Skip to main content
NCBI home page
The Gerontologist logoLink to The Gerontologist
. 2026 May 20;66(7):gnag103. doi: 10.1093/geront/gnag103

Mindfulness and rumination: stress pathways for caregivers of persons living with dementia

Jyoti Savla 1,2,✉, Marrium Mansoor 3, Nahyun Kim 4, Lauren Hagemann 5, Katherine Luci 6,7, Mamta Sapra 8,9
Editor: Joseph E Gaugler
PMCID: PMC13334642  PMID: 42159557

Abstract

Background and Objectives

Family caregivers of persons living with dementia face caregiving demands that heighten risk for anxiety, depressive symptoms, and perceived stress. Although caregiving distress is well documented, less is known about the cognitive processes that link anxiety to psychological distress, or whether mindfulness-based interventions affect these processes. This study examined rumination as a cognitive mediator linking anxiety with these outcomes and tested whether mindfulness training moderated this association.

Research Design and Methods

Using data from a randomized controlled trial comparing the Practice of Awareness, Acceptance, and Compassion in Caregiving (PAACC) program with Resources for Enhancing All Caregivers Health Veterans Affairs Program, we conducted a secondary mechanistic analysis using structural equation models (N = 133). Models tested whether post-intervention rumination mediated associations between post-intervention anxiety and depressive symptoms, perceived stress, and caregiver burden, while adjusting for baseline variables, and whether the anxiety–rumination association differed by intervention group.

Results

Rumination mediated associations between anxiety and both depressive symptoms and perceived stress, but not caregiver burden. Mindfulness moderated the anxiety–rumination association, with stronger coupling at higher anxiety in PAACC participants. No direct group differences in post-intervention outcomes emerged. Alternate model specifications showed weaker fit than the hypothesized model.

Discussion and Implications

Findings highlight rumination as a cognitive process linking anxiety with psychological distress and suggest that mindfulness training may influence how anxiety is cognitively elaborated, rather than uniformly reducing distress. Future research in diverse caregiver populations and with longer timeframes can clarify temporal ordering and variability in responsiveness to mindfulness-based approaches.

Clinical Trial Registration Number

NCT03447860.

Keywords: Dementia caregiving, Anxiety, Psychological distress, Cognitive mechanisms, Veterans

Family caregivers of persons living with dementia shoulder a complex mix of practical, emotional, and existential demands. Beyond hands-on assistance with daily living, they manage unpredictable behavioral symptoms, cope with the gradual loss of the person they know, and contend with the erosion of their own time, relationships, and health. These caregiving demands are often conceptualized within a stress process framework that distinguishes between primary and secondary stressors (Pearlin et al., 1990). Primary stressors include assistance with activities of daily living (ADL) and responding to behavioral symptoms that require constant vigilance (Cerejeira et al., 2012; National Alliance for Caregiving & AARP, 2020; Sörensen & Conwell, 2011). Secondary strains arise when these demands spill over into other roles such as employment, parenting, household finances, and erode sleep, leisure, and social connection (Aneshensel et al., 1995; Bauer & Sousa-Poza, 2015; Gao et al., 2019; Liu et al., 2019).

Together, these stressors accumulate and help explain the elevated levels of anxiety, depressive symptoms, perceived stress, and caregiver burden (Puga et al., 2023; Watson et al., 2019) relative to non-caregiving peers (Pinquart & Sörensen, 2003; Walter & Pinquart, 2020). Within this cluster of outcomes, anxiety occupies a distinctive position. It is both unusually prevalent with estimates around 25%–44% among dementia caregivers (Cooper et al., 2007; Sallim et al., 2015) and differs from the other outcomes in how it is experienced. Conceptually, anxiety reflects an appraisal-based response to perceived threat or anticipated harm and is characterized by worry and tension (Eysenck & Fajkowska, 2018; Lazarus & Folkman, 1984). In contrast, depressive symptoms and perceived stress reflect broader psychological responses that tend to consolidate over time under prolonged strain: depressive symptoms manifest as low mood and diminished motivation, whereas perceived stress reflects an appraisal of life demands as overwhelming. Consistent with this distinction, a systematic review by Cooper et al. (2007) found that while depressive symptoms and anxiety frequently co-occur, the overlap is asymmetric, wherein most depressed caregivers report experiencing anxiety, whereas many anxious caregivers do not show depressive symptoms. This pattern suggests that anxiety may be operating, at least in part, as an antecedent to depressive symptoms rather than a mere correlate. For the purposes of this paper, we use the term psychological distress to refer to depressive symptoms and perceived stress, while recognizing their substantial empirical overlap with anxiety. Caregiver burden, by contrast, reflects a subjective appraisal of the caregiving role itself and is treated here as a related but conceptually distinct outcome.

Despite this potential antecedent role, the cognitive processes linking anxiety to depressive symptoms, perceived stress, and caregiver burden have received less attention. The present study examines rumination as one such process and tests whether embedding mindfulness practices within a caregiver psychoeducational intervention is associated with differences in this pathway.

From anxiety to broader distress: the role of rumination

Stress process models emphasize that objective stressors do not affect well-being directly; rather, their impact is shaped by appraisal of the stressors and coping responses (Lazarus & Folkman, 1984; Pearlin et al., 1990). Rumination, characterized by a repetitive, passive preoccupation with the causes, meanings, and consequences of negative emotional experiences (Nolen-Hoeksema et al., 2008), is one cognitive response pattern that can prolong and intensify emotional distress rather than resolve it. Anxiety can elicit repetitive negative thinking as individuals attempt to manage uncertainty and worry, while ruminative engagement can, in turn, sustain or exacerbate anxious affect over time.

In dementia caregiving specifically, where stressors are chronic and unresolvable, anxiety may function as a recurring input into ruminative cycles. In this context, rumination may arise not from internal cognitive vulnerability alone, but from external conditions that continuously re-elicit it. Rumination has nonetheless often been conceptualized as an upstream cognitive vulnerability contributing to depressive symptoms and a range of psychopathologies, including anxiety (Nolen-Hoeksema et al., 2008). However, growing evidence suggests that anxiety and rumination are likely dynamically and reciprocally related (McLaughlin & Nolen-Hoeksema, 2011; Tamm et al., 2024) rather than strictly ordered. In this paper, we focus on the anxiety-to-rumination direction as one plausible specification.

Prior work has documented associations between rumination and anxiety, depressive symptoms, and broader distress in dementia caregivers (Keune et al., 2023; Romero-Moreno et al., 2016). Yet this literature has typically positioned rumination as a mechanism through which caregiving stressors contribute to anxiety and distress (Romero-Moreno et al., 2016), leaving open the question of whether rumination also explains how anxiety is linked to depressive symptoms, perceived stress, and burden. The present study reorients the analytic focus accordingly, modeling anxiety as the focal predictor and rumination as the linking process.

Mindfulness as a potential disruptor of ruminative cycles

Mindfulness is commonly defined as the intentional and non-judgmental awareness of present-moment experience (Bishop et al., 2004), and it is best understood as a trainable cognitive-emotional skill rather than solely as a programmatic intervention label. Mindfulness practice may cultivate decentering, defined as the capacity to observe thoughts and emotions as transient mental events rather than accurate reflections of the self or reality (Hawley et al., 2014). This decentered stance is theorized to reduce cognitive reactivity by interrupting anxious thought patterns and limiting the escalation of momentary worry into rumination (Guendelman et al., 2017). This implies a theoretically distinctive prediction: mindfulness training should weaken the association between anxiety and rumination and not just shift mean levels of either construct.

Because the stressors of dementia caregiving persist beyond what problem-solving can address, skills that alter internal responses, rather than external circumstances, may be especially well-suited to this population. Meta-analyses and randomized trials indicate that mindfulness-based interventions can reduce depressive symptoms, perceived stress, and rumination among caregivers (Galante et al., 2023; Kor et al., 2021; Oh et al., 2022), although effects are often comparable to those observed in strong active controls such as cognitive-behavioral or psychoeducational approaches. However, equivalent outcomes do not require equivalent mechanisms, leaving open the question of whether mindfulness distinctively alters the anxiety–rumination link.

The present study

Using data from the Practice of Awareness, Acceptance, and Compassion in Caregiving (PAACC) randomized controlled trial, we conducted a secondary mechanistic analysis to examine whether post-intervention rumination mediated the associations between post-intervention anxiety and three outcomes—depressive symptoms, perceived stress, and caregiver burden—while adjusting for baseline levels of anxiety, rumination, and each outcome. We further examined whether mindfulness training moderated the anxiety–rumination association. Figure 1 presents the conceptual model guiding these analyses. This approach is consistent with residualized change modeling and allows examination of cognitive processes associated with psychological outcomes over the intervention period, while acknowledging the limits of causal inference with two assessment time points. Although caregiving stressors initiate this process, the present analysis focuses on downstream cognitive and emotional processes measured in the trial and does not directly model the primary and secondary stressors.

Figure 1.

Path diagram of a mediated moderation model. Three boxes are connected by arrows in a left-to-right sequence: Post-intervention anxiety, Post-intervention rumination, and Post-intervention outcomes, the last of which lists three outcomes: depressive symptoms, perceived stress, and caregiver burden. A separate box labeled Intervention Group, comparing REACH-VA versus PAACC, sits above and points downward to the arrow connecting Anxiety to Rumination, indicating that intervention group moderates the anxiety-to-rumination path. A dashed box in the upper-left labelled Primary and Secondary Caregiving Stressors notes that these are contextual influences not modeled directly.

Open in a new tab

Hypothesized mediated moderation model linking post-intervention anxiety, rumination, and caregiver outcomes. Note. Baseline anxiety, rumination, and outcome variables were included as covariates (not shown). Primary and secondary caregiving stressors are shown as contextual influences but were not modeled directly. REACH-VA = Resources for Enhancing All Caregivers Health–VA Program; PAACC = Practice of Acceptance, Awareness, and Compassion in Caregiving.

We hypothesized that (1) rumination would mediate the associations between post-intervention anxiety and depressive symptoms, perceived stress, and caregiver burden, and (2) compared with the well-established psychoeducational intervention, Resources for Enhancing All Caregivers Health Veterans Affairs Program (REACH-VA), exposure to mindfulness practices in PAACC would alter how anxiety is cognitively elaborated rather than uniformly reducing distress, resulting in weaker coupling between anxiety and rumination and, in turn, smaller indirect associations with the outcomes.

Method

Participants and procedure

Data were drawn from the PAACC randomized controlled trial conducted at the Salem Veterans Affairs (VA) Health Care System. The study enrolled 133 family caregivers of veterans living with dementia who reported moderate to severe burden and were randomly assigned to either the PAACC intervention (n = 67) or to the REACH-VA comparison group (n = 66). Randomization used block sizes of 2–4 in a 1:1 ratio, and both participants and outcome assessors were blinded to group assignment.

Participants were eligible if they were actively providing daily care, assisting with at least one ADL or two Instrumental Activities of Daily Living (IADLs), and were either co-residing or providing more than 5 hours of care per week at the person living with dementia’s home. Additional inclusion criteria included elevated caregiver burden (Zarit Burden Interview [ZBI] ˃40) and normative cognitive functioning (Montreal Cognitive Assessment ≥23). Exclusion criteria included a diagnosis of psychosis or bipolar I disorder, psychiatric hospitalization within the prior 3 months, recent medication changes, ongoing psychotherapy for mood disorders, active substance use within the previous year, or institutionalization of the care recipient. All caregivers provided written informed consent under protocols approved by the VA Central Institutional Review Board (ClinicalTrials.gov Identifier: NCT03447860). Participants were recruited through VA memory and primary care clinics via flyers and clinician referrals, and they received compensation for their participation.

Interventions

Both interventions consisted of four total sessions; each session lasting approximately 60 min, plus baseline and post-intervention assessments. REACH-VA, a cognitive-behavioral caregiver program, provided dementia education, structured problem-solving, cognitive restructuring, relaxation training (e.g., signal breath), and pleasant-activity planning to help caregivers manage stress.

PAACC was developed as an augmentation of REACH-VA, retaining comparable dementia education and behavior-management components, while replacing cognitive restructuring and relaxation training with structured mindfulness practices. Mindfulness practices were embedded within each session rather than delivered as stand-alone modules. Caregivers were introduced to brief, structured mindfulness exercises, such as mindful breathing, mindful eating, loving-kindness, and the RAIN and STOP practices, with one formal practice incorporated into each session. These practices were integrated into caregivers’ individualized behavior-management plans to support more flexible, less reactive responses to dementia-related stressors and were reinforced through between-session home practice assignments, and caregivers were encouraged to incorporate mindfulness informally in daily life. Interventionists in both programs followed standardized manuals and fidelity procedures with ongoing supervision. For complete intervention protocols and session content, see Sapra et al. (2025).

Measures

Treatment group

Participants were randomized to PAACC (coded as 1) or REACH-VA (coded as 0).

Anxiety

The Generalized Anxiety Disorder Questionnaire (GAD-7; Spitzer et al., 2006) is a 7-item self-report scale assessing anxiety symptoms over the past 2 weeks. Items are rated from 0 (not at all) to 3 (nearly every day), yielding total summary scores from 0 to 21. The GAD-7 has demonstrated strong reliability and validity across caregiver populations, including those caring for individuals with dementia (Joling et al., 2015).

Rumination

The Rumination–Reflection Questionnaire (RRQ; Trapnell & Campbell, 1999) consists of 24 items divided into two subscales—self-rumination and self-reflection—where questions were rated from 1 (strongly disagree) to 5 (strongly agree) and summed to provide a total score. Only the 12-item rumination subscale was used in the present analyses (higher scores = greater rumination). The RRQ has been used in both nonclinical and caregiving contexts (Segerstrom et al., 2010; Waldman-Levi et al., 2020), demonstrating excellent internal consistency and construct validity for assessing repetitive self-focused thought.

Depressive symptoms

The Patient Health Questionnaire–9 (PHQ-9; Kroenke et al., 2001) is a brief, 9-item scale derived from the Primary Care Evaluation of Mental Disorders (PRIME-MD) diagnostic instrument. It evaluates the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV) criteria for major depression over the previous 2 weeks on a 4-point scale from 0 (not at all) to 3 (nearly every day) that were added to provide a total score. Scores of 5, 10, 15, and 20 correspond to mild, moderate, moderately severe, and severe depressive symptom levels. The PHQ-9 has been extensively validated among family caregivers, including those supporting persons with dementia (Alfakhri et al., 2018; Lou et al., 2015; Pinyopornpanish et al., 2021).

Perceived stress

The Perceived Stress Scale (PSS; Cohen et al., 1983) is a 10-item measure capturing the degree to which life circumstances are appraised as unpredictable, uncontrollable, or overwhelming. Responses are rated on a 5-point Likert scale from 0 (never) to 4 (very often) and summed to provide a total score, with higher scores indicating greater perceived stress. The PSS has shown strong psychometric properties in caregiving and dementia-related studies (Deeken et al., 2018; Peavy et al., 2022).

Caregiver burden

The Zarit Burden Interview (ZBI; Zarit et al., 1980) assesses subjective burden associated with caregiving tasks and personal strain. The 22-item version used in this study includes items rated on a 5-point scale from 1 (never) to 5 (nearly always), resulting in total scores ranging from 22 to 110, with higher scores indicating greater burden. The ZBI remains one of the most widely validated measures in caregiving research (Hébert et al., 2000; Seng et al., 2010).

Consistent with the conceptual framing of this study (see Figure 1), anxiety was modeled as the focal predictor, rumination as the mediator, and depressive symptoms, perceived stress, and caregiver burden as outcomes. All measures demonstrated acceptable to high internal consistency in this sample (Cronbach’s α ranging from .79 to .93; see Supplementary Table S1).

Analytic strategy

Data were screened for normality and outliers; none were problematic. Group differences at baseline were evaluated using t-tests for continuous variables and chi-square tests for categorical variables (see Table 1). Bivariate correlations are provided in the Supplementary Table S1.

Table 1.

Baseline descriptive statistics of demographics and measures.

Variables Overall (n = 133)
 REACH-VA (n = 66)
 PAACC (n = 67)
 Comparison
M (SD) N (%) M (SD) N (%) M (SD) N (%) t/χ2 (p)
Age 67.17 (9.84) 68.32 (9.80) 66.03 (9.81) 1.34 (.18)
Sex
 Male 20 (15) 10 (15.2) 10 (14.9) 0.001 (.97)
 Female 113 (85) 56 (84.8) 57 (85.1)
Race
 African American 23 (17.6) 12 (18.2) 11 (16.9) 1.05 (.59)
 White/Caucasian 107 (81.7) 54 (81.8) 53 (81.5)
 Hispanic 1 (0.8) 0 (0) 1 (1.5)
Education
 Less than high school 3 (2.3) 1 (1.6) 2 (3) 0.68 (.71)
 High school 27 (20.8) 12 (18.8) 15 (22.7)
 More than high school 100 (76.9) 51 (79.7) 49 (74.2)
Relationship to the care recipient
 Spouse 92 (69.7) 46 (70.8) 46 (68.7) 7.08 (.13)
 Partner 1 (0.8) 0 (0) 1 (1.5)
 Child 30 (22.7) 13 (20) 17 (25.4)
 Sibling 2 (1.5) 0 (0) 2 (3)
 Other 7 (5.3) 6 (9.2) 1 (1.5)
Living situation
 Live in the same house 119 (90.2) 61 (92.4) 58 (87.9) 2.01 (.37)
 Lives in own home (<20 min from care recipient) 10 (7.6) 3 (4.5) 7 (10.6)
 Lives in own home (>20 min from care recipient) 3 (2.3) 2 (3) 1 (1.5)
Years living with care recipient 32.55 (20.37) 33.01 (21.42) 32.07 (19.40) 0.25 (.80)
Current occupation
 Always been a homemaker 7 (5.4) 3 (4.6) 4 (6.2) 5.13 (.53)
 Retired 80 (62) 43 (66.2) 37 (57.8)
 Employed (full-time) 21 (16.3) 10 (15.4) 11 (17.2)
 Employed (part-time) 8 (6.2) 5 (7.7) 3 (4.7)
 Unemployed but looking for a job 1 (0.8) 1 (1.5) 0 (0)
 Disability  8 (6.2) 2 (3.1) 6 (9.4)
 Other 4 (3.1) 1 (1.5) 3 (4.7)
Perceived stress (baseline) 20.93 (6.96) 20.11 (6.75) 21.76 (7.11) −1.37 (.17)
Caregiver burden (baseline) 74.13 (10.97) 72.74 (10.69) 75.49 (11.14) −1.45 (.15)
Ruminative thinking (baseline) 37.22 (10.84) 36.73 (9.94) 37.69 (11.71) −0.50 (.62)
Depressive symptoms (baseline) 9.65 (6.00) 8.56 (5.14) 10.73 (6.60) −2.11 (.04)
Anxiety symptoms (baseline) 8.49 (5.44) 7.42 (5.12) 9.54 (5.58) −2.27 (.02)
Open in a new tab

Note. M = mean; SD = standard deviation; χ2 = chi-square; REACH-VA = Resources for Enhancing All Caregivers Health Veterans Affairs Program; PAACC = Practice of Acceptance, Awareness, and Compassion in Caregiving.

To test the hypothesized model, moderated-mediation analyses were conducted in Stata (Version 18) using structural equation modeling (SEM) with full-information maximum likelihood (FIML) to handle missing data under the missing-at-random assumption. Post-intervention anxiety served as the independent variable (X), rumination as the mediator (M), and depressive symptoms, perceived stress, and caregiver burden as the dependent variables (Y). Baseline levels of anxiety, rumination, and each outcome were included as covariates, allowing post-intervention associations to be interpreted as residualized change over the intervention period. Given the use of two assessment time points and concurrent measurement of post-intervention anxiety, rumination, and outcomes, this approach permits examination of hypothesized mechanistic associations but cannot establish temporal or causal ordering. This is consistent with recommendations for mechanism-focused research when ideal tests of mechanism change over time are not feasible (Science of Behavior Change, n.d.).

Intervention group (0 = REACH-VA, 1 = PAACC) was specified as a moderator of the path between anxiety (X) and rumination (M), allowing examination of whether the association between anxiety and rumination differed by intervention group. Living alone, relationship to the care recipient (spouse/partner vs non-spouse/partner), and sex were considered as additional covariates but were not retained because they were not significant predictors. A double-moderated mediation path testing the moderation of rumination on the study outcomes association by intervention group was also examined, but was trimmed when found to be nonsignificant.

To address concerns about ordering of constructs, we estimated alternate models reversing the ordering of anxiety and rumination, as well as a parallel-outcomes specification in which rumination simultaneously predicted anxiety, depressive symptoms, perceived stress, and caregiver burden. These models were evaluated alongside the hypothesized model using standard fit indices such as the chi-square statistic (χ2), the root mean square error of approximation (RMSEA), the Comparative Fit index (CFI), the Tucker–Lewis index (TLI), the Akaike information criterion (AIC), and the Bayesian information criterion (BIC).

Indirect effects were estimated using 5,000 bootstrap samples, with bias-corrected accelerated (BCa) 95% confidence intervals (CIs). Indirect effects were considered statistically significant when BCa CIs did not include 0. All statistical tests were two-tailed with α = .05. Table 2 reports path coefficients for the mediator and outcomes equations, and Table 3 presents conditional indirect effects for participants in REACH-VA (W0) and in PAACC (W1), and the between‑intervention group difference in the indirect effects (i.e., the index of moderated mediation) with inference based on its BCa 95% CI.

Table 2.

Moderated mediation models predicting rumination and psychological distress outcomes (N = 133).

Path A: Mediator model—rumination (post-intervention)
Depressive symptoms model
 Perceived stress model
 Caregiver burden model
Predictors Rumination (post-intervention)
 Rumination (post-intervention)
 Rumination (post-intervention)
b z 95% CI p b z 95% CI p b z 95% CI p
Anxiety (post) 0.39 1.64 [−0.08, 0.86] .101 0.39 1.62 [−0.08, 0.86] .106 0.38 1.58 [−0.09, 0.85] .113
Group (1 = PAACC) −6.01 −2.50 [−10.71, −1.30] .012 −5.91 −2.46 [−10.61, −1.21] .014 −6.09 −2.54 [−10.78, −1.40] .011
Anxiety × Group 0.91 2.16 [0.09, 1.73] .030 0.91 2.18 [0.09, 1.74] .029 0.93 2.23 [0.11, 1.76] .026
Rumination (baseline) 0.40 5.90 [0.27, 0.53] <.001 0.40 5.93 [0.27, 0.54] <.001 0.40 5.94 [0.27, 0.53] <.001
Anxiety (baseline) 0.03 0.18 [−0.28, 0.33] .856 0.03 0.19 [−0.27, 0.33] .847 0.02 0.13 [−0.28, 0.32] .900
Intercept 15.26 5.40 [9.72, 20.80] <.001 15.13 5.36 [9.60, 20.66] <.001 15.36 5.46 [9.85, 20.87] <.001
Path B: Outcome models
Predictor Depressive symptoms
 Perceived stress
 Caregiver burden
b z 95% CI p b z 95% CI p b z 95% CI p
Rumination (post) 0.10 2.39 [0.02, 0.18] .017 0.12 2.50 [0.03, 0.21] .012 0.23 1.80 [−0.02, 0.48] .071
Anxiety (post) 0.40 4.29 [0.22, 0.59] <.001 0.81 7.32 [0.59, 1.03] <.001 0.79 2.64 [0.20, 1.38] .008
Group (1 = PAACC) −0.27 −0.44 [−1.45, 0.92] .661 0.64 0.89 [−0.77, 2.05] .371 2.05 1.05 [−1.79, 5.90] .295
Outcome (baseline) 0.20 2.65 [0.05, 0.35] .008 0.20 2.85 [0.06, 0.34] .004 0.47 4.68 [0.28, 0.67] <.001
Rumination (baseline) −0.04 −1.19 [−0.11, 0.03] .232 −0.01 −0.18 [−0.09, 0.07] .861 −0.16 −1.48 [−0.37, 0.05] .140
Anxiety (baseline) −0.03 −0.38 [−0.20, 0.13] .706 −0.04 −0.45 [−0.24, 0.15] .656 0.21 0.90 [−0.24, 0.66] .370
Intercept 0.38 0.31 [−2.04, 2.80] .760 3.98 2.31 [0.61, 7.35] .021 15.86 2.00 [0.30, 31.41] .046
Open in a new tab

Note. Models were estimated using structural equation modeling with full-information maximum likelihood (FIML). Group was coded 0 = REACH-VA and 1 = PAACC. Post = post-intervention; Path A reports mediator models predicting post-intervention rumination; Path B reports outcome models predicting post-intervention depressive symptoms, perceived stress, and caregiver burden. All models adjust for baseline levels of anxiety, rumination, and the corresponding outcome. PAACC = Practice of Acceptance, Awareness, and Compassion in Caregiving; REACH-VA = Resources for Enhancing All Caregivers Health Veterans Affairs Program. Columns present b, z, 95% bias-corrected accelerated (BCa) CIs, and p values. Effects are considered statistically different from 0 when the BCa CI does not include 0, and are shown in bold.

Table 3.

Conditional indirect effects of anxiety on psychological distress via rumination by intervention group.

Outcome W 0 = REACH-VA
 W 1 = PAACC
 Index of moderated mediation (i.e., difference between W1 and W0)
Indirect, b BSE 95% BCa CI Indirect, b BSE 95% BCa CI b BSE 95% BCa CI
Depressive symptoms 0.038 0.036 [−0.005, 0.144] 0.127 0.077 [0.015, 0.321] 0.089 0.062 [0.003, 0.264]
Perceived stress 0.046 0.038 [−0.005, 0.163] 0.155 0.077 [0.041, 0.370] 0.109 0.070 [0.003, 0.289]
Caregiver burden 0.088 0.081 [−0.007, 0.383] 0.305 0.195 [0.014, 0.836] 0.217 0.173 [−0.020, 0.683]
Open in a new tab

Note. Conditional indirect effects and bootstrapped standard errors (BSE) were estimated using 5,000 bootstrap resamples. CIs are bias-corrected and accelerated (BCa). Indirect effects are considered statistically significant when the BCa 95% CI does not include 0, and are shown in bold. W0 = REACH-VA; W1 = PAACC. The Index of Moderated Mediation represents the difference between indirect effects for PAACC and REACH-VA. REACH-VA = Resources for Enhancing All Caregivers Health Veterans Affairs Program; PAACC = Practice of Acceptance, Awareness, and Compassion in Caregiving.

Results

Sample characteristics

Caregivers had a mean age of 67.2 years (SD = 9.8). The majority were female (85%) and White (82%); approximately 70% were spouses, and 90% co-resided with the care recipient. Nearly two-thirds were retired, and 16% were employed full-time. Most caregivers had completed at least high school (89%). Dementia type was reported by the caregiver. Alzheimer’s disease was the most commonly reported diagnosis (47.4%), followed by vascular dementia (20.3%). Other reported diagnoses included Parkinson’s disease-related dementia (5.3%), Lewy-body dementia (3.8%), traumatic brain injury-related dementia (1.5%), frontotemporal dementia (0.8%), corticobasal degeneration (0.8%); in 20.3% of cases the dementia type was unspecified. Caregivers reported an average of 9.2 memory or behavior problems occurring at least once during the prior week (SD = 4.2, range = 0–22). Care recipients required assistance with an average of two basic ADL (SD = 1.83; range = 0–6).

Preliminary analyses

Baseline characteristics were largely comparable across the intervention groups (see Table 1). PAACC participants reported slightly higher baseline anxiety and depressive symptoms. Correlations among primary variables were in the expected directions: baseline anxiety was positively associated with rumination, depressive symptoms, perceived stress, and caregiver burden (r ranging from .406 to .745, all ps < .01), and baseline rumination was positively associated with depressive symptoms and perceived stress (r ranging from .298 to .398, all ps < .05; see Supplementary Table S1).

Moderated mediation analyses

Table 2 presents unstandardized path coefficients for each outcome, and Table 3 provides the conditional indirect effects. In the mediator equation model, which has rumination as the outcome, the main effect of post-intervention anxiety on rumination was not significant; however, the anxiety × group interaction was consistently significant in every mediator model, indicating that the anxiety-to-rumination association was stronger in PAACC than REACH-VA (interaction bs ≈ 0.91–0.93, ps = .026 to .030).

To clarify this interaction, we examined the simple slopes for each intervention group (see Figure 2). Among REACH-VA participants, post-intervention anxiety showed a small, nonsignificant association with rumination (b = 0.35, SE = 0.26, p = .180), whereas among PAACC participants, anxiety showed a significantly stronger positive association with rumination (b = 1.31, SE = 0.39, p < .001). Overall model fit was acceptable for each outcome-specific SEM (Depressive Symptoms χ2[2] = 0.58, p = .75; Perceived Stress χ2[2] = 0.63, p = .73; Caregiver Burden χ2[2] = 0.26, p = .88).

Figure 2.

Line graph showing predicted post-intervention rumination on the y-axis (ranging from 14 to 58) as a function of post-intervention anxiety on the x-axis (ranging from 0 to 19), plotted separately for two intervention groups. The REACH-VA group, shown as a solid black line, has a gentle positive slope, rising from approximately 30 at low anxiety to approximately 38 at high anxiety. The PAACC group, shown as a dashed black line, has a markedly steeper positive slope, rising from approximately 24 at low anxiety to approximately 49 at high anxiety. The two lines cross at an anxiety value near 6.6, indicating that PAACC participants had lower predicted rumination than REACH-VA participants at lower anxiety levels but higher predicted rumination at higher anxiety levels.

Open in a new tab

Simple slopes for the anxiety × intervention group interaction predicting post-intervention rumination. Note. Predicted post-intervention rumination is plotted as a function of post-intervention anxiety, separately for each intervention group, holding baseline rumination and baseline anxiety at their sample means. The simple slope was nonsignificant for REACH-VA, b = 0.39, SE = 0.24, p = .101, and significant for PAACC, b = 1.30, SE = 0.37, p = .001. REACH-VA = Resources for Enhancing All Caregivers Health–VA Program; PAACC = Practice of Acceptance, Awareness, and Compassion in Caregiving.

Depressive symptoms

In the outcome equation, both post-intervention anxiety (b = 0.40, p < .001) and rumination (b = 0.10, p = .017) predicted higher depressive symptoms after adjusting for baseline depressive symptoms. The conditional indirect effect via rumination was significant for PAACC (b = 0.127, BCa 95% CI [0.015, 0.321]) but not for REACH‑VA (b = 0.038, BCa 95% CI [−0.005, 0.144]). The index of moderated mediation was significant (Δ  =  0.089, BCa 95% CI [0.003, 0.264]), indicating stronger statistical mediation of the anxiety–depressive symptoms association via rumination in PAACC than in REACH-VA.

Perceived stress

Post‑intervention anxiety (b = 0.81, p < .001) and rumination (b = 0.12, p = .012) predicted higher perceived stress after adjusting for baseline stress. The conditional indirect effect was significant for PAACC (b = 0.155, BCa 95% CI [0.041, 0.370]) but not for REACH‑VA (b = 0.046, BCa 95% CI [−0.005, 0.163]); the index of moderated mediation was significant (Δ  =  0.109, BCa 95% CI [0.003, 0.289]). Thus, the anxiety–perceived stress association was more strongly statistically mediated by rumination among PAACC participants.

Caregiver burden

Rumination showed a marginal effect on burden (b = 0.23, p = .07), whereas post‑intervention anxiety remained significant (b = 0.79, p = .008) after adjusting for baseline burden. The conditional indirect effect via rumination was significant for PAACC (b = 0.305, BCa 95% CI [0.014, 0.836]) but not for REACH‑VA (b = 0.088, BCa 95% CI [−0.007, 0.383]); however, the index of moderated mediation was not statistically significant (Δ  =  0.217, BCa 95% CI [−0.020, 0.683]).

Alternate models

Results from the alternate specifications that reversed the ordering of anxiety and rumination (i.e., post-rumination predicting post-anxiety), as well as a parallel-outcomes specification in which rumination predicted anxiety, depressive symptoms, perceived stress, and caregiver burden are presented in Supplementary Tables S2–S4 and Figure S1. Although these models demonstrated acceptable absolute fit, they consistently exhibited higher AIC and BIC values than the hypothesized anxiety-to-rumination model. Also, the rumination × group interaction term was nonsignificant, indicating no moderated pathway in these alternate specifications.

Discussion

The present study examined how anxiety is associated with depressive symptoms, perceived stress, and caregiver burden among dementia caregivers, and whether embedding mindfulness practices within a psychoeducational program (PAACC) alters rumination-based cognitive-emotional linkages compared with an established psychoeducational intervention (REACH-VA). As hypothesized, rumination emerged as a robust cognitive process that statistically mediated the associations between anxiety and both depressive symptoms and perceived stress, but not caregiver burden. This pattern is consistent with prior work identifying repetitive negative thinking as a key driver of emotional distress (Keune et al., 2023; Ruscio et al., 2015).

The moderation findings, however, were more complex than anticipated. Contrary to the hypothesis, mindfulness did not uniformly attenuate the anxiety–rumination association; the pattern differed by baseline anxiety levels. Simple-slopes analysis helped clarify this pattern. At low-to-moderate anxiety levels, caregivers in PAACC reported lower rumination than those in REACH-VA. In contrast, among PAACC caregivers with higher anxiety, stronger associations between anxiety and rumination were observed, alongside higher depressive symptoms and perceived stress. This pattern is consistent with prior work suggesting that mindfulness-based approaches may be differentially associated with outcomes depending on baseline distress levels, readiness, or engagement capacity, particularly in caregiver populations (Galante et al., 2023; Kor et al., 2021; Shim et al., 2021; Stone et al., 2025).

The stronger coupling of anxiety and rumination among highly anxious caregivers in PAACC allows for two complementary interpretations. One possibility is that mindfulness practices may be most beneficial when introduced before anxiety becomes overwhelming, when caregivers still have the emotional and cognitive bandwidth to disengage from emerging ruminative cycles. Once repetitive negative thinking becomes entrenched, brief mindfulness practices may be insufficient to shift automatic responses. Alternatively, mindfulness training explicitly cultivates non-judgmental awareness of thoughts and emotions, which may increase caregivers’ recognition and reporting of ruminative thinking that was previously minimized or overlooked. In this context, higher self-reported rumination may not necessarily indicate worsening but rather increased metacognitive awareness of ongoing thought processes or reflective processing, a distinction that cannot be resolved with self-report measures alone.

Prior research frequently conceptualizes rumination as a cognitive vulnerability linked to emotional distress, although evidence also suggests that rumination and anxiety are dynamically and reciprocally related (McLaughlin & Nolen-Hoeksema, 2011; Tamm et al., 2024). Consistent with this view, our supplementary analyses compared models specifying anxiety as upstream of rumination against models specifying rumination as upstream of anxiety. The hypothesized anxiety-to-rumination model showed better model fit than the alternate specifications and was the only model in which the moderated pathway was supported.

Despite these differences in cognitive pathways, overall psychological distress at post-intervention was comparable across the two intervention programs. This is consistent with meta-analytic evidence indicating that mindfulness-based interventions produce outcomes comparable to cognitive-behavioral or psychoeducational approaches when evaluated against strong active controls (Li et al., 2021). These findings suggest that caregivers may benefit from either approach and could be matched to the intervention whose underlying processes best align with their needs, preferences, and capacity to engage.

This equivalence across conditions, however, did not extend to all three outcome measures. Unlike depressive symptoms and perceived stress, which appear more tightly coupled to caregivers’ cognitive and emotional responses to stressors, burden may be less amenable to change through cognitive mechanisms alone. This distinction is consistent with the stress-process model (Pearlin et al., 1990), which frames burden as an intrapsychic strain shaped by both objective caregiving conditions and subjective appraisal. As a composite of these factors, burden may therefore be less responsive to interventions that primarily target internal coping processes, particularly when external caregiving demands continue to remain high (Bauer & Sousa-Poza, 2015; Zarit et al., 1980).

Implications

Clinically, these findings suggest the potential value of routinely screening caregivers for ruminative tendencies alongside more traditional indicators, such as care hours or behavioral symptom severity. Brief questions about “going over the same worries” or “replaying difficult moments” can identify caregivers at elevated risk for depression and perceived stress and help guide referral to appropriate supports.

The PAACC model illustrates one feasible approach: embedding short, contextually anchored mindfulness practices within dementia education and behavioral management content. This integrated format may lower barriers to engagement compared with stand-alone mindfulness programs, particularly for caregivers facing substantial time constraints.

Timing also appears important. Introducing mindfulness-based components early, when anxiety is still low to moderate, may help caregivers develop a more flexible relationship to ruminative thoughts. For caregivers with severe or long-standing anxiety, mindfulness remains useful but may need to be paired with additional support such as individual therapy, peer support, pharmacotherapy when indicated, or concrete structural relief.

Ultimately, psychoeducational intervention alone cannot offset the objective demands families shoulder. Programs like PAACC may help relieve the emotional toll of caregiving, but they cannot create respite hours, flexible workplaces, or affordable home-care services. Health systems and aging-service networks should consider incorporating brief mindfulness elements into standard caregiver programming. Families may also benefit from policies that address their practical needs, such as expanded Medicaid respite benefits, employer-based caregiver leave policies, and Medicare-covered caregiver skills training.

Limitations and future directions

The sample was predominantly White and female, limiting generalizability across cultural, gender, and socioeconomic contexts in which norms around emotional expression, family obligation, and help-seeking may differ. Most caregivers were also spouses and co‑resided with veterans receiving care within a VA health system, which may differ from community-based caregiving contexts.

As noted in the analytic strategy, the two-timepoint design limits inference about temporal dynamics or causal sequencing among anxiety, rumination, and outcomes. The trial also evaluated whether mindfulness augments an established psychoeducational intervention rather than outperforming it outright, and the strong active comparator set a high bar for detecting incremental benefits. Anxiety, depressive symptoms, and perceived stress were also correlated in this sample, so the conceptual distinctions among them should be understood as analytic rather than strictly empirical. The analytic models did not incorporate care-recipient clinical severity indices or daily caregiving stressors, which may further shape caregivers’ cognitive and emotional responses. Reliance on self-report measures raises an interpretive ambiguity noted earlier: mindfulness training may have influenced caregivers’ reporting or awareness of internal experiences as much as their underlying cognitive or emotional processes, a distinction that multimethod designs would be better positioned to address.

Future studies with larger and more diverse samples, longitudinal designs, and multimethod assessments would be better positioned to evaluate the temporal ordering and discriminant validity of these constructs. Incorporating ecological momentary assessment and physiological indicators such as heart-rate variability could help distinguish changes in awareness from changes in maladaptive perseveration. Trials with no-treatment or usual-care controls and longer follow-up periods could further clarify the unique and sustained contributions of mindfulness components.

Conclusion

Rumination appears to be a key cognitive pathway through which anxiety is statistically linked to psychological well-being among dementia caregivers in this VA sample. Mindfulness-based approaches do not uniformly reduce distress, nor can they eliminate the substantial objective burdens inherent to caregiving. Instead, they may shape how anxiety is noticed, interpreted, and mentally elaborated, especially when introduced before distress becomes overwhelming. Future interventions should pair psychological skills training with burden-reduction components, including concrete caregiver relief, to address both the emotional and objective demands of the caregiving role.

Supplementary material

Supplementary material is available at The Gerontologist online.

Supplementary Material

gnag103_Supplementary_Data
gnag103_supplementary_data.pdf (255.5KB, pdf)

Contributor Information

Jyoti Savla, Department of Human Development and Family Science, Virginia Tech, Blacksburg, Virginia, United States; Center for Gerontology, Virginia Tech, Blacksburg, Virginia, United States.

Marrium Mansoor, Department of Human Development and Family Science, Virginia Tech, Blacksburg, Virginia, United States.

Nahyun Kim, Department of Human Development and Family Science, Virginia Tech, Blacksburg, Virginia, United States.

Lauren Hagemann, Center for Aging and Neurocognitive Services, VA Salem Health Care System, Salem, Virginia, United States.

Katherine Luci, Center for Aging and Neurocognitive Services, VA Salem Health Care System, Salem, Virginia, United States; Department of Psychiatry and Behavioral Medicine, Virginia Tech Carilion School of Medicine, Roanoke, Virginia, United States.

Mamta Sapra, Center for Aging and Neurocognitive Services, VA Salem Health Care System, Salem, Virginia, United States; Department of Psychiatry and Behavioral Medicine, Virginia Tech Carilion School of Medicine, Roanoke, Virginia, United States.

Funding

This work was supported by the U.S. Department of Defense’s Peer Reviewed Alzheimer’s Research Program (PRARP)—Quality of Life Research Award (QUAL): W81XWH-17-1-0326. The content is the sole responsibility of the authors and does not necessarily represent the official views of the funding agency. M.M. and N.K. were funded by the Interdisciplinary Graduate Education and Research Program in Translational Obesity and Healthspan Research at Virginia Tech.

Conflicts of interest

None declared.

Data availability

The study data are not available because the study investigators have not completed their original work with the data set. The specific hypotheses for this study were preregistered.

Author contributions

Jyoti Savla (Conceptualization [lead], Data curation [lead], Formal analysis [lead], Funding acquisition [equal], Investigation [lead], Methodology [lead], Supervision [equal], Writing—original draft [lead], Writing—review & editing [lead]), Marrium Mansoor (Data curation [supporting], Formal analysis [supporting], Writing—original draft [supporting]), Nahyun Kim (Data curation [supporting], Writing—original draft [supporting], Writing—review & editing [supporting]), Lauren Hagemann (Funding acquisition [equal], Methodology [equal], Project administration [equal], Supervision [equal], Writing—review & editing [supporting]), Katherine Luci (Funding acquisition [supporting], Methodology [equal], Writing—review & editing [supporting]), and Mamta Sapra (Funding acquisition [equal], Methodology [equal], Project administration [lead], Supervision [lead], Writing—review & editing [supporting])

Consent statement

Written informed consent was obtained from all participants before enrollment in the study.

References

  1. Alfakhri A. S., Alshudukhi A. W., Alqahtani A. A., Alhumaid A. M., Alhathlol O. A., Almojali A. I., Alotaibi M. A., Alaqeel M. K. (2018). Depression among caregivers of patients with dementia. Inquiry: A Journal of Medical Care Organization, Provision and Financing, 55, 46958017750432–46958017750437. 10.1177/0046958017750432 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aneshensel C. S., Pearlin L. I., Mullan J. T., Zarit S. H., Whitlatch C. J. (1995). Profiles in caregiving: The unexpected career. Academic Press. [Google Scholar]
  3. Bauer J. M., Sousa-Poza A. (2015). Impacts of informal caregiving on caregiver employment, health, and family. Journal of Population Ageing, 8, 113–145. 10.1007/s12062-015-9116-0 [DOI] [Google Scholar]
  4. Bishop S. R., Lau M., Shapiro S., Carlson L., Anderson N. D., Carmody J., Segal Z. V., Abbey S., Speca M., Velting D., Devins G. (2004). Mindfulness: A proposed operational definition. Clinical Psychology: Science and Practice, 11, 230–241. 10.1093/clipsy.bph077 [DOI] [Google Scholar]
  5. Cerejeira J., Lagarto L., Mukaetova-Ladinska E. B. (2012). Behavioral and psychological symptoms of dementia. Frontiers in Neurology, 3, 73. 10.3389/fneur.2012.00073 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cohen S., Kamarck T., Mermelstein R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385–396. 10.2307/2136404 [DOI] [PubMed] [Google Scholar]
  7. Cooper C., Balamurali T. B. S., Livingston G. (2007). A systematic review of the prevalence and covariates of anxiety in caregivers of people with dementia. International Psychogeriatrics, 19, 175–195. 10.1017/S1041610206004297 [DOI] [PubMed] [Google Scholar]
  8. Deeken F., Häusler A., Nordheim J., Rapp M., Knoll N., Rieckmann N. (2018). Psychometric properties of the Perceived Stress Scale in a sample of German dementia patients and their caregivers. International Psychogeriatrics, 30, 39–47. 10.1017/S1041610217001387 [DOI] [PubMed] [Google Scholar]
  9. Eysenck M. W., Fajkowska M. (2018). Anxiety and depression: Toward overlapping and distinctive features. Cognition & Emotion, 32, 1391–1400. 10.1080/02699931.2017.1330255 [DOI] [PubMed] [Google Scholar]
  10. Galante J., Friedrich C., Aeamla-Or N., Arts-de Jong M., Barrett B., Bögels S. M., Buitelaar J. K., Checovich M. M., Christopher M. S., Davidson R. J., Errazuriz A., Goldberg S. B., Greven C. U., Hirshberg M. J., Huang S.-L., Hunsinger M., Hwang Y.-S., Jones P. B., Medvedev O. N., White I. R.; Collaboration of Mindfulness Trials (CoMinT). (2023). Systematic review and individual participant data meta-analysis of randomized controlled trials assessing mindfulness-based programs for mental health promotion. Nature Mental Health, 1, 462–476. 10.1038/s44220-023-00081-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gao C., Chapagain N. Y., Scullin M. K. (2019). Sleep duration and sleep quality in caregivers of patients with dementia: A systematic review and meta-analysis. JAMA Network Open, 2, e199891. 10.1001/jamanetworkopen.2019.9891 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Guendelman S., Medeiros S., Rampes H. (2017). Mindfulness and emotion regulation: Insights from neurobiological, psychological, and clinical studies. Frontiers in Psychology, 8, 220. 10.3389/fpsyg.2017.00220 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hawley L. L., Schwartz D., Bieling P. J., Irving J., Corcoran K., Farb N. A. S., Anderson A. K., Segal Z. V. (2014). Mindfulness practice, rumination and clinical outcome in mindfulness-based treatment. Cognitive Therapy and Research, 38, 1–9. 10.1007/s10608-013-9586-4 [DOI] [Google Scholar]
  14. Hébert R., Bravo G., Préville M. (2000). Reliability, validity, and reference values of the Zarit Burden Interview for assessing informal caregivers of community-dwelling older persons with dementia. Canadian Journal on Aging/La Revue Canadienne du Vieillissement, 19, 494–507. 10.1017/S0714980800012484 [DOI] [Google Scholar]
  15. Joling K. J., van Marwijk H. W. J., Veldhuijzen A. E., van der Horst H. E., Scheltens P., Smit F., van Hout H. P. J. (2015). The two-year incidence of depression and anxiety disorders in spousal caregivers of persons with dementia: Who is at the greatest risk?  The American Journal of Geriatric Psychiatry, 23, 293–303. 10.1016/j.jagp.2014.05.005 [DOI] [PubMed] [Google Scholar]
  16. Keune P. M., Bernard F., Meister R., Keune J., Springer R., Arlt R., Holmerová I., Auer S., Oschmann P., Hansen S. (2023). Mindfulness in persons with mild dementia and their caregivers: Exploring trait rumination as a clinical outcome measure. Mindfulness, 14, 1908–1917. 10.1007/s12671-023-02192-7 [DOI] [Google Scholar]
  17. Kor P. P. K., Liu J. Y. W., Chien W. T. (2021). Effects of a modified mindfulness-based cognitive therapy for family caregivers of people with dementia: A randomized clinical trial. The Gerontologist, 61, 977–990. 10.1093/geront/gnaa125 [DOI] [PubMed] [Google Scholar]
  18. Kroenke K., Spitzer R. L., Williams J. B. W. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16, 606–613. 10.1046/j.1525-1497.2001.016009606.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lazarus R. S., Folkman S. (1984). Stress, appraisal, and coping. Springer Publishing Company. [Google Scholar]
  20. Li J., Cai Z., Li X., Du R., Shi Z., Hua Q., Zhang M., Zhu C., Zhang L., Zhan X. (2021). Mindfulness-based therapy versus cognitive behavioral therapy for people with anxiety symptoms: A systematic review and meta-analysis of randomized controlled trials. Annals of Palliative Medicine, 10, 7596–7612. 10.21037/apm-21-1212 [DOI] [PubMed] [Google Scholar]
  21. Liu Y., Dokos M., Fauth E. B., Lee Y. G., Zarit S. H. (2019). Financial strain, employment, and role captivity and overload over time among dementia family caregivers. The Gerontologist, 59, e512–e520. 10.1093/geront/gnz099 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lou Q., Liu S., Huo Y. R., Liu M., Liu S., Ji Y. (2015). Comprehensive analysis of patient and caregiver predictors for caregiver burden, anxiety, and depression in Alzheimer’s disease. Journal of Clinical Nursing, 24, 2668–2678. 10.1111/jocn.12870 [DOI] [PubMed] [Google Scholar]
  23. McLaughlin K. A., Nolen-Hoeksema S. (2011). Rumination as a transdiagnostic factor in depression and anxiety. Behaviour Research and Therapy, 49, 186–193. 10.1016/j.brat.2010.12.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. National Alliance for Caregiving & AARP. (2020). Caregiving in the U.S. 2020. National Alliance for Caregiving & AARP. https://www.aarp.org/ppi/info-2020/caregiving-in-the-united-states.html
  25. Nolen-Hoeksema S., Wisco B. E., Lyubomirsky S. (2008). Rethinking rumination. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 3, 400–424. 10.1111/j.1745-6924.2008.00088.x [DOI] [PubMed] [Google Scholar]
  26. Oh V. K. S., Sarwar A., Pervez N. (2022). The study of mindfulness as an intervening factor for enhanced psychological well-being in building the level of resilience. Frontiers in Psychology, 13, 1056834. 10.3389/fpsyg.2022.1056834 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pearlin L. I., Mullan J. T., Semple S. J., Skaff M. M. (1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30, 583–594. 10.1093/geront/30.5.583 [DOI] [PubMed] [Google Scholar]
  28. Peavy G. M., Mayo A. M., Avalos C., Rodriguez A., Shifflett B., Edland S. D. (2022). Perceived stress in older dementia caregivers: Mediation by loneliness and depression. American Journal of Alzheimer’s Disease & Other Dementias, 37, 1–10. 10.1177/15333175211064756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pinquart M., Sörensen S. (2003). Differences between caregivers and non-caregivers in psychological health and physical health: A meta-analysis. Psychology and Aging, 18, 250–267. 10.1037/0882-7974.18.2.250 [DOI] [PubMed] [Google Scholar]
  30. Pinyopornpanish M., Pinyopornpanish K., Soontornpun A., Tanprawate S., Nadsasarn A., Wongpakaran N., Wongpakaran T. (2021). Perceived stress and depressive symptoms—not neuropsychiatric symptoms—predict caregiver burden in Alzheimer’s disease: A cross-sectional study. BMC Geriatrics, 21, 180. 10.1186/s12877-021-02136-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Puga F., Wang D., Rafford M., Poe A., Pickering C. E. Z. (2023). The relationship between daily stressors, social support, depression and anxiety among dementia family caregivers: A micro-longitudinal study. Aging & Mental Health, 27, 1291–1299. 10.1080/13607863.2022.2116392 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Romero-Moreno R., Losada A., Márquez-González M., Mausbach B. T. (2016). Stressors and anxiety in dementia caregiving: Multiple mediation analysis of rumination, experiential avoidance, and leisure. International Psychogeriatrics, 28, 1835–1844. 10.1017/S1041610216001009 [DOI] [PubMed] [Google Scholar]
  33. Ruscio A., Gentes E., Jones J., Hallion L., Coleman E., Swendsen J. (2015). Rumination predicts heightened responding to stressful life events in major depressive disorder and generalized anxiety disorder. Journal of Abnormal Psychology, 124, 17–26. 10.1037/abn0000025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Sallim A. B., Sayampanathan A. A., Cuttilan A., Ho R. C.-M. (2015). Prevalence of mental health disorders among caregivers of patients with Alzheimer disease. Journal of the American Medical Directors Association, 16, 1034–1041. 10.1016/j.jamda.2015.09.007 [DOI] [PubMed] [Google Scholar]
  35. Sapra M., Hagemann L., Luci K., Savla J. (2025). A mindfulness-based multicomponent caregiver intervention (PAACC): Objectives, study design, and cohort description. Alzheimer’s & Dementia (New York, N. Y.), 11, e70064. 10.1002/trc2.70064 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Science of Behavior Change. (n.d.). CLIMBR tool: Checklist for investigating mechanisms in behavior-change research. https://scienceofbehaviorchange.org/climbr-tool/
  37. Segerstrom S. C., Roach A. R., Evans D. R., Schipper L. J., Darville A. K. (2010). The structure and health correlates of trait repetitive thought in older adults. Psychology and Aging, 25, 505–515. 10.1037/a0019456 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Seng B. K., Luo N., Ng W. Y., Lim J., Chionh H. L., Goh J., Yap P. (2010). Validity and reliability of the Zarit Burden Interview in assessing caregiving burden. Annals of the Academy of Medicine, Singapore, 39, 758–763. 10.47102/annals-acadmedsg.V39N10p758 [DOI] [PubMed] [Google Scholar]
  39. Shim M., Tilley J. L., Im S., Price K., Gonzalez A. (2021). A systematic review of mindfulness-based interventions for patients with mild cognitive impairment or dementia and caregivers. Journal of Geriatric Psychiatry and Neurology, 34, 528–554. 10.1177/0891988720957104 [DOI] [PubMed] [Google Scholar]
  40. Sörensen S., Conwell Y. (2011). Issues in dementia caregiving: Effects on mental and physical health, intervention strategies, and research needs. The American Journal of Geriatric Psychiatry: Official Journal of the American Association for Geriatric Psychiatry, 19, 491–496. 10.1097/JGP.0b013e31821c0e6e [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Spitzer R. L., Kroenke K., Williams J. B., Löwe B. (2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166, 1092–1097. 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
  42. Stone S. M., Giraldo-Santiago N., Miller A., O’Donnell A., Travis A., Ritchie C. S., Vranceanu A.-M. (2025). Development of a mindfulness and self-compassion (MASC) stress reduction program for caregivers of persons with dementia with behavioral symptoms. The Gerontologist, 65, gnaf058. 10.1093/geront/gnaf058 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tamm G., Koster E. H. W., Hoorelbeke K. (2024). Multiple paths to rumination within a network analytical framework. Scientific Reports, 14, 10874. 10.1038/s41598-024-61469-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Trapnell P. D., Campbell J. D. (1999). Private self-consciousness and the five-factor model of personality: Distinguishing rumination from reflection. Journal of Personality and Social Psychology, 76, 284–304. 10.1037/0022-3514.76.2.284 [DOI] [PubMed] [Google Scholar]
  45. Waldman-Levi A., Finzi-Dottan R., Cope A. (2020). Mother–child joint play: The role of maternal caregiving and reflective function. Journal of Child and Family Studies, 29, 94–104. 10.1007/s10826-019-01638-8 [DOI] [Google Scholar]
  46. Walter E., Pinquart M. (2020). How effective are dementia caregiver interventions? An updated comprehensive meta-analysis. The Gerontologist, 60, e609–e619. 10.1093/geront/gnz118 [DOI] [PubMed] [Google Scholar]
  47. Watson B., Tatangelo G., McCabe M. (2019). Depression and anxiety among partner and offspring carers of people with dementia: A systematic review. The Gerontologist, 59, e597–e610. 10.1093/geront/gny049 [DOI] [PubMed] [Google Scholar]
  48. Zarit S. H., Reever K. E., Bach-Peterson J. (1980). Relatives of the impaired elderly: Correlates of feelings of burden. The Gerontologist, 20, 649–655. 10.1093/geront/20.6.649 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gnag103_Supplementary_Data
gnag103_supplementary_data.pdf (255.5KB, pdf)

Data Availability Statement

The study data are not available because the study investigators have not completed their original work with the data set. The specific hypotheses for this study were preregistered.

Articles from The Gerontologist are provided here courtesy of Oxford University Press

RESOURCES