionnaire [
51], in which participants state valued directions for each of ten life areas, for example, employment. Participants then rate from 1 to 10 (least to most) the absolute importance of each valued direction and how consistently they are living in accordance with the valued direction. Finally, participants rank the valued directions from 1 to 10 according to their relative importance.
Feasibility and acceptability
Feasibility was first assessed by recording the time taken to complete the assessments and rates of completion. In order to evaluate acceptability, a semi-structured interview schedule was derived from the PRODIGY trial schedule [
38,
39]. We retained questions regarding experiences of completing the research assessments and removed questions relating to specific PRODIGY trial procedures. We added specific questions to explore the process of completing the assessments, for example; “What was it like for you when we asked you about social groups that you belong to?”
Procedure
After providing written informed consent, participants engaged in a combined assessment and interview session conducted by the first author in the presence of an interpreter. Sessions were conducted in a private location convenient to the participant; in clinic or meeting rooms on NGO premises, in the participant’s home or place of work. The duration of the assessments is reported below. Qualitative interviews lasted between 16:53 and 41:31 min (Mean = 26:20, Standard Deviation = 8:23). Interpreters (N = 6; 5 female and 1 male) were staff members (therapists and/or programme directors) from the mental health arm of the collaborating NGO to allow for signposting and provision of support services to participants if necessary. Interpreters had received a one-day training session on the study aims, social recovery approach, and assessment procedures. Assessments were not translated in advance but were administered by the first author in the English language. Interpreters provided interpretation as needed for participant comprehension. The interviewer checked understanding of interpreted questions and responses with all parties through further questioning and additional interpretation was conducted as needed. Interpreters variably used first, second, or third person pronouns within and across interviews. For ARMS screening purposes, all participants were asked to complete the PQ-16 and any participant scoring 6 or more was asked to then complete the CAARMS assessment. All sessions were audio-recorded using a digital recorder with participant permission and the English content was transcribed verbatim.
Qualitative analysis
A deductive thematic analysis [
52,
53] approach was used to cross-validate themes identified in the previous UK PRODIGY studies [
38,
39]. The thematic analysis was conducted using six of Braun & Clarke’s seven steps [
53]; transcription, familiarisation, coding, searching for themes, reviewing themes, and defining themes. The seventh step, naming themes, was not performed. Coding focused on coding units of text which appeared to reflect the presence of themes from the previous accounts. Searching for and reviewing themes focused on reviewing the ‘fit’ of present data with these previous themes, analysing the thematic content of the coded excerpts, and identifying manifestations of the respective ‘central organising concepts’ [
53]. These steps also involved re-reading and re-familiarisation with the previous themes [
38,
39] to ensure continual reflection on the ‘fit’ of present data. At least two authors independently coded 80% of transcripts to ensure reliability in coding and identified themes.
Discussion
This pilot study assessed the feasibility and acceptability of a social recovery approach in a youth mental health setting in Malaysia. Young Malaysian participants from varied ethnic and cultural backgrounds, all of whom were vulnerable young people from low-income families, completed core social recovery assessments and discussed their experiential process reflections in a qualitative interview. Our findings suggest that, as in the UK PRODIGY trial [
38,
39], the assessment of core social recovery variables appears feasible with vulnerable young people from Malay, Chinese, Indian and Indigenous populations. The time taken to administer social recovery assessments was very favourable and the mean total assessment time of just under one and a half hours is similar to what would be expected when conducted with UK participants, in English language and without interpretation. This corroborates participants’ reflections that the assessment tools were comprehensible.
The rate of completion is also favourable with a minimum completion by three quarters of the sample for the CAARMS assessment. It is notable that main reasons for non-completion related to fatigue and practical issues rather than to specific feasibility challenges presented by individual assessment tools. The present study did not allow for flexibility in dividing the assessment into multiple sessions as has been found useful in the UK context [
39]. Moreover, fatigue may have been exacerbated due to the need for interpretation during assessments. In addition, current participants were not incentivised to complete assessments, i.e. there was no financial reimbursement nor potential provision of an intervention, which again may have inflated the non-completion rate.
Our findings also point towards the acceptability and the cultural validity of social recovery assessments. Participants appeared to find the assessment of time use acceptable and valuable and they engaged readily with qualitative questions around valued activities and barriers to engagement. Despite some instances of potential mild discomfort, especially relating to assessment questions about worry and anxiety, participants also valued assessments of their mental health. Participants expressed particular interest in questions about unusual experiences, such as hearing voices, with many participants suggesting a psychoeducational value to completing these assessments. Participants reported that answering assessment questions could aid in self and other reflection and help them monitor change in their emotions and experiences. Participants also expressed appreciation for the experience of reflecting on their values and social groups. For many participants, the act of completing the assessment tools appeared to give rise to an increased sense of self-agency and ability to consider and plan for a desired future. Thus, our findings suggest that the experiences of Malaysian young people echo those from our previous UK samples and perhaps underscore the intuitiveness of social recovery concepts, and the potential utility and possible universality of related clinical research tools across diverse contexts. Moreover, the essence of social recovery appeared to have some resonance for current participants insofar as they seemed to share a sense of structured activity as personally meaningful and facilitative of social connection—and reflected that engagement in such activity can be complicated by individual, psychological and systemic barriers. Our findings corroborate those of Byrne and Morrison [
54], who explored participant experiences of symptom and functioning monitoring within a UK trial of early detection and prevention of psychosis, in which research engagement facilitated normalisation and ‘opening up’ around unusual psychological experiences and other difficulties. Furthermore, our findings fit with a model in which assessment itself is considered a therapeutic task rather than purely an information-gathering exercise [
48].
The potential therapeutic value of being asked about unusual experiences is a particularly notable finding. Young people in Malaysia may tend to underestimate the seriousness of their own problems and set a very high threshold for help-seeking [
55]. A lack of knowledge about mental health problems is considered to underpin the high level of mental health stigma in Malaysia, and education and awareness generation are therefore key activities for stigma reduction [
56]. Current findings suggest that broad use of psychosis screening tools such as the PQ-16 [
42] or CAARMS [
57], for example in NGO or educational contexts, could in itself facilitate increased knowledge regarding unusual experiences. This could encourage engagement in mental health services at an earlier point and potentially contribute to reducing the long average DUP in Malaysia [
13]. Furthermore, SRT assessment tools. Such as values-based and social group mapping exercises, could additionally provide young people with an enhanced sense of self-agency, which may also promote help-seeking and mental health service engagement.
Nevertheless, our findings also suggest that sensitivity is needed when exploring activities and engagement with family, cultural and religious groups. In asking about ‘your’ values and ‘your’ social groups there is an embedded individualism which may represent an invitation to challenge the dominant relatively collectivistic culture in Malaysia. The privileged position afforded to independence, self-enhancement, and explicit communication within Western cognitive-based therapies may also require further consideration in a Malaysian context [
24‐
26], with reference made to locally-developed guidance around exploring spiritual or religious beliefs, resources and duties [
16].
Limitations and future research directions
Whilst efforts were made to represent young people from different ethnic and cultural communities, the qualitative methodology and small convenience sample limit the generalisability of our findings. Furthermore, no participants explicitly identified themselves as having experienced mental health problems per se. Nevertheless, current participants represented the groups that have been found to be particularly vulnerable to mental health problems in Malaysia; namely young adults from low-income families, including people from indigenous backgrounds [
18]. Actual assessment scores also suggested reduced structured activity compared to the normative level in the UK [
9] and revealed variance in experiences of mood, anxiety, and psychotic-like phenomena. The mean total difficulties SDQ score was in the borderline mental health problems range [
58]. Moreover, the mean total PQ-16 score was just below the psychosis risk threshold [
42]—with nearly half of current participants scoring in excess of this range—and one screened participant met full CAARMS criteria for at risk mental states for psychosis. Qualitative data also indicated the presence of subjective mental distress among a proportion of current participants. Additionally, for some participants, it seemed that emotional or psychological problems were preventing or reducing engagement in structured activity. Previous research has suggested that Malaysian people have limited knowledge about mental health problems, tend to underestimate their own problems and specifically do not tend to label mood and anxiety symptoms as mental health problems [
55,
59]. Therefore, we cautiously suggest that our findings have relevance for young people in Malaysia experiencing mental health problems and support the acceptable use of core social recovery assessment measures within screening initiatives for early detection of young people with emerging social disability and psychological difficulties. Nevertheless, further testing in Malaysia would usefully involve young people with confirmed serious mental health problems including psychosis. Replicating the present study with a larger sample of young people would generate more robust evidence regarding the time taken to administer assessment measures. This could help to facilitate the formal translation and validation of social recovery tools in Malaysia.
A further limitation relates to the fact that the same researcher administered both assessments and interviews, with the same interpreter present, which may have impacted on responses. Identified instances of mild discomfort do nevertheless suggest that participants felt able to divulge candid reflections on the assessment process. Furthermore, whilst the present study provides preliminary evidence of feasibility and acceptability of Western assessment and therapeutic tools, their use should be further supported with an indigenisation-from-within approach. This should involve the local review of measures translated from English to consider supplementing appropriate colloquial terminology in place of explicit translations and testing the validity of these amendments. Furthermore, acceptability of measures of psychotic or psychotic-like phenomena – translated and/or locally developed – does not preclude cultural differences in the phenomenology of experiences. Measurement structures of Western constructs such as psychotic experiences may differ in a Malaysian setting [
60], therefore, future research should continue to empirically explore the fit of measurement models on which Western assessments are predicated. Furthermore, explorations of intra-associations between core social recovery assessment scores, for example assessing relevant clinical time use thresholds, would also further inform a Malaysian social recovery approach. Finally, future research could evaluate practitioner perspectives on using a social recovery approach and of promoting valued structured activities with young people in Malaysia and potential for optimisation through integration with non-Western philosophies [
28] – in addition to assessing potential individual and structural barriers to uptake and sustained use of screening, outcome and therapeutic formulation tools in LMIC settings.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.