Cognitive impairment has been widely reported among cancer survivors, significantly impacting their quality of life. Mindfulness interventions are increasingly used to alleviate cognitive impairment in patients with cancer. This study aimed to assess the effects of mindfulness interventions on cognitive impairment in patients with cancer post-intervention and at follow-up.
MethodsUp until February 2024, five English databases (PubMed, Web of Science, Embase, Cochrane, and PsycINFO) and three Chinese databases (CNKI, Wan Fang, and CBM) were searched to identify relevant studies. To determine the effect size, we used random effects model to compute the standardized mean difference and 95 % confidence intervals.
ResultsWe included 23 randomized controlled trials and seven non-randomized controlled trials. Mindfulness interventions significantly improved patients’ subjective cognitive function post-intervention (SMDbetween-group=0.81, 95 %CI: 0.58 to 1.03; SMDwithin-group=1.12, 95 %CI: 0.71 to 1.52) and at follow-up (SMDbetween-group=0.39, 95 %CI: 0.09 to 0.68; SMDwithin-group=0.59, 95 %CI: 0.35 to 0.82). Subgroup analysis indicated significantly larger effect of the interventions in developing countries than those in developed countries (pbetween-group=0.014; pwithin-group=0.008), and of the interventions without additional home practice than those with home practice in within-group comparisons (pbetween-group=0.217; pwithin-group=0.018). There were no significant differences in the effects between interventions lasting ≥eight weeks and < eight weeks (pbetween-group=0.093; pbetween-group=0.303). However, no significant effects were observed on objective cognitive function.
ConclusionsMindfulness-based interventions can effectively improve the subjective cognitive function in cancer patients both post-intervention and at follow-up. Future intervention research should take into account regions, home practice, and intervention duration.
The incidence of cancer has risen dramatically with increasing life expectancy and early detection. The number of new cancer cases globally is projected to exceed 35 million by 2050, a 77 % increase from the estimated 20 million cases in 2022 (Bray et al., 2024). At the same time, cancer survival rates have improved, with the 5-year relative survival rate increasing from 49 % to 69 % over a 40–years period (Siegel et al., 2024). The increased prevalence and survival rates resulted in a growing number of cancer survivors. Among these survivors, cognitive impairment has become one of the most feared complications, drawing considerable attention over the past three decades (Ahles & Root, 2018). Cognitive impairment in cancer patients primarily results from various mechanisms across anticancer therapies. Chemotherapy disrupts the blood-brain barrier, induces oxidative stress, and triggers neuroinflammation, causing neuronal damage. Immunotherapies, endocrine therapies, and targeted therapies exacerbate cognitive deficits through cytokine release, reduced neuroprotective hormones, and altered cerebral blood flow (Fleming et al., 2023). The prevalence of self-reported cognitive impairment ranges from 21 % and 83 %, while objective neuropsychological measures show a prevalence of 17 % to 34 % (Dijkshoorn et al., 2021; Whittaker et al., 2022). Cognitive impairment can persist for months to years, significantly affecting patients’ ability to return to work, perform daily activities, live independently, and overall quality of life (Lange et al., 2024; Von Ah et al., 2013; Yang & Von Ah, 2024). It also increases healthcare burden by driving higher resource utilization and costs (Park & Hwang, 2012). Therefore, focusing on and seeking appropriate interventions to improve cognitive function is necessary.
To provide patient-centered care, complementary therapies have been integrated into comprehensive treatment plans to mitigate cognitive impairment among patients with cancer (Myers, 2015). These therapies are generally well-accepted among the patients, with a prior systematic review reporting that more than half of them have used complementary and alternative treatments (Keene et al., 2019). Many of them, including cognitive training, cognitive behavioral therapy, mindfulness-based interventions, and physical activity (Zeng et al., 2020), have been proposed to improve cognitive function. Mindfulness-based interventions, in particular, teach participants to focus on present-moment experiences with compassion and without judgment (Creswell, 2017). Compared to cognitive training and cognitive behavioral therapy, mindfulness interventions require less professional and field training, making them a potentially more accessible and cost-effective option (Singh & Gorey, 2018). Research shows that mindfulness interventions improve cognitive function by reducing chronic stress, regulating glucocorticoid levels, and decreasing neuroinflammation, thereby preserving hippocampal function. Additionally, they can improve metabolic disturbances, including insulin resistance and oxidative stress, thereby promoting neuroprotection and slowing cognitive decline (Larouche et al., 2015; Malinowski & Shalamanova, 2017).
However, the results of the effect of mindfulness intervention on cognitive function in cancer patients were inconsistent. Some studies reported improvements in both subjective and objective cognitive function (Johns et al., 2016; Milbury et al., 2013), while others found benefits only in subjective cognition, with no significant effects on objective performance (Duval et al., 2022; Van der Gucht et al., 2020). Additionally, one study reported no significant effects on either (Reich et al., 2017). These conflicting findings highlight the need for a comprehensive review to clarify the cognitive benefits of mindfulness in cancer patients. Several systematic reviews have attempted to evaluate it. The first comprehensive systematic review on mindfulness interventions and cognitive function in cancer patients was published in 2018 (Cifu et al., 2018), focusing specifically on breast cancer survivors. This review analyzed six studies using a narrative synthesis and reported mixed findings. A subsequent review analyzed the effects of mindfulness interventions on cognitive function in patients with various types of cancers (Flynn et al., 2023). This meta-analysis of 11 studies on self-reported cognition showed a significant positive effect, while a narrative synthesis of objective cognition yielded mixed results, indicating the need for further research. Additionally, these reviews have not assessed the effects of mindfulness interventions on cognition at both after-intervention and at follow-up. While some studies have analyzed the impact of mindfulness interventions on stress and anxiety across different regions, intervention durations, and the inclusion of additional home practice, their effects on cognitive function remain unclear (Li et al., 2023; Ling et al., 2021; Lloyd et al., 2018). Therefore, we conducted a subgroup analysis based on these factors to better elucidate the effects of mindfulness interventions on cognitive function across different subgroups.
The quality of life in patients with cancer is a crucial outcome in both research and clinical practice (Iconomou et al., 2004). Cognitive impairment has been shown to correlate with general quality of life (Song et al., 2023; Williams et al., 2017), and many studies investigating the effects of mindfulness interventions on cognitive function usually evaluate their influence on quality of life (Bo et al., 2023; Jang et al., 2016; Liu et al., 2019; Vadiraja et al., 2009). In this study, our primary objective was to summarize the effects of mindfulness interventions on cognitive function in patients with cancer both after-intervention and at follow-up. Additionally, we aimed to explore whether these effects varied across different regions, intervention durations, and the presence of additional home practice. Our secondary objective was to examine the impact of mindfulness interventions on quality of life.
MethodsThis meta-analysis was registered with the PROSPERO, number CRD42022362034, and adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.
Data sources and search strategiesFive English databases (PubMed, Web of Science, Embase, Cochrane, and PsycINFO) were searched, with the original search completed by August 2022, and the updated search completed by February 2024. During the updated search, three Chinese databases (China National Knowledge Infrastructure, Wan Fang, and China Biology Medicine Database) were included, covering the relevant literature from inception to February 2024. A personalized search strategy was developed for each database (Supplementary Table 1).
Inclusion and exclusion criteriaArticles that strictly met to the eligibility criteria were included: 1) patients with non-central nervous system cancer who are 18 years or older; 2) mindfulness as the core component of the intervention and must be delivered directly to patients; 3) randomized control trials(RCTs), quasi-experimental design trials, and single-group pre-post-test studies, with comparisons involving wait-list control, treatment-as-usual, or other alternative interventions; 4) the primary outcome was cognitive function, assessed using both subjective and objective measures, including subdomains such as memory, attention, and executive function. Chinese and English-language articles were included. Two researchers independently screened the literature, and in case of disagreements, a third researcher participated to reach a consensus.
Data extraction and risk of bias assessmentWe extracted information on publication details (e.g., author, year, country), study design (e.g., RCTs or non-RCTs, sample size, age and sex distribution, cancer type), intervention details (e.g., control group conditions, intervention form, duration, and inclusion of additional home practice), and outcomes related to cognitive function and quality of life. For RCTs, we assessed risk of bias using the Revised Cochrane Risk of Bias, version 2 (RoB2) tool (Sterne et al., 2019). For non-RCTs, we used the Risk Of Bias In Non-randomized Studies–of Interventions (ROBINS-I) tool (Sterne et al., 2016). Two reviewers independently assessed and rated the studies, with any disagreements resolved through consultation with a third reviewer.
Data analysisDue to the use of various scales across studies to assess cognitive function and quality of life, effect sizes were reported as standardized mean differences (SMDs) with 95 % confidence intervals (CIs). Heterogeneity was assessed using the I² statistic, with values <25 %, 25 %−75 %, and above 75 % indicating low, moderate, and high heterogeneity, respectively (Higgins et al., 2003). A random-effects model was used to estimate the overall effect size.
To comprehensively summarize the effects of mindfulness interventions on cognitive function, we included both RCTs and non-RCTs. A between-group meta-analysis was conducted for RCTs to estimate the effects of mindfulness interventions compared to control conditions. SMDs were calculated as the difference in mean values between the treatment and control groups, divided by the pooled standard deviation (Cuijpers et al., 2017). This analysis provides high-quality causal evidence (Jones & Podolsky, 2015). Additionally, a within-group meta-analysis was conducted to assess changes within all included studies (both RCTs and non-RCTs). For this analysis, SMDs were calculated by dividing the difference between post-intervention and pre-intervention mean values by the pre-intervention standard deviation (Becker, 1988). While this approach captures potential subjective benefits, it is susceptible to confounding factors and does not allow for causal inference (Fendel et al., 2024). Together, these two approaches offer a balanced evaluation, capturing both robust causal effects and broader intervention trends.
Subgroup analyses were conducted for the primary outcome to explore potential sources of heterogeneity: (a) regions of the included studies (developed vs. developing countries), (b) intervention duration (≥8 weeks vs. <8 weeks), and (c) intervention format (with or without additional home practice). To assess the stability of the results, we conducted sensitivity analyses: (a) the leave-one-out method, where one study was sequentially removed at a time to test the robustness of the results; (b) exclusion of studies whose 95 % confidence intervals did not overlap with the pooled effect size’s 95 % confidence interval; (c) exclusion of studies with a high risk of bias to ensure the robustness of the effect estimates. In addition, we conducted a subgroup analysis based on study design, considering the heterogeneity in within-group analyses. We also used funnel plots and Egger’s test to detect potential publication bias (Sterne et al., 2001). All statistical analyses were performed using R version 4.3.3 with the ‘meta’ package.
ResultsAs shown in Fig. 1, our electronic search strategy identified 1106 English articles published until August 2022, of which eight met the inclusion and exclusion criteria (Dobos et al., 2015; Duval et al., 2022; Eyles et al., 2015; Janelsins et al., 2016; Johns et al., 2016; Rahmani et al., 2014; Reich et al., 2017; Vadiraja et al., 2009). We subsequently conducted an updated search up to February 2024 and included Chinese articles from the database establishment to February 2024. Additionally, based on previously published literature reviews, we have included a total of 30 articles (An et al., 2020; Bo et al., 2023; Carlson & Garland, 2005; Chen et al., 2017; Cheng et al., 2022; Ding et al., 2021; Ding et al., 2018;Guo et al., 2023; Hao et al., 2019; Henneghan et al., 2020; Jang et al., 2016; Ji & Wang, 2018; Lengacher et al., 2018; Liu et al., 2019; Milbury et al., 2013; Peng et al., 2022; Shan, 2014; Van der Gucht et al., 2020; Wang et al., 2022; Xu et al., 2021;Zhu et al., 2023; Zou & Peng, 2020).
Characteristics of the included studies
Table 1 summarizes the characteristics of the included studies. Among the 30 studies, 11 were in Chinese and 19 were in English. Three were quasi-experimental, four were single-arm, and 23 were randomized controlled trials. Regarding control groups, 22 used non-active controls, and four used active controls (fatigue education, brief supportive therapy, music listening, and metacognition treatment). Sixteen studies were conducted in China, six in the United States, two in Canada, and one each in South Korea, Iran, Germany, Belgium, India and the United Kingdom. More than half of the studies focused on breast cancer (19 trials, 63.3 %), with mindfulness-based stress reduction (MBSR) being the most common intervention (13 studies, 43.3 %). Follow-up studies primarily focused on the 1–6 months post-intervention. Four interventions lasted more than eight weeks, 10 lasted less than eight weeks, and 15 lasted exactly eight weeks. The most commonly used tool for assessing patients’ subjective cognitive function was the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30). There were two studies excluded from the meta-analysis as one only reported post-intervention data (Ji & Wang, 2018) and the other one only reported pre- and post-intervention differences (Eyles et al., 2015). Therefore, we performed a narrative synthesis of these studies.
Characteristics of studies included in the systematic review and meta-analysis.
| Author, year | country | study design | Name of program | Cancer category | Age (years) | Sex | Education | duration, wk | home practice | CG(control group) |
|---|---|---|---|---|---|---|---|---|---|---|
| Janelsins et al., 2016 | USA | RCT | YOCAS(yoga) | breast cancer | IG: 55.24±10.94, CG: 53.97±8.72 | Female: IG:95 %, CG: 96 % | 4 | no | standard care | |
| Van der Gucht et al., 2020 | Belgium | RCT | Mindfulness based stress reduction | breast cancer | IG: 43.89±6.03, CG: 47.4 ± 5.45 | IG&CG:Female (100 %) | total: secondary school:45 %;a higher education degree: 45 %; never finished secondary school:10 % | 8 | yes | usual care |
| Duval et al., 2022 | Canada | RCT | Mindfulness based stress reduction | breast cancer | IG: 49.20±10.02, CG: 53.47±8.55 | IG&CG:Female (100 %) | bachelor’s degress: total: 24 (40 %); IG: 9 (30 %); CG: 15 (50 %) | 8 | yes | usual care |
| Milbury et al., 2013 | USA | RCT | Tibetan Sound Meditation (TSM) | breast cancer | IG: 53.0 ± 6.6, CG: 54.1 ± 8.6 | IG&CG:Female (100 %) | Some college or higher: IG: 22 (95.6 %); CG: 18 (74.9 %) | 6 | yes | usual care |
| Johns et al., 2016 | USA | RCT | Mindfulness based stress reduction | breast cancer; colorectal cancer | IG: 56.9 ± 9.9, CG: 56.4 ± 12.7 | Female: IG:94.3 %, CG: 86.1 % | College degress: IG: 15(42.9 %), CG:16(44.4 %) | 8 | yes | fatigue education and support (ES) intervention |
| Vadiraja et al., 2009 | India | RCT | Yoga | breast cancer | IG: 46.7 ± 9.3, CG: 48.5 ± 10.2 | IG&CG:Female (100 %) | ≥high school education (100 %) | 6 | yes | Brief supportive therapy |
| Reich et al., 2017 | USA | RCT | Mindfulness based stress reduction | breast cancer | IG: 56.5 ± 10.2,CG: 57.6 ± 9.2 | IG&CG:Female (100 %) | College graduate and above: IG: 75(44.9 %); CG: 65(42.2 %) | 6 | yes | usual care |
| An et al., 2020 | China | RCT | Mindfulness based stress reduction | cervical cancer | IG:48.52±1.28;CG:48.49±1.22 | IG&CG:Female (100 %) | Primary school: IG:12(21.82 %), CG:13(24.07 %); Junior high school or technical secondary school: IG:27(49.09 %), CG:25(46.30 %); High school: IG:8(14.55 %), CG:7(21.96 %); Junior college and above: IG:8(14.55 %), CG:9(16.67 %); | 6 | yes | usual care |
| Henneghan et al., 2020 | USA | RCT | Mindfulness-Based Art Therapy | breast cancer | IG:50±9.52;CG:48.93±10.69 | IG&CG:Female (100 %) | Bachelors or higher: IG:9(56.25 %); CG:13(86.67 %); | 8 | no | Music Listening |
| Jang et al., 2016 | Korea | RCT | Mindfulness-Based Art Therapy | breast cancer | IG:51.75 ± 5.32; CG:51.42 ± 6.33 | IG&CG:Female (100 %) | Low (<10 years)/Middle(10–13years)/high: IG:2(16.7 %),7(58.3 %),3(25 %); CG:3(25 %),7(58.3),2(16.7 %) | 12 | no | usual care |
| Liu et al., 2019 | China | RCT | Mindfulness based stress reduction | thyroid cancer | IG:43.32±10.99/CG:42.38±12.60 | Female: IG:69 %, CG: 72 % | Middle school or lower/High school or higher: IG:16(32.7 %),33(67.3 %); CG:19(35.8 %)/34(64.2 %) | 8 | no | usual care |
| Bo et al., 2023 | China | RCT | mindfulness meditation | Leukaemia/Lymphoma/Myeloma/Others | IG:47.16±8.40/CG:36.49±8.34 | Female: IG:40 %, CG: 36.67 % | Junior and below/Senior and secondary school /Senior and secondary school: IG:13(43.33 %)/11(36.67 %)/6(20 %); CG:12(40 %)/12(40 %)/6(20 %) | 4 | no | conventional care |
| Wang et al., 2022 | China | RCT | Internet-Delivered Mindfulness-Based Cancer Recovery | breast cancer | IG:45.37 ± 7.59; CG:48.17 ± 8.05 | IG&CG:Female (100 %) | College or university/High school or vocational/Secondary /≤Primary:IG:18(35 %)/18(35 %)/9(18 %)/6(12 %);CG:9(17 %)/21(40 %)/15(29 %)/7(14 %) | 4 | yes | usual care |
| Peng et al., 2022 | China | RCT | online mindfulness-based intervention | breast cancer | 30≤age<40 / 40≤age<50/age≥50:IG:6/15/7;CG:6/15/8 | IG&CG:Female (100 %) | junior high school /high school /university and above:IG:6(21.4 %)/15(53.6 %)/7(25 %);CG:5(17.2 %)/13(44.8 %)/11(37.9 %) | 6 | yes | usual care |
| Chen et al., 2017 | China | RCT | Group Mindfulness Cognitive Therapy | gastric cancer | 8 | yes | conventional care | |||
| Cheng et al., 2022 | China | RCT | Mindfulness Rehabilitation Training Combined with Ear Acupressure Therapy | colon cancer | IG:60.53±7.2;CG:60.16±7.59 | Female: IG:45 %, CG: 49 % | 8 | no | conventional care | |
| Shan, 2014 | China | RCT | mindfulness intervention therapy | breast cancer | IG:44.11±8.69;CG:46.92±8.20 | IG&CG:Female (100 %) | Junior high school and below/Senior high school/college or higher:IG:7(18.4 %)/12(31.6 %)/19(50 %);CG:7(16.6 %)/13(31.0 %)/22(52.4 %) | 12 | yes | conventional care |
| Ding et al., 2021 | China | RCT | Integrating Team Management with Mindfulness-Based Stress Reduction | glioma patients | IG: 45.25±12.19;CG:42.81±11.65 | Female: IG:53 %, CG: 49 % | 7 | yes | conventional care | |
| Ding et al., 2018 | China | RCT | Mindfulness-Based Behavioral Training | lung cancer | IG: 56.32±6.12;CG:55.45±6.24 | Female: IG:47 %, CG: 42 % | middle school or below/high school/college or higher:IG:21(32.8 %)/28(43.8 %)/15(23.4 %);CG:19(29.7 %)/26(40.6 %)/19(29.7 %) | 8 | no | conventional care |
| Guo et al., 2023 | China | RCT | Integrated Healthcare and Nursing Mindfulness Training | lung cancer | IG:58. 17±6.53;CG:57. 24±6. 47 | Female: IG:44 %, CG: 48 % | 8 | no | conventional care | |
| Hao et al., 2019 | China | RCT | Group Mindfulness Cognition | breast cancer | IG: 43.8 6 ± 9.17;CG:44.3 6 ± 9.0 5 | IG&CG:Female (100 %) | 8 | no | conventional care | |
| Zhu et al., 2023 | China | RCT | Psychological Intervention Guided by Mindfulness Acceptance and Commitment Therapy | liver cancer | IG:58.49±8.32; CG:59.75±8.16 | Female: IG:40 %, CG: 47 % | 9 | no | conventional care | |
| Zou and Peng, 2020 | China | RCT | Mindfulness based stress reduction | breast cancer | IG:46.30±7.57;CG:46.16±7.12 | IG&CG:Female (100 %) | 8 | yes | conventional care | |
| Rahmani et al., 2014 | Iran | quasi-experimental design | Mindfulness based stress reduction | breast cancer | IG: 43.25±3.08, CG1: 44.92±1.83,CG2: 44.08±3.28 | IG&CG:Female (100 %) | junior high and above (100 %),20 % of them were graduated from university | 8 | yes | CG1: metacognition treatment; CG2:conventional nursing care |
| Ji and Wang, 2018 | China | quasi-experimental design | Mindfulness-Based Behavioral Training | Lung cancer | IG:60.12±3.57; CG:60.25±3.52 | Female: IG:36 %, CG:40 % | no | conventional care | ||
| Xu et al., 2021 | China | quasi-experimental design | Motivational Care Combined with Mindfulness-Based Stress Reduction | colorectal cancer | IG:58.12 ± 6.83; CG:58.63 ± 7.01 | Female: IG:40 %, CG:44 % | 8 | no | conventional care | |
| Dobos et al., 2015 | Germany | single-arm cohort study | Kliniken Essen-Mitte mindfulness-based day care clinic group program | breast cancer | 53.91±10.7 | Female:91 % | with A-level and higher (59/117, 45.8 %) | 11 | yes | |
| Eyles et al., 2015 | UK | single-arm cohort study | Mindfulness based stress reduction | breast cancer | 37–65 | IG&CG:Female (100 %) | 8 | yes | ||
| Lengacher et al., 2018 | USA | single-arm cohort study | mobile Mindfulness based stress reduction | breast cancer | 57±9 | Female (100 %) | High school graduate:2(13 %);Some college:7(47 %);College graduate:4(27 %);Graduate or professional school:2(13 %) | 6 | yes | |
| Carlson and Garland, 2005 | Canada | single-arm cohort study | Mindfulness based stress reduction | breast cancer, followed by prostate, ovarian, and Non-Hodgkins lymphoma | 32–78(mean:54) | Female:78 % | 8 | yes |
RCTs and non-RCTs were evaluated separately due to differences in study design. For non-RCT studies, 14.3 % (n = 1) had low risk, and 85.7 % (n = 6) had serious risk, with confounding being the most common issue. Among the RCTs, 8.7 % (n = 2) were rated as poor quality, 56.5 % (n = 13) as fair quality, and 34.8 % (n = 8) as excellent quality. Common issues included intervention deviations and missing data (Supplementary Figure 1).
Primary outcomeEffects of mindfulness intervention on subjective cognitive functionThe results of both between-group and within-group analyses were shown in Fig. 2 and 3 (between-group: post-intervention: SMD=0.81, 95 %CI:0.58 to 1.03; follow-up: SMD=0.39, 95 %CI: 0.09 to 0.68; within-group: post-intervention: SMD=1.12, 95 %CI: 0.71 to 1.52; follow-up: SMD=0.59, 95 %CI: 0.35 to 0.82). Mindfulness interventions significantly enhanced subjective cognitive performance both after-intervention and at follow-up. Sensitivity analyses, including the leave-one-out method, removal of outliers, and exclusion of low-quality studies, confirmed that the effect size remained significant. The funnel plot suggested potential publication bias, but Egger’s test did not identify statistically significant publication bias in the between-group comparison (post-intervention, p = 0.951; follow-up, p = 0.582). Additionally, the two studies excluded from the meta-analysis used the EORTC QLQ-C30 scale to assess cognitive function. One study found that Mindfulness-Based Stress Reduction significantly improved cognitive function in breast cancer, with a mean score increase of 18.42 (p = 0.015) from baseline to follow-up, indicating a decrease in cognitive difficulties. Another study reported that short-term mindfulness behavioral training improved cognitive function compared to the control group (intervention group: 84.41±6.15; control group: 70.52±5.62, p < 0.001). For subjective cognitive domains, mindfulness interventions did not significantly improve subjective memory difficulties (Supplementary Figure 6 and Figure 7).
We conducted a subgroup analysis based on study design, region, intervention duration, and the inclusion of additional home practice, as shown in Table 2. For between-group studies, interventions conducted in developing countries showed a significantly lager effect size (SMD=0.93, 95 %CI:0.68 to 1.18) compared to those in developed countries (SMD=0.41, 95 %CI:0.09 to 0.74) (p = 0.014 for subgroup differences). Similarly, within-group analysis revealed greater improvements in developing countries (SMD = 1.40, 95 % CI: 0.81 to 2.00) compared to developed countries (SMD = 0.55, 95 % CI: 0.34 to 0.77) (p = 0.008 for subgroup differences). While longer interventions (≥8 weeks) generally exhibited larger effect sizes than shorter ones (<8 weeks) in both analyses, the differences were not statistically significant (p > 0.05). The inclusion of additional homework had no significant impact in between-group studies (p = 0.217). However, within-group analyses indicated that studies without additional homework reported greater improvements (SMD=1.76, 95 %CI:0.89 to 2.64) than those with it (SMD=0.69, 95 %CI:0.48 to 0.89) (p = 0.018 for subgroup differences). Within-group analyses also showed that mindfulness interventions consistently improve cognitive function across different study design.
Subgroup analysis of the effects on subjective cognitive function post intervention.
Only three studies evaluated subdomains of objective cognitive function (Supplementary Figures 8 to 11), which may have impacted the reliability of the results. Both within-group and between-group analyses indicated that mindfulness interventions did not significantly improve objective executive function either after-intervention or at follow-up (between-group: post-intervention: SMD=0.30, 95 %CI: −0.23 to 0.84; follow-up: SMD=0.35, 95 %CI: −0.08 to 0.78; within-group: post-intervention: SMD=0.08, 95 %CI: −0.74 to 0.90; follow-up: SMD=0.15, 95 %CI: −0.63 to 0.93). Additionally, follow-up assessments revealed no significant improvement in objective attention (between-group: SMD=0.42, 95 %CI: −0.04 to 0.89; within-group: SMD=0.24, 95 %CI: −0.23 to 0.71).
Secondary outcomeWe further analyzed the impact of mindfulness interventions on the quality of life among cancer patients (Supplementary Figures 12 and Figure 13). The results from RCTs indicated that mindfulness interventions significantly improved quality of life compared to the control group post-intervention (SMD=0.94, 95 %CI:0.35 to 1.53). However, the funnel plot revealed potential publication bias (Egger’s test, p = 0.043). The within-group analysis showed no significant improvement in quality of life at post-intervention (SMD=1.02, 95 %CI: −0.08 to 2.12). Furthermore, both between-group and within-group analyses consistently demonstrated that mindfulness interventions did not significantly improve quality of life during follow-up (between-group: SMD=0.37, 95 %CI: −0.28 to 1.03; within-group: SMD=0.06, 95 %CI: −0.61 to 0.73).
DiscussionOur study provides a comprehensive evaluation of the effects of mindfulness interventions on cognitive function among cancer patients at both post-intervention and follow-up stages. The results showed that mindfulness interventions improve subjective cognitive function in cancer patients at both stages, with a smaller effect size during the follow-up than post-intervention. However, no significant effects were observed in objective cognitive function. To our knowledge, this is the first study to assess the intervention effects across multiple subgroups. We found that interventions conducted in developing countries demonstrated significantly greater effects compared to those in developed countries; and interventions without additional home practice were significantly effective than those with additional practice. Although interventions lasting eight weeks or longer had a larger effect size than those less than eight weeks, the difference was not statistically significant.
Cifu et al.’s systematic review provided preliminary evidence of the effectiveness of mindfulness intervention on cognitive function in cancer patients through narrative analysis (Cifu et al., 2018). The results of our study confirmed their findings by adding quantitative evidence, and also aligned with Flynn et al.’s conclusions regarding the effects on subjective cognitive function at post-intervention (Flynn et al., 2023). Furthermore, we also found the beneficial effects of mindfulness interventions at follow-up. The results can be explained by cognitive load theory, which suggests that reducing unnecessary cognitive load enhances cognitive performance. Mindfulness promotes present-moment awareness and reshapes situated conceptualizations, thereby alleviating unnecessary cognitive load (Papies, 2017; Sweller, 2011; Takhdat et al., 2024). This reduction in cognitive burden may explain the significant improvements observed in self-reported cognitive function. In contrast, objective cognitive function is primarily influenced by neurobiological mechanisms. Mindfulness interventions have demonstrated neurobiological effects, including changes in brain imaging and functional connectivity (Melis et al., 2022). However, traditional neuropsychological assessments were originally designed to evaluate severe cognitive impairments (e.g., from stroke or traumatic brain injury). Cognitive impairment in cancer patients is typically more subtle and may not be adequately captured by these assessments. Moreover, these assessments are conducted in highly controlled settings, which may not reflect the real-world cognitive challenges faced by cancer patients. As a result, even if mindfulness interventions confer cognitive benefits, these improvements may not be captured by traditional neuropsychological test scores (Horowitz et al., 2018; Hutchinson et al., 2012). Therefore, future studies should integrate neurobiological measures (e.g., fMRI, biomarkers) to enhance cognitive impairment assessment in cancer patients. Additionally, adopting a multimodal assessment approach—incorporating objective tests, subjective self-reports, and real-world cognitive performance measures—may provide a more comprehensive evaluation of the cognitive benefits of mindfulness interventions.
Mindfulness interventions have shown positive effects in developed and developing countries. The widespread applicability can be attributed to their inherent flexibility, allowing them to be tailored to diverse culture contexts, healthcare settings and patient needs. Despite these universal benefits, subgroup analyses revealed that mindfulness interventions were more effective in developing countries than in developed countries. This result may be influenced by cultural attitudes and healthcare disparities. Mindfulness has deep historical and spiritual roots in developing countries like China and India in our study where it is often integrated into traditional spiritual and healing practices, making it more culturally embedded and widely accepted. In contrast, in developed countries, mindfulness is typically viewed in a more secular and individualistic context, which may influence patient engagement and intervention outcomes (Schmidt, 2011). Furthermore, disparities in postoperative rehabilitation management also play a crucial role. An international survey has shown that developed countries possess superior medical resources and multiple postoperative rehabilitation interventions for cancer patients compared to developing countries (Signorelli et al., 2024). As a result, mindfulness interventions, as a cost-effective and accessible approach, have played a significant role in improving cognitive function in developing countries (Ling et al., 2021; Pillay & Eagle, 2021). This finding highlights the importance of considering socioeconomic and culture factors when implementing mindfulness interventions across different regions.
The traditional duration of mindfulness interventions typically takes approximately eight weeks (Carmody & Baer, 2009). In our study, although interventions lasting eight weeks or longer tended to show greater effect than those lasting less than eight weeks, the difference was not statistically significant. This suggests that the potential benefits of longer interventions should be interpreted with caution. Longer interventions allow patients to develop mindfulness skills, integrate them into daily life, and enhance self-awareness, leading to more lasting benefits over time. Our findings differ from prior studies focused on improving depression and anxiety in lung cancer patients (Li et al., 2023). This discrepancy may be attributed to differences in the outcomes being measured. Emotional outcomes like depression and anxiety may respond more quickly to mindfulness interventions, whereas improvements in cognitive function may require a longer duration of intervention.
Interestingly, the interventions were effective regardless of whether additional home practice was included, with interventions that did not include additional home practice showing better effects. The effectiveness of additional home practice likely depends on patients’ mindfulness proficiency and their ability to apply these skills independently. If participants have not fully mastered these skills, additional practice may increase their psychological burden, reduce adherence, and ultimately hinder cognitive improvements (Lloyd et al., 2018). Furthermore, due to the lack of fidelity data on additional home practice, it is possible that participants assigned to additional sessions may have experienced intervention fatigue or burnout, leading to lower completion rates and diminished intervention fidelity. These findings suggest that reducing additional home practice may enhance cognitive outcomes. However, this does not imply that additional home practice is unimportant for consolidating long-term effects. Future research should systematically assess patient adherence and fidelity to clarify the relationship between additional home practice and cognitive improvements.
Our findings on quality of life in between-group post-intervention were consisted with previous studies (Huang et al., 2016; Xunlin et al., 2020). However, we did not find a significant improvement in follow-up. This suggests that the benefits of mindfulness interventions on quality of life may be more immediate and diminish over time without continued practice or reinforcement. Additionally, our analysis may be subject to potential publication bias, as we included studies that primarily focused on cognitive function, with quality of life often reported as secondary outcomes. This could affect the reliability of our conclusions.
Several factors should be considered when interpreting these results. In RCTs, blinding participants and outcome assessors was challenging due to the nature of the interventions and the reliance on self-reported cognitive outcomes, leading to a moderate to high risk of bias. Additionally, three studies reported participant dropout, which may undermine the reliability of the results (Hao et al., 2019; Henneghan et al., 2020;Shan, 2014). Non-RCT studies, while providing valuable insights by including a broader range of participants, were more susceptible to confounding bias due to their limited ability to control for external factors such as individual differences and temporal variables. To address these bias, future research should focus on improving blinding procedures in RCTs. If full blinding is not feasible, assessor blinding or the use of objective evaluation tools could be adopted to reduce bias. Additionally, intervention designs should consider participant adherence strategies to minimize attrition, such as providing more support or incentives. Our findings also indicate that interventions lasting ≥8 weeks do not show significantly greater effects than those lasting <8 weeks. Therefore, shorter interventions may be considered in future studies to reduce burden and enhance patient adherence. For non-RCT studies, techniques such as propensity score matching could be applied to better control for confounders. In conclusion, future studies should adopt rigorous methodologies to enhance the reliability, generalizability, and clarity of findings on mindfulness interventions.
LimitationThe results of this study should be interpreted with caution in light of these limitations. First, we conducted a within-group meta-analysis for both RCT and non-RCT studies, which may introduce bias (Cuijpers et al., 2017). However, non-RCT studies were included to provide a more comprehensive assessment of the impact of mindfulness interventions on cognitive function. Second, this study only included research published in Chinese and English, which may have introduced a selection bias. However, as Chinese and English are widely used languages in academic research, the inclusion of these studies likely captured a significant proportion of relevant findings. Third, although we conducted subgroup analyses to explore the heterogeneity, there remain other factors were not fully considered, such as the total hours of intervention, the structure of sessions, and the additional communication during the intervention. Finally, due to limitations in the study design of the included articles, there was a moderate to high risk of bias, which may have impacted the reliability of our findings.
ConclusionsThe findings of this systematic review and meta-analysis indicate that mindfulness-based interventions can effectively improve subjective cognitive function in cancer patients both after the intervention and during short-term follow-up, but not objective cognitive function. Specifically, interventions conducted in developing countries and without additional home practice yield larger effects than those in developed countries and with additional practice. Furthermore, there are no differences in the beneficial effects between interventions lasting eight weeks or longer and those less than eight weeks. Additionally, these interventions can improve quality of life post-intervention. Based on the findings and limitations, future research should focus on conducting higher-quality trials with larger sample sizes and more rigorous methodologies to confirm the effectiveness. Additionally, exploring the biological mechanisms, such as stress reduction, hormonal modulation, and neural connectivity, incorporating neurobiological measures will help provide a more comprehensive understanding of the cognitive benefits of mindfulness interventions. Lastly, tailoring mindfulness interventions to different cultural and socioeconomic contexts will be important for enhancing their broader applicability and effectiveness.
