Elsevier

Journal of Affective Disorders

Volume 202, 15 September 2016, Pages 39-45
Journal of Affective Disorders

Research paper
The relationship of self-compassion and depression: Cross-lagged panel analyses in depressed patients after outpatient therapy

https://doi.org/10.1016/j.jad.2016.05.032Get rights and content

Highlights

  • (A lack of) self-compassion significantly predicted subsequent depressive symptoms.

  • Depressive symptoms did not significantly predict subsequent levels of self-compassion.

  • These patterns were also found for the presence of a major depressive episode.

  • Patterns remained for either only positive or negative facets of self-compassion.

Abstract

Background

Previous cross-sectional studies suggest that self-compassion and depressive symptoms are consistently negatively associated. Although it is often implicitly assumed that (a lack of) self-compassion precedes depressive symptoms, so far no study has tested whether (lack of) self-compassion is a cause or a consequence of depressive symptoms, or both.

Method

To examine such reciprocal effects, we used data of 125 depressed outpatients after a time limited cognitive-behavioral psychotherapy. We assessed self-compassion and depressive symptoms via self-report measures and the presence of a major depressive episode directly after therapy, as well as 6 and 12 months later.

Results

Cross-lagged panel analyses indicated that (lack of) self-compassion significantly predicted subsequent depressive symptoms while controlling for autoregressive effects, whereas depressive symptoms did not predict subsequent levels of self-compassion. This was also the case for the relationship between self-compassion and the presence of a major depressive episode. The same patterns also occurred when we separately tested the reciprocal effects for two composite sub-measures of either positive or negative facets of self-compassion.

Limitations

Causality cannot be inferred from our results. Depressive symptoms and self-compassion could still be causally unrelated, and a third variable could account for their negative association.

Conclusions

These findings support the notions that (a lack of) self-compassion could serve as a vulnerability factor for depression and that cultivating self-compassion may deserve a focus in depression prevention programs or treatments.

Introduction

Depression is one of the most common mental disorders (Kessler et al., 2005) and its associated burden represents a major public health problem that affects depressive patients as well as society (Kupfer et al., 2012, Üstün et al., 2004). Depression is often marked by recurrent nature (Bockting et al., 2015), and although many patients can benefit from depression treatments, relapse and recurrence rates even after a successful therapy are considerable (Vittengl et al., 2007). Additionally, residual depressive symptoms have been found to be the most consistent and strongest predictor of depression relapse (e.g., Judd et al., 1998).

Self-compassion describes a mindful and benevolent attitude towards oneself when challenged with failure, personal weaknesses or facing physical pain. A recent meta-analysis showed that self-compassion is positively associated with well-being (Zessin et al., 2015). Moreover, a fast growing body of research suggests that self-compassion and its cultivation deserves closer attention also in clinical and non-clinical populations (for an overview see Barnard and Curry, 2011, Galante et al., 2014, Gilbert and Procter, 2006, Hofmann et al., 2011, Neff, 2015).

Many studies have investigated the cross-sectional association of self-compassion and depression. A meta-analysis by MacBeth and Gumley (2012) including about 4000 subjects has shown a mean effect size of r=−.51 between self-compassion and depressive symptoms. Moreover, a study by Krieger and colleagues (Krieger et al., 2013) indicated that people suffering from a major depressive episode reported significantly lower levels of self-compassion than never-depressed people even when controlling for depressive symptoms. Together, these studies suggest a close association between depression and self-compassion. However, most of the studies that investigated the relationship between depression or depressive symptoms with self-compassion used a cross-sectional design. One exception is a study by Raes (2011) that investigated the longitudinal association between self-compassion and depressive symptoms. A sample of 347 first-year psychology students completed measures of self-compassion and depressive symptoms at two assessments separated by a 5-month period. Results showed that levels of self-compassion at baseline were significantly negatively associated with depressive symptoms. In line with this result, Zeller and colleagues (Zeller et al., 2014) found in an at-risk youth sample that self-compassion prospectively predicted not only reduced levels of depression but also posttraumatic stress, panic, and suicidal ideation, as well as wellbeing outcomes over six months.

However, a limitation of both studies is that they did not preclude a possible reciprocal effect, i.e., that change in (depressive) symptoms may lead to a change of self-compassion. As a consequence, the cross-sectional negative correlation between self-compassion and depressive symptoms may be the result of different mechanisms. It may be that (1) (a lack of) self-compassion causes depressive symptoms, (2) depressive symptoms cause (a lack) of self-compassion, (3) depressive symptoms and (a lack of) self-compassion cause each other, or (4) depressive symptoms and self-compassion are causally unrelated and a third variable accounts for their negative association. Multi-wave longitudinal data are needed to test reciprocal relations. Our study uses a 12-month, three-wave longitudinal design that tests the interplay between self-compassion and depressive symptoms over time. Thus, the main aim of the present study was to test the possible mechanisms outlined above. Consequently, the present study combines a multi-wave longitudinal design with a cross-lagged analysis (Finkel, 1995), that allows for disentangling the cause-effect relationships.

To arrive at a better understanding of the association between depression and self-compassion, we may also need to look more closely into the assessment of the self-compassion construct. Self-compassion is typically measured by the Self-Compassion Scale (Neff, 2003a). Conceptually, Neff (2003b) defined self-compassion in terms of three bipolar components: (a) self-kindness (vs. self-judgment), (b) common humanity (vs. isolation), and (c) mindfulness (vs. overidentification). Self-kindness refers to the ability of treating oneself with care and understanding as opposed to harsh self-judgment. Common humanity refers to the recognition that imperfection is a shared aspect of the human experience, as opposed to feeling isolated and alone by one's failures and imperfections. Mindfulness involves holding and accepting one's present-moment experience as opposed to getting involved with the emotion. The SCS includes one scale for each component and their negative opposites. Although this 6-factor-factor structure finds empirical support in different samples, there has been a controversial discussion about the assessment of self-compassion (Neff, 2015). Particularly relevant for the current research is the question whether an empirical association between (lack of) self-compassion and psychopathology may be inflated by the above-mentioned negative components of the SCS. As recently argued by Muris (2015) the composite of the negative components of the SCS, i.e., “Self-Coldness” (Gilbert et al., 2011, Körner et al., 2015) or “Uncompassion” (Neff, 2015), not only taps the true, protective nature of self-compassion but also non-compassionate mechanisms, such as rumination or self-criticism, that do not fit with the original definition of self-compassion. To investigate such possible differential contributions of the positive and negative components of the SCS to the association with depressive symptoms, we perform separate analyses for the positive and negative subscales and compare the results with those obtained with the SCS total score.

The present study aimed to investigate the longitudinal relationship of self-compassion with depressive symptoms and depressive episodes. Based on previous research, our hypothesis was that there will be a negative time-lagged relationship of self-compassion with both depressive symptoms and depressive episodes. For the reversed time-lagged relationship of depressive symptoms and depressive episodes with self-compassion we had no specific hypothesis, since this time-lagged relationship has never been investigated before. Similarly, we had no specific hypothesis for the time-lagged associations between the self-compassion subscales and depressive symptoms/depressive episodes.

Section snippets

Participants

The sample consisted of 125 participants (54% female) who had completed treatment in a depression treatment efficacy study. The study protocol for the treatment study and the follow-up assessments used in the present study were approved by the local ethics committee, and all patients signed informed-consent forms before the initial inclusion in the study. At intake all participants completed a diagnostic interview, the Structured Clinical Interview for the DSM–IV Axis I (SCID; First et al., 1995

Results

Table 1 shows the means, standard deviations, internal consistencies, and correlations of the measures used at all time points of assessment. Self-compassion as well as SCS-POS and depressive symptoms were negatively associated, both within and across measurement occasions. SCS-NEG and depressive symptoms were positively associated, both within and across measurement occasions.

Discussion

The present study investigated the reciprocal effects of self-compassion and depression in a sample of former psychotherapy patients who had originally been diagnosed with major depression after the end of treatment in a prospective study with three waves over 12 months. Results indicate that (lack of) self-compassion predicts depressive symptoms, while depressive symptoms do not predict levels of self-compassion. This pattern of results was corroborated when depression was measured by the

Conclusion

The present study showed that self-compassion predicts depression, whereas depression does not predict self-compassion. Although the results of the present study indicate that self-compassion is a rather stable trait with stability coefficients of about .80, an increasing number of intervention studies in clinical and subclinical populations (Albertson et al., 2014, Held and Owens, 2015, Jazaieri et al., 2014, Neff and Germer, 2012, Shahar et al., 2012, Shapira and Mongrain, 2010) show that

References (54)

  • E.R. Albertson et al.

    Self-compassion and body dissatisfaction in women: a randomized controlled trial of a brief meditation intervention

    Mindfulness

    (2014)
  • A.B. Allen et al.

    Self-compassion, stress, and coping

    Soc. Personal. Psychol. Compass

    (2010)
  • L.K. Barnard et al.

    Self-compassion: conceptualizations, correlates, and interventions

    Rev. Gen. Psychol.

    (2011)
  • A.T. Beck et al.

    Manual for Beck Depression Inventory – Second Edition

    (1996)
  • T. Brockmeyer et al.

    Preliminary evidence for a nexus between rumination, behavioural avoidance, motive satisfaction and depression

    Clin. Psychol. Psychother.

    (2014)
  • D.A. Cole et al.

    Testing mediational models with longitudinal data: questions and tips in the use of structural equation modeling

    J. Abnorm. Psychol.

    (2003)
  • J.C. Coyne et al.

    Effects of recent and past major depression and distress on self-concept and coping

    J. Abnorm. Psychol.

    (1998)
  • S.E. Finkel

    Causal analysis with panel data

    (1995)
  • M.B. First et al.

    Structured Clinical Interview for DSM-IV (SCID)

    (1995)
  • J. Galante et al.

    Effect of kindness-based meditation on health and well-being: a systematic review and meta-analysis

    J. Consult. Clin. Psychol.

    (2014)
  • P. Gilbert et al.

    Fears of compassion: development of three self-report measures

    Psychol. Psychother.: Theory, Res. Pract.

    (2011)
  • P. Gilbert et al.

    Compassionate mind training for people with high shame and self-criticism: overview and pilot study of a group therapy approach

    Clin. Psychol. Psychother.

    (2006)
  • P. Held et al.

    Effects of self-compassion workbook training on trauma-related guilt in a sample of homeless veterans: a pilot study

    J. Clin. Psychol.

    (2015)
  • L. Hu et al.

    Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives

    Struct. Equ. Modeling: A Multidiscip. J.

    (1999)
  • J. Hupfeld et al.

    Validierung einer deutschen Version der Self-Compassion Scale (SCS-D) [Validation of a German version of the Self-Compassion Scale (SCS-D)]

    Z. Klin. Psychol. Psychother.

    (2011)
  • H. Jazaieri et al.

    A randomized controlled trial of compassion cultivation training: Effects on mindfulness, affect, and emotion regulation

    Motiv. Emot.

    (2014)
  • R.C. Kessler et al.

    Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication

    Arch. Gen. Psychiatry

    (2005)
  • Cited by (91)

    • Six-month stability of individual differences in sports coaches’ burnout, self-compassion and social support

      2022, Psychology of Sport and Exercise
      Citation Excerpt :

      In a study of female US athletes over a short time span of five days (pre- and post-competition), Killham et al. (2018) reported a rank-order stability of r = 0.81; to our knowledge, there are no other empirical data on the stability of self-compassion in coaches (or athletes). In other studies of adult depressive outpatients and US college students, self-compassion exhibited rank-order stabilities between r = 0.51 and r = 0.80 over periods ranging from 12 months to four years (Donald et al., 2018; Krieger et al., 2016; Stutts et al., 2018). In a sample of Chinese adolescents, self-compassion exhibited a lower rank-order stability of r = 0.34 over a three-year period (Yang et al., 2021).

    View all citing articles on Scopus
    View full text