Abstract
Asian American and Pacific Islanders are one of the fastest growing and most diverse groups in the United States. Yet, they are often aggregated as a single group, masking within-group differences in rates of disease and demographic characteristics commonly associated with elevated health risk. While more than four decades have passed since the Khmer Rouge genocide, Cambodians continue to experience trauma-related psychiatric disorders, including post-traumatic stress disorder and major depression. Funded by the California Department of Public Health Office of Health Equity, the Community Wellness Program (CWP) aimed to reduce mental health disparities among Cambodians in Long Beach and Santa Ana, California, using community-defined approaches. The 6-month program comprised community outreach, educational workshops, strengths-based case management, and social and spiritual activities. Our study aimed to examine the effects of the CWP on trauma symptoms. Program evaluation followed an incomplete stepped wedge waitlist design with two study arms. A linear mixed models analysis revealed that participants reported fewer trauma symptoms as a result of participation in the CWP and that participants experienced fewer symptoms over time. This is an especially important finding, as trauma can lead to long-term individual health effects and to social and health repercussions on an entire cultural group by way of intergenerational trauma. As the number of refugees and displaced individuals continues to grow, there is an urgent need for programs such as the CWP to prevent the lasting effects of trauma.
Keywords
Among racially and ethnically minoritized groups in the United States, Asian American and Pacific Islanders (AAPIs) are one of the fastest growing (Budiman & Ruiz, 2021) and most diverse (Sharif et al., 2019). Yet, they are often aggregated as a single group, masking within-group differences in rates of disease and demographic characteristics commonly associated with elevated health risk (Holland & Palaniappan, 2012; Kiang et al., 2017). The presumed homogeneity also glosses over the Western colonization of (Booth, 2007) and political unrest in many AAPI nations, and their effect on the socioeconomic trajectory of inhabitants (Sangalang et al., 2019) that led to differential starting points in the United States, resulting in significant socioeconomic, educational, and health disparities within the AAPI population. This article presents findings from the evaluation of the Community Wellness Program (CWP), a California Reducing Disparities Project funded by the California Department of Public Health Office of Health Equity to reduce mental health disparities among Cambodian immigrants and refugees in Long Beach and Santa Ana, California. We discuss the CWP within the context and needs of the Cambodian community as they relate to the development and delivery of program activities.
Cambodians In The United States
In human developmental terms, Cambodian immigrants and refugees are among the most traumatized populations in the United States (Berthold et al., 2014). Following the Vietnam War and the withdrawal of U.S. troops from Southeast Asia in 1975, a radical communist group, the Khmer Rouge, was determined to demolish any Western influence and begin a new classless society (Kiernan, 2008). From 1975 to 1979, an internal genocide extinguished nearly 2 million people, or one third of the Cambodian population (Heuveline, 2015; Quintiliani, 2009). Survivors experienced forced separation from family, and confinement at work camps, where they faced starvation, beatings, torture, and disease (Abueg & Chun, 1996). Cambodian refugees suffered additional trauma from the indeterminacy of flight and camp life (Mollica et al., 1993, 1997; Savin et al., 1996), and from resettling in a foreign country, such as loss of a familiar way of life, social network, and social status, and a lack of marketable skills among men forced women to join the workforce, thereby disrupting traditional roles within the family structure (Chung et al., 1998). Downward mobility and loss of family support (Chung & Kagawa-Singer, 1993), resettlement in poor, and high-crime neighborhoods (D’Anna et al., 2018) further add to existing trauma. More than four decades later, many Cambodian refugees still experience the effects of the Khmer Rouge. A study of Cambodian refugees in Long Beach, California, found that Cambodians continued to have high rates of trauma-related psychiatric disorders; 69% met the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) diagnosis criteria for post-traumatic stress disorder (PTSD) and/or major depression (59% with PTSD only and 37% with both PTSD and depression) (Wong et al., 2015).
Despite having poor mental health, few Cambodians utilize mental health services due to stigma that stems from the belief that mental illness is a reflection of personal weakness and an inability to exercise willpower, and therefore, seeking help can bring shame to oneself and one’s family (Aberdein & Zimmerman, 2015; Thikeo et al., 2015). Other barriers to mental health services include lack of transportation, high cost, cultural and linguistic discordance of existing services, lack of awareness of available services, and the inability to navigate the U.S. health care system (Sue, 1999; Wycoff et al., 2011). Thus, more culturally responsive programs are needed to address mental health needs of the Cambodian community.
Method
Community Wellness Program
The aim of the CWP was twofold: (1) reduce mental health disparities among Cambodians in Long Beach and Santa Ana, California, and (2) generate evidence on the effectiveness of a culturally centered program that incorporated community-defined practices. The CWP drew upon the strengths of previous programs that serve the Cambodian community (e.g., community engagement in program design, participant empowerment activities; Berkson et al., 2014; Grigg-Saito et al., 2008, 2010; Lee et al., 2016), but also recognized the effectiveness of long-standing spiritual and traditional practices that have yet to be examined scientifically. The CWP followed a strength-based approach that empowered participants through shared knowledge and capacity-building activities, such as case management and goal setting. Central to this approach is resilience and the recognition that refugees are active survivors with an innate ability to heal and thrive (Silove, 2013). The CWP was implemented by a collaborative of five Cambodian-serving community-based organizations, Cambodian Association of America, Families in Good Health, Khmer Parent Association, The (Cambodian Family, United Cambodian Community), allowing the program team to leverage the strengths and resources of each organization. Four organizations implemented program activities, while one organization was responsible for outreach and engagement only. Program content was based on the Theory of Planned Behavior (Ajzen, 1991) and the Social Cognitive Theory (Bandura, 1986), in that activities were designed to increase knowledge and awareness of mental health, reduce mental health stigma, and increase self-efficacy to change individual attitude and community subjective norms concerning mental health and perceived control to improve their physical and mental wellness.
We developed the CWP with input from community stakeholders composed of community-based organizations that have been serving Cambodians for more than 40 years. Key personnel (e.g., Executive Directors and project managers) at each organization met to discuss gaps in services and challenges to utilizing existing services. Community partners emphasized the importance of spirituality in Cambodian culture and the history of its use to promote wellness. As such, the CWP included traditional practices, oral history, and use of spiritual healers to more appropriately serve the Cambodian community. The CWP was sensitive to past and present trauma specific to Cambodian immigrants and refugees, and integrated key cultural elements (e.g., collectivism) into its strategies. Community outreach and engagement activities centered on social, physical, and emotional wellness (Figure 1) because partners shared that a focus on mental health alone would be a barrier to recruitment due to the stigma that exists within the Cambodian culture (Aberdein & Zimmerman, 2015; Thikeo et al., 2015).

Community Wellness Program Model
To be eligible to enroll in the CWP, individuals had to (1) self-identify as Cambodian, (2) be 18 years and older, (3) live in the greater Long Beach and Santa Ana areas, and (4) commit to the program for 6 months. Participants were enrolled in CWP in consecutive cohorts of approximately 100 participants (25 at each partner organization) for a period of 6 months.
We hired bilingual and bicultural Community Health Workers (CHWs) from within the Cambodian community to facilitate all CWP activities. CHWs received training on community outreach and engagement, workshop facilitation, trauma-informed care, case management, and data collection activities. CHW training took place in 2- to 3-hour sessions every 4 to 6 weeks during the first 2 months of the program. Training was delivered by the CWP lead agency, Cambodian Association of America, and/or a third-party technical assistance provider. Additional CHW training and/or support was provided throughout the CWP to address unforeseen issues (loss of participant due to suicide, staff burnout, etc.). As required by the California State University, Long Beach (CSULB) Institutional Review Board (IRB), CHWs also had to complete an online CITI training on human subjects research.
The CWP included four components: (1) outreach and engagement, (2) educational workshops, (3) strength-based case management, and (4) social and spiritual activities. Community outreach took place at Buddhist temples, community centers, cultural and community events, and at Cambodian-owned businesses to engage the larger Cambodian community in a dialogue about the importance of overall wellness and to dispel myths and misconceptions about mental illness. CHWs also utilized social media for outreach, which enabled the CWP team to reach an even greater number of individuals. While in the program, participants took part in a series of six educational workshops that integrated culturally relevant references and examples, and culturally appropriate strategies to promote physical and mental wellness. CHWs provided case management and worked with each participant to set short- and long-term goals. CHWs checked in with participants weekly to provide support and referrals, as needed. Social activities such as field trips, in-house events, arts & crafts, dance classes, and weekly visits to a community garden decreased social isolation and increased social connectedness while serving as opportunities for participants to share personal stories. Most Cambodian refugees follow Buddhism (Hsu et al., 2004), and as such, CHWs organized monthly in-house water blessings and visits to Buddhist temples to promote spiritual wellness. For non-Buddhist participants, yoga and group meditation were additional activities to promote spiritual wellness. Although attendance was not mandatory, we encouraged participants to attend at least one social/spiritual activity per week, in addition to bi-monthly workshops and weekly case management sessions.
CWP Evaluation
The Center for Health Equity Research at California State University, Long Beach (CSULB) led the CWP evaluation and worked collaboratively with CWP partners on evaluation design. CWP partners played an instrumental role in developing data collection instruments and in identifying the appropriate methods of data collection. The evaluation integrated the community’s view of mental health, linguistic needs, literacy level, history of trauma, and trust in research, as these factors informed assessment measures and participants’ ability and/or willingness to share information. The CSULB IRB reviewed and approved the study protocol and all data collection instruments.
The overall CWP evaluation examined how participation in the program contributed to changes in physical and mental health, and in alleviating symptoms of trauma, although this article presents changes in trauma symptoms only. Program implementation was staggered across four waves, and as such, the evaluation followed an incomplete stepped wedge waitlist design. Half of the participants engaged in program activities upon enrollment, while the other half waited 3 months before beginning the intervention (Figure 2).

Schematic Illustration of Study Design
Participants completed a baseline assessment upon enrollment, and a follow-up assessment at 3, 6, and 9 months. Participants who waited 3 months before engaging in program activities completed two baseline assessments, one upon enrollment and another at 3 months, which added an additional baseline control measurement point and increased the number of crossover points, thereby increasing the power of the design (Baio et al., 2015). Follow-up assessments for this group occurred at 6, 9, and 12 months after enrollment. Baseline and follow-up surveys were available in English and Khmer, and were self-administered on paper in a group setting. CHWs were present to clarify survey items, to ensure minimal participant interaction during survey completion, and to provide one-on-one assistance to participants with low literacy or visual impairment. The baseline assessment consisted of 82 items and took approximately 90 to 120 minutes to complete, while the follow-up assessment consisted of 61 items that took approximately 45 to 60 minutes to complete.
Measures
Demographic characteristics included gender, age, marital status, and educational level. English fluency, nativity, years in the United States, refugee camp experience, and social isolation were also included in the baseline assessment. We used the Harvard Trauma Questionnaire (HTQ; Mollica et al., 1992) to assess trauma. However, to prevent re-traumatization, we used Part 4 of the questionnaire to assess trauma symptoms only and not the specific events that participants experienced. Part 4 of the HTQ comprises 16 items, each assessed using a 4-point Likert-type scale, ranging from not at all to extremely, to examine how often participants are bothered by individual symptoms of trauma (recurring nightmares, trouble sleeping, feeling jumpy, etc.) in the past week. Mollica et al. suggest an average score of 2.5 and above as indicative of symptomatic for PTSD (Mollica et al., 1992). However, our goal for using Part 4 of the questionnaire was not to examine prevalence of PTSD among participants but rather to examine changes in trauma symptoms as a result of program participation. As such, we operationalized “trauma symptoms” as a continuous variable instead of a dichotomous variable.
Data Analysis Plan
Our analysis only included participants age 50 and older, as those younger than 50 may not have directly experienced the Khmer Rouge or were too young to remember. Doing so did not significantly reduce the sample size, as 77.2% of CWP participants (N = 304) were 50 years and older.
We used SPSS V.28 to perform independent-samples t tests to examine the relationship between trauma symptoms and dichotomous variables, and bivariate correlations to examine the association between trauma symptoms and continuous variables. The study data were clustered or hierarchical by cohort and partner organization, thereby violating the assumption of independence among data points in traditional linear regression models. As such, we used linear mixed-effect modeling (LMM) to analyze the effects of the intervention on trauma symptoms. In LMM, random effects are used to model the correlation between individuals in the same cluster, thereby accounting for clustering effects (Hussey & Hughes, 2007). We entered cohort number and partner organization into the model as random effects, as participants were enrolled in one of four different cohorts and took part in program activities at one of four different organizations. We used an AR(1) covariance structure for participants, and a variance compound structure for organizations and cohorts. We entered time into the model as a fixed effect to control for secular trends (Hemming et al., 2015). A condition variable was coded 0 if participants received no intervention prior to the 3-month assessment and 1 if they received at least 3 months of the intervention prior to the assessment. The final assessment was administered 3 months after program completion, and as such, we coded each completed assessment as 0.
Results
The analytic sample comprised 304 CWP participants. Sample characteristics are presented in Table 1.
Sample Characteristics (N = 304)
Note. HTQ = Harvard Trauma Questionnaire; PTSD = post-traumatic stress disorder.
A score >2.5 on the HTQ is indicative of symptomatic PTSD.
The independent-samples t tests and bivariate correlations revealed significant differences in marital status and time in the United States in relation to trauma symptoms, such that those who were married reported fewer trauma symptoms, t(274.9) = 2.20, p < .05, and more time in the United States was associated with increased trauma symptoms (r = .35, p < .001). In addition, social isolation was positively correlated with trauma symptoms (r = .43, p < .001; Table 2).
Key Variables Associated With Trauma Symptoms
Note. NS = not significant.
Bivariate correlation
When examining the effects of the intervention on trauma symptoms, findings from the LMM revealed that the condition variable (i.e., program participation) was found to predict trauma symptoms (B = −0.10, p < .05), suggesting that participants reported fewer trauma symptoms as a result of participation in the CWP (see Table 3). Time was also found to be a statistically significant predictor of trauma symptoms (B = −0.05, p < .05),. This finding suggests a dose-response relationship, such that participants experienced fewer symptoms over time while enrolled in the CWP.
Linear Mixed-Effect Model Examining Effectiveness of Intervention on Trauma Symptoms (N = 304)
Note. NS = not significant.
Discussion
Results from our baseline assessment of trauma indicate that older Cambodians are in particular need of mental health services. We found that almost one quarter of participants in our study met the cutoff score indicative of PTSD, compared with 6% of U.S. adults (Kessler et al., 2005). In addition, almost one third of participants spent time in a refugee camp prior to immigrating to the United States, three quarters of participants never completed high school, and the vast majority did not speak English well or at all. These factors contribute to the difficulty in engaging this population in mainstream mental health services and warrant the need for more culturally centered programs.
Results from our study showed that participation in the CWP contributed to fewer trauma symptoms and that participants experienced fewer symptoms over time while enrolled in the program. We attribute these findings to two unique aspects of our program. Our CHWs focused on trust building with program participants during the initial stages of the program, which led to a slow unfolding of participants’ trauma and an increased willingness to engage in program activities over time. The inclusion of CHWs for the provision of mental health services is a recommended strategy to address disparities in mental health treatment (Barnett et al., 2021). Our CHWs delivered program activities using a trauma-informed approach, which involves awareness of trauma in the population, its effects on mental health, and the need to avoid re-traumatization during care delivery (Javakhishvili et al., 2020). CWP activities were situated within the context of pre- and postmigration experiences. Activities and services were held in safe spaces that allowed participants to share their experiences with past and current trauma openly and honestly. These spaces fostered shared storytelling and collective healing. This is especially important, as unresolved trauma can be passed on to subsequent generations and have lasting repercussions, often manifesting as social and health disparities within specific cultural groups (Ballard-Kang, 2020; Kirmayer et al., 2014; Wieling et al., 2020).
Lessons Learned
Implementation of the CWP provides valuable insight for future programs aimed at reducing health disparities within immigrant and refugee communities, particularly those who have experienced extreme trauma. Providing services and program material by trusted bilingual and bicultural staff using a trauma-informed approach was vital to program success. Transportation was a significant barrier to program engagement. We addressed this barrier by encouraging participants to carpool and by providing transportation via agency vans or private vehicles. The transition of program services to online platforms during the height of the COVID-19 pandemic eliminated the transportation barrier. However, low computer literacy was a new barrier that we had to address with alternative methods of service delivery (e.g., providing case management and educational workshops by telephone).
Isolation increased during the pandemic and even more so for older participants, who were not able to use social media or Zoom to connect with friends and family. The rise in Asian hate crimes during the COVID-19 pandemic caused many AAPIs to live in fear (Tessler et al., 2020) and we observed the same among older participants, as many refused to leave their homes when encouraged to go for walks. For participants who lived alone, the CWP was their only social connection, and as such, CHWs often received telephone calls outside of work hours. While CHWs were instructed to set boundaries for their own well-being, they continued to answer evening and weekend telephone calls to avoid damaging the trusting relationship that they worked so hard to build. We recommend that future programs establish guidelines and expectations at the onset to avoid this problem.
Strengths and Limitations
The evaluation of the CWP is not without limitations. In designing the evaluation, it was important for us to be mindful of the unique characteristics of the Cambodian community. CWP partners provided valuable insight such as the challenges of conducting a randomized controlled study due to the close-knit nature of the Cambodian community, the mistrust of government authority among Cambodian refugees that would prevent accurate data collection using computer tablets (i.e., participants would be wary of how their data are used), and the need that exists within the community that would make withholding services unethical. While a quasi-experimental design may limit our ability to conclude a causal association between program participation and outcome variables, we believe that it was appropriate for our study, as this design honors and serves the community by ensuring that everyone could have participated in and benefited from the CWP if they wished to enroll. Another limitation of the evaluation relates to the COVID-19 pandemic, during which all program activities were delivered online. This required participants to connect to online platforms such as Zoom. While every effort was made to find alternative methods of service delivery (e.g., workshops via phone), inability to connect online may have affected the evaluation, as not all participants had equal exposure to the intervention.
A strength of the CWP is the number of organizations that came together to leverage their strengths and resources to serve the Cambodian community. This was the first time that all five organizations worked collaboratively and the team continues to work together to advocate for on behalf of the Cambodian community. Perhaps the greatest strength of the CWP was its inclusion of CHWs who shared similar lived experiences as CWP participants, thereby allowing CHWs to better identify and address barriers to program engagement. In addition, having CHWs assist with data collection strengthened the evaluation, as the participants were more comfortable answering sensitive questions. During the height of the COVID-19 pandemic, CHWs worked tirelessly to keep participants engaged in the CWP while also helping to meet immediate needs by distributing food and personal protective equipment. Doing so helped to gain participants’ trust, which further contributed to program engagement.
Conclusion
The CWP is the first project of its kind to examine the effects of community-defined practices on mental health among the Cambodian community. Using an evaluation design that took into consideration the unique characteristics of the Cambodian community, we found that culturally relevant, trauma-informed services provided by trusted program staff led to statistically significant improvements in trauma symptoms. The CWP filled a gap in service for the Cambodian community and may be adapted for other underserved immigrant and refugee populations who experience similar needs. As the number of refugees and displaced individuals forced to migrate worldwide due to war and political instability continues to grow, there is urgent need to find effective, culturally relevant ways to address trauma to prevent long-term individual health effects and to break the cycle of intergenerational trauma and its link to poor social and health outcomes in subsequent generations.
Footnotes
Authors’ Note:
The authors would like to thank members of the Cambodian community in Southern California who participated in data collection activities as part of the Community Wellness Program. Your strength and resilience continues to inspire us all. This study is part of an evaluation of the Community Wellness Program, a California Reducing Disparities Project funded by the California Department of Public Health Office of Health Equity.
