Exploring the associations between social support and symptoms of posttraumatic stress disorder among Malaysian and Australian trauma survivors

ABSTRACT

Background: Social support is an important feature in understanding posttraumatic stress disorder (PTSD) and its treatment. Non-clinical research has identified distinct profiles of culturally appropriate social support. Despite this, little research has examined cultural influences on social support in the context of PTSD.

Objective: This study examined cultural differences in the associations between social support and symptoms of PTSD.

Method: The study employed a cross-sectional design. Australian (n = 91) and Malaysian (n = 91) trauma survivors completed an online survey assessing PTSD symptomatology and social support (explicit and implicit social support, perceived helpfulness of support provider, attitudes towards professional help-seeking). A quasi-experimental paradigm assessed the influence of mutual (i.e. the sharing of support between relationship partners) and non-mutual support (i.e. where one person constantly receives support, while the other person constantly provides support) on negative emotion and subjective distress.

Results: First, explicit social support was negatively associated with PTSD symptoms for the Australian group but not the Malaysian group. Second, perceived helpfulness of support from family was negatively associated with PTSD symptoms for the Malaysian group but not the Australian group. Third, the Malaysian group reported significantly greater distress for non-mutual support and significantly fewer negative emotions and distress for mutual support than the Australian group. Fourth, the Malaysian group reported that they were significantly more open to acknowledging psychological problems and the possibility of seeking professional help for these problems than the Australian group.

Conclusions: As the PTSD social support literature continues to evolve, it is essential that cultural influences are considered given the important theoretical and clinical implications.

HIGHLIGHTS

• Social support is an important feature in understanding posttraumatic stress disorder (PTSD). While non-clinical research has identified distinct profiles of culturally appropriate social support, little research has examined cultural influences on social support in the context of posttraumatic stress disorder.

• Disclosing the trauma to others and explicating requesting assistance was negatively associated with PTSD symptoms for the Australian group but not the Malaysian group. Support from family was negatively associated with PTSD symptoms for the Malaysian group but not the Australian group. The Malaysian group reported significantly greater distress for non-mutual support and significantly less negative affect and distress for mutual support than the Australian group.

• As the psychotraumatology literature continues to evolve, it is essential that cultural influences on social support are considered given the important theoretical and clinical implications.

Antecedentes: El apoyo social es una característica importante para comprender el trastorno de estrés postraumático (TEPT) y su tratamiento. La investigación no clínica ha identificado distintos perfiles de apoyo social culturalmente apropiados. A pesar de ello, pocas investigaciones han examinado las influencias culturales sobre el apoyo social en el contexto del TEPT.

Objetivo: Este estudio examinó las diferencias culturales en las asociaciones entre el apoyo social y los síntomas del TEPT.

Método: El estudio empleó un diseño transversal. Sobrevivientes de trauma australianos (n = 91) y malasios (n = 91) completaron una encuesta en línea que evaluaba la sintomatología del TEPT y el apoyo social (apoyo social explícito e implícito, la ayuda percibida del proveedor de apoyo, actitudes hacia la búsqueda de ayuda profesional). Un paradigma cuasi experimental evaluó la influencia del apoyo mutuo (es decir, el apoyo compartido entre los miembros de una relación) y no mutuo (es decir, cuando una persona recibe apoyo constantemente, mientras que la otra lo proporciona constantemente) sobre la emoción negativa y el malestar subjetivo.

Resultados: En primer lugar, el apoyo social explícito se asoció negativamente con los síntomas de TEPT en el grupo australiano, pero no en el grupo malasio. En segundo lugar, la ayuda percibida de la familia se asoció negativamente con los síntomas de TEPT en el grupo malasio, pero no en el australiano. En tercer lugar, el grupo malasio informó de una perturbación significativamente mayor en el caso del apoyo no mutuo y significativamente menos emociones negativas y perturbación en el caso del apoyo mutuo que el grupo australiano. En cuarto lugar, el grupo malasio declaró estar significativamente más abierto a reconocer problemas psicológicos y la posibilidad de buscar ayuda profesional para estos problemas que el grupo australiano.

Conclusiones: A medida que la literatura sobre el apoyo social del TEPT sigue evolucionando, es esencial que se tengan en cuenta las influencias culturales, dadas las importantes implicaciones teóricas y clínicas.

背景：社会支持是理解创伤后应激障碍 (PTSD) 及其治疗的一个重要特征。 非临床研究已经确定了文化上适当的社会支持的不同特征。 尽管如此，很少有研究考查文化对创伤后应激障碍背景下社会支持的影响。

目的：本研究考查了社会支持与 PTSD 症状之间关联的文化差异。

方法：本研究采用横断面设计。澳大利亚 (n = 91) 和马来西亚 (n = 91) 创伤幸存者完成了一项评估 PTSD 症状和社会支持（显性和隐性社会支持、支持提供者的感知帮助感、对专业求助的态度）的在线调查。准实验范式评估了相互支持（即关系伙伴之间分享支持）和非相互支持（即一个人不断获得支持，而另一个人不断提供支持）对负性情绪和主观痛苦的影响 。

结果：首先，显性社会支持与澳大利亚组而非马来西亚组的 PTSD 症状呈负相关。 其次，对于马来西亚组而非澳大利亚组而言，家庭支持的感知帮助与 PTSD 症状呈负相关。 第三，马来西亚组相较于澳大利亚组，报告了非相互支持显著更大的痛苦，相互支持显著更少的负性情绪和痛苦。 第四，马来西亚组相较于澳大利亚组，报告自己更愿意承认心理问题和更可能寻求专业帮助。

结论：随着 PTSD 社会支持文献的不断发展，鉴于重要的理论和临床意义，必须考虑文化影响。

KEYWORDS:

• Social support
• culture
• Malaysia
• Australia
• trauma
• posttraumatic adjustment

PALABRAS CLAVE:

• Apoyo social
• cultura
• Malasia
• Australia
• trauma
• adaptación postraumática

关键词:

• 社会支持
• 文化
• 马来西亚
• 澳大利亚
• 创伤
• 创伤后调整

Posttraumatic stress disorder (PTSD) is a chronic disabling psychiatric condition observed in most societies and cultures (Kessler et al., 2017). PTSD is characterised by re-living symptoms, negative alterations in mood and cognition, avoidance of trauma-related reminders, and increased physiological arousal (American Psychiatric Association, 2013). Social support is an important feature in understanding PTSD and its treatment (Hansford & Jobson, 2022). Considerable research indicates that limited social support is associated with greater PTSD symptoms (Brewin et al., 2000; Ozer et al., 2003; Trickey et al., 2012), social support and PTSD reciprocally predict each other over time (Wang et al., 2021), and social support is a key mechanism in the prevention and treatment of PTSD (Whealin et al., 2008). While this body of literature is extensive, it mainly comprises studies conducted in Western cultural contexts, with most samples comprising Western participants (Hansford & Jobson, 2022). Consequently, research and current PTSD psychological treatments targeting social support tend to be based on Western cultural norms and values (Hansford & Jobson, 2022). This is concerning as the non-clinical literature has clearly identified distinct profiles of culturally appropriate social support, which has important implications for psychological adjustment (Hansford & Jobson, 2022). However, psychotraumatology research lags concerningly behind. Therefore, the current study aimed to investigate culture and social support in the context of PTSD.

1. Social support and PTSD

Social support is a multidimensional construct with many definitions (Wang et al., 2021). As in previous research, we conceptualised social support as actual or available social resources in times of need (Wang et al., 2021). Social support is emphasised in theoretical accounts of PTSD. For instance, cognitive models posit that the responses of others following trauma can influence maladaptive appraisals about others and the world, which consequently maintain PTSD symptoms (Ehlers & Clark, 2000). Other models posit that social support modulates stress by influencing the individual’s cognitive processing of the trauma, their coping mechanisms and emotional adjustment (Joseph et al., 1997; Lepore, 2001; Williams & Joseph, 1999). More recent PTSD models highlight the need to consider PTSD within a trauma survivor’s interpersonal and socio-cultural context (Jobson, 2009; Maercker & Horn, 2013). Despite these theoretical advances, social support research in the context of PTSD has predominately occurred within Western cultural frameworks (Hansford & Jobson, 2022; Yeomans & Forman, 2009).

2. Social support and culture

Understanding trauma and its effects on an individual within their socio-cultural context is essential (Kirmayer et al., 2010). One of the most recognised and studied cultural dimensions relates to the value placed on the self in relation to others (Markus & Kitayama, 2010). Those from Western individualistic cultures (e.g. Australia, United States) perceive the self as independent, unique, and valuing personal goals (Markus & Kitayama, 2010). In contrast, those from Asian collectivistic cultures (e.g. Malaysia, China) perceive the self as interdependent, interconnected with others and prioritize harmony with close others (Markus & Kitayama, 2010). These differences effect interpersonal interactions (Markus & Kitayama, 2010) and consequently influence whether support is sought, how it is sought, whom it is sought from, and the benefits and costs of seeking support (Kim et al., 2012). Whilst there is limited cross-cultural PTSD and social support research, there is research examining the relationship between social support and wellbeing. Campos and Kim note that while relationship characteristics are important, it is the cultural representations of relationships that influence the appraisal of interpersonal interactions and when appraisals align with cultural expectations they contribute to wellbeing (Campos & Kim, 2017).

3. Implicit versus explicit support

Implicit social support is defined as ‘the emotional comfort one obtains from social networks without disclosing specifics of the stressful event, such as enjoying pleasant activities together or being in the company of close others without discussing problems’ (p. 1503) (Yang et al., 2015). In contrast, explicit social support is the ‘active disclosure and discussion of problems as well as request for assistance, and can take the form of advice, instrumental aid, or emotional comfort recruited from social networks’ (p. 1502) (Yang et al., 2015). Those from Asian cultural backgrounds use explicit support less frequently and view implicit support more positively, than those with European heritage (Ishii et al., 2017; Kim et al., 2006; Mojaverian et al., 2013; Taylor et al., 2004). Additionally, those from Asian cultures benefit more from implicit support and explicit support can exacerbate emotional distress, while those from Western cultures benefit from explicit support (Ishii et al., 2017; Kim et al., 2006; Mojaverian et al., 2013; Taylor et al., 2004). For instance, Taylor et al. (2007) using a laboratory stressor paradigm found that Asian Americans showed greater increases in subjective distress and cortisol reactivity when seeking explicit support, but were more comforted by implicit support. In contrast, European Americans benefited more from requesting support explicitly (Taylor et al., 2007). Additionally, explicit support is more closely associated with health outcomes for those with European cultural backgrounds than those with Asian cultural backgrounds (Chiang et al., 2013).

These differences are proposed to reflect concerns about explicit social support negatively impacting interpersonal relationships for members of Asian cultures (Wang & Lau, 2015). Those with European cultural heritage view negative feelings as something to be examined and explicitly explored in interpersonal discussions and explicit support may be more ‘beneficial for individualists who prioritize agency in “getting one’s needs met”’ (p. 918) (Wang & Lau, 2015). Those from Asian cultures, however, are more likely to focus on managing the problem in ways that preserve relationships and the ‘effects of seeking support are likely shaped by the broader context of the relationship, with concern for nurturing harmonious social relationships’ (p. 918) (Wang & Lau, 2015). Given the cultural emphasis on group harmony and mutual obligation, members of Asian collectivistic cultures may be less likely to recruit explicit social support to avoid burdening others or losing face (Kim et al., 2008; Wang & Lau, 2015). In contrast, among individualistic cultures the independent self is less obligated within interpersonal relationships, and relationships are generally perceived as freely chosen (Wang & Lau, 2015; Yang et al., 2015).

4. Support provider

Next it is worth considering who provides the support. Social support may come from a significant other, family, friends, community members or professional health providers (Kim et al., 2006; Wang & Lau, 2015). Despite the benefits of social support, asking for support can come with costs (Kim et al., 2006). Wang and Lau (Wang & Lau, 2015) showed that Asian Americans were more inclined to seek help from discretionary (i.e. peers) than obligatory (i.e. family) ties. Similarly, Kim et al. (2006) found that Asian Americans were less willing to seek social support when the support provider was relationally close, while for European Americans there was no difference. It has been proposed that to maintain group harmony, save face and not burden those close to an individual, members of Asian cultures may be less likely to find help from close others (e.g. family) as beneficial (Kim et al., 2006; Wang & Lau, 2015). However, some research has found that family plays a significant role in care and treatment of those from Asian cultural backgrounds (Lin et al., 1991; Mojaverian et al., 2013; Lin & Cheung, 1999; Park and Chelsa (2010).

5. Mutual versus non-mutual support

One relationship dynamic that is especially relevant to social harmony is whether the relationship is characterised by mutual (i.e. the sharing of comfort and help between relationship partners) or non-mutual support (i.e. where one person constantly receives support and the other person constantly provides the support) (Wang & Lau, 2015). Wang and Lau (2015) note that costs to the self are potentially reduced when relationship partners both reciprocally provide and receive support. Thus, in collectivistic cultures, participating in mutual support may be perceived as nurturing the welfare of the relationship and counteracting any potential burden that may arise from requesting explicit support (Wang & Lau, 2015). To examine this, Wang and Lau conducted a laboratory stressor quasi-experiment exploring whether European Americans and Asian Americans experience explicit support seeking differently depending on whether the participant was primed to perceive the support as mutual or non-mutual. They found that Asian Americans experienced the support as less stressful and more helpful when the support was perceived as mutual (rather than non-mutual), while European Americans in the mutual and non-mutual social support conditions did not differ significantly. This extends the above work on explicit support, as the stress associated with explicit support seeking for those from Asian cultural backgrounds may be offset by the perception of mutuality in a relationship (Wang & Lau, 2015). Whereas, for European Americans the broader context of the relationship may not significantly influence experiences of that support (Wang & Lau, 2015).

6. Professional support

Cultural differences in professional help-seeking is another important consideration in PTSD. To receive assistance from health professionals, trauma survivors need to actively solicit their help (Mojaverian et al., 2013). Substantial research highlights systematic cultural differences in seeking professional assistance. Those from Asian cultural backgrounds are less likely than those from European cultural backgrounds to seek professional help (Henderson et al., 2013; Hwang, 2006). Again, cultural differences in independence and interdependence have been used to explain these differences. Mojaverian et al. (2013) note that in many Asian cultural contexts help-seeking from an out-group source (e.g. mental health professionals) may create discord and problems within the in-group. Given the focus on interdependence in Asian cultural contexts (including a strong sense of familial obligation) mental health concerns of a family member may be perceived as a disruption of family balance, eliciting the entire family’s involvement in the care of the individual (Mojaverian et al., 2013). Thus, in Asian collectivistic cultures professional help-seeking could be interpreted as a result of dysfunction within one’s in-group and a threat to in-group relational functioning, fostering reluctance to seek professional help (Mojaverian et al., 2013). Another factor is concern for the stigma surrounding professional help-seeking (Henderson et al., 2013). Research has found that stigma relating to mental health can be high in Asian cultures (Hanafiah & Van Bortel, 2015; Lauber & Rössler, 2007). Another aspect that influences attitudes toward professional help-seeking is psychological openness to mental health concerns (Masuda et al., 2009). Mojaverian et al. (2013) examined cultural differences in attitudes related to professional help-seeking and found that Japanese participants were more reluctant than Americans to seek support from professionals and reported less propensity for help-seeking and less indifference to stigma. Moreover, the relationship between cultural group and professional help-seeking attitudes was mediated by indifference to stigma (Mojaverian et al., 2013).

7. Current study

This study aimed to examine cultural differences in social support (implicit vs explicit; support provider; mutual vs. non-mutual provider; professional help-seeking) in the context of PTSD. To investigate this aim, we selected trauma survivors from an Asian, collectivistic cultural group (Malaysia) and from a Western individualistic cultural group (Australia) (Church et al., 2012; Hofstede et al., 2010; Timothy et al., 2012). Contemporary Malaysian society tends to hold interdependent notions of self and values social harmony and the honouring of cultural norms and traditions (Ward, 1987). In contrast, Australia is a highly individualistic society (Hofstede et al., 2010).

We hypothesised that the Malaysian group would report using implicit social support significantly more and explicit social support significantly less than the Australian group (Hypothesis 1a). We also hypothesised that implicit support would be associated with fewer PTSD symptoms among the Malaysian group, while explicit social support would be associated with fewer PTSD symptoms among the Australian group (Hypothesis 1b). Second, we hypothesised that Malaysians would perceive social support as significantly less helpful when the support provider was relationally close (i.e. significant others, family) than the Australian group (Hypothesis 2). We conducted exploratory analyses examining the associations between perceived helpfulness of social support provider and PTSD symptoms, and whether cultural group moderated these associations. Third, we predicted that Malaysians in the mutual support condition would experience less negative affect and psychological distress than those in the non-mutual condition, while Australians in the mutual and non-mutual social support conditions would not differ significantly in terms of negative affect and distress (Hypothesis 3). Finally, we hypothesised that that the Malaysian group would be more reluctant than the Australian group to seek support from professionals; reporting less propensity for help-seeking, openness to psychological concerns and indifference to stigma (Hypothesis 4a). Based on the findings of Mojaverian and colleagues (Kim et al., 2006), we also predicted that the relationship between cultural group and professional help-seeking attitudes would be mediated by indifference to stigma (Hypothesis 4b). We conducted exploratory analyses examining the associations between professional help-seeking tendencies and PTSD symptoms, and whether cultural group moderated these associations.

8. Method

8.1. Research design

This study employed a cross-sectional, cross-country design. It also included a quasi-experimental paradigm to assess cultural differences in the influence of mutual and non-mutual support on negative affect and subjective distress. Australian and Malaysian researchers co-designed the study to ensure cultural appropriateness of the design, data analysis and interpretation, and dissemination of findings. All measures selected for use have routinely been used in cross-cultural research. The survey was available in both English and Malay. For any measures not available in Malay, English versions were translated into Malay using gold-standard procedures (Ghazy et al., 2022).

8.2. Participants

Participants included 91 Australian (M_age= 34.11 years, SD = 9.18; 45 men, 43 women, 3 non-binary/gender diverse) and 91 Malaysian (M_age= 26.42 years, SD = 7.12; 27 men, 60 women, 4 non-binary/gender diverse) trauma survivors. Participants indicated that their highest level of education was: secondary school (Australian n = 15; Malaysian n = 4), post-secondary school qualification (Australian n = 39; Malaysian n = 9), and undergraduate or post-graduate degree (Australian n = 36; Malaysian n = 75)(see Table 1).

Table 1. Participant characteristics and group means (SD) for social support measures.

Variable	Australian group	Malaysian group
Age	34.11 (9.18)	26.42 (7.12)
Gender (Man: Woman: Non-binary/Gender diverse)	45:43:3	27:60:4
Education^a	15:39:29:7:0:1	4:9:60:15:2:1
Trauma type^b	23:4:30:14:0:4:9:7	25:4:12:19:1:6:14:10
Number of Traumatic Life Events	8.40 (4.98)	8.36 (4.50)
Time since trauma (years)	7.75 (12.13)	6.55 (6.10)
Independent self-construal	71.58 (12.65)	72.82 (12.16)
Interdependent self-construal	69.37 (13.52)	73.00 (10.47)
PTSD Symptoms	36.03 (18.00)	35.99 (18.79)
Depression Symptoms	34.77 (9.51)	33.27 (12.16)
Anxiety Symptoms	21.78 (6.09)	19.77 (7.03)
Implicit Social Support	19.97 (5.87)	22.53 (5.66)
Explicit Social Support	39.26 (9.77)	41.65 (10.83)
Family Support	4.09 (1.87)	4.32 (2.28)
Significant Other Support	4.96 (2.09)	4.55 (2.24)
Friend Support	4.15 (1.51)	4.87 (1.68)
Tendency to Seek Support	29.13 (6.37)	29.33 (5.37)
Psychological Openness	22.71 (7.19)	24.82 (5.75)
Indifference to Stigma	22.77 (6.31)	19.93 (6.45)

^a Secondary: Post-Secondary: Undergraduate: Post-graduate: Other: Prefer not to say.

^b Trauma type refers to the event participants indicated as the worst trauma that still bothers them the most. The PCL-5 was answered in relation to this trauma event. Trauma types included: Accident, injury, assault:natural disaster: sexual assault: war: unexpected death and life-threatening illness. There were no cultural group differences in trauma type distribution, χ²(7,182) = 11.57, p = .12.

To obtain our sample, we collected data from 276 participants. Inclusion criteria were: (a) having experienced a criterion A trauma experience (as indexed by the Life Events Checklist (Weathers et al., 2013), (b) Malaysian participants identifying as having Malay, Chinese or Indian heritage (i.e. all four grandparents of Malay, Chinese or Indian heritage) and Australian participants identifying as having European heritage (i.e. all four grandparents of Western European heritage), (b) being aged between 18 and 65 years, and (c) able to complete the online survey in either English or Malay. Exclusion criteria included rapid responders (i.e. those who completed the survey in under 10 minutes), scoring below the conscientious response cut-off (those who did not score a minimum of three correct responses on the Conscientious Responder Scale (Marjanovic et al., 2014)), and completing the survey more than once. Based on our exclusion criteria, 61 participants were excluded (no criterion A trauma n = 30, duplicate responses n = 4, did not met cultural inclusion criteria n = 18, failed to meet cut-off on conscientious responder scale n = 9). Following exclusion of these participants, we had 124 Australian participants and 114 Malaysian participants.

Given the importance of our two cultural groups reporting similar levels of PTSD symptomatology, we examined group differences in PTSD symptomatology and found the Australian group reported significantly higher levels of PTSD symptomatology than the Malaysian group, t(236) =  3.56, p < .001. Hence, the groups were matched for PTSD symptomatology (Zwarun, 2017) and our final sample comprised of 91 Australian and 91 Malaysian participants.¹

8.3. Sampling selection

Participants were recruited from the general community in Australia and Malaysia using social media adverts (Facebook, Gumtree). G*Power was used to estimate sample size. Given the novelty of this work, we were guided by sample sizes used in previous non-clinical studies (Kim et al., 2006; Yang et al., 2015; Wang & Lau, 2015). Using small to moderate effect sizes (f²= .10), alpha of 0.05, and 80% power, it was estimated that the study required 77–84 participants per group.

8.4. Measures

8.4.1. Trauma exposure and PTSD symptoms

8.4.1.1. PTSD checklist for the DSM-5 with life events checklist for DSM-5 and criterion A (PCL-5) (Weathers et al., 2013)

The LEC is a self-report measure that screens for potentially traumatic events in a participant’s lifetime. It assesses exposure to 16 traumatic events known to potentially result in PTSD symptoms. The PCL-5 is a 20-item self-report measure of PTSD symptomatology in response to the worst trauma (i.e. the trauma that still bothers the participant the most) reported on the LEC. Items are scored on 5-point Likert scales (0 = not at all to 5 = extremely) and responses are summed to provide a total PTSD severity score, which can range from 0–80. Higher scores indicate greater PTSD symptom severity (Weathers et al., 2013). A PCL-5 cut-point score of 33 indicates a provisional PTSD diagnosis (Blevins et al., 2015). The PCL-5 has good psychometric properties (Blevins et al., 2015), including in cross-cultural research (Jobson et al., 2022). In the present study, the PCL-5 yielded excellent internal consistency for the Australian (McDonald’s Omega = .95) and Malaysian (McDonald’s Omega = .95) groups.

8.4.1.2. Hospital anxiety and depression scale (HADS) (Zigmond & Snaith, 1983)

Given the comorbidity between PTSD and depression and anxiety, to provide greater details about our sample we used the HADS to assess symptoms of depression and anxiety (Zigmond & Snaith, 1983). The HADS has been found to have good validity and reliability in Australian (Lloyd et al., 2023) and Malaysian samples (Yahya & Othman, 2015). In the current study internal consistency was good for anxiety (Malaysia McDonald’s Omega = .91; Australia McDonald’s Omega = .84) and depression (Malaysia McDonald’s Omega = .94; Australia McDonald’s Omega = .90).

8.5. Measures of social support

8.5.1. Explicit and implicit social support

As in Yang et al. (2015), explicit and implicit social support were assessed using the Medical Outcome Study-Social Support Scale (MOS-SSS) (Sherbourne & Stewart, 1991). The questionnaire assesses perceived availability of support (‘How many times in the last 3 months have the following types of support been available to you if you need it?’) with a five-point scale from 0 (none of the time) to 5 (all of the time). The MOS-SSS includes four subscales: (1) emotional/informational support (eight items); (2) tangible support (four items); (3) affectionate support (three items); and (4) positive social interaction (four items). As in Yang et al., explicit social support was characterised by the first two subscales (emotional/informational support and tangible support) and implicit support was characterised by the final two subscales (affectionate support and positive social interaction). The MOS-SSS has been found to have good validity and reliability (Sherbourne & Stewart, 1991), including with Malaysian (Norhayati et al., 2015) and Australian (Holden et al., 2014) samples. In the current study internal consistency was good for implicit social support (Malaysia McDonald’s Omega = .93; Australia McDonald’s Omega = .91) and explicit social support (Malaysia McDonald’s Omega = .93; Australia McDonald’s Omega = .90).

8.5.2. Support provider

Following the approach of Kim and colleagues (Kim et al., 2006), participants reported how helpful their significant other, family and friends were in supporting them deal with the trauma on scales from 1 (not at all helpful) to 7 (very helpful). While the validity and reliability of these items have not been established, this approach has been used in past cross-cultural research with participants with Asian and European heritage and thus, was included in the current study (Kim et al., 2006).

To examine mutual vs non-mutual support provider, we adapted the quasi-experimental approach of Wang and Lau (Wang & Lau, 2015). However, unlike Wang and Lau, where participants completed a lab stressor paradigm (mental arithmetic and speech task), in this study participants focused on their index trauma. Participants were randomly assigned to one of two writing conditions – mutual support or non-mutual support. In both conditions, participants completed a writing task. They spent three minutes writing a letter to a close other (of their choosing) asking for help with dealing with the trauma. In the non-mutual support condition, participants were asked to write the letter for help to someone they had received help or support from in the past. In the mutual support condition, participants were specifically instructed to write the letter for help to someone they had provided help or support to in the past. Participants then wrote about their trauma experience for five minutes. Immediately following this, participants completed the Positive And Negative Affect Scale (PANAS) (Watson et al., 1988). The PANAS is a widely used 20-item self-report measure of affect (positive and negative) (Watson et al., 1988). Participants rate how they are feeling now in relation to different positive (10 items) and negative (10 items) emotions on 5-point Likert scales (Watson et al., 1988). The reliability and validity of the PANAS has been found to be good, including with Malaysian and Australian samples (Ke et al., 2022; Mackinnon et al., 1999). Given our research aims and hypotheses, we were particularly interested in the PANAS Negative Affect subscale and item relating to subjective distress.

8.5.3. Professional support

Based on Mojaverian et al. (2013), we used the Inventory of Attitudes toward Seeking Mental Health Services (IASMHS) (Mackenzie et al., 2004) to assess professional help-seeking tendencies. The IASMHS includes 24 items which are responded to on 5-point Likert scales (1 = disagree to 5 = agree). The IASMHS includes three subscales; (1) willing and able to seek professional help (tendency to seek help); (2) open to acknowledging psychological problems and the possibility of seeking professional help for these problems (openness to psychological problems); and (3) lack of concern about what various important others might think about seeking professional help (indifference to stigma) (Mackenzie et al., 2004). The validity and reliability of the IASMHS has been found to be good (Mackenzie et al., 2004), including with Malaysian (Yee, 2017) and Australian (Rickwood et al., 2012) samples. In the current study internal consistency was adequate (tendency to seek help Malaysia McDonald’s Omega = .73; Australia McDonald’s Omega = .70; openness to psychological problems Malaysia McDonald’s Omega = .68; Australia McDonald’s Omega = .85; indifference to stigma Malaysia McDonald’s Omega = .81; Australia McDonald’s Omega = .78).

8.5.4. Self-construal scale (SCS) (Singelis, 1994)

In order to assess group differences in self-construal we also included the SCS, which is a 30-item scale that assess how people view themselves in relation to others. It includes two sub-scales; independent self-construal (15 items) and interdependent self-construal (15 items). Participants respond to statements about themselves on 7-point rating scales. Scores are totalled for each respondent providing an independent subscale score and an interdependent subscale score. This scale is widely used in cross-cultural research, including with Malaysian and Australian samples (Jobson et al., 2022; Singelis, 1994). In the current study, internal consistency was good (Australia independent McDonald’s Omega = .83 interdependent McDonald’s Omega = .86; Malaysian independent McDonald’s Omega = .84, interdependent McDonald’s Omega = .86).

8.5.5. Conscientious responder scale (CRS) (Marjanovic et al., 2014)

The CRS was used to differentiate between conscientious and indiscriminate response on a survey. It includes 5 items that are randomly inserted throughout a survey, with each item instructing participants how to respond (e.g.: ‘Please answer this question by choosing option number two, ‘disagree.’”). Correct responses are scored as ‘1’, while incorrect responses are scored as ‘0’. Item scores are summed, with higher scores indicating more conscientious responses. Previous literature has classified responses of 0–2 as indicative of indiscriminate responses (Jobson et al., 2022; Marjanovic et al., 2014). Thus, the cut-off score for conscientious responders in the present study was a minimum of 3 correct responses.

8.6. Procedure

Ethical approval was obtained from Monash University Human Research Ethics Committee (ID: 29275). Upon expressing interest, prospective participants were emailed a link to the online Qualtrics survey. The survey first presented the explanatory statement and participants were clearly informed that by commencing the survey they were providing consent to participate. Questionnaires took approximately 30-40 min to complete. The questionnaire included the PCL-5 (with LEC), measures of social support, CRS and SCS. For the quasi-experimental task, Qualtrics randomly allocated participants into the mutual or non-mutual condition. Participants were compensated for their time (A$25 voucher or equivalent in Malaysia).

8.7. Data analysis

Prior to hypothesis-testing, data cleaning was conducted using Microsoft Excel. All analyses were conducted using IBM SPSS Statistics 27. As several variables were not normally distributed and transformations did not improve normality, bootstrapping (5000 bootstrapped samples) was used for all analyses. Due to group differences in age, gender, and education (see below), these variables were included as covariates in all analyses. Following current recommendations (Miller & Chapman, 2001), anxiety and depression were not included as covariates but were used to provide further details about the sample. Given increased exposure to trauma is associated with PTSD symptomatology (Frans et al., 2005), we used participants responses on the LEC as an index of lifetime trauma exposure. We calculated the number of trauma types each participant had been exposed to. As the LEC has ‘no formal scoring protocol or interpretation per se’ (PTSD: National Center for PTSD, 2020), our analyses were exploratory in examining whether a similar pattern of results emerged when lifetime trauma exposure was also included as a covariate. For each Hypothesis, a similar pattern of results, to that outlined below, emerged. A summary of these findings is documented in Supplemental Material.

To assess Hypothesis 1a, we conducted a one-way (Cultural group; Australian, Malaysian) multivariate analysis of covariance (MANCOVA), with explicit and implicit support as dependent variables. To assess Hypothesis 1b, we used two moderation analyses (PROCESS model 1) to assess whether cultural group moderated the association between implicit and explicit support and PTSD symptoms. To test Hypothesis 2, we used a one-way (Cultural group; Australian, Malaysian) MANCOVA with perceived helpfulness of support from specific providers (significant other, family, friends) as the dependent variables. We also conducted exploratory moderation analyses (PROCESS model 1) examining whether cultural group moderated the associations between perceived helpfulness of support providers and PTSD symptoms. To test Hypothesis 3, we conducted two 2 (Cultural group: Asian vs Australian) × 2 (Support type: mutual vs non-mutual) ANCOVAs, with either negative emotion or subjective distress as the dependent variable (here we included age, gender, education, and PTSD symptoms as covariates²). Finally, to test Hypothesis 4, we conducted a one-way (cultural group; Malaysians, Australians) MANCOVA with tendency to seek professional help, indifference to stigma and openness to psychological problems as the dependent variables. We also assessed the mediation model assessed by Mojaverian et al. (2013), whereby we assessed the indirect pathway from cultural group to tendency to seek professional help through indifference to stigma. We also explored whether cultural group moderated the associations between professional help tendencies and PTSD symptoms using PROCESS (model 1). For all moderation analyses, confidence intervals were used to determine significance of results, with confidence intervals not including 0 being considered significant.

9. Results

9.1. Participant characteristics

Participant characteristics and social support variables are presented in Table 1. There were significant group differences in age, F(1,179) = 39.57, p < .001, η_p²= .18, gender, χ²(2, 182) = 7.45, p = .02, and education level, χ²(5,182) = 40.83, p < .001. Thus, these variables were included as covariates in the below analyses. The two cultural groups did not differ significantly in number of traumas exposed to during lifetime, F(1,180) = .002, p = .96, η_p²<.001, time since trauma, F(1,177) = .69, p = .41, η_p²= .004, PTSD symptomatology, F(1,180) < .001, p = .99, η_p²<.001, or depression and anxiety symptomatology, Wilks Lambda = .97, F(2,174) = 2.45, p = .09, η_p²= .03. In each cultural group 50 participants (54.95%) met clinical cut-off for probable PTSD diagnosis. While the two groups did not differ significantly for independent self-construal, F(1,180) =  0.46, p = .50, η_p²= .003, the Malaysian group reported greater interdependent self-construal than the Australian group, F(1,180) = 4.09, p = .045, η_p²= .02.

9.2. Hypothesis 1: implicit and explicit support

The MANCOVA for social support type was not significant, Wilks Lambda = .98, F(2,175) = 1.45, p = .24, η_p²= .02. However, it is worth noting that when age was not included as a covariate the MANCOVA was significant, Wilks Lambda = .95, F(2,177) = 4.34, p = .01, η_p²= .05; the Malaysian group reported significantly greater implicit support than the Australian group, F(1,178) = 5.12, p = .03, η_p²= .03, and the two groups did not differ significantly for explicit support, F(1,178) = .41, p = .52, η_p²= .002. There was no evidence that explicit support was associated with PTSD symptoms, B = −.19, SE = .14, t = 1.43, p = .16, 95%CI[-.46,.12]. However, cultural group moderated this association, B = −.66, SE = .27, t = 2.43, p = .02, 95%CI[−1.20,−0.12]. The association was significant for the Australian group, B = −.58, SE = .21, t = 2.79, p = .01, 95%CI[−.99,−.17], but not the Malaysian group, B = .08, SE = .18, t = .45, p = .65, 95%CI[−.27,.43]. Simple slopes are presented in Figure 1. There was no evidence that implicit social support was associated with PTSD symptoms, B = −0.39, SE = .23, t = 1.67, p = .10, 95%CI[−0.83, 0.08], nor that cultural group moderated this association, B = −.67, SE = .50, t = 1.32, p = .19, 95%CI[−1.66,0.33].

Figure 1. Simple slopes for the explicit social support moderation analyses.

9.3. Hypothesis 2: support provider

A MANCOVA found no significant cultural group differences in perceived helpfulness of specific providers (significant other, family, friends), Wilks Lambda = .96, F(3,174) = 2.33, p = .08, η_p²= .04. Support from family was significantly associated with PTSD symptoms, B = −1.29, SE = 0.69, t = 1.98, p = .049, 95%CI[−2.65,0.05], but cultural group moderated this association, B = 3.40, SE = 1.33, t = 2.55, p = .04, 95%CI[0.77, 6.04]. The association was significant for the Malaysian group, B = −2.53, SE = .84, t = 3.02, p = .003, 95%CI[−4.19,−0.88], but not the Australian group, B = .87, SE = 1.03, t = .84, p = .40, 95%CI[−1.16,2.90]. Simple slopes are presented in Figure 2. There was no evidence that support from significant other was associated with PTSD symptoms, B = −0.29, SE = 0.63, t = 0.44, p = .66, 95%CI[−1.51,0.94], or that cultural group moderated this association, B = .16, SE = 1.31, t = .12, p = .90, 95%CI[−2.42,2.75]. The association between support from friends and PTSD symptoms was approaching significance, B = −1.59, SE = 0.88, t = 1.91, p = .058, 95%CI[−3.33,0.22]. Cultural group did not moderate this association, B = −0.58, SE = 1.74, t = 0.34, p = .74, 95%CI[−4.02, 2.85].

Figure 2. Simple slopes for support from family moderation analyses.

9.4. Hypothesis 3: mutual support provider

As shown in Figure 3a, for negative affect the cultural group x support type interaction was significant, F(1,163) = 5.32, p = .02, η_p²= .03. Follow-up analyses showed that for the non-mutual condition the two cultural groups did not differ significantly, F(1,94) = .66, p = .42, η_p²= .01. However, for the mutual condition, the Malaysian group reported significantly less negative affect than the Australian group, F(1,65) = 11.44, p = .001, η_p²= .15. The two Australian groups, F(1,80) =  2.99, p = .09, η_p²= .04, and two Malaysian groups did not differ significantly, F(1,79) = 2.25, p = .14, η_p²= .03.

Figure 3. Mean negative affect (Figure 3a) and mean subjective distress (Figure 3b) for the Malaysian and Australian groups in the non-mutual social support and mutual social support conditions.

In terms of subjective distress, as shown in Figure 3b, the interaction was significant, F(1,163) = 10.70, p = .001, η_p²= .06. Follow-up analyses showed that for the non-mutual condition the Malaysian group reported significantly greater distress than the Australian group, F(1,94) = 3.97, p = .049, η_p²= .04. However, for the mutual condition, the Malaysian group reported significantly less distress than the Australian group, F(1,65) = 6.83, p = .011, η_p²= .10. Additionally, for the Malaysian groups those in the mutual reported significantly less distress than those in the non-mutual condition, F(1,79) = 11.75, p = .001, η_p²= .13. The two Australian groups did not differ significantly, F(1,80) = 2.52, p = .12, η_p²= .03.

9.5. Hypothesis 4: professional support

The MANCOVA was significant, Wilks Lambda = .95, F(2,172) = 2.99, p = .03, η_p²=  .05. In contrast to our hypothesis, the Malaysian group reported that they were significantly more open to psychological problems than the Australian group, F(1,174) =  4.13, p = .04, η_p²=  .02, and the two groups did not differ significantly in terms of tendency to seek professional help, F(1,174) = .002, p = .97, η_p²<.001, or indifference to stigma, F(1,174) = 1.07, p = .30, η_p²= .01. Indifference to stigma was significantly associated with greater PTSD symptoms, B = .46, SE = .23, t = 2.20, p = .03, 95%CI[0.01, 0.91], and the association between tendency to seek support and PTSD symptoms was approaching significance, B = .44, SE = .23, t = 1.93, p = .056, 95%CI[−0.20,0.93]. There was no evidence that openness to psychological problems was associated with PTSD symptoms, B = .19, SE = .23, t = 0.92, p = .36, 95%CI[−0.23,0.66]. Cultural group did not moderate these associations; openness to psychological problems, B = −.15, SE = .44, t = .34, p = .73, 95%CI[−1.03,0.72], tendency to seek support, B = −.08, SE = .48, t = .17, p = .87, 95%CI[−1.02,0.86], indifference to stigma, B = .55, SE = .43, t = 1.28, p = .20, 95%CI[−0.30,0.40]. When we assessed the mediation model, we found that the indirect pathway from cultural group to tendency to seek professional help through indifference to stigma was significant, B = −0.46, SE = .25, 95%CI[−1.03,−0.05] (see Supplemental Table 1).

10. Discussion

This study examined cultural differences in social support in the context of PTSD. First, while the two cultural groups did not differ significantly in implicit and explicit social support, cultural group moderated the association between explicit support and PTSD symptoms. The association between greater explicit support and lower levels of PTSD symptomatology was observed for the Australian group but not the Malaysian group. Second, there were no significant cultural differences in perceived helpfulness of significant other, family and friends in supporting participants deal with the trauma. While perceived helpfulness of support from family and friends tended to be associated with lower PTSD symptomatology, cultural group moderated the association between perceived helpfulness of family support and PTSD symptoms – the association was significant for the Malaysian group but not the Australian group. Third, Malaysians in the non-mutual condition reported significantly greater distress than the Australian non-mutual group. However, Malaysians in the mutual condition reported significantly less negative affect and distress than the Australian group. Additionally, the Malaysian mutual condition reported significantly less distress than those in the non-mutual condition, but the two Australian groups did not differ significantly. Finally, contrary to our prediction, the Malaysian group reported that they were significantly more open to psychological problems than the Australian group. Indifference to stigma and tendency to seek support tended to be associated with greater PTSD symptomatology. Replicating Mojaverian et al. (2013), the indirect pathway from cultural group to tendency to seek professional help through indifference to stigma was significant.

There was evidence of cultural differences in the benefits of explicit social support. Consistent with past research (Ishii et al., 2017; Kim et al., 2006; Mojaverian et al., 2013; Taylor et al., 2004; Taylor et al., 2007), Australians appeared to benefit more from explicit social support than Malaysians. Past research indicates that explicit support is more strongly associated with health outcomes for those with European cultural backgrounds than those with Asian cultural backgrounds (Chiang et al., 2013). This study extends these findings to PTSD. This finding is important as clinicians often encourage trauma survivors to gain explicit support from their social networks (Yang et al., 2015). While this has benefits for those from European cultural backgrounds, it may have limited benefits for those from Asian cultural contexts. This may reflect Malaysians, perceiving explicit social support as negatively impacting on interpersonal relationships, burdening others or associated with losing face (Kim et al., 2008; Wang & Lau, 2015). However, further research is needed to explore these proposals, particularly as the current study was cross-sectional.

While explicit support was not associated with PTSD symptomatology for the Malaysian group, there was evidence that explicit support sought from a mutual (vs. non-mutual) source may be less distressing. Our findings align with the findings of Wang and Lau (Wang & Lau, 2015). In Asian cultures, participating in mutual support may be perceived as contributing to the welfare of the relationship and counteract any burden on one’s partner that may result from requesting explicit support (Wang & Lau, 2015). Thus, the distressing nature of explicit support seeking for Malaysians may be buffered by the perception of mutuality in a relationship, while for Australians the broader context of the support relationship may not substantially influence experiences of that support (Wang & Lau, 2015).

In contrast to previous research, we found no evidence to indicate that Malaysians found help from close others as less beneficial than Australians (Kim et al., 2006; Wang & Lau, 2015). Our finding supports past research that family plays an influential role in care and treatment of those from Asian cultural backgrounds (Lin et al., 1991; Lin & Cheung, 1999; Park & Chelsa, 2010). Additionally, while substantial research highlights systematic cultural differences in seeking professional assistance, we found no evidence that Malaysians reported lower help-seeking tendencies than Australians (Hwang, 2006). Indeed, Malaysians reported greater psychological openness than Australians. Our mediation analyses replicated previous research (Mojaverian et al., 2013), thereby providing further support that attitudes to help-seeking may play a role in understanding cultural differences in willingness to seek professional help (Mojaverian et al., 2013). Further research is needed to examine cultural influences on professional help-seeking in the context of PTSD, with studies exploring additional constructs and considering further cultural variables underpinning professional help-seeking tendencies.

Theoretically, PTSD models that highlight the key role of social support in the development, maintenance and treatment of PTSD should consider cultural influences. These models have been developed independent of cross-cultural models and non-clinical research that highlights culture influences whether support is sought, how it is sought, whom it is sought from, and the benefits and costs of seeking support (Kim et al., 2012). This study demonstrated that these points need to be considered in the context of PTSD. Thus, as PTSD models continue to evolve, cultural influences need to be included. There are also emerging clinical implications. First, for Australians explicit support appears beneficial but suggesting Malaysians seek explicit support may have limited benefits. Second, it is important to consider the importance of family in the posttraumatic recovery of Malaysians. Third, while for Australians requesting support from mutual or non-mutual providers may have little relevance, it seems more beneficial for Malaysians to seek support from those who they have previously supported.

Limitations include the study being cross-sectional and thus, causality cannot be inferred. Given the cross-sectional design there are other possible interpretations of our findings, such as higher levels of PTSD being a motivator for support seeking and the bi-directional relationship between PTSD and social support (i.e. PTSD can also erode social support). Thus, further studies assessing causality and longitudinal research are needed. Second, the study included a community sample, and the generalizability of findings to a clinical sample still needs to be examined. Nevertheless, over half of each cultural group met PTSD clinical cut-off. Third, while we considered self-construal, it is important to recognise that Malaysian and Australian cultures differ in several other respects (e.g. religion, holism/analytic thinking) that could influence findings and should be considered in further research. Additionally, while the current study replicated methodological approaches used in the non-clinical cross-cultural literature, further research should consider the specific mechanisms underpinning cultural differences. Finally, while the groups did not differ in identified index traumas, trauma type (e.g. childhood, interpersonal) may have influenced findings.

In sum, this study found that cultural group moderated the association between explicit support and PTSD symptoms, whereby the association between greater explicit support and lower levels of PTSD symptomatology was observed for the Australian group but not the Malaysian group. Second, cultural group moderated the association between perceived helpfulness of support from family and PTSD symptoms, whereby the association was significant for the Malaysian group but not the Australian group. Third, for Malaysians non-mutual support appeared to be associated with significantly greater distress than the Australian group, but for Malaysians mutual support was associated with significantly less negative affect and distress than the Australian group. Finally, the Malaysian group reported that they were significantly more open to psychological problems than the Australian group. This initial study examining social support and culture in the context in PTSD highlights a need for further research in this area.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are openly available in Open Science Framework at https://osf.io/z8ekf/.

Additional information

Funding

This research was supported by a grant from Mental Research Institute.

Notes

1 A similar pattern of results to that presented emerged when we used the complete data set.

2 PTSD symptoms were included as a covariate given the association between PTSD symptoms and negative affect. A similar pattern of results emerged when the analyses were conducted without the covariates.