Associations between Spirituality, Mindfulness, and Psychological Symptoms among Advanced Lung Cancer Patients and Their Spousal Caregivers

Abstract

Context:

Metastatic lung cancer (LC) patients and their spousal caregivers are at high risk of psychological symptoms. Mindfulness may improve psychological symptoms via spiritual well-being (SW); yet, this mediation model has not been examined in a dyadic context.

Objectives:

We examined the mediating role of two dimensions of SW (meaning/peace, faith) in the mindfulness-symptoms link in stage IV LC patients and their spousal caregivers.

Methods:

We examined the actor-partner interdependence model of mediation (APIMeM) using multivariate multilevel modeling with 78 couples. Four APIMeM analyses were conducted to examine: 1 predictor (mindfulness) × 2 mediators (meaning/peace, faith) × 2 psychological symptoms (depressive symptoms, cancer distress). We also tested four alternative models in which mindfulness mediates the associations between SW and psychological symptoms.

Results:

The alternative model (SW→ Mindfulness→ Psychological symptoms) was preferred than the original model (Mindfulness→ SW→ Psychological Symptoms). For patients, meaning/peace was directly associated with their own psychological symptoms, while faith was only indirectly associated with their own psychological symptoms via mindfulness. For spouses, meaning/peace was both directly and indirectly associated with their own psychological symptoms, while faith was only directly associated with their own depressive symptoms (but not cancer distress). Moreover, spouses’ faith was indirectly associated with patients’ psychological symptoms through patients’ mindfulness.

Conclusions:

SW are associated with patients and spouses’ psychological symptoms both directly and indirectly through mindfulness. Thus, interventions that target SW, particularly meaning and peace, along with mindfulness may be beneficial to the psychological management of patients facing a terminal disease and their spousal caregivers.

Introduction

Non-small cell lung cancer (LC) patients’ stage at diagnosis is their primary predictor of survival; yet, 57% of new LC cases are detected with distant metastases [1]. Patients with metastatic LC are at high risk of physical and psychiatric symptom burden and overall quality of life (QoL) deficits [2]. Spouses/romantic partners (spouses from this point forward), who are patients’ most important source of support and care [3–6], play critical roles in cancer patients’ symptom management, including providing psychosocial and physical support, monitoring and assessing symptoms, and making care decisions [7]. Regrettably, spouses report high rates of distress, fatigue and sleep disturbances, which may compromise the quality of care they are able to provide to the patient [8–10]. Indeed, interdependence in cancer patients’ and spouses’ outcomes (i.e., spouses’ outcomes influence patients’ outcomes and vice versa [8, 11–14]) has been demonstrated in various symptoms and QoL dimensions [15]. Thus, managing symptom burden with couple-focused approach may optimize cancer care.

Mindfulness-based interventions are a promising strategy to improve psychological symptoms and enhance QoL among cancer patients [16–18]. Mindfulness is a form of attention regulation in which thoughts, feelings, and physical sensations are observed at present-moment and accepting these experiences as non-judgmental ways [19]. Meta-analyses of a mindfulness training program, mindfulness-based stress reduction (MBSR; [20]), report that MBSR had moderate-to-high effects to enhance mental health among cancer patients [18, 21]. Relatively unknown is the mechanisms of how MBSR or overall mindfulness influences one’s psychological health.

The positive effects of mindfulness on mental health may partly be explained via enhanced spiritual well-being (SW) including meaning/peace and faith [22]. Mindfulness can cultivate spiritual growth such as sense of inner peace and transcendence in that mindfulness fosters disengagement from a narrow self-focus and engagement much broader view of interconnectedness in which oneself is not seen as separate from everyday activities, other people, or the world [23–25]. Studies showed that MBSR increased SW [26–28], particularly meaning in life [29]. However, the direction is unclear because SW may also increase mindfulness alternatively; a study that investigated mechanisms of MBSR program among adults reported that mindfulness partly explained the association between increased spirituality and improved mental health following the 8-week, MBSR program [30].

Enhanced SW is an important QoL index in itself and it may also in turn influence physical, mental and social health outcomes [31–33]. For patients with terminal cancer, it was found that SW protects against the desire for a hastened death, hopelessness, and suicidal ideation independent of depression, social support, physical function and cancer symptoms [34]. Furthermore, SW is interdependent in couples coping with cancer as spouses’ SW is associated with patients’ QoL outcomes [11]. Thus, while spouses’ SW is important in its own right as it is associated with their own’ QoL outcomes, it may also have important implications for patients’ QoL management.

Despite the interrelationships between mindfulness, SW and health outcomes only a few studies have explicitly examined the mediational role of SW in the mindfulness-health link [26, 30, 35], and none of them have been conducted in advanced cancer patients. More importantly, the interrelatedness between mindfulness, SW, and psychological health outcomes in cancer patient-spouse dyads remains untested, although the interdependence of mindfulness [36] and SW [11] was separately investigated in cancer patient-spouse dyads.

Accordingly, we examined the mediating role of two dimensions of SW (meaning/peace, faith) in the mindfulness-psychological symptoms link in stage IV LC patient-spouse dyads. We hypothesized that spouses’ mindfulness would positively influence patients’ symptoms (depressive symptoms, cancer distress) through their own or patients’ SW, and vice versa (see Figure 1 for a hypothesized model). Due to the possibility that SW may increase mindfulness, we also explored an alternative model in which mindfulness mediates the SW-symptoms link and compared the hypothesized and alternative models.

Figure 1. Hypothesized Interdependent Mediation Model of Associations between Mindfulness, SW and Symptoms.

Note. SW=Spiritual well-being. a1 indicates the direct effect from patients’ mindfulness to patients’ SW; b1 indicates the direct effect from patients’ SW to patients’ symptoms; c’1 indicates the direct effect of patients’ mindfulness to patients’ symptoms; a2 indicates the direct effect from spouses’ mindfulness to spouses’ SW; b2 indicates the direct effect from spouses’ SW to spouses’ symptoms; c’1 indicates the direct effect of spouses’ mindfulness to spouses’ symptoms; A=actor effect; P=partner effect; E=error variance.

Methods

Participants

The present study is a baseline data analysis of a feasibility study of a couple-focused, mindfulness-based intervention for metastatic LC patient-spouse dyads [37, 38]. Inclusion criteria for patients were: (1) diagnosed with stage IV non-small cell LC; (2) currently receiving treatment (e.g. radiotherapy, chemotherapy); (3) having an Eastern Cooperative Oncology Group (ECOG) performance status of ≤ 2; and (4) having a romantic partner with whom they have resided for a minimum of 6 months. Patients and spouses must be: (1) ≥18 years old; (2) able to read and speak English; and (3) able to provide informed consent. Exclusion criteria for both patients and spouses were: (1) not oriented to time, place, or person as deemed by the clinical team and (2) regular (self-defined) participation in psychotherapy or a formal cancer support group.

Procedures

Research staff identified potential participants via the institution’s computerized appointment system for the thoracic clinics. Potential participants were approached during patients’ clinic visits, screened for eligibility, and consented. If spouses were not present during the initial contact, we asked the patient for permission to contact the spouse via phone to obtain consent. We approached 134 eligible dyads of which 81 consented to participate (60.4%) and 78 completed baseline assessments (96.3%). Refusal reasons were primarily coping well (n=29 dyads) and too busy (n=16 dyads). Couples were given the initial assessments to be returned prior to randomization or any other intervention related procedures. Each participant received a $20 gift card for completing the baseline assessment. The MD Anderson Institutional Review Board approved the study.

Measures

Demographic and medical factors.

Both patients and spouses were asked demographic information such as age, sex, race/ethnicity, marital status/length, level of education and income, and smoking history. Patients were also asked to report their level of overall distress using the National Comprehensive Cancer Network distress thermometer. Patients’ medical information (e.g., cancer diagnosis, treatment) was obtained from electronic medical records.

Mindfulness.

Mindfulness was measured with the 15-item Mindful Attention Awareness Scale (MAAS; [39]) that assesses a core characteristic of dispositional mindfulness, open or receptive awareness of and attention to what is taking place in the present. Participants were asked to rate each item from 1 (almost always) to 6 (almost never). Mean scores were used and higher scores indicates higher mindfulness (Cronbach’s α=.90).

Spiritual well-being.

Two dimensions of SW—meaning/peace and faith—were assessed with the 12-item (four items per dimension) Functional Assessment of Cancer Therapy-Spiritual Well-Being Scale (FACIT-SP) [22]. Participants were asked to rate each item from 0 (not at all) to 4 (very much). Mean scores were calculated and higher scores indicate higher meaning/peace and faith (αs=.87 for both meaning/peace and faith).

Psychological symptoms.

Depressive symptoms were assessed with the 20-item Center for Epidemiologic Studies Depression Scale (CESD; [40]). Higher scores indicate higher depressive symptoms. Cancer-related distress symptoms were assessed with the Impact of Events Scale (IES), a 15-item scale assessing three aspects of distress including avoidant and hyperarousal behaviors and intrusive thoughts [41]. Spouses were asked to report distress caused by their partners’ cancer. We used the total IES scores and higher scores indicate higher cancer-related distress. Cronbach’s α was .88 for the CESD and .92 for the IES.

Analytic Plan

Firstly, we conducted descriptive analyses and paired t-tests (or chi-square tests for categorical variables) to examine differences in each variable between patients and spouses. Secondly, we conducted between- and within-person bivariate correlational analyses of mindfulness, SW, and psychological symptoms. Finally, we conducted actor-partner interdependence mediation models (APIMeM) [42] using multilevel modeling approach. We tested four models: one predictor (mindfulness) × two mediators (meaning/peace, faith) × two psychological symptoms (depressive symptoms, cancer distress) separately. With double-entry structure for dyadic data, two multilevel models [43] were applied—(1) actors’ and partners’ predictors and their interactions with the indicator variable for distinguishable dyads on actors’ mediator variable and (2) actors’ and partners’ predictors and mediators with the same interactions on actors’ outcome variable. With the interactions with the indicator variable, each of the actor and partner effects could be differentiated into patients’ and spouses’ effects, and a random intercept was incorporated in each model for within-dyad nonindependence. The covariates for outcome included patients’ gender, age, and ECOG score, and couples’ relationship length based on previous dyadic studies [11, 12]. To simultaneously fit the two multilevel models, multivariate multilevel modeling for mediation [44] that stacks the mediator and outcome variables, fits a single model with a selection variable to distinguish the two, and allows heteroscedastic residual variances was used.

Each model estimates the direct effects from mindfulness to SW (a), the direct effects from SW to symptom (b) and the direct effects from mindfulness to symptom (c’), in both patients and spouses. Also, each effect can be classified into either actor (A) or partner (P) effect. The actor effect indicates that individual’s scores are associated with his/her own scores. The partner effect indicates that individual’s scores are associated with his/her partner’s scores. Moreover, per an individual’s outcome, four specific indirect effects were calculated, two through their own SW (e.g., a1A·b1A, a2P·b1A) and the other two through partner’s SW (e.g., a1P·b2P, a2A·b2P) (see Figure 1). The coefficients, standard errors (SEs), and bias-corrected confidence intervals (CIs) for the indirect and total (i.e., direct + indirect) effects were obtained from 5,000 bootstrapped resamples [45]. Because the indirect and total effects estimated from bootstrapping are not under any distributional assumptions, CIs were provided.

For each APIMeM analysis, we also tested an alternative model, a reverse pathway from SW via mindfulness. We compared the original model with the alternative model with model fit indices regarding Akaike Information Criteria and Bayesian Information Criteria. The number of missing values in the variables ranged from 3 to 6 for both patients and spouses. For covariates, the length of relationship had 9 missing values. All these missing data in both patients and spouses were imputed with missForest method [46], which was shown to outperform other methods (e.g., k-nearest-neighbors imputation or parametric multivariate-imputation-by-chained-equations) [47]. All the analyses were conducted using R 3.6.3.

Results

Descriptions of Mindfulness, SW and Psychological Symptoms

A total of 156 participants (i.e., 78 couples) completed the baseline assessment. Mean age of participants was 64.42 years (SD=10.36, range=30-93). Approximately half of patients (51%) and spouses (51%) were females and there were three female same-sex couples. Most patients (76%) and spouses (69%) were non-Hispanic whites and 55% of patients and 58% of spouses had at least a college degree. Fifty nine percent of spouses reported that they never smoked, and 49% of patients reported that they never smoked. The proportion of patients’ ECOG scores was: 0 (33.3%), 1 (55.1%) and 2 (11.5%).

Table 1 shows mean score of mindfulness, SW, and symptoms. Patients and spouses differed only in the IES score; Spouses reported significantly higher cancer distress than patients.

Table 1.

Differences between Patients and Spouses with regard to Demographics, Mindfulness, Spiritual Well-Being and Psychological Symptoms

Variable	Patients (n=78)		Spouses (n=78)		χ2 or Paired t (2-tailed)
	M or %	SD	M or %	SD
Gender					.01
Male	49.35%	-	48.72%	-
Female	50.65%	-	51.28%	-
Age, years	65.00	10.44	63.85	10.32	3.46***
Mindfulness	4.51	.90	4.58	.82	−.78
Meaning/peace	25.06	5.78	24.89	5.49	.30
Faith	11.90	4.32	11.99	4.03	−.29
Depressive symptoms	12.59	9.47	12.78	8.98	−.26
Cancer distress	19.32	16.65	23.55	17.42	−2.48*

Note.

^*p<.05,

^***p<.001.

Correlations between Mindfulness, SW and Psychological Symptoms

Table 2 shows within-person bivariate correlations of mindfulness, SW, and psychological symptoms in either patients or spouses. Both patients’ and spouses’ mindfulness was negatively correlated with their own symptoms. Patients’ mindfulness was positively correlated with their own meaning/peace and faith. Spouses’ mindfulness was only positively correlated with their own meaning/peace. Both patients’ and spouses’ meaning/peace but not faith was negatively correlated with their own symptoms.

Table 2.

Within-Person Bivariate Correlations between Mindfulness, Spiritual Well-Being and Psychological Symptoms

Variable	Mindfulness	Meaning/Peace	Faith	Depressive symptoms	Cancer distress
Mindfulness	-	.33**	−.09	−.50***	−.54***
Meaning/Peace	.47***	-	.30**	−.60***	−.44***
Faith	.34**	.52***	-	−.00	.09
Depressive symptoms	−.40***	−.72***	−.18	-	.70***
Cancer distress	−.38**	−.58***	−.10	.62***	-

Note. Patients correlates are on lower diagonal and shaded in light grey and spouses correlations on upper diagonal in darker grey.

^*p<.05,

^**p<.01,

^***p<.001.

Table 3 shows the between-person bivariate correlations of mindfulness, SW, and symptoms between patients and spouses. Patients’ meaning/peace was correlated with spouses’ mindfulness and cancer distress. Patients’ faith was correlated with spouses’ faith.

Table 3.

Between-Person Bivariate and Partial Correlations between Mindfulness, Spiritual Well-Being, and Psychological Symptoms

Variable	Patients’ mindfulness	Patients’ meaning/peace	Patients’ faith	Patients’ depressive symptoms	Patients’ cancer distress
Spouses’ mindfulness	.17	.24*	.15	−.18	−.20
Spouses’ meaning/peace	−.04	.11	−.05	−.03	.02
Spouses’ faith	−.11	.08	.38***	.16	.13
Spouses’ depressive symptoms	−.17	−.22	.07	.32**	.07
Spouses’ cancer distress	−.17	−.27*	.03	.34**	.15

Note.

^*p<.05,

^**p<.01,

^***p<.001.

APIMeM

The hypothesized model was that SW mediates the mindfulness-symptoms link and the alternative model was mindfulness mediates the SW-symptoms link. Contrary to our hypothesis, the alternative model resulted in a better fit than the hypothesized model (see Supplementary Table 1). Thus, we present results from the alternative models below (see Figure 2 for depressive symptoms; Figure 3 for cancer distress).

Figure 2. Indirect effects Model of Mindfulness in the SW-Depressive Symptoms Link.

Note. SW=Spiritual well-being. Numbers indicate unstandardized coefficients. Patients’ and spouses’ gender, age, length of relationship and patients’ level of overall distress were controlled. Bold lines indicate significant paths. *p<.05, **p<.01, ***p<.001.

Figure 3. Indirect effects Model of Mindfulness in the SW-Cancer Distress Link.

Note. SW=Spiritual well-being. Numbers indicate unstandardized coefficients. Patients’ and spouses’ gender, age, length of relationship and patients’ level of overall distress were controlled. Bold lines indicate significant paths. *p<.05, **p<.01, ***p<.001.

Direct actor effects (S➔Psychological symptoms).

Patients’ higher meaning/peace was associated with patients’ lower depressive symptoms (b=−1.02, SE=.15, p<.001) and patients’ lower cancer distress (b=−1.09, SE=.32, p<.001). Spouses’ higher meaning/peace was associated with spouses’ lower depressive symptoms (b=−.70, SE=.14, p<.001) and spouses’ lower cancer distress (b=−.80, SE=.30, p=.007). Spouses’ higher faith was associated with spouses’ lower depressive symptoms (b=−.48, SE=.24, p=.046).

Direct actor effects (S➔Mindfulness).

Patients’ higher meaning/peace (b=.08, SE=.02, p<.001) and faith (b=.10, SE=.02, p<.001) were associated with patients’ higher mindfulness in a model when depressive symptoms was the outcome. Likewise, patients’ higher meaning/peace (b=.08, SE=.02, p<.001) and faith (b=.10, SE=.02, p<.001) were associated with patients’ higher mindfulness in a model when cancer distress was the outcome. Similarly, spouses’ higher meaning/peace was associated with spouses’ higher mindfulness in a model when depressive symptoms was the outcome (b=.04, SE=.02, p=.008) and when cancer distress was the outcome (b=.04, SE=.02, p=.008). There was no direct actor effect of faith among spouses.

Direct actor effects (Mindfulnes➔Psychological symptoms).

Patients’ higher mindfulness was associated with patients’ lower depressive symptoms (faith as predictor: b=−3.69, SE=1.01, p<.001) and lower cancer distress (meaning/peace as predictor: b=−4.09, SE=1.94, p=.037; faith as predictor: b=−7.03, SE=1.94, p<.001). Spouses’ higher mindfulness was associated with spouses’ lower depressive symptoms (meaning/peace as predictor: b=−2.68, SE=1.03, p=.010; faith as predictor: b=−4.41, SE=1.15, p=<.001) and lower cancer distress (meaning/peace as predictor: b=−7.85, SE=2.23, p<.001; faith as predictor: b=−10.31, SE=2.21, p<.001).

Direct partner effects (S➔Psychological symptoms).

None of the direct partner effect from SW to psychological symptoms was significant.

Direct partner effects (S➔Mindfulness).

Higher spouses’ faith was associated with lower patients’ mindfulness in a model when depressive symptoms was the outcome (b=−.07, SE=.02, p=.008) and cancer distress was the outcome (b=−.07, SE=.02, p=.008).

Direct partner effects (Mindfulnes➔Psychological symptoms).

Spouses’ higher mindfulness was associated with patients’ lower cancer distress in a model when faith was the predictor (b=−4.55, SE=2.22, p=.042).

Indirect effects.

The indirect effects from patients’ faith to patients’ depressive symptoms (b=−.35, SE=.15, p=.021, 95% CI=−.72,−.14) and cancer distress (b=−.67, SE=.30, p=.031, 95% CI=−1.38,−.24) through their own mindfulness were significant; patients’ higher faith was associated with patients’ higher mindfulness, which further was associated with patients’ lower depressive symptoms and lower cancer distress.

For spouses’ SW, the indirect effects from spouses’ meaning/peace to spouses’ depressive symptoms (b=−.12, SE=.05, p=.039, 95% CI=−.28,−.04) and cancer distress (b=−.34, SE=.16, p=.041, 95% CI=−.78,−.11) through their own mindfulness were significant; spouses’ higher meaning/peace was associated with spouses’ higher mindfulness, which further was associated with spouses’ lower depressive symptoms and lower cancer distress. The indirect effects from spouses’ faith to patients’ depressive symptoms (b=.24, SE=.12, p=.043, 95% CI=.07,.53) and cancer distress (b=.46, SE=.19, p=.021, 95% CI=.15,.91) through patients’ mindfulness were significant; spouses’ higher faith was associated with patients’ lower mindfulness, which in turn, was associated with patients’ higher depressive symptoms and higher cancer distress.

Discussions

We examined the dyadic associations between mindfulness, SW and psychological symptoms among metastatic LC patient-spouse dyads including the mediating role of SW in the mindfulness-symptoms link. Contrary to our hypothesis expecting SW to mediate the association between mindfulness and psychological symptoms, we revealed support for an alternative model in which mindfulness mediates the SW-symptoms link. Our findings were consistent with a previous study in which mindfulness partly explained the association between increased spirituality and improved mental health following the 8-week, MBSR program [30]. However, it remains unclear why the alternative model was preferred to the originally hypothesize model and future research based on experimental designs will be needed to better understand the interrelatedness.

Although the alternative model was favored in all analyses, the two SW, meaning/peace and faith, had associations with mindfulness and psychological symptoms in different ways among patients and spouses. Specifically, the direct effect between meaning/peace and psychological symptoms was found among both patients and spouses; higher meaning/peace was associated with lower depressive symptoms and lower cancer distress. However, the indirect effect between meaning/peace and psychological symptoms was found only among spouses; the positive effects of meaning/peace on psychological symptoms among spouses were partly mediated by their own mindfulness. For faith, the direct effect between faith and psychological symptoms was found only among spouses; spouses’ faith was inversely associated with their own depressive symptoms. While patients’ faith was not directly associated with their own depressive symptoms and cancer distress, we revealed indirect effects through their own mindfulness. That is, patients’ faith was positively associated with patients’ mindfulness, which in turn, was inversely associated with their own depressive symptoms and cancer distress. Taken together, for patients, meaning/peace was directly associated with their own psychological symptoms, while faith was only indirectly associated with their own psychological symptoms. For spouses, meaning/peace was both directly and indirectly associated with their own psychological symptoms, while faith was only directly associated with their own depressive symptoms.

Although spouses’ faith seemed to be helpful to deal with their own depressive symptoms, its effect may not be the same for patients’ psychological symptoms; spouses’ faith was inversely associated with patients’ mindfulness, which was relayed to patients’ depressive symptoms and cancer distress. That is, spouses’ higher faith was associated with patients’ lower mindfulness, which in turn, was associated with patients’ higher depressive symptoms and higher cancer distress. Therefore, while faith was helpful to manage their own psychological symptoms, spouses’ higher faith may indirectly increase patients’ psychological symptoms.

The divergent findings for the meaning/peace and faith subscales of SW may indicate that faith is a distinct dimension from meaning/peace [22]. Previous studies conducted among cancer patients with the FACIT-SP have shown that the beneficial effects of SW on health were mostly evidenced with meaning/peace, but inconsistent with regard to the association between faith and health [48, 49]. Note that spirituality is a multicomponent construct encompassing beliefs, search for meaning, spiritual coping, a connection with the transcendent, a sense of personal wholeness, and an awareness of relatedness to others [50]. The faith subscale of the FACIT-SP (e.g., “I find strength in my faith/spiritual beliefs”, “My illness has strengthened my faith/spiritual beliefs”) may be more related to a coping strategy [51] to reduce their distress, which may be either positively or negatively associated with health outcomes. While faith may help manage their own symptoms, spouses’ faith potentially as a religious coping may hinder patients to focus on the presence (i.e., be mindful). Yet, future research regarding spouses’ faith and religious coping using a longitudinal assessment approach is warranted to elucidate an in-depth understanding of these associations.

Note that the partner effects existed only from spouses to patients, and not vice versa. Therefore, interventions targeting spouses may be additionally helpful for patients’ symptom management. Particularly, given the direct partner effect from spouses’ mindfulness to patients’ cancer distress, supportive care programs for metastatic LC patients may consider providing MBSR or other mindfulness-based interventions to both patients and their spouses. The parent trial of this study has revealed that a couple-based program may be a promising approach to increase SW in patients and spouses [38]. Moreover, interventions that aim to enhance meaning/peace such as meaning-focused group therapy for patients may be helpful to reduce patients’ psychological symptoms and improve QoL [52], although it remains unclear whether providing interventions targeting meaning/peace to spouses will be additionally beneficial to patients.

This study has several limitations. First, causality between SW, mindfulness, and symptoms is not guaranteed in this cross-sectional, observational study. Although the alternative model was favored, this does not rule out the path from mindfulness to SW. Longitudinal studies are needed for in-depth understanding of the interrelatedness. Second, sample size was small and thus, some of the non-significant results might be due to lack of power. Third, participants were patient-spouse dyads from a large, National Cancer Institute (NCI) designated comprehensive cancer center. Results from the present study may not be generalizable to all patient-spouse dyads. Fourth, the MAAS focuses on awareness. However, mindfulness encompasses other aspects beyond the awareness, such as non-judgmental inner experience [53, 54]. If we had comprehensively assessed various aspects of mindfulness, we might have observed stronger associations between mindfulness and psychological symptoms. Likewise, we focused on SW, meaning/peace and faith. However, spirituality is a multidimensional construct, which is beyond SW [50]. We might have missed including other important dimensions of spirituality. Finally, the within-person correlations between the two outcomes (depressive symptoms and cancer distress) were high as .62 in patients and .70 in spouses. We performed separate analyses for each of these outcome measures because they are conceptually distinguishable. For future research, a confirmatory factor analysis to estimate the common variance of these two outcome measures may be needed.

In conclusion, to our knowledge, this is the first study that investigated interrelatedness between mindfulness, SW, and psychological symptoms among advanced cancer patient-spousal caregiver dyads. Our findings suggest that one’s meaning/peace and faith help manage their own psychological symptoms both directly and indirectly through mindfulness, although spouses’ faith may indirectly exacerbate patients’ psychological symptoms by reducing patients’ mindfulness. Spouses’ mindfulness may help manage their own and additionally patients’ psychological symptoms.

Clinical Implications

Given the direct and indirect effects of SW on psychological symptoms via mindfulness, interventions that target SW particularly meaning/peace along mindfulness may be provided during patients’ cancer treatment to manage spouses’ and patients’ psychological symptoms. As we identified several dyadic associations between stage IV LC patients and their spouses, there is reason to believe that a dyadic symptom management approach that not only targets both patients and spouses but capitalizes on their interdependence may be an effective strategy. However, it remains uncertain if dyadic approach is in fact superior to individual-focused supportive care.