International Journal of Clinical and Health Psychology 21 (2021) 100229 International Journal of Clinical and Health Psychology www.elsevier.es/ijchp ORIGINAL ARTICLE Impact of mindfulness and self-compassion on anxiety and depression: The mediating role of resilience Adrián Pérez-Aranda a,b,c , Javier García-Campayo a,d , Francisco Gude e,∗ , Juan V. Luciano c,‡ , Albert Feliu-Soler b,c,‡ , Arturo González-Quintela f , Yolanda López-del-Hoyo a,g , Jesus Montero-Marin h,‡ a Aragon Institute for Health Research (IIS Aragón), Miguel Servet University Hospital, Spain b Faculty of Psychology, Universitat Autònoma de Barcelona, Spain c AGORA Research Group, Teaching, Research & Innovation Unit, Parc Sanitari Sant Joan de Déu, Spain d Primary Care Prevention and Health Promotion Research Network, RedIAPP, Spain e Department of Clinical Epidemiology, Complejo Hospitalario Universitario, Health Research Institute of Santiago de Compostela (IDIS), Spain f Department of Internal Medicine, Complejo Hospitalario Universitario, Health Research Institute of Santiago de Compostela (IDIS), Spain g Department of Psychology and Sociology, University of Zaragoza, Spain h Department of Psychiatry, University of Oxford, Warneford Hospital, United Kingdom Received 25 October 2020; accepted 2 February 2021 KEYWORDS Abstract Background/Objective ‘Third-wave’ psychotherapies have shown effectiveness for Mindfulness; treating psychopathological symptoms such as anxiety and depression. There is burgeon- Self-compassion; ing interest in examining how these therapies’ core constructs produce their therapeutic Resilience; benefits. This study explores the hypothetical mediating effect of resilience in the impact Psychopathological of mindfulness and self-compassion on anxiety and depressive symptoms. Method: Cross- symptoms; sectional study design. The sample consisted of 860 Spanish general population participants. Ex post fact study The measures included the Mindful Attention Awareness Scale (MAAS), the Self-Compassion Scale (SCS-12), the Connor-Davidson Resilience Scale (CD-RISC) and the Goldberg Anxiety and Depression Scale (GADS). Bivariate correlations were calculated, and path analy- sis models were performed. Results: Significant correlations were found between the study variables, always in the expected direction (all p values <.001). The path analy- sis models showed significant direct effects of mindfulness and self-compassion on anxiety and depression symptoms, but the only significant indirect effects through resilience were ∗ Corresponding author: Department of Clinical Epidemiology, Complejo Hospitalario Universitario, Health Research Institute of Santiago de Compostela (IDIS), Spain. E-mail address: francisco.gude.sampedro@sergas.es (F. Gude). ‡ These authors share senior authorship. https://doi.org/10.1016/j.ijchp.2021.100229 1697-2600/© 2021 Asociación Española de Psicologı́a Conductual. Published by Elsevier España, S.L.U. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). A. Pérez-Aranda, J. García-Campayo, F. Gude et al. found on depression (MAAS: ␤ = -.05, 95% CI = -.11 to -.02; SCS-12: ␤ = -.06, 95% CI = -.33 to -.07). Conclusions: Resilience might partially mediate the effect of mindfulness and self-compassion on depression, but not on anxiety. © 2021 Asociación Española de Psicologı́a Conductual. Published by Elsevier España, S.L.U. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4. 0/). PALABRAS CLAVE Impacto del mindfulness y de la autocompasión en la ansiedad y la depresión: rol Mindfulness; mediador de la resiliencia Autocompasión; Resumen Antecedentes/Objetivo: Las terapias de ‘‘tercera generación’’ han probado ser efi- Resiliencia; caces para tratar síntomas psicopatológicos como la ansiedad y la depresión. Hay un interés Síntomas creciente en examinar cómo los constructos nucleares de estas terapias producen beneficios psicopatológicos; terapéuticos. Este estudio explora el hipotético efecto mediador de la resiliencia en la relación Estudio ex post fact del mindfulness y la autocompasión sobre los síntomas de ansiedad y depresión. Método: Estudio de diseño transversal. Muestra formada por 860 participantes españoles de la población gen- eral. Las medidas incluidas fueron: MAAS, SCS-12, CD-RISC y GADS. Se calcularon correlaciones bivariadas y se realizaron modelos de análisis del camino. Resultados: Se hallaron correlaciones significativas entre las variables, siempre en la dirección esperada (p < 0,001). El modelo de análisis del camino mostró efectos directos significativos de mindfulness y autocompasión sobre síntomas de ansiedad y depresión, pero el único efecto indirecto significativo a través de la resiliencia se encontró en depresión (MAAS: ␤ = -0,05, 95% CI = -0,11 --- -0,02; SCS-12: ␤ = -0,06, 95% CI = -0,33 --- -0,07). Conclusiones: La resiliencia puede mediar parcialmente el efecto de mindfulness y autocompasión sobre la depresión, pero no sobre la ansiedad. © 2021 Asociación Española de Psicologı́a Conductual. Publicado por Elsevier España, S.L.U. Este es un artı́culo Open Access bajo la licencia CC BY (http://creativecommons.org/licenses/ by/4.0/). Depression and anxiety are the most common mental dis- Mindfulness was defined as the ‘awareness that emerges orders in the general population, with a lifetime prevalence through paying attention on purpose, in the present of 20.60% in the case of major depressive disorder (Hasin moment, and nonjudgmentally to the unfolding of experi- et al., 2018), and a current global prevalence of anxiety dis- ence’ (Kabat-Zinn, 2003). Although it has been described as orders of 7.30% (Stein et al., 2017). These conditions, which a construct that can be trained through practice, mindful- not seldom are experienced together, often imply a signifi- ness has also been studied as a trait, and has been observed cant loss of quality of life and a certain degree of impairment to play a significant role in emotional self-regulation and in different life areas (Khansa et al., 2020). depression vulnerability (Guendelman et al., 2017). Other Among the many different approaches that have been studies have also found a significant mediating effect of proposed for treating anxiety and depressive symptoms, mindfulness in psychological wellbeing, as well as depres- ‘third-wave’ psychotherapies have been granted much sion and anxiety (Pagnini et al., 2019; Takahashi et al., 2019, attention in the last two decades. These represent an inno- 2020). Self-compassion, on its part, is defined as ‘being vation within cognitive-behavior therapy, as they do not touched by and open to one’s own suffering, not avoiding aim at symptom improvement as their only objective and or disconnecting from it, generating the desire to allevi- underline themes such as mindfulness, compassion, cog- ate one’s suffering and to heal oneself with kindness’ (Neff, nitive fusion, acceptance, and spirituality (Jahoda et al., 2003). It is a strong predictor of reduced depression and trait 2017; Pérez-Aranda et al., 2019). Mindfulness-based inter- anxiety, and greater life satisfaction (Van Dam et al., 2011), ventions, Compassion-Focused Therapy, and Acceptance and and different studies have found a mediating effect of self- Commitment Therapy are some examples of ‘third-wave’ compassion in anxiety and depression (Mehr & Adams, 2016; psychotherapies that have proven efficacy for treating dif- Takahashi et al., 2019). ferent conditions, mainly depressive and anxiety disorders Thus, the relation between mindfulness and self- (O’Connor et al., 2018; Pardos-Gascón et al., 2021; Wilson compassion with anxiety and depression seems to be well et al., 2019). documented (Conversano et al., 2020). However, this rela- The increasing body of evidence of these interventions’ tion could be mediated by some mechanistic variables, such effectiveness has stimulated analyses on how ‘third-wave’ as resilience. Resilience is defined as a dynamic and flexible variables may influence the individual’s mental health. In process of adaptation to life changes that enables an indi- this regard, two possible mediating factors have been stud- vidual to cope with and recover from stress, and to flourish ied with growing interest: mindfulness and self-compassion. when faced with adversity (Rutter, 1985), and its protective 2 International Journal of Clinical and Health Psychology 21 (2021) 100229 effect on mental disorders is widely accepted (Southwick jects (68%) agreed to participate in the study. In a second & Charney, 2012). Different studies have observed signifi- stage, two out of three individuals (N = 1,010) were invited cant associations between resilience and mindfulness (Joyce to participate in the survey, of which 860 subjects com- et al., 2018; Kemper et al., 2015; Montero-Marin et al., pleted the questionnaires (58%). This sample size fulfilled 2015). Moreover, there are many studies conducted with the Nunnally and Bernstein (1994) recommendations---a min- clinical and non-clinical samples supporting the mediating imum of 10 participants per variable---for performing path role of resilience in the impact of mindfulness on different analytic approximations. outcomes related to subjective wellbeing (Bajaj & Pande, The present study was approved by the Regional Ethics 2016; Wang et al., 2016). To our knowledge, the potential Committee (code 2012-025). All participants gave their writ- mediating role of resilience has not been explored yet in ten informed consent for data collection, and their provided the impact of self-compassion on psychological outcomes, data were completely anonymized. The study was per- but significant associations between the two variables, along formed in accordance with the Helsinki Declaration. with mindfulness and quality of life, have been reported (Asensio-Martínez et al., 2019; Kemper et al., 2015; Neff & McGehee, 2010; Sünbül & Güneri, 2019). Considering the abovementioned findings, it seems pos- Measures sible that resilience is somehow promoted by the practice of mindfulness skills and self-compassion. These imply prac- The following battery of paper-and-pencil self-report mea- ticing emotion and attention regulation abilities (Diedrich sures was administered to the participants along with a et al., 2014; Hanley et al., 2017) which, in turn, are closely sociodemographic questionnaire asking about age, gender, related to the capacity of recovering from stressful situa- marital status, level of education, and employment. tions (Kay, 2016; Mayordomo et al., 2016), to the point that The Goldberg Anxiety and Depression Scale (GADS; some authors consider ‘emotional resilience’ to be a poten- Goldberg et al., 1988) is an 18-item questionnaire that tial mechanism of mindfulness (Hayes & Feldman, 2006; contains 2 subscales (i.e., Anxiety and Depression), each Polizzi et al., 2018). composed of 9 binary items (yes/no). The aim of this scale In the present study, our main objective was to explore is to detect ‘‘probable cases’’, orienting the clinician in the the potential differential mediation role of resilience in the diagnosis (Montón et al., 1993). Scores for each subscale effect of mindfulness and self-compassion on anxiety and range from 0 to 9, where higher scores indicate more sever- depression. Our hypotheses were: (1) mindfulness and self- ity. Previous studies have proposed cut-off points (≥4 for compassion would have significant direct effects on anxiety anxiety and ≥2 for depression) for considering ‘‘probable and depression, and considering the findings of previous cases’’ (Reivan-Ortiz et al., 2019). The Spanish version of works (Quist-Møller et al., 2018; Van Dam et al., 2011), the GADS presented 82% of specificity and 83% of sensibil- self-compassion was expected to have a stronger effect than ity, next to an adequate concurrent validity (Montón et al., mindfulness; and (2) both mindfulness and self-compassion 1993). would have a significant indirect effect on anxiety and The Mindful Attention Awareness Scale (MAAS; Brown & depression through the mediating role of resilience. Ryan, 2003) is a 15-item scale which measures mindfulness as a trait. Each item is scored using a 6-point Likert scale. The total score is calculated by the mean of the 15 items Method and ranges from 1 to 6, with higher scores indicating greater levels of mindfulness. The Spanish version of the MAAS (Soler Participants and procedure et al., 2012) has shown appropriate convergent validity with other mindfulness measures, internal consistency (˛ = .89), The subjects taking part in this cross-sectional study were and test-retest reliability (r = .82). a subset of the participants in the A Estrada Glycation The Self-Compassion Scale-short form (SCS-SF; Raes and Inflammation Study (AEGIS; trial NCT01796184) (Gude et al., 2011) is a 12-item questionnaire designed to assess et al., 2017). A multistage sampling was carried out in overall self-compassion and three dimensions: Common the municipality of A Estrada, Galicia (Spain), with an humanity, Mindfulness, and Self-kindness. For the present adult population of 18,897 residents. From November 2012 study only the total score, which is calculated by the mean through March 2015, all subjects were successively convened of the 12 items, was considered (Neff et al., 2019). Total for 1 day at the Primary Care Center for evaluation. The scores range from 1 to 5, with higher ones indicating greater inclusion criteria were: (1) aged 18 years and older; (2) levels of self-compassion. The Spanish version of the SCS-12 proficiency in spoken Spanish; and (3) provide informed con- (Garcia-Campayo et al., 2014) has shown good internal con- sent. The exclusion criterion was the presence of severe sistency (˛ = .86) and very high convergence with the long physical disease. Participants were randomly selected from form (26-items version) of the scale (r ≥ .97). the Healthcare Registry. First, a computer program (sample The Connor-Davidson Resilience Scale (CD-RISC; function in R) generated a random sample of 3,500 subjects, Campbell-Sills & Stein, 2007) consists of 10 items measuring stratified by age group (in 7 categories, every 10 years). Of resilience. Each item is scored in a 5-point Likert scale, these, 639 could not be contacted, 134 lived outside of A and the total score, which ranges 1-5, is calculated by Estrada, 19 did not have healthcare coverage, and 84 were averaging the scores of the items; higher scores indicate deceased. Of the remaining eligible subjects (N = 2,624), 394 greater resilience. The CD-RISC has shown good internal were excluded due to failure to meet the inclusion criteria, consistency (˛ = .89) and test-retest reliability (r = .87) in and 714 subjects refused to participate. A total of 1,516 sub- people with anxiety or stress-related disorders. The Spanish 3 A. Pérez-Aranda, J. García-Campayo, F. Gude et al. version (Soler Sánchez et al., 2016) presented good internal Table 1 Demographic and clinical characteristics of the consistency (˛ = .86) and test-retest reliability (r = .87). sample. Total sample Statistical analysis (N = 860) Descriptive data analyses were performed for describing Demographic characteristics the sample, reporting frequencies and percentages for - Sex, n women (%) 489 (56.90%) categorical data and means and standard deviations for con- - Age, M (SD) 48.87 (16.91) tinuous variables. Bivariate analyses were carried out by  18-29 102 (11.90%) calculating correlation between variables, using Pearson’s  30-39 144 (16.70%) ␸ coefficient, the point-biserial correlation (rb ), and Pear-  40-49 172 (20%) son’s r coefficient when appropriate. Path analysis models  50-59 164 (19.10%) were performed to test the study hypothesis; path analysis  60-69 142 (16.50%) includes mediation effects and simultaneous estimation of  70-79 87 (10.10%) the relationships among variables in order to estimate these  ≥80 49 (5.70%) relationships in an unbiased way (Lockhart et al., 2011; - Marital status, n of married (%) 496 (57.70%) MacKinnon, 2008). Figure 1 shows a generic path analytic - Educational status, n of university 129 (14.90%) model with two (correlated) independent variables and one studies (%) mediator. For our study, we computed two models consid- - Employment status, n of currently 378 (44%) ering MAAS and SCS-12 as independent variables, CD-RISC working (%) as a mediator, and GADS subscales as outcomes (i.e. proba- Clinical characteristics [range] ble case vs. improbable case of anxiety/depressive disorder - GADS-Anxiety, M (SD) [0-9] 1.60 (2.50) using the abovementioned cut-off criteria). Standardized Probable cases, n (%) 192 (22.3%) regression coefficients (␤) of bias-corrected bootstrapped - GADS-Depression, M (SD) [0-9] 1.14 (2.11) indirect effects based on 10,000 bootstrap samples were cal- Probable cases, n (%) 213 (24.9%) culated as well as their standard error and 95% confidence - MAAS, M (SD) [1-6] 4.51 (0.86) interval (CI). Parameters of indirect effects were consid- - SCS-12, M (SD) [1-5] 3.16 (0.67) ered statistically significant when the 95% CI did not include - CD-RISC, M (SD) [1-5] 3.67 (0.73) 0 (Lockhart et al., 2011). The statistical packages used for Note. GADS: Goldberg Anxiety and Depression Scale; MAAS: the present study were SPSS v27.0 and Mplus v8.4. Mindful Attention Awareness Scale; SCS-12: Self-Compassion Scale, Short form; CD-RISC: Connor-Davidson Resilience Scale. Results Descriptive analysis Path analysis for probable cases of anxiety The study sample was composed of 860 participants, among disorders which 489 (56.90%) were women. Age ranged from 18 to 88, with a mean of 48.90 years (SD = 16.91). The age The path analysis model showed significant direct effects of distribution was normal and similar to the Spanish popu- the two independent variables on the outcome, but no sig- lation’s (Kolmogorov-Smirnov test’s p = .052). Most of the nificant effect was found of the mediator on the outcome (p participants were married (n = 496, 57.70%) and had at = .619) and, therefore, the indirect effects were not sta- least primary studies (n = 723, 84.10%). Almost half of tistically significant. Table 3 presents the unstandardized them were employed (n = 378, 44%). The GADS-Anxiety coefficients, standard errors, and the statistical significance presented a mean score of 1.60 (SD = 2.50), and the GADS- of the direct and indirect effects, and the paths are repre- Depression a score of 1.14 (SD = 2.11). Considering the cutoff sented in Figure 2 including the corresponding standardized points mentioned above, 192 participants (22.30%) were coefficients. ‘‘probable cases’’ of anxiety disorders, and 214 (24.90%) were ‘‘probable cases’’ of depression. These results are summarized in Table 1. Path analysis for probable cases of depression Bivariate analysis A partial mediation effect of resilience was observed; the GADS-anxiety and GADS-depression were the most associ- direct effects were statistically significant (all p values < ated variables (see Table 2; Pearson’s ␸ = .51, p < .001). .001), including the effect of CD-RISC on the outcome (p Significant moderately low and negative correlations were = .002), and the indirect effects were also statistically sig- found between the GADS subscales and the other study mea- nificant. Table 4 presents the unstandardized coefficients, sures, with rb ranging between -.22 and -.32. MAAS, SCS-12 standard errors, and the statistical significance of the direct and CD-RISC presented moderate relationships, with r rang- and indirect effects, and the paths are represented in ing from .31 to .48. All correlations remained significant (p Figure 3 including the corresponding standardized coeffi- < .001) when controlling for CD-RISC. cients. 4 International Journal of Clinical and Health Psychology 21 (2021) 100229 Figure 1 Generic example of a path analysis model with two correlated independent variables (IV) and one mediator. Table 2 Correlations between psychological variables. Between brackets, partial correlations controlling for CD-RISC. GADS-Depression MAAS SCS-12 CD-RISC GADS-Anxiety .51* (.49*) -.30* (-.22*) -.31* (-.21*) -.22* GADS-Depression -.32* (-.24*) -.31* (-.19*) -.25* MAAS .37* (.27*) .31* SCS-12 .48* Note. * means p value <.001. Table 3 Direct and Bootstrap indirect effects in the multiple mediational models for GADS-anxiety. Direct effects Path Coeff. SE p value MAAS→CD-RISC a1 1.30 0.30 <.001 MAAS→GADS-Anxiety c1 -0.52 0.11 <.001 SCS-12→CD-RISC a2 4.62 0.34 <.001 SCS-12→GADS-Anxiety c2 -0.75 0.15 <.001 CD-RISC→GADS-Anxiety b -0.01 0.01 .619 Indirect effects Path Boots. SE 95% CI MAAS→CD-RISC→GADS-Anxiety a1 ×b -0.01 0.02 -0.05 to 0.03 SCS-12→CD-RISC→GADS-Anxiety a2 ×b -0.03 0.06 -0.15 to 0.09 Total indirect effects -0.04 0.15 -0.20 to 0.12 Note. GADS = Goldberg Anxiety and Depression Scale; MAAS = Mindful Attention Awareness Scale; SCS-12=Self-Compassion Scale, Short form; CD-RISC = Connor-Davidson Resilience Scale. These are non-standardized results; standardized results are reported in Figure 2. Table 4 Direct and Bootstrap indirect effects in the multiple mediational models for GADS-depression. Direct effects Path Coeff. SE p value MAAS→CD-RISC a1 1.30 0.30 <.001 MAAS→GADS-Depression c1 -0.59 0.11 <.001 SCS-12→CD-RISC a2 4.62 0.34 <.001 SCS-12→GADS-Depression c2 -0.61 0.15 <.001 CD-RISC→GADS-Depression b -0.04 0.01 .002 Indirect effects Path Boots. SE 95% CI MAAS→CD-RISC→GADS-Depression a1 ×b -0.05 0.02 -0.11 to -0.02 SCS-12→CD-RISC→GADS-Depression a2 ×b -0.19 0.06 -0.33 to -0.07 Total indirect effects -0.25 0.16 -0.41 to -0.09 Note. GADS = Goldberg Anxiety and Depression Scale; MAAS = Mindful Attention Awareness Scale; SCS-12=Self-Compassion Scale, Short form; CD-RISC = Connor-Davidson Resilience Scale. These are non-standardized results; standardized results are reported in Figure 3. 5 A. Pérez-Aranda, J. García-Campayo, F. Gude et al. Figure 2 Path analysis model for GADS-anxiety. Note. * means p value < .001; R2 represents the proportion of the variance for a dependent variable that is explained by an independent variable or variables. All coefficients are standardized. 95% CI are detailed between brackets. Figure 3 Path analysis model for GADS-depression. Note. * means p value < .001; R2 represents the proportion of the variance for a dependent variable that is explained by an independent variable or variables. All coefficients are standardized. 95% CI are detailed between brackets. Discussion positive correlations with resilience (Kemper et al., 2015; Neff & McGehee, 2010) and was negatively associated with The present work aimed at exploring the intermediary anxiety and depression (Mehr & Adams, 2016; Van Dam et al., role of resilience between ‘third-wave’-related constructs 2011), although the effect sizes were smaller than expected. (mindfulness and self-compassion) and psychopathological The results of the path analysis model supported our first symptoms (anxiety and depression). Our results bear out hypothesis, as both mindfulness and self-compassion had the expected relationships between the variables, con- significant direct effects on anxiety and depressive symp- sidering previous works: mindfulness presented significant tomatology; however, contrary to what was hypothesized positive associations with self-compassion (Baer et al., 2012) following previous findings (Quist-Møller et al., 2018; Van and resilience (Kemper et al., 2015; Montero-Marin et al., Dam et al., 2011), self-compassion’s direct effect was not 2015), and negative correlations with anxiety and depression significantly stronger than mindfulness’. (Takahashi et al., 2019). Similarly, self-compassion showed The second hypothesis was not supported by our results in the case of anxiety symptomatology: resilience showed a 6 International Journal of Clinical and Health Psychology 21 (2021) 100229 non-significant effect on anxiety after controlling for mind- mine a causal relationship, so the results are exploratory and fulness and self-compassion, which implied that the indirect should be interpreted with caution. Second, despite using paths were not statistically significant. Thus, and contrary a large sample of Spanish general population individuals, to expectations, resilience was not a significant mediator all of them were from the same region and, therefore, the of the effect of mindfulness and self-compassion on anxi- results could not be completely representative of the Span- ety symptomatology in our sample. Although resilience had ish population. Third, the use of self-reported measures in been identified as a mediator of the effect of mindfulness, this study, particularly the GADS, undermines the strength it was on other outcomes such as emotion regulation (Wang of the outcomes, as it is only an indicator of ‘‘probable et al., 2016) and positive affect (Bajaj & Pande, 2016), which cases’’ of anxiety disorders and depression, and previous can be related to anxiety symptoms, but would probably be studies have identified specificity and sensitivity issues with more directly associated to depression. Previous works have this scale. It is likely that the use of standardized psy- observed that other ‘third-wave’ constructs mediate the chiatric interviews (SCID, CIDI, MINI, etc.) conducted by effect of mindfulness on anxiety, such as non-attachment trained interviewers would have yielded a lower prevalence (Whitehead et al., 2019), which is defined as freedom from of anxiety and depression cases. Fourth, this study considers unhealthy cognitive fixations on objects and others (Deits- mindfulness as a unidimensional trait, but there is consistent Lebehn et al., 2019), and also decentering, defined as the evidence that it could be a multi-facet construct and, thus, capacity to observe items that arise in the mind as mere psy- future studies should replicate our work examining the role chological events (Hoge et al., 2015). Further studies should of the different facets of mindfulness. Finally, it needs to replicate these findings and include other ‘third-wave’ core be considered that the data were collected between 2012 constructs, such as psychological flexibility, which has been and 2015, which implies that the results should be framed found to mediate effects of ‘third-wave’ psychotherapies in the socioeconomic context of that period; the impact of on outcomes such as stress and anxiety in clinical sam- the global economic crisis on mental health was particularly ples (Montero-Marín et al., 2018; Pérez-Aranda et al., 2019; noteworthy in Spain (Bartoll et al., 2014; Gili et al., 2013), Wicksell et al., 2010) and is defined as the ability to feel which could justify the high proportion of ‘‘probable cases’’ and think with an open mind while forging habits that allow of anxiety and depressive disorders among our sample. Thus, us to live in a way that is consistent with our values and future studies should replicate our model using more recent aspirations (Hayes, 2020; page 17). data and considering other designs such as longitudinal stud- On the other hand, our hypothesis was partially supported ies. in the case of depressive symptomatology: both direct and indirect paths were significant. That implies a possible par- Conclusions tial mediation of resilience that goes in line with the results reported by previous studies: Wang et al. (2016) found The results of the present study suggest that partial media- that emotional resilience partially mediated the effect of tion of resilience is playing a significant role on the effect of mindfulness on emotion regulation in college students in mindfulness and self-compassion on depression, but not on China; similarly, Bajaj and Pande (2016) reported a par- anxiety symptoms. These latter could be mediated by other tial mediation effect of resilience on mindfulness effect ‘third-wave’ variables such as decentering, non-attachment on life satisfaction, positive and negative affect in univer- or psychological flexibility, as previous studies have sug- sity students in India. In this regard, it is noteworthy that gested. some authors consider emotional resilience to be a potential mechanism of mindfulness (Hayes & Feldman, 2006; Polizzi et al., 2018); mindfulness may promote the generation of Funding and acknowledgements positive emotions and the ability to recover from negative emotions, helping the individual to maintain a decentered This project was supported by grants (PI16/01404 and attitude toward difficult situations (Bajaj & Pande, 2016). PI16/01395), the Spanish Network for Prevention and These mechanisms would help the person to regulate emo- Health Promotion in Primary Care (RD16/0007/0005, tions, increase life satisfaction and reduce negative affect RD16/0007/0006 and RD16/0007/0012), the Network for (Wang et al., 2016), making them less likely to present Addictive Disorders (RD16/0017/0018), ---- Instituto de Salud depressive symptomatology. Carlos III (ISCIII) of the Spanish Ministry of Economy and Com- For what concerns to self-compassion, no other work petitiveness, and DGA group (B17-17R), co-financed with has studied how its effect on anxiety and depression could European Union ERDF funds. A.P.-A. has a ‘Sara Borrell’ be mediated by resilience; however, different studies have research contract from the ISCIII (CD20/00181). J.M.-M. is reported significant associations between resilience and supported by the WellcomeTrust Grant (104908/Z/14/Z/). self-compassion (Kemper et al., 2015; Neff & McGehee, J.V.L. has a ‘‘Miguel Servet’’ research contract from the 2010), and considering its tight relation with mindful- ISCIII (CPII19/00003). A.F.S acknowledges the funding from ness (Neff, 2003), it could be expected that similar the Serra Húnter program (Generalitat de Catalunya; refe- mechanisms---i.e. generating positive emotions, in this case rence number UAB-LE-8015). through the desire to alleviate one’s suffering and to heal oneself with kindness---could explain the impact of self- References compassion on depression through the mediating effect of psychological resilience. Asensio-Martínez, Á., Oliván-Blázquez, B., Montero-Marín, J., Some limitations of this work must be acknowledged; Masluk, B., Fueyo-Díaz, R., Gascón-Santos, S., Gudé, F., first, the cross-sectional design of this study cannot deter- Gónzalez-Quintela, A., García-Campayo, J., & Magallón-Botaya, 7 A. Pérez-Aranda, J. García-Campayo, F. Gude et al. R. (2019). Relation of the psychological constructs of resilience, growth. Psychological Trauma: Theory, Research, Practice, and mindfulness, and self-compassion on the perception of physical Policy, 9, 526---536. https://doi.org/10.1037/tra0000208 and mental health. Psychology Research and Behavior Manage- Hasin, D. S., Sarvet, A. L., Meyers, J. L., Saha, T. D., Ruan, ment, 12, 1155---1166. https://doi.org/10.2147/PRBM.S225169 W. J., Stohl, M., & Grant, B. F. (2018). Epidemiology of Baer, R. A., Lykins, E. L. B., & Peters, J. R. (2012). Mind- adult DSM-5 major depressive disorder and its specifiers in fulness and self-compassion as predictors of psychological the United States. JAMA Psychiatry, 75, 336---346. https://doi. wellbeing in long-term meditators and matched nonmeditators. org/10.1001/jamapsychiatry.2017.4602 Journal of Positive Psychology, 7, 230---238. https://doi.org/ Hayes, A. M., & Feldman, G. (2006). Clarifying the Construct of 10.1080/17439760.2012.674548 Mindfulness in the Context of Emotion Regulation and the Process Bajaj, B., & Pande, N. (2016). Mediating role of resilience in the of Change in Therapy. Clinical Psychology: Science and Practice, impact of mindfulness on life satisfaction and affect as indices 11, 255---262. https://doi.org/10.1093/clipsy.bph080 of subjective well-being. Personality and Individual Differences, Hayes, S. C. (2020). Una mente liberada: la guía esencial de la 93, 63---67. https://doi.org/10.1016/j.paid.2015.09.005 terapia de aceptación y compromiso (ACT). Paidós. Bartoll, X., Palència, L., Malmusi, D., Suhrcke, M., & Borrell, C. Hoge, E. A., Bui, E., Goetter, E., Robinaugh, D. J., Ojserkis, R. (2014). The evolution of mental health in Spain during the eco- A., Fresco, D. M., & Simon, N. M. (2015). Change in Decen- nomic crisis. European Journal of Public Health, 24, 415---418. tering Mediates Improvement in Anxiety in Mindfulness-Based https://doi.org/10.1093/eurpub/ckt208 Stress Reduction for Generalized Anxiety Disorder. Cogni- Brown, K. W., & Ryan, R. M. (2003). The Benefits of Being tive Therapy and Research, 39, 228---235. https://doi.org/ Present: Mindfulness and Its Role in Psychological Well-Being. 10.1007/s10608-014-9646-4 Journal of Personality and Social Psychology, 84, 822---848. Jahoda, A., Stenfert Kroese, B., Pert, C., Jahoda, A., Stenfert https://doi.org/10.1037/0022-3514.84.4.822 Kroese, B., & Pert, C. (2017). Mindfulness and Third Wave Thera- Campbell-Sills, L., & Stein, M. B. (2007). Psychometric analy- pies. In A. Jahoda, B. Stenfert Kroese, & C. Pert (Eds.), Cognitive sis and refinement of the Connor-Davidson Resilience Scale Behaviour Therapy for People with Intellectual Disabilities (pp. (CD-RISC): Validation of a 10-item measure of resilience. 181---212). Palgrave Macmillan. Journal of Traumatic Stress, 20, 1019---1028. https://doi.org/ Joyce, S., Shand, F., Tighe, J., Laurent, S. J., Bryant, R. 10.1002/jts.20271 A., & Harvey, S. B. (2018). Road to resilience: A sys- Conversano, C., Ciacchini, Ŕ., Orrù, G., Di Giuseppe, M., Gemignani, tematic review and meta-analysis of resilience training A., & Poli, A. (2020). Mindfulness, compassion, and self- programmes and interventions. BMJ Open, 8 https://doi. compassion among health care professionals: What’s new? A org/10.1136/bmjopen-2017-017858 systematic review. Frontiers in Psychology, 11, Article 1683 Kabat-Zinn, J. K. (2003). Mindfulness-based interventions in https://doi.org/10.3389/fpsyg.2020.01683 context: Past, present, and future. Clinical Psychology- Deits-Lebehn, C., Smith, T. W., Grove, J. L., Williams, P. G., & Science And Practice, 10(2), 144---156. https://doi. Uchino, B. N. (2019). Dispositional mindfulness, nonattachment, org/10.1093/clipsy.bpg016 and experiential avoidance in the interpersonal circumplex. Kay, S. A. (2016). Emotion regulation and resilience: Over- Personality and Individual Differences, 151, Article 109522 looked connections. Industrial and Organizational Psychology, https://doi.org/10.1016/j.paid.2019.109522 9, 411---415. https://doi.org/10.1017/iop.2016.31 Diedrich, A., Grant, M., Hofmann, S. G., Hiller, W., & Berking, M. Kemper, K. J., Mo, X., & Khayat, R. (2015). Are mindfulness and self- (2014). Self-compassion as an emotion regulation strategy in compassion associated with sleep and resilience in health pro- major depressive disorder. Behaviour Research and Therapy, 58, fessionals? Journal of Alternative and Complementary Medicine, 43---51. https://doi.org/10.1016/j.brat.2014.05.006 21, 496---503. https://doi.org/10.1089/acm.2014.0281 Garcia-Campayo, J., Navarro-Gil, M., Andrés, E., Montero-Marin, Khansa, W., Haddad, C., Hallit, R., Akel, M., Obeid, S., Had- J., López-Artal, L., & Demarzo, M. M. P. (2014). Validation dad, G., Soufia, M., Kheir, N., Abi Elias Hallit, C., Khoury, R., of the Spanish versions of the long (26 items) and short Salameh, P., & Hallit, S. (2020). Interaction between anxiety (12 items) forms of the Self-Compassion Scale (SCS). Health and depression on suicidal ideation, quality of life, and work and Quality of Life Outcomes, 12, Article 4 https://doi. productivity impairment: Results from a representative sample org/10.1186/1477-7525-12-4 of the Lebanese population. Perspectives in Psychiatric Care, Gili, M., Roca, M., Basu, S., McKee, M., & Stuckler, D. (2013). The 56, 270---279. https://doi.org/10.1111/ppc.12423 mental health risks of economic crisis in Spain: Evidence from Lockhart, G., MacKinnon, D. P., & Ohlrich, V. (2011). Medi- primary care centres, 2006 and 2010. European Journal of Public ation analysis in psychosomatic medicine research. Health, 23, 103---108. https://doi.org/10.1093/eurpub/cks035 Psychosomatic Medicine, 73, 29---43. https://doi.org/ Goldberg, D., Bridges, K., Duncan-Jones, P., & Grayson, 10.1097/PSY.0b013e318200a54b D. (1988). Detecting anxiety and depression in general MacKinnon, D. P. (2008). Introduction to statistical mediation anal- medical settings. British Medical Journal, 297, 897---899. ysis. Erlbaum Psych Press. https://doi.org/10.1136/bmj.297.6653.897 Mayordomo, T., Viguer, P., Sales, A., Satorres, E., & Melén- Gude, F., Díaz-Vidal, P., Rúa-Pérez, C., Alonso-Sampedro, M., dez, J. C. (2016). Resilience and Coping as Predictors Fernández-Merino, C., Rey-García, J., Cadarso-Suárez, C., of Well-Being in Adults. Journal of Psychology: Inter- Pazos-Couselo, M., García-López, J. M., & Gonzalez-Quintela, disciplinary and Applied, 150, 809---821. https://doi.org/ A. (2017). Glycemic Variability and Its Association with 10.1080/00223980.2016.1203276 Demographics and Lifestyles in a General Adult Population. Mehr, K. E., & Adams, A. C. (2016). Self-Compassion as a Mediator Journal of Diabetes Science and Technology, 11, 780---790. of Maladaptive Perfectionism and Depressive Symptoms in Col- https://doi.org/10.1177/1932296816682031 lege Students. Journal of College Student Psychotherapy, 30, Guendelman, S., Medeiros, S., & Rampes, H. (2017). Mindfulness 132---145. https://doi.org/10.1080/87568225.2016.1140991 and emotion regulation: Insights from neurobiological, psycho- Montero-Marín, J., Navarro-Gil, M., Puebla-Guedea, M., Luciano, logical, and clinical studies. Frontiers in Psychology, 8, Article J. V., Van Gordon, W., Shonin, E., & García-Campayo, J. 220 https://doi.org/10.3389/fpsyg.2017.00220 (2018). Efficacy of ‘‘attachment-based compassion therapy’’ Hanley, A. W., Garland, E. L., & Tedeschi, R. G. (2017). Relating in the treatment of fibromyalgia: A randomized controlled dispositional mindfulness, contemplative practice, and positive trial. Frontiers in Psychiatry, 8, Article 307 https://doi.org/ reappraisal with posttraumatic cognitive coping, stress, and 10.3389/fpsyt.2017.00307 8 International Journal of Clinical and Health Psychology 21 (2021) 100229 Montero-Marin, J., Tops, M., Manzanera, R., Piva Demarzo, M. Depression Scale (GADS) in Ecuadorian population. Inter- M., De Mon, M. Á., & García-Campayo, J. (2015). Mind- national Journal of Psychological Research, 12, 41---48. fulness, resilience, and burnout subtypes in primary care https://doi.org/10.21500/20112084.3745 physicians: The possible mediating role of positive and Rutter, M. (1985). Resilience in the face of adversity. British negative affect. Frontiers in Psychology, 6, Article 1895 Journal of Psychiatry, 147, 598---611. https://doi.org/ https://doi.org/10.3389/fpsyg.2015.01895 10.1192/bjp.147.6.598 Montón, C., Pérez-Echevarría, M., Campos, R., García, J., & Lobo, Soler, J., Tejedor, R., Feliu-Soler, A., Pascual, J. C., Cebolla, A., A. (1993). Escalas de ansiedad y depresión de Goldberg: guía Soriano, J., Alvarez, E., & Perez, V. (2012). Psychometric pro- de entrevista eficaz para la detención del malestar psíquico. prieties of Spanish version of Mindful Attention Awareness Scale Atencion Primaria, 12, 345---349. (MAAS). Actas Espanolas de Psiquiatria, 40, 19---26. Neff, K. D. (2003). The development and validation of a scale Soler Sánchez, M. I., Meseguer de Pedro, M., & García to measure self-compassion. Self and Identity, 2, 223---250. Izquierdo, M. (2016). Propiedades psicométricas de la ver- https://doi.org/10.1080/15298860309027 sión española de la escala de resiliencia de 10 ítems de Neff, K. D., & McGehee, P. (2010). Self-compassion and Connor-Davidson (CD-RISC 10) en una muestra multiocupa- psychological resilience among adolescents and young cional. Revista Latinoamericana de Psicologia, 48, 159---166. adults. Self and Identity, 9, 225---240. https://doi.org/10. https://doi.org/10.1016/j.rlp.2015.09.002 1080/15298860902979307 Southwick, S. M., & Charney, D. S. (2012). The science of resilience: Neff, K. D., Tóth-Király, I., Yarnell, L. M., Arimitsu, K., Castilho, Implications for the prevention and treatment of depression. P., Ghorbani, N., Guo, H. X., Hirsch, J. K., Hupfeld, J., Hutz, Science, 338, 79---82. https://doi.org/10.1126/science.1222942 C. S., Kotsou, I., Lee, W. K., Montero-Marin, J., Sirois, F. M., De Stein, D. J., Scott, K. M., de Jonge, P., & Kessler, R. C. (2017). Souza, L. K., Svendsen, J. L., Wilkinson, R. B., & Mantzios, M. Epidemiology of anxiety disorders: From surveys to nosology (2019). Examining the factor structure of the Self-Compassion and back. Dialogues in Clinical Neuroscience, 19, 127---136. Scale in 20 diverse samples: Support for use of a total score www.dialogues-cns.org and six subscale scores. Psychological Assessment, 31, 27---45. Sünbül, Z. A., & Güneri, O. Y. (2019). The relationship between https://doi.org/10.1037/pas0000629 mindfulness and resilience: The mediating role of self compas- Nunnally, J. C., & Bernstein, I. (1994). Psychometric theory (3rd sion and emotion regulation in a sample of underprivileged ed.). McGraw-Hill. Turkish adolescents. Personality and Individual Differences, 139, O’Connor, M., Munnelly, A., Whelan, R., & McHugh, L. (2018). 337---342. https://doi.org/10.1016/j.paid.2018.12.009 The Efficacy and Acceptability of Third-Wave Behavioral and Takahashi, T., Kawashima, I., Nitta, Y., & Kumano, H. (2020). Cognitive eHealth Treatments: A Systematic Review and Meta- Dispositional Mindfulness Mediates the Relationship Between Analysis of Randomized Controlled Trials. Behavior Therapy, 49, Sensory-Processing Sensitivity and Trait Anxiety, Well-Being, 459---475. https://doi.org/10.1016/j.beth.2017.07.007 and Psychosomatic Symptoms. Psychological Reports, 123, Pagnini, F., Cavalera, C., Rovaris, M., Mendozzi, L., Molinari, 1083---1098. https://doi.org/10.1177/0033294119841848 E., Phillips, D., & Langer, E. (2019). Longitudinal associations Takahashi, T., Sugiyama, F., Kikai, T., Kawashima, I., Guan, S., between mindfulness and well-being in people with multiple Oguchi, M., Uchida, T., & Kumano, H. (2019). Changes in sclerosis. International Journal of Clinical and Health Psychol- depression and anxiety through mindfulness group therapy ogy, 19, 22---30. https://doi.org/10.1016/j.ijchp.2018.11.003 in Japan: The role of mindfulness and self-compassion as Pardos-Gascón, E. M., Narambuena, L., Leal-Costa, C., & van-der possible mediators. BioPsychoSocial Medicine, 13, Article 4 Hofstadt-Román, C. J. (2021). Differential efficacy between https://doi.org/10.1186/s13030-019-0145-4 cognitive-behavioral therapy and mindfulness-based therapies Van Dam, N. T., Sheppard, S. C., Forsyth, J. P., & Earleywine, for chronic pain: Systematic review. International Jour- M. (2011). Self-compassion is a better predictor than mindful- nal of Clinical and Health Psychology, 21, Article 100197 ness of symptom severity and quality of life in mixed anxiety https://doi.org/10.1016/j.ijchp.2020.08.001 and depression. Journal of Anxiety Disorders, 25, 123---130. Pérez-Aranda, A., Feliu-Soler, A., Montero-Marín, J., García- https://doi.org/10.1016/j.janxdis.2010.08.011 Campayo, J., Andrés-Rodríguez, L., Borràs, X., Rozadilla- Wang, Y., Xu, W., & Luo, F. (2016). Emotional resilience Sacanell, A., Peñarrubia-Maria, M. T., Angarita-Osorio, N., mediates the relationship between mindfulness and emo- McCracken, L. M., & Luciano, J. V. (2019). A randomized tion. Psychological Reports, 118, 725---736. https://doi.org/ controlled efficacy trial of mindfulness-based stress reduc- 10.1177/0033294116649707 tion compared with an active control group and usual care Whitehead, R., Bates, G., Elphinstone, B., Yang, Y., & Mur- for fibromyalgia: The EUDAIMON study. Pain, 160, 2508---2523. ray, G. (2019). Nonattachment Mediates the Relationship https://doi.org/10.1097/j.pain.0000000000001655 Between Mindfulness and Psychological Well-Being, Subjec- Polizzi, C., Gautam, A., & Lynn, S. J. (2018). Trait Mindfulness: A tive Well-Being, and Depression, Anxiety and Stress. Jour- Multifaceted Evaluation. Imagination, Cognition and Personal- nal of Happiness Studies, 20, 2141---2158. https://doi.org/ ity, 38, 106---138. https://doi.org/10.1177/0276236618774321 10.1007/s10902-018-0041-9 Quist-Møller, S. A., Sami, S., & Shapiro, L. S. (2018). Health Benefits Wicksell, R. K., Olsson, G. L., & Hayes, S. C. (2010). Psycholog- of (Mindful) Self-Compassion Meditation and the Potential Com- ical flexibility as a mediator of improvement in Acceptance plementarity to Mindfulness-Based Interventions: A Review of and Commitment Therapy for patients with chronic pain follow- Randomized-Controlled Trials. OBM Integrative and Complemen- ing whiplash. European Journal of Pain, 14, 1059.e1---1059.e11. tary Medicine, 4 https://doi.org/10.21926/obm.icm.1901002 https://doi.org/10.1016/j.ejpain.2010.05.001 Raes, F., Pommier, E., Neff, K. D., & Van Gucht, D. (2011). Wilson, A. C., Mackintosh, K., Power, K., & Chan, S. W. Y. (2019). Construction and factorial validation of a short form of the Self- Effectiveness of Self-Compassion Related Therapies: A Sys- Compassion Scale. Clinical Psychology and Psychotherapy, 18, tematic Review and Meta-analysis. Mindfulness, 10, 979---995. 250---255. https://doi.org/10.1002/cpp.702 https://doi.org/10.1007/s12671-018-1037-6 Reivan-Ortiz, G., Pineda-Garcia, G., & Parias, B. D. L. (2019). Psychometric properties of the Goldberg Anxiety and 9