Introduction

People living with advanced and metastatic cancer have unique psychosocial needs and, with longer survival times due to treatment advancements, a greater focus on survivorship care is needed. Having advanced and metastatic cancer involves facing the possibility of incurable disease, with continued treatment and monitoring until the end of life [1]. The number of individuals living with metastatic cancer will increase to nearly 700,000 by 2025 in the USA alone [2]. Medically underserved or understudied populations which include racial and ethnic minority groups, people with lower socioeconomic status, underserved rural communities, sexual and gender minority groups, and people with disabilities are at higher risk for diagnosis at advanced and metastatic stages [3, 4]. Advanced and metastatic cancer survivors experience physical and psychosocial challenges resulting from long-term symptoms related to prolonged treatment [5]. The psychological impact of later-stage disease is different from earlier-stage cancer patients, including greater risks for adverse outcomes such as suicidality [6, 7]. Adverse outcomes may be prevented with psychosocial or behavioral interventions defined as “interpersonal or informational activities, techniques, or strategies that target biological, behavioral, cognitive, emotional, interpersonal, social, or environmental factors” to improve “health functioning and well-being” [8]. Effective psychosocial and behavioral interventions are needed to improve patient-reported outcomes, including quality of life (QOL), which can predict progression-free survival above and beyond tumor markers in advanced cancer [9].

Culture is vital in designing and implementing effective health-related interventions [10]. Culture is the dynamic framework underlying all human endeavors that shapes how members of a group collectively see and experience the world around them [11]. For all people, culture is a lens that shapes one’s priorities and goals as well as one’s experience and evaluation of others’ actions and social events [12]. Culture helps individuals navigate relationships with others, organizations, and the environments around them [13, 14]. Therefore, culture can significantly influence how individuals experience and live with cancer. For example, studies have shown the heightened importance of maintaining hope and including family in care discussions among Latino cancer patients compared to US-born White non-Latinos. However, healthcare providers face challenges initiating discussions about prognosis and palliative care with Latino patients and caregivers due to cultural preferences to focus on hope and optimism for a cure rather than palliative and comfort care discussions [15, 16]. Such challenges can be ameliorated by culturally aligned psychosocial interventions, leading to positive participant experience, and improved outcomes [17]. Further, previous meta-analyses have broadly demonstrated the benefit of cultural adaptations in psychological treatments and interventions such as psychotherapy [18,19,20]. Considering the growing number of advanced and metastatic cancer survivors, effective and culturally appropriate psychosocial or behavioral interventions are critical [9].

Cultural Incorporation in Psychosocial or Behavioral Intervention Research

Approaches to cultural incorporation in psychosocial or behavioral intervention research involve diverse methods. The most well-known methods use theoretically based adaptation models/frameworks including Bernal’s ecological validity model [21], and, generally, there are recommendations for the “what” and “how” of culturally adapting an existing intervention. Some examples for “what” to adapt in interventions include language, persons, metaphors, content, concepts, goals, methods, context [22], and concepts of distress, treatment components, and delivery [21]. Other frameworks describe the manner in which the interventionist would adapt to the recipient as part of the therapeutic interaction [23]. Cultural adaptations also involve iterative steps such as building a collaboration with the community of focus, assessing intervention interest and needs with community leaders, developing and piloting the adapted intervention, observing adaptations and collecting feedback [24], and incorporating community-based knowledge from interested parties [25].

Despite variation across suggestions for cultural incorporation, adaptation, and tailoring [26], there are commonalities. Firstly, intervention development can be conducted top-down or bottom-up. A top-down design is based on existing mainstream interventions and utilizes a universal approach assuming the original intervention is applicable to all subgroups as is or with cultural adaptations. Conversely, a bottom-up design utilizes a culture-specific approach which considers emic cultural theories of problem and change, utilizes community-generated data or tradition-based practices, and at times informs a novel intervention that may not yet be evidence-based [27]. Secondly, the cultural congruence considered may be at a surface or deep level [28]. Surface-level cultural match increases the acceptability or feasibility by focusing on the "observable social or cultural" behaviors [29], for example, the language or setting in which the intervention is conducted. On the other hand, deep-level cultural match increases engagement by focusing on shared understandings including how social, cultural, and historical environments influence behavior [29], for example, taking into consideration a cultural group’s distinctive practices for maintaining health, strong ties to religion, or historically negative relationship with the government or healthcare system. Thirdly, there are specific elements of the intervention outlined for adaptation such as content (presented as part of the intervention), process (how interventionist interacts with the recipient), and delivery (how intervention is delivered).

Cultural incorporation and adaptation are particularly relevant for understudied and underrepresented populations; however, the majority of previous behavioral health interventions have been developed predominantly through a European American lens or with Western, Educated, Industrialized, Rich, Democratic (WEIRD) populations based in North America [11, 30,31,32]. A 2020 National Institutes of Health (NIH) portfolio review of cancer survivorship research revealed a dearth of focus on advanced and metastatic cancer populations. Of the twenty-five studies focused on advanced and metastatic cancer survivorship, only nine studies centered on minoritized or medically underserved populations such as African Americans and Hispanic/Latinos, as well as older adults and low-income and socioeconomically disadvantaged groups [33]. Given the limited representation of understudied groups and non-WEIRD populations in behavioral health research, the goal of the present review is to focus on advanced and metastatic cancer populations outside of well-researched WEIRD countries while still prioritizing medically underserved and understudied groups (e.g., ethnoracial minorities, adolescents and younger adults (AYA), rural) in WEIRD countries [32].

This systematic review summarizes psychosocial and behavioral interventions incorporating culture, including those designed for understudied and underserved populations living with advanced and metastatic cancer. The specific aims of this systematic review are to (1) identify psychosocial or behavioral interventions with intentional cultural adaptations/components, (2) describe the approaches to cultural adaptation or developing culturally based interventions, (3) describe the progress for understudied/underserved populations and which groups remain understudied, (4) identify gaps and opportunities for psychosocial or behavioral intervention science for those living with advanced and metastatic cancer, and (5) examine the impact of these interventions on intended psychosocial outcomes if availability of data allowed so. This review is registered with the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42022375667).

Methods

Search Strategy

A psycho-oncology researcher (JK), public health researcher (KL), and medical librarian (HPG) developed a comprehensive search strategy based on the research question to identify relevant literature. Keywords were selected by using the PICO framework, medical subject headings database, and relevant literature during search preparation. A broad search strategy was used to identify all potentially relevant studies to the cancer population of focus. The search was primarily based on the “population” and “intervention” search terms as “comparison” and “outcome” search terms limited the results and exemplary studies would be missed. Final search terms consisted of MESH terms and free-text keywords combined with Boolean operators related to advanced and metastatic cancer, psychosocial and behavioral interventions, and cultural adaptation/tailoring.Footnote 1 The search was conducted November to December 2022 by HPG in PubMed, CINAHL, Scopus, PsycInfo, and Cochrane Library and completed on December 21, 2022. Review articles published in the past two years were manually searched for any missed articles. Results were imported into Covidence and duplicates were removed.

Inclusion and Exclusion Criteria

Manuscripts were screened using the following inclusion criteria: (1) centered on advanced and metastatic cancer survivors or a majority sample of advanced and metastatic cancer survivors, (2) described a psychosocial intervention design/adaptation intended to improve psychosocial or behavioral components related to QOL in advanced and metastatic cancer, (3) evidence of incorporating non-WEIRD cultural context OR understudied/underserved populations’ contexts in WEIRD societies in the intervention design/adaptation efforts, and (4) was available in English.

Screening and Selection

Screening was conducted by a group of trained reviewers composed of students, research specialists, and faculty. Titles and abstracts were independently screened for inclusion by at least two members of the reviewer group. Potentially relevant sources identified in the title/abstract stage were moved to the full text review stage. The full texts were assessed in detail against the inclusion criteria by one member of the authorship group and one additional reviewer. Reasons for exclusion at the full text level were recorded. Any disagreements that arose between the reviewers at each stage of the selection process were resolved by a third reviewer or discussed as a team with JK and BC until consensus was reached.

Data Extraction and Coding

Study design, sample characteristics, quantitative data, intervention details (e.g., format, setting, delivery), and cultural incorporation methods (i.e., use of adaptation model or framework, top-down or bottom-up design, and surface- and deep-level cultural congruence) were extracted by two independent reviewers within Covidence. Conflicts were resolved by a third reviewer or discussed with JK and BC until consensus was reached. Quantitative data points for extraction were identified in consultation with a biostatistician (ME).

Data Analysis

Qualitative Synthesis

Qualitative thematic synthesis was conducted per usual standards in health-related systematic reviews [34, 35]. All text under the “Results/Findings”, “Discussion” and “Conclusion” sections for 23 qualitative studies were imported into Atlas.ti for qualitative coding. Qualitative data from non-patient participants were included if the focus was on patient-oriented interventions. Two trained researchers (AP, JK) independently coded the imported qualitative data and evaluated the quality of each manuscript for sensitivity analysis [36, 37] simultaneously [38, 39]. Coders focused on answering the question “What are the cultural considerations/adaptations suggested for psychosocial or behavioral interventions in advanced and metastatic cancer?” and resolved coding disagreements with discussion. Codes were grouped into themes and subthemes. The stability of themes was re-examined with and without poorer quality studies determined by sensitivity analyses [36, 37] (see Supplementary Appendix B).

Meta-analytic Methods

Although limited, the study team identified an adequate number of studies reporting patient-focused quantitative data for exploratory meta-analyses. Two types of meta-analyses were completed: (1) the effect size for how much the main intervention group benefited for a psychosocial outcome vs. the control group with RCT and quasi-experimental reports and (2) the effect size for single-arm trials from pre-intervention to post-intervention. The first data point available post-intervention was utilized, as psychosocial changes are the most significant closest to the end of an intervention [40]. Analyses were conducted using R (v.4.2.1). A random-effects estimation model was used with Hedge’s g, which adjusts for small sample bias. Only studies adequately reporting means and standard deviations across groups and time points were included. Corresponding authors were contacted to obtain missing data points, but 12 studies were excluded from analyses due to nonresponse. Heterogeneity among studies was assessed using Cochran’s Q and I2 [41]. Publication bias was evaluated using Egger’s test of funnel plot asymmetry for outcomes with 10 or more studies, as recommended [42]. We conducted exploratory moderator analyses comparing the effect sizes of studies based on types of cultural incorporation, when there was at least four studies per subgroup as recommended [43].

Risk of Bias for Included RCTs and Single-Arm Trials

Two trained reviewers (KL, AP) independently assessed the methodological quality of the included studies using the JBI Critical Appraisal Checklist for RCTS [44] and the methodological index for non‐randomized studies (MINORS) [45]. Disagreements were resolved through discussion or by a third author (JK). No studies were excluded based on the quality appraisal score as recommended [44]. Ratings are in Supplementary Appendix B.

Results

A total of 86 reports were identified and included in the review; 35 randomized controlled trials, 14 qualitative studies, 11 single-arm studies, 8 quasi-experimental studies, and 7 protocols, 7 feasibility/acceptability trials, and 4 mixed-methods studies (see Fig. 1). Over half (n = 49, including 3 protocols) of the reports reviewed were conducted with populations in non-WEIRD countries, and 38 reports were conducted in WEIRD countries with ethnic minority and medically underserved populations. A total of 73 interventions were described, twelve of which were described in multiple reports. A majority of the studies were conducted in China (n = 20), USA (n = 17), Japan (n = 10), and South Korea (n = 4), and the most commonly used languages were Chinese, English, Spanish, and Japanese (see Supplementary Appendix C for full list).

Fig. 1
figure 1

PRISMA flow diagram

Full size image

Participant Characteristics

A total of 5981 participants were examined across the 80 reports,Footnote 2 and 4620 participants were identified as having stage 3, stage 4, terminal/end-stage cancer, and advanced and metastatic cancer. Four hundred fifty-six cancer survivors’ cancer staging was not described, and eight reports with a majority sample of advanced and metastatic cancer also included 152 participants with non-advanced and non-metastatic cancer. The remaining 753 participants included caregivers (n = 477), lay health workers (n = 195), healthcare professionals (n = 67), and community stakeholders (n = 14). The intervention and control groups had mean sample sizes of 37 (range 7–112) and 39 (range 10–111), respectively. Participants with a cancer diagnosis were 59.38 years old on average; 53.62% were women.Footnote 3 Caregivers were 53.81 years old on average; 75.98% were women. From available healthcare professional participant data,Footnote 4 84.00% were women. Twenty-three reports included race and ethnicity of participants with cancer where 815 were White (36.63%), 451 were Latino/a/x or Hispanic (20.27%), 442 were African American/Black (19.87%), 289 were Asian (12.89%) and 151 were reported as an aggregate Asian/Pacific Islander group (6.74%). The most common cancer types were lung, breast, gastrointestinal, and prostate; see Supplementary Appendix C for all cancer types. In line with prior research on advanced and metastatic cancer populations, the majority of reports in this review (n = 50) focused on people nearing end of life or in palliative care.

Intervention Characteristics (Settings, Type, Format)

The majority of reports recruited from clinical (e.g., palliative care centers, outpatient clinics) and non-clinical community settings (n = 56, 69.14%), followed by clinical inpatient settings (n = 25, 30.86%, e.g., hospitalization), and five studies (6.17%) from both outpatient and inpatient settings. Interventions conducted in non-clinical settings included the participant’s home (n = 15), web-based (n = 11), community center (n = 1), and 12 studies conducted in more than one setting.

The psychosocial and behavioral interventions included in this review were grouped into 12 intervention approaches (see Table 1), with the three most common being (1) dignity therapy, life review, and narrative-based (n = 23), (2) a mix of two or more approaches (n = 15), and (3) education/psychoeducation (n = 14).

Table 1 Intervention and patient participant characteristics
Full size table

Interventions were delivered face-to-face (n = 66, 76.74%), remotely (n = 10, 11.63%), and a hybrid of both (n = 6, 6.98%). Four qualitative studies (4.59%) evaluated community needs and preferences prior to adaptations and did not describe any intervention delivery methods. The most common remote methods of delivery were telemedicine (n = 6), instant messaging (n = 3, e.g., WeChat), and one mobile game.

Interventions were delivered by clinicians (n = 22, e.g., physicians, psychiatrists, nurses, social workers, dieticians), trained facilitators (n = 52, e.g., music therapists, yoga specialists, traditional Chinese medicine practitioners, mindfulness practitioners, research students, lay navigators, lay health workers), trained peer mentors and community volunteers (n = 5), and four studies included a team of clinicians and trained facilitators. For intervention details including refusal/attrition rates and intervention dosage, see Supplementary Appendix D.

Adaptation Models/Frameworks

Seven studies applied an established model/framework to guide cultural adaptation of existing interventions. The models/frameworks used include the ecological validity model (n = 5), cultural adaptation process model (n = 2), and sunrise model (n = 1). One report applied both the ecological validity model and the cultural adaptation process model.

Cultural Considerations/Adaptations (Table 2 and Supplementary Appendix E)

Table 2 Included psychosocial or behavioral interventions in advanced and metastatic cancer by study type
Full size table

There were 48 reports originating from non-WEIRD countries (46 from Asian countries, one from Nigeria, and one from Puerto Rico). Of these studies, 28 utilized interventions developed within WEIRD cultural societies. Over half of those studies (n = 16) adapted the intervention beyond language to incorporate local cultural elements, while all others only modified the language in which the intervention was delivered. Across all included studies, 44 interventions described additional cultural modifications compared to 28 interventions focused solely on language match. Majority of interventions were top-down (n = 61) with 11 interventions using a bottom-up approach. Bottom-up studies include interventions such as combined Naikan and Morita therapies, which are grounded in Buddhism and Eastern psychology [46].

The most common surface-component adaptations to match the target population were (1) language (e.g., written materials in native language, bilingual speakers; top 10 languages used were Chinese, English, Spanish, Japanese, Italian, Danish, Korean, Dutch, Hindi, and Portuguese); (2) delivery (e.g., face-to-face vs. telephone calls, WeChat); and (3) culturally relevant content (e.g., acupressure and tai chi in traditional Chinese medicine, music). The most prevalent examples of deep structure adaptations of culture were (1) highlighting the importance of family inclusion in caregiving and decision-making (filial piety, familismo); (2) the inclusion of Asian values and cultural norms of Eastern philosophies (e.g., Confucianism, yin and yang) and religions (e.g., Buddhism, Taoism), such as avoiding taboo topics (e.g., death) or concealing a terminal diagnosis; and (3) incorporating Hispanic/LatinoFootnote 5 values of establishing trust (confianza), personal relationships (personalismo), and about death and religion (fatalismo/spiritualismo).

Qualitative Synthesis Results

Eleven descriptive subthemes formed four overarching themes about how to incorporate culture in psychosocial interventions for advanced and metastatic cancer starting from introduction of the intervention to long-term implementation: (1) promoting engagement through trust; (2) matching the intervention language and content; (3) cultural attunement at the individual level; and (4) preparing for successful implementation (Fig. 2). Themes remained the same (i.e., low risk of bias) without three lower quality studies from sensitivity analysis (Supplementary Appendix B).

Fig. 2
figure 2

Qualitative synthesis themes

Full size image

Promoting Engagement Through Trust

To introduce psychosocial interventions for advanced and metastatic cancer populations less aware of psychological distress or need, a basic foundation of trust for both the intervention and the interventionist must first be established. People more easily overcame stigma related to their cancer diagnoses and difficult topics when building rapport with their interventionist face-to-face (for Hispanic/Latino and US Chinese, Taiwanese, Singaporean patients) (Fig. 3).

Fig. 3
figure 3

a Forest plot of RCT effect sizes (QOL). b Forest plot of RCT effect sizes (eudaimonic well-being). c Forest plot of RCT effect sizes (distress). d Forest plot of RCT effect sizes (anxiety)

Full size image

After trust had been built, the intervention could be introduced and preferably in a structured manner. Focusing on solving physical symptoms earlier in the intervention allowed for further discussions of additional mental health support, especially in Asian populations where the expected outcome may be reducing symptom burden. A trusted figure is recommended for discussing mental health-related topics, such as a nurse in the Singaporean context. Once participants agree to engage in an intervention, studies suggest strategically timing the introduction of taboo topics to maintain rapport:

If you have not build rapport with the patient yet and they are just coming to terms with their diagnosis I think advanced care planning may be just a bit too much to handle cause they don't know where they are headed now… later on maybe after third or fifth follow-up…then it's fine to bring it up. [Healthcare Worker, Singapore] [48]

Family or caregiver involvement was seen as a positive method to increase receptivity across many cultures, but even in family-centric cultures, family involvement could be a barrier for patients to participate or be able to freely express their experience with cancer. In a life review program, a participant described how having a family member present influenced how much they wanted to share:

My daughter’s presence influenced our talks to a degree. I didn’t want to talk about the difficulties of my life in front of her. These experiences would make her feel sorry for me. [Participant, Hong Kong] [49]

Matching the Intervention Language and Content

When it came to the core components of the intervention itself, there were suggestions across all studies to adjust the wording and content to match the recipient’s cultural norms and beliefs. For some interventions for Asian populations, this meant couching the intervention content in philosophical traditions and religious beliefs (Confucianism (filial piety and social roles), Taoism, Buddhism, Catholicism). In a meaning-centered psychotherapy intervention for Chinese in the USA, infusing Confucian virtues like the respect for parents and elders was critical:

They have this sense of duty to take care of their parents and to respect them. If you don’t emphasize those values, it will be very hard for [MCP] to be culturally relevant and to get buy-in. [Community Leader, U.S. Chinese] [50]

Revising the wording not only refers to the language of the intervention but presenting less familiar concepts in a manner that matches the participant’s cultural context. In a dignity intervention for Danish patients, participants had difficulty understanding terms for dignity and personal success due to the unacceptable nature of self-praise within their culture [51]. More culturally acceptable phrasing was needed to discuss meaning, end-of-life, or death.

Cultural Attunement at the Individual Level

Beyond direct efforts to adjust the language and content of the intervention itself, studies described the importance for interventionists to be broadly aware of cultural group beliefs in order to be attuned with individual-level variations in how much cultural norms affect cognitive and behavioral responses during the intervention. For example, studies focusing on interventions for Asian populations commented on the cultural value of not burdening family members. By understanding the pervasiveness of this cultural value and assessing its importance to each recipient, interventionists may recognize cases where less expressivity does not equal a lack of supportive needs:

Chinese cancer patients…just bear the burden…on their own. They don’t want to create trouble for other people. But in this process, they bury things in their heart, and it’s not good for their health or for their family. [Participant, US Chinese] [52]

Cultural norms also suggest how individuals interact with authority figures, such as a doctor or the interventionist, especially for Asian or Hispanic/Latino cultural groups’ norms of when to engage respectfully or what topics to bring up; the degree to which this affects the intervention should be assessed on an individual basis. Similarly, culturally sensitive topics, such as sexuality in Singapore, could occur earlier for younger patients if it is relevant during interactions. Additionally, interventionists may need to tailor for each family. For instance, in Singaporean culture, concealing the gravity of diagnosis from the patient is normative for preventing worry [48]. Understanding how much the participant actually knows of their diagnosis and their family wishes will help the interventionist tailor the intervention.

Interventionists can also leverage upon the recipient’s cultural contexts to further enhance the effect of the intervention. For interventions in immigrant communities, this may mean being prepared to acknowledge each participant’s unique immigration history and discuss it in a strength-based manner. For example, in a meaning-centered psychotherapy intervention, interventionists were advised to be aware of patients’ immigration histories to reframe attitudes toward end of life:

It’s a survivor culture. Chinese participants have gone through lots of wars, lost lots of things. We try to frame [end-of-life preparation] as hoping for the best, planning for the worst. [Interviewee, US Chinese] [50]

Preparing for Successful Implementation

The authors described what they observed or thought was important for interventions to be feasible and better implemented in the future. Practical aspects of working with people with advanced and metastatic cancer were described, such as not always being able to complete the full intervention as intended due to its length, physical fatigue, or the impracticality of a more intensive intervention. Caregivers or family members were often not able to afford the time to provide reliable transportation for the patient to fully participate. In other cases where interventions were digitally adapted, such as using an electronic tablet to deliver a palliative care patient navigator and counseling intervention, some Latino/a participants did not have access to technology or did not know how to use the technology [53].

Overall, the authors described a need to design interventions that are more accessible and flexible to those most underserved and facing more barriers (limited language proficiency, lower education/literacy, lower health awareness, limited finances/insurance) to retain participation. Suggestions included keeping the duration of the intervention more flexible and making adjustments to the method of intervention delivery such as providing video conferencing, telephone intervention, or home-based activities.

The authors also mentioned that some of their interventions lacked the full support needed from their organizational structure (most often in healthcare) and that larger-scale implementation may not be possible. Many healthcare professionals, including nurses, who were already limited in their time with patients found it difficult to also deliver an intervention:

In fact, our routine workload is already very busy and I (now) have to spend more time to do this [advance care planning]. I am actually in the dilemma of how long I should engage in the dialogue while at the same time I have to complete all my other daily assignments. [Hospice Nurse, China] [54]

This suggested that healthcare setting policies and care structures need to change to benefit advanced and metastatic cancer patients; leveraging existing healthcare workers may not be the best solution within current systems.

Meta-analysis Results

Of the 86 reports included in the review, 19 RCTs, 4 quasi-experimental studies, and 10 single-arm trials reported data sufficient for meta-analysis. Table 3 parts a and b present mean effect sizes for the intervention outcomes. Outcomes showing improvement compared to the control condition were QOL (g = 0.84) or the “subjective measurement of an individual’s sense of well-being and ability to enjoy life” [55], eudaimonic well-being (g = 0.53) or the positive experience of living one’s life with a sense of meaning and to one’s fullest potential [56], distress (g = −0.49), and anxiety (g = −0.37). When examining RCT effect sizes between studies with only a language adaptation vs. those evidencing more cultural adaptations, the latter contributed to significantly better outcomes compared to controls only for QOL. Similarly, studies incorporating both surface and deep structure cultural incorporation vs. only surface structure resulted in significantly better outcomes compared to controls only for QOL (Table 3 part c). For the single-arm trials, pre-to-post-intervention, significant reductions occurred for anxiety, physical symptoms, and depression. Only one single-arm study was identified having a medium risk of bias (Supplementary Appendix B); however, all studies with available data were included as recommended [44]. Egger’s test results (Table 3 part a) point to some risk of publication bias; however, publication bias is less likely because small studies generally contributed to lowering meta-analysis effect size compared to larger studies, or there was no significant effect [57, 58]. See Supplementary Appendix F for additional forest plots and list of studies.

Table 3 Meta-analysis effect sizes by psychosocial outcome
Full size table

Discussion

This systematic review synthesized the literature on psychosocial and behavioral interventions in advanced and metastatic cancer that incorporate culture, particularly for understudied and underserved populations. Various approaches for incorporating culture into interventions were identified, in addition to the gaps and opportunities for psychosocial and behavioral intervention science in advanced and metastatic cancer. Similar to prior reviews reporting the benefits of psychosocial or behavioral interventions in advanced and metastatic cancer [59], this systematic review demonstrates the overall benefit of psychosocial interventions that incorporated the culture of their targeted intervention population, particularly for QOL. The present review bolsters what is known by extensively documenting the forms of cultural incorporation in psychosocial and behavioral interventions in advanced and metastatic cancer and provides directions for moving forward.

Past reviews in cancer-related psychosocial or behavioral interventions were limited to specific types of interventions and outcomes and/or were limited to RCTs. Due to the importance of this work and how understudied these populations are, the present review applied inclusive methods to identify studies that might otherwise have been missed. First, the definition of psychosocial or behavioral interventions was broadened to include “nonmedical interventions intended to modify psychological, social, and behavioral processes, including cognition and emotions” [59] (e.g., complementary alternative medicine, rehabilitation medicine, native/ethnic traditions-based). A similar approach was applied in the conceptualization of the population of interest with the goal of including advanced and metastatic cancer populations globally and attending to research on medically underserved and understudied groups (e.g., ethnoracial minorities, younger/AYA, rural in WEIRD countries) [32]. Although comprehensive, a consequence of an inclusive search strategy is the high level of heterogeneity across the included studies. As such, substantial caution is needed when interpreting the present meta-analytic results. For example, heterogeneity across studies potentially shrouds notable effects that may be found in future reviews. This can be demonstrated via sensitivity analysis in our meta-analysis for depression, where exclusion of a tai chi study [60] results in a significant effect size between control and intervention conditions because most other studies were psychosocial or spiritual in nature that likely shift more proximal mechanisms of depression. Further, the available data is likely underpowered to detect moderation effects across psychosocial outcomes. Nonetheless, cultural modifications beyond language match appeared beneficial for QOL. Although heterogeneity of studies and underpowered data limit our quantitative findings, there is still much to learn from examining the body of work conducted thus far on this topic.

To the authors’ knowledge, no other systematic reviews have coded for cultural consideration involving understudied/underserved groups and international populations in cancer survivorship, with theories/models, methods, and components of adaptation/tailoring outlined. This work is a significant step forward in this regard and provides a variety of possibilities for future research. One direction that could greatly impact the field moving forward is clarity in modifications in design for cultural fit. In conceptualization of this review, the author group intended to summarize “systematic modification[s] of a protocol to consider language, culture, and context in such a way that it is compatible with the client’s cultural patterns, meanings, and values” [61]. However, in conducting this review, the author group found the included reports often contained little detail on what cultural incorporations were made systematically.

Another opportunity for further growth is the inclusion of deep-level cultural congruence in intervention development. A preponderance of studies included in this review focused adaptation efforts on language, to translate intervention materials for the target population. However, studies that demonstrated cultural incorporation beyond surface structure (i.e., language) such as addressing historical events when conducting a life review with Chinese chemotherapy patients [62] appeared more beneficial for QOL compared to interventions describing a language consideration only. Moving forward, a greater focus on deep culture structure is needed. Evidence from psychology suggests consideration of deep cultural structure need not be arduous. For example, a study examining the effect of relational savoring in parenting suggests a greater intervention engagement for Latina mothers due to the compatibility of relational savoring with cultural values of familism and simpatia [63]. Although not originally designed specifically for Latina mothers, deep structure cultural fit allowed for positive parental outcomes such as increased parental pride, gratitude, and feelings of closeness with their children.

The utility of applying deep cultural structure in intervention work aligns with most theoretical models of cultural adaptation; however, only 8 of the 86 reports utilized a model/framework for cultural incorporation, limiting the ability to evaluate the usefulness of these tools. Future studies may consider use of such models/frameworks in intervention development. This review initially intended to code mentions of the (1) fidelity of the intervention, (2) rationale for tailoring or adapting an intervention, and (3) future considerations for dissemination/implementation. However, these details were dropped from extraction because they were sparse and unavailable. As such, the results in this review are limited to details available in the final dataset of reports and past published work referenced in those reports. Scarce data also limited opportunities to examine non-self-report biobehavioral outcomes such as performance status in activities of daily living (n = 4), actigraphy-measured physical activity (n = 3), actigraphy-measured sleep (n = 2), heart rate (n = 2), BMI (n = 1), and tumor progression/survival (n = 1). Only 2 studies examined biomarkers related to immune function and inflammation. The potential impact of interventions with cultural incorporation on biobehavioral outcomes is ripe for future investigation.

A limitation of this review is that articles could only be included if they were available in English. Although effort was made to obtain English-translated versions of non-English publications from authors, seven reports were ultimately excluded from the present review. It is also important to note that for non-US research, a greater number of studies were conducted in Asia compared to other regions such as Africa and South America. It is unclear whether this is due to the scarcity of research on this topic in those regions or if research reports are in non-English outlets inaccessible via the databases searched in this review. Nonetheless, future work incorporating culture in psychosocial or behavioral interventions advanced and metastatic cancer can benefit from the more expansive global perspective sought out by this review.

Additionally, important cultural considerations in the reports may have been missed because the coding process relied on explicit written description. Much of the cultural incorporation may occur in unscripted interactions between interventionists and recipients. Future research should report such unscripted cultural incorporation; this will facilitate a stronger understanding of the importance of cultural congruence in intervention work. Additionally, more details such as cultural adaptation rationale as well as whether culture was considered earlier in the design for dissemination/implementation will be beneficial.

Despite the limitations of this review, the qualitative findings also highlight important future directions for incorporating culture in psychosocial and behavioral interventions for advanced and metastatic cancer patients. Notably, the qualitative themes identified in this review suggest cultural consideration should permeate the multiple stages of intervention development. Further, there is a clear opportunity for understudied/underserved populations to be involved in the research. To broaden the availability of culturally appropriate psychosocial interventions for advanced and metastatic cancer survivors, it is necessary to gain an understanding of the varying manifestations of culture, particularly beyond stereotypical values often associated with different ethnic/racial minorities (i.e., collectivism, familism). This can be achieved through community-based participatory research conducted among underserved subgroups of advanced and metastatic cancer survivors, their families/caregivers, and healthcare providers.

Literature on cultural adaptations of interventions demonstrate significant benefits to diverse populations including five times greater odds for symptom remission than unadapted intervention conditions [64], and concerns about fidelity have been debunked as most cultural adaptations maintain core aspects of an intervention [29]. While culturally appropriate interventions expand access, relevance, and engagement of overlooked cancer survivor groups, more investigation is needed on how to best navigate the dichotomy between generalizing and individualizing an intervention. “Dynamic sizing” and flexibility are imperative to prevent overgeneralizations and to increase cultural competence [65, 66]. Further, increasing the personal relevance of interventions may improve outcomes among underserved groups [67].

Conclusion

Intentional cultural incorporation and adaptation of psychosocial and behavioral interventions are valuable for addressing the survivorship needs of underserved and understudied populations with advanced and metastatic cancer. Thoughtful consideration of deep cultural structures influencing the population of focus and the use of cultural adaptation models and frameworks can improve the impact of future interventions. This work will ensure minoritized groups are not "left behind" as long-term survivorship needs come to be universally addressed.