Evaluating reliable and clinically significant changes in health outcomes of a mindfulness-based cognitive defusion training program among older adults with mild cognitive impairment: a randomized controlled trial

Abstract

Background

Recognizing perceived stress as a modifiable risk factor, mindfulness-based programs show promise for stress mitigation in older adults with mild cognitive impairment (MCI).

Objective

To assess the efficacy of a mindfulness-based contextual cognitive defusion training (M-bCCDT) program on perceived stress and other health outcomes, and to examine the reliable and clinically significance of these improvements at individual-level among older adults with MCI.

Design

A two-arm, assessor-blinded randomized controlled trial.

Settings and participants

102 community-dwelling older adults with MCI.

Methods

Participants were randomly allocated to either a M-bCCDT program (weekly 60-minute sessions for 8 weeks, followed by 12 weeks of unsupervised practice) or health promotion classes. Measures of perceived stress, memory function, global cognitive function, psychomotor speed and mindfulness awareness were collected at baseline (T0), 8-week (T1) and 20-week (T2). Intervention effects were assessed at a group level (Generalized Estimating Equation, GEE) and individual level (Reliable and Clinically Significant Changes, RCSC).

Results

The M-bCCDT program demonstrated significant interaction effects in perceived stress compared to the wait-list control group by GEE analysis (β_T1 = −3.686, 95% CI [−5.397, −1.976]; β_T2 = −7.608, 95% CI [−9.387, −5.829]). Furthermore, this program also showed significant efficacy in memory function, psychomotor speed and mindfulness awareness. RCSC indicated that 30 participants (59%) in the intervention group showed statistically significant improvement in perceived stress at 8-week, with 7 (14%) clinically significant. This increased to 38 (75%) with 20 (39%) clinically significant at 20-week. Secondary outcomes also showed statistically and clinically significant improvements over time, but no improvement in global cognitive function at the individual level.

Conclusions

The M-bCCDT program positively impacted perceived stress and mindfulness awareness in older adults with MCI, facilitating the improvements in memory and psychomotor speed, with these benefits sustained for 20 weeks. It offers a systematic approach for community healthcare providers in MCI stress management.

Key Points

• The mindfulness-based contextual cognitive defusion training (M-bCCDT) impacts on stress and other outcomes in older mild cognitive impairment (MCI) adults.

• Individual effect by Reliable and Clinically Significant Change (RCSC) of M-bCCDT was less significant than the population effect shown by Generalized Estimating Equation, despite an increasing trend.

• Clinical significance cutoffs for stress and other outcomes were set by RCSC, offering benchmarks for intervention goals.

Introduction

Older adults with mild cognitive impairment (MCI) exhibit diminished responsiveness to external stimuli due to impaired cognitive function, making them more prone to experiencing negative emotions and being highly susceptible to stress, contributing to their elevated perceived stress levels [1]. Perceived stress arises from an individual’s cognitive appraisal of environmental threats, manifesting as psychological distress [2]. Excessive perceived stress may lead to heightened tension and emotional dysregulation, impeding the development of cognition and undermining capacities for self-management [3]. Notably, there is an independent association between perceived stress and both prevalent and incident cognitive impairment [4]. High perceived stress is more common in older adults with MCI, who face 91% of stressors compared to 65% in normal older adults [5]. These stressors primarily arise from memory loss and difficulties with cognitive tasks essential for daily life, such as managing medications and finances [5, 6]. When stress levels exceed their coping capacity, it triggers significant perceived stress, commonly associated with psychological issues such as anxiety and depression, along with physiological responses like elevated heart rates [7]. Given that perceived stress is a modifiable risk factor for MCI, it is imperative for clinicians and policymakers to incorporate stress-reduction strategies into broader initiatives aimed at preserving cognitive function in aging populations.

Previous studies have indicated that mindfulness-based interventions (MBIs), like mindfulness-based stress reduction (MBSR) [8] and mindfulness-based cognitive therapy (MBCT) [9], yield positive outcomes in reducing stress indicators, improving cognitive performance and enhancing brain plasticity [8–10]. Research on mindfulness and its impact on MCI has also explored various interventions, including MBSR, MBCT, meditation and Yoga, etc., to mitigate cognitive decline and enhance mental well-being in individuals with MCI [8, 9, 11, 12]. However, recent meta-analyses [13, 14] have found limited significant effects of MBIs on key health-related outcomes, likely due to methodological limitations. These include the absence of tailored strategies to address specific stress contexts faced by individuals with MCI, such as memory loss and the challenges of managing daily tasks. Furthermore, most intervention studies have relied heavily on statistical comparisons of group means, which often fail to capture clinically significant changes at the individual level. These gaps underscore the need for alternative, context-specific approaches to improve the effectiveness of MBIs, supported by more precise and refined analytical techniques to accurately assess their true impacts.

Cognitive defusion, a core component of Acceptance and Commitment Therapy (ACT), is grounded in relational frame theory (RFT) [15]. It involves altering the way individuals relate to their thoughts, encouraging them to perceive thoughts as transient mental events rather than absolute truths [16]. The efficacy of cognitive defusion in reducing the impact of negative thoughts on emotional wellbeing has been supported by previous studies [17]. This technique can mitigate negative thinking patterns’impact on behaviour and emotions, which could be particularly relevant for individuals with MCI who may experience cognitive rigidity or distressing thoughts related to their cognitive decline. Conversely, mindfulness, also an ACT component, emphasizes cultivating present-moment awareness and non-judgmental acceptance of experiences. It is grounded in principles of Buddhist meditation practices and contemporary psychological frameworks, such as MBSR and MBCT [8, 9]. These practices aim to enhance psychological flexibility and emotional regulation by encouraging individuals to fully engage with their current experiences without avoidance or over-engagement [18]. While both techniques converge on the goal of enhancing psychological flexibility, their pathways could be complementary. MBSR/MBCT have established efficacy across various populations, and emerging evidence suggests that combining mindfulness with cognitive defusion may offer additional benefits [10, 19]. Prior studies have also shown that cognitive defusion is associated with more approach- and less avoidance-oriented coping strategies for managing stress [20]. Furthermore, research indicated that mindfulness and cognitive defusion techniques tailored to specific stressors or stress contexts can more effectively alleviate stress and improve mental well-being [19]. Older adults with MCI often face specific stressors related to memory loss and the management of daily tasks, which can heighten distress and perceived stress levels [6]. However, no studies have yet combined mindfulness with cognitive defusion to address perceived stress in these specific stress contexts in this population, and the efficacy of this integrated approach requires further evaluation.

Additionally, in the evaluation of intervention effects of MBIs, the reliance on statistical comparisons of group means in most intervention studies often fails to capture clinically significant changes at the individual level, underscoring the need for more sensitive and precise analytical approaches to evaluate both population-average effects and individual-level changes. The Generalized Estimating Equation (GEE) accounts for intra-individual correlations and evaluates population-averaged effects, making it suitable for longitudinal intervention studies [21]. Reliable and Clinically Significant Change (RCSC) analysis is a robust approach that assesses both statistical and clinical significance at the individual level, addressing criticisms of traditional methods that focus primarily on statistical significance and lack sensitivity to individual variations [22].

To address these challenges, this study aimed to develop and validate a mindfulness-based contextual cognitive defusion training (M-bCCDT) program tailored to the specific stress contexts of individuals with MCI. Using the RCSC analysis, the study sought to answer two key questions: (i) What are the impacts of the M-bCCDT program on health outcomes in older adults with MCI? Hypothesis 1: This program would have positive effects on perceived stress, cognitive function and mindfulness awareness compared to a wait-list control group. (ii) Whether participation in the M-bCCDT program results in reliable and clinically significant improvements in health outcomes among older adults with MCI. Hypothesis 2: The magnitude of changes observed in the intervention group will exceed that of the wait-list control group as determined by RCSC analysis, with demonstrating both statistical and clinical significance.

Method

Study design and ethical consideration

This study was an assessor-blinded, randomized, controlled trial, for which a total of 102 older adults with MCI were recruited from September 2020 to March 2023. This study was approved by the Institutional Review Board of the third affiliated hospital of Huzhou University in accordance with the guidelines of the Declaration of Helsinki and registered at Chictr.org.cn (ChiCTR2000033013). Written informed consent was obtained from each participant. The study report complied with the CONSORT statement [23].

Participants

Older adults were recruited at community health centers and senior centers. Recruitment leaflets were distributed by study staff, and local healthcare providers also helped making referrals to our study. The inclusion criteria: (i) age ≥ 60 years old; (ii) diagnosis of MCI; (iii) absence of self-reported visual or auditory impairment; (iv) perceived stress performance as evidenced by Chinese Perceived Stress Scale (CPSS) score ≥ 26; (v) able to make an informed consent. The operational criteria adopted for MCI screening included [24]: (i) report of a relative decline in cognitive function during the past year by the participant or informant; (ii) objective cognitive impairment inconsistent with age and education-Montreal Cognitive Assessment-Basic (MoCA-B) scores 18 for those with elementary education level, scores 21 for those with secondary education level, scores 23 for those with tertiary education level [25]; (iii) intact activities of daily living (Lawton-Brody Activities of Daily Living score ≤ 16); (iv) absence of dementia.

Participants were excluded if they reported: (i) a history of neurological, psychiatric and other severe medical issues that may affect brain function; (ii) a history of alcohol dependency or other addiction within 10 years; (iii) taking any medications in the past 6 months which may cause impaired or improved cognitive performance (e.g. memantine and antipsychotic); (iv) participation in other research projects. Participants were allowed, or required, to withdraw from the trial based on the following: development of a serious disease preventing continuation in the trial; request to be withdrawn from the trial.

Individuals expressing interest were invited for an in-person eligibility screening conducted by three trained staff, followed by a diagnostic assessment from a trained neurologist-psychiatrist to confirm MCI. A total of 376 individuals with subjective memory impairment underwent eligibility assessments.

Sample size estimation

Based on our pilot study, there was a large effect (Cohen’s d = 2.04, effect size = 0.71) on perceived stress status. With α set at 0.05 and β at 0.95, the power analysis indicated that 44 participants were required for each group to detect a large effect (N = 88). Considering the expected attrition, the decision was made to oversample. Finally, the sample size was 102, which was calculated by the 15% rate of lost follow up.

Randomization, allocation concealment and blinding

After informed consent and baseline assessment, eligible participants were randomly assigned 1:1 to either the intervention or wait-list control group using a computer-generated random number table (simple randomization), resulting in 51 participants in each group. The random allocation sequence was managed by a statistician blinded to recruitment, evaluation and intervention. Assignments were sealed in opaque, sequentially numbered envelopes and stored by a study assistant uninvolved in recruitment or outcome assessments. Due to the nature of the intervention, participants and intervention coordinators were not blinded, but outcome assessors remained blinded to group allocation. Participants were informed of their group allocation by the study coordinator. To minimize experimental contamination, the interventions for each group were delivered at separate sites.

Intervention

The M-bCCDT program was developed by a multidisciplinary team consisting of a nursing specialist, neuropsychologist, psychotherapist, and occupational therapists. It is characterized by focusing on common stressful contexts in older adults with MCI, integrating mindfulness and cognitive defusion techniques into stress management training to enhance coping skills, build mental resilience and facilitate cognitive restructuring, ultimately aiming to improve cognitive and emotional well-being. The program sessions are exhibited in Figure 1.

Wait-list control group

Participants in the wait-list control group attended a 30-minute weekly health education program for the 8 weeks, delivered by a general practitioner at the community health service center in groups of up to six people. This program included information on cognitive impairment, risk factors related to MCI, healthy eating and living habits and emotion regulation strategies. Participants were also informed that they would receive the M-bCCDT program after follow-up assessment. For any questions inquired by the wait-list control group participants, general advice, but not information relating to mindfulness-based cognitive defusion training, was given for ethical consideration.

Intervention group

In addition to the same health education program given to the wait-list control group, intervention group participants attended supervised M-bCCDT sessions/week, 60 min/session for 8 weeks in small groups with a maximum number of six people per group, practicing at a community healthcare center. This program was divided into three parts. The first part involved theoretical and techniques learning on mindfulness and cognitive defusion techniques. The second part involved practicing stress coping training using mindfulness and cognitive defusion techniques in three stressful contexts: medication management, financial management and memory training. The third part was a summary session where the researcher assessed participants’mastery of mindfulness and cognitive defusion techniques. Participants discussed their experiences and changes resulting from the M-bCCDT program, and collaboratively developed a plan for coping with potential future stressors. The details of this program were exhibited in Supplementary eTable 1.

All training commences with in-person instruction provided and supervised by one researcher and two coordinators following a standardized intervention protocol. The study coordinator informed participants of the next training time and contacted them 2 days prior to ensure on-time attendance. Those unable to attend were offered an alternative appointment. After completion of the 8 weeks training sessions, participants then were asked to practice unsupervised M-bCCDT sessions/week, 60 min/session for the following 12 weeks at home following the intervention protocol. A detailed instruction manual and guided audio were provided to assist participants in practicing the program. Additionally, participants received a training logbook to record weekly logs of training methods, durations, difficulties encountered and insights gained. Study coordinators conducted weekly phone calls to monitor progress and address any issues. If participants did not meet the prescribed duration or frequency of training, in-person or telephone was arranged to offer additional assistance.

All participants in both groups were asked to maintain their routine lifestyle, and were advised not to seek any other psychological intervention during the study period. All participants were asked to report if they changed their routines (including supplement and medication use, attending a recreation club, hospital admissions and illnesses condition).

Outcomes and measuring instruments

An investigator-developed questionnaire was used to collect basic demographic and medical characteristics of participants.

Primary outcome

Perceived stress was assessed by the 14-item Chinese Perceived Stress Scale (CPSS), validated by Yang et al. [26]. This scale measures two dimensions (sense of losing control and sense of tension), with 26 or above indicating a significant health risk due to stress. The Cronbach’s alpha of this scale was 0.88 [27].

Secondary outcomes

• (1)

  Global cognitive function The Chinese version of Montreal Cognitive Assessment Basic (MoCA-B) was used to assess global cognitive function. The optimal cut-off scores for MCI screening: 19 for individuals with ≤ 6 years of education, 22 for those with 7-12 years and 24 for those with > 12 years, and the reported Cronbach’s α was 0.807 [25].

• (2)

  Psychomotor speed The Finger Tapping Test (FTT) and the Purdue Pegboard Test (PPT) were used to measure the psychomotor speed. The FTT (Torrance, CA 90503–5124) requires participant quickly and accurately tapping a 1.5 cm square keyboard from 20 cm above with their dominant index finger within 15 seconds, and the total of accurate taps as the test result. The reliability coefficient for the FTT was 0.78 [28]. The PPT (Model 32020A, Lafayette, US) includes tasks using right hand (PPTR), left hand (PPTL) and bimanual (PPTB), as well as assembling equipment (PPTA). Scores are based on the accurate placement of nails for each method. Previous study has supported the reliability and validity of PPT [29].

• (3)

  Memory function Memory function was assessed by the Chinese version of Auditory Verbal Learning Test (AVLT-H) [30]. A 12-word list is read by the assessor at pace of one word per second. Participant was then required to memorize and promptly recall these words across three consecutive tests, with the recall counts noted as N1, N2, and N3. After a 20-minute interval, participant need to perform a long-delayed recall of these words without prompts, with the number of correct recalls recorded as N4. AVLT immediate recall: N1 + N2 + N3, AVLT delayed recall: N4 [31]. The reported Cronbach’s α was 0.99 [30].

• (4)

  Mindfulness awareness The Five Facet Mindfulness Questionnaire (FFMQ) was used to assess participants’emotional condition during the mindfulness-based cognitive defusion training. The previous study confirmed its reliability in older adults, with a Cronbach’s α of 0.848 [32].

Data collection

The basic characteristics were measured at baseline (T0); CPSS, MoCA-B, FTT, PPTA, PPTB, PPTL, PPTR, AVLT-immediate/delayed recall and FFMQ were measured at baseline, 8 weeks (T1) and 20 weeks after baseline (T2). Trained assessors conducted all outcome measures in community healthcare centers.

Statistical analysis

Data were analysis in SPSS 27.0 (SPSS Inc., USA) and R version 4.3.1. Exploratory analysis was used to determine the data’s distribution characteristics. Baseline characteristics differences between groups were analysed through chi-square tests for categorical variables, t-tests for variables conforming to normal distribution and Wilcoxon Rank Sum Tests for variables deviating from normal distribution.

Intervention effects validation

GEE was utilized to quantify variations (mean differences) in the primary variable and secondary outcomes between two groups at T1 and T2, using T0 as the reference point. An unstructured correlation matrix was chosen for data analysis.

Statistical and clinical significance analysis

Furthermore, the RCSC were calculated and plotted using the Jacobson-Truax (JT) method [22].

• (1) Reliable change index (RCI): comparison of individual the standardized difference (change) score with the 1.96 to assess whether improvements observed in participant over the intervention course are statistically significant [33]. The RCI and standardized differences were calculated using the following formula [34], which includes the standard deviation (SD) of the control group, and the coefficient α (reliability) for the measuring tool. Group differences in the proportions of participants exhibiting reliable improved (RCI < -1.96), deteriorated (RCI > 1.96), or no change were tested using the chi-square test.

  $$ {\mathbf{S}}_{\boldsymbol{Diff}}=\sqrt{\mathbf{2}\ast{\left(\boldsymbol{SD}\ast \sqrt{\left(\mathbf{1}-\boldsymbol{reliability}\right)}\right)}^{\mathbf{2}}} $$

• (2) Clinically significant change (CSC): individual improvement is considered clinically significant when post-intervention score exceeds the clinical cut-off, determined by RCI (statistically significant individual improvement) [34]. The clinical Cutoff A criterion, defined as the point 2 SDs beyond the range of the pre-intervention mean was used [35].

Result

Recruitment, attrition and adherence

As shown in the CONSORT flow diagram (Figure 2), 102 participants were randomly assigned into two groups at 1:1 ratio. Eight participants dropped out from the intervention group and seven from the wait-list control group due to family care, health issue and other reasons, resulting in 43 and 44 participants remaining, respectively. The compliance rate in the intervention group was 84.3%. There were no study-related injuries. Following the intention-to-treat principle, missing data were imputed with the last observation carried forward, analysing data from 102 participants.

Descriptive data of participant characteristics

Of the 102 participants, the mean age was 72.74 ± 4.68 years and 67 (65.69%) were female. Table 1 presents the baseline characteristics of participants by group, which were comparable between two groups. The between-group mean differences (95% CI) for age was −0.059 (−1.917, 1.799), further indicating comparability.

Impact of the M-bCCDT program on outcomes

Study outcomes for GEE analysis

GEE analysis showed significant group*time effects for the intervention group on CPSS total score (T1: β = −3.686, 95% CI [−5.397, −1.976]; T2: β = −7.608, 95% CI [−9.387, −5.829]), feeling of tension dimension (T1: β = −1.765, 95% CI [−2.663, −0.866]; T2: β = −3.765, 95% CI [−4.679, −2.850]) and feeling of losing control dimension (T1: β = −1.902, 95% CI [−2.847, −0.957]; T2: β = −3.824, 95% CI [−4.817, −2.830]) compared to the wait-list control group, demonstrating the M-bCCDT program’s efficacy on perceived stress over 20 weeks (Table 2).

Significant group*time effects on MoCA-B score for the intervention group at T2 (β = 2.529, 95% CI [1.530, 3.529]), with a marginal effect at T1 (β = 1.020, 95% CI [−0.018, 2.057]) (Table 3), indicating a gradual improvement in global cognitive function over time. Compared to the wait-list control group in psychomotor speed, the intervention group showed significant improvements in FTT (T1: β = 1.020, 95% CI [0.123, 1.917]; T2: β = 3.000, 95% CI [2.027, 3.973]), PPTA (T1: β = 0.549, 95% CI [0.004, 1.094]; T2: β = 1.510, 95% CI [0.948, 2.071]) and PPTB (T1: β = 0.784, 95% CI [0.165, 1.403]; T2: β = 2.059, 95% CI [1.378, 2.740]) after GEE analysis (Table 3), suggesting the efficacy of M-bCCDT program in enhancing psychomotor speed with sustained benefits for 20 weeks. Despite the group*time effects on PPTL and PPTR at T1, significance emerged by T2, indicating a delayed but positive impact. As shown in Table 3, the effect of group*time on AVLT immediate recall (T1: β = 1.098, 95% CI [0.542, 1.654]; T2: β = 2.176, 95% CI [1.542, 2.811]) and AVLT delayed recall (T1: β = 0.667, 95% CI [0.038, 1.295]; T2: β = 1.373, 95% CI [0.697, 2.048]) were significant at T1 and T2 in the intervention group, meaning that the M-bCCDT program had effects on memory function and these effects lasted for 20 weeks.

Table 3 indicates that, compared to the wait-list control group, the group*time effects on FFMQ (T1: β = 3.510, 95% CI [1.401, 5.618]; T2: β = 7.020, 95% CI [4.901, 9.138]) in the intervention group at T1 and T2 reached was significant, demonstrating the benefit of the M-bCCDT program on mindfulness awareness maintained for up to 20 weeks.

In the wait-list control group, MoCA-B scores improved at T1 (β = 0.235, 95% CI [0.067, 0.404]) but not at T2, indicating temporary cognitive improvement. Similarly, improvements in the FTT (β = 0.137, 95% CI [0.028, 0.246]), PPTL (β = 0.078, 95% CI [0.005, 0.152]), PPTR (β = 0.176, 95% CI [0.047, 0.306]) and AVLT delayed recall (β = 0.157, 95% CI [0.057, 0.257]) at T1 did not persist at T2. With no improvement in PPTA, PPTB, AVLT immediate recall and FFMQ at any time.

Study outcomes for RCSC analysis

Figure 3A and Figure 3B demonstrate RCSC analysis applied to assess the impact of the M-bCCDT program on the perceived stress. The RCI was established at 4.18 and the CSC cutoff was set at 23.64. In the intervention group, 30 (59%) demonstrated statistically significant improvement at T1, with 7 (14%) achieving clinically significant improvements. Additionally, 21 (41%) remained unchanged, and none experienced deterioration. By T2, 38 (75%) showed statistically significant improvement, 20 (39%) also met the criteria for clinical significance; 13 (25%) remained unchanged, with no deterioration observed. In the wait-list control group, no participant showed reliable improvement at T1 and T2, respectively significantly less than those in the intervention group (T1: χ² = 42.500, P < .001; T2: χ² = 60.563, P < .001).

For the RCSC analysis of AVLT immediate recall, AVLT delayed recall, PPTA, PPTL and FTT exhibited statistically significant at T1 and T2 compared to the wait-list control group. In MoCA-B, while none experienced statistically significant at T1 and four participants exhibited statistically significant at T2, these were not significantly different from those in the wait-list control group. Conversely, for the RCSC analysis of FFMQ, PPTB, and PPTR no statistically significant improvement was observed at T1, but a significantly higher number of participants showed improvement at T2 compared to the wait-list control group. RCSC analysis of secondary outcomes were detailed in Supplementary eFigures 1–4.

Discussion

Summary of the findings

To our knowledge, this is the first study used the RCSC analyses to evaluate the intervention effects of a M-bCCDT program at the individual level among older adults with MCI. This study supports Hypothesis 1, demonstrating that the M-bCCDT program has significant and sustainable benefits for improving perceived stress, memory function, psychomotor speed and mindfulness awareness, with benefits lasting for 20 weeks. Hypothesis 2 was also confirmed, as the intervention group demonstrated a higher proportion of participants achieving statistically and clinically meaningful improvements in perceived stress, memory function and psychomotor speed at 8 and 20 weeks, and mindfulness awareness at 20 weeks compared to the wait-list control group. These findings offer a new approach to evaluating intervention efficacy using RCI and CSC cutoffs to measure individual-level changes reliably.

Comparison with other studies

Health-related benefits of the M-bCCDT program

Previous studies have demonstrated that mindfulness-based cognitive defusion leads to improvements in well-being [36], which is consistent with this study. The M-bCCDT program significantly reduced perceived stress compared with the wait-list control, with sustained benefits at 20 weeks. This effect may reflect the unique integration of mindfulness and cognitive defusion techniques within the M-bCCDT program, which offers two complementary pathways to stress reduction. The program helps individuals detach from thoughts and emotions by viewing them as temporary mental events, fostering present-moment awareness and acceptance. By combining immediate emotional relief with long-term coping strategies, it may potentially reduce the impact of negative thoughts, enhance cognitive flexibility and resilience and improve the ability to adapt to stressors [36]. These mechanisms likely underpin the efficacy of this integrated approach.

Compared to the wait-list control group, the M-bCCDT had a positive impact on improving memory function and psychomotor speed, with gains sustained at 20 weeks, consistent with prior findings on MBIs [37]. This cognitive benefit might be explained by mindfulness’s ability to enhance attentional control and executive function while reducing cognitive interference [38, 39], allowing for more efficient allocation of brain resources, optimizing information processing and facilitating action planning in complex tasks [40], which could potentially accelerate psychomotor speed. Furthermore, mindfulness-based practices incorporating cognitive defusion have been verified to increase grey matter density in the prefrontal cortex and hippocampus [41], regions critical for executive functions like attention control and cognitive flexibility [42]. These structural changes may provide a potential physiological mechanism underlying improvements in psychomotor speed. Improvements in psychomotor speed may facilitate more efficient cognitive processing, enhancing memory encoding, storage and retrieval. These findings may explain the efficacy of the M-bCCDT program in enhancing specific cognitive domains. Additionally, reduced perceived stress, anxiety and depression associated with higher mindfulness levels may further enhance psychomotor speed and memory function [43]. However, repeated testing may have induced practice effects, possibly overestimating the results.

Notably, although immediate post-intervention changes in global cognitive function were non-significant, significant improvement were observed at 20 weeks, indicating the M-bCCDT program may requires time to enhance global cognitive function by improving attention regulation and reducing stress. Previous research demonstrated that MBIs have an ameliorative effect on cognitive function through specific mechanisms, such as increased functional connectivity in the default mode network and reduced hippocampal atrophy [44]. Additionally, stress can lead to systemic inflammation, which negatively affects cognitive function. Reducing stress may mitigate the inflammatory response, potentially protecting cognitive health over time [45]. Additionally, a meta-analysis found that more frequent mind-body exercises sessions yielded greater cognitive gains than less frequent ones [46]. This study further validates this conclusion, indicating that increasing the frequency of interventions could enhance cognitive gains in the future.

Additionally, the M-bCCDT program may enhance participants’ mindfulness awareness, with effect persisting at the 20 weeks, extending previous research that shows MBIs improve mindfulness awareness [47]. This may be related to the trait of mindfulness produced by the training. Trait mindfulness could promote inner peace, emotional regulation and non-judgmental awareness of negative thoughts [48], thereby reducing emotional reactions and avoidance [49]. These benefits lead to fewer psychological symptoms. Worth mentioning is that the observed benefits of the M-bCCDT program may be influenced by the supportive group environment, which could reduce stress and enhance cognitive function.

Reliable and clinically significance effects of M-bCCDT program

The RCI is used to confirm whether an individual’s change is statistically significant, and CSC is used to further confirm is the change is of practical or real-world importance, not just statistical significance [22]. At a methodological level, this study demonstrates that tests for clinical significance can be complementary to inferential statistics and calculated with relative ease to evaluate intervention effectiveness. GEE analysis revealed positive group-average effects of the M-bCCDT program. However, RCSC analysis showed limited individual-level improvements, despite a gradual upward trend over time. For instance, left hand dexterity measured by PPTL, only eight participants showed reliable or clinically significant changes at T1, increasing to 24 at T2 and 13 showed clinically significant improvement. Inter-individual differences may accumulate and magnify over time, leading to greater disparities between individual change analysis and group average changes. This emphasizes the need for greater precise and refined analytical techniques to assess intervention effects. Additionally, this study also established RCI and CSC values for health outcomes, which healthcare providers can use these metrics to dynamically adjust interventions based on observed outcomes [34].

Study strengths and implications

Two strengths of this study are as follows. First, use of a randomized control design allowed us to make causal effects about differences in perceived stress, global cognitive function, psychomotor speed, memory function and mindfulness awareness between the groups. The second strength is the application of RCSC analysis, which ensures statistically reliable and practically meaningful changes, highlighting individual-level improvements in health outcomes from the M-bCCDT program. By distinguishing meaningful progress from random fluctuations, this analysis provides a robust evaluation of the intervention’s real efficacy on participants.

This study could provide strong clinical implications for community healthcare practice, though its practical implementation requires further validation. The M-bCCDT program retains the core elements of mindfulness and cognitive defusion while incorporating specific stress contexts and coping techniques, offering a targeted and practical approach to managing health in older adults with MCI. Effective implementation strategies, including practitioner training on mindfulness and MCI-specific stress coping, integration of the M-bCCDT program into community health services and tailoring the intervention to diverse populations by addressing socioeconomic and cultural contexts, will enhance its relevance and provide practical guidance for real-world application. Additionally, RCSC analysis with RCI and CSC value guides healthcare practitioners to tailor interventions based on each individual’s response. The method helps distinguish genuine improvements in health outcomes, validating the efficacy of the M-bCCDT program. Furthermore, RCSC analysis allows for further exploration of clinically significant population traits, assisting in directing resources towards individuals likely to respond effectively to maximize health outcomes for older adults with MCI.

Limitations and future research directions

This study has several limitations that must be considered. (1) The study’s focus on one location and its relatively small sample size limit the generalizability of the findings. Future studies involving larger and more diverse populations across multiple locations are necessary to enable subgroup analysis (socioeconomic and cultural factors) and enhance clinical relevance. (2) The use of self-reported measures in in-person interviews in this study might have introduced response biases. Future studies should incorporate neurobiochemical measures to provide a more comprehensive evaluation of the intervention’s effects. (3) In our study, participants in both groups who reached clinical significance may have undergone a clinically meaningful change, either positive or negative. However, factors such as demographic and socioeconomic status, and comorbidities could affect the estimation of clinical significance [50], further investigation is needed to identify clinically significant population characteristics and other potential conditions that contribute to improved or worsened health outcomes, so that tailored cognitive defusion programs for specific MCI populations to enhance clinical relevance. (4) Long-term outcomes are important for clinical significance. The 20 weeks follow-up may not fully capture long-term effects. Future studies should implement multi-stage follow-up protocols and extend the duration of follow-up to determine the durability and delayed effects of the M-bCCDT program. (5) LOCF method for imputing missing data, assuming no change after the last observation, may bias estimates if data are non-random or intervention-related, requiring cautious interpretation. Future studies should use sophisticated methods, like multiple imputation or mixed-effects modelling, and improve retention through online training, remote assessments and enhanced follow-up protocols to ensure robust findings. Additionally, although assessor blinding and intention-to-treat analysis minimized bias, the lack of participant blinding remains a limitation, as sham interventions were unfeasible due to differences in M-bCCDT and health education content. Future studies should use active controls (established stress reduction techniques) to better evaluate the relative efficacy of M-bCCDT.

Conclusions

This study demonstrated that the M-bCCDT program improved perceived stress, memory, mindfulness and psychomotor speed in older adults with MCI, with cognitive function improving over time. The program provides healthcare professionals with a structured intervention to support the mental and cognitive health of older adults with MCI in community settings. Future studies will focus on assessing cost-effectiveness and scalability on broader implications (like quality of life and caregiver burden) to comprehensively evaluate the feasibility and practical implementation of M-bCCDT program.

Declaration of Conflicts of Interest:

None.

Declaration of Sources of Funding:

This work was supported by the National Natural Science Foundation of China (grant number 72174061, 71704053), application author: Lina Wang; China Scholarship Council foundation (grant number 202308330251), application author: Lina Wang; and the Zhejiang Health and Science and Technology Project (grant number 2022KY370, 2024KY1662), application authors: Chen Zhang, Mengna Zhu.

Data availability:

The datasets generated or analysed during the current study are not publicly available but will be available from the corresponding author upon reasonable request.

References

Author notes