Research Article: Mediators linking insecure attachment to eating symptoms: A systematic review and meta-analysis

Date Published: March 7, 2019

Publisher: Public Library of Science

Author(s): Laura Cortés-García, Bahi Takkouche, Gloria Seoane, Carmen Senra, Kenji Hashimoto.


In the last two decades, the number of studies focused on the mediators connecting insecure attachment with Eating Disorders (EDs), at both clinical and sub-clinical level, has considerably increased. However, there has not been a systematic synthesis of this literature to date. To fill this gap, the current meta-analytic review aimed at identifying and quantifying the extent to which mediators contribute to the explanation of this relationship. The present study was registered with PROSPERO (CRD42017076807). A comprehensive search process in seven different electronic databases retrieved 24 studies that examined how insecure attachment leads to ED symptoms through mediation analysis. Standardized regression coefficients of the indirect and total paths of 21 mediation models were pooled. Studies were coded and ranked for quality. We found evidence to show that maladaptive emotion regulation and depressive symptoms had the highest effect size for mediation (mediation ratio [PM] = 0.71). Further, body dissatisfaction, neuroticism, perfectionism, mindfulness and social comparison had significant, but moderate to low mediating effects (PM = 0.21–0.58). The methodological quality of these studies was mostly low to moderate and potential areas for development were highlighted. Our findings support the direct targeting of these psychological constructs in prevention programs and treatment of EDs. Future investigations addressing the time sequence between the variables will provide valuable clues to untangle the prospective contribution of each variable on the development and maintenance of eating pathology.

Partial Text

Broad research has found evidence of the relationship between insecure attachment style and mental disorders, including Eating Disorders (EDs) [1–3]. It is accepted that individuals with EDs have a higher prevalence of insecure attachment than healthy controls as it has been unanimously established in previous reviews [3–6]. In addition, insecure attachment seems to be a risk factor for the subsequent emergence of disordered eating in non-clinical populations [7,8].

The aim of this study was to provide a review and meta-analysis of studies investigating the mediators connecting insecure attachment with eating psychopathology at both clinical and sub-clinical level. Our results build on the already existent evidence that insecure attachment and ED symptoms could be explained by the indirect effect of various mechanisms. Consistent with previous reviews [5,8] and partly with our hypothesis, the mediators with the largest effect size were emotional dysregulation at both clinical and sub-clinical level, and depressive symptoms at sub-clinical level. In addition, body dissatisfaction, neuroticism, perfectionism, mindfulness and social comparison yielded significant impact also, but their effect size ranged from moderate to low.

The present study has several limitations worth noting. First, due to the scarce number of studies in relation to some mediators, only a subset of the studies could be included in this meta-analysis. Second, the majority of the findings proved in this meta-analysis were provided by normal population (predominantly younger Caucasian women), thus reducing the possibility for generalization; caution is needed in extrapolating the results to other sociodemographic groups. Third, the inclusion of cross-sectional data in the present meta-analysis did not make it possible to draw definitive conclusions regarding the development of the ED psychopathology and reflects the need for future studies to implement prospective designs. Fourth, some of the included studies [42,49,52,59] used exclusively Sobel test to prove the significance of mediation despite the fact that this test has been overcome [65]. Fifth, the heterogeneity of effects between studies was high. Nevertheless, as it is claimed by experts, heterogeneity in a meta-analysis should be viewed as expected rather than inevitable and not as a nuisance [111]. Therefore, we decided to interpret our results taking random-effects estimates as it might be the most appropriate way to deal with this issue [112]. Sixth, the mediation ratio [67] was used as a summary of the effect size for each mediator despite the fact that it is a measure that suffers from several limitations [35,37]. However, as the data included in the present meta-analysis reported larger total effects than the indirect effects and of the same sign, we followed recommendations of its application [35]. In addition, it should be noted that so far it is the most widely used measure of effect size for mediation models and a method relatively unaffected by sample size. Finally, most studies relied only on self-reports. Hence, some response biases may have affected the results.

Future studies should explore prospectively mediating mechanisms implementing designs with a temporal sequence ascertaining the precedence of the independent variable on the mediator and, by the same token, of the mediator on the dependent variable [38]. Because of the possible relationship between the main mediators found in this review, it would be interesting to explore the interplay among them through sequential multiple mediation models and moderated mediation models to understand the contribution of each mediator. Further, future mediation studies should apply powerful statistical techniques such as SEM with bootstrapping in order to strengthen conclusions and reporting the magnitude of the mediated effect [65]. Additionally, given the paucity of studies, it will be useful in the future to ascertain more precisely whether the mediational effects of such variables differ by specific diagnoses of ED and also by age distribution. Lastly, the control for confounding variables merits particular attention to rule out possible spurious effects [70].

A consistent finding in the literature is the influence of insecure attachment on the poorer therapeutic outcomes among patients with ED [13,113]; as insecurely attached patients tend to change or to abandon therapy frequently [45,114]. As suggested by Tasca and Balfour [5], clinicians should assess patient’s quality and level of attachment in order to adapt the therapy and to guarantee better therapeutic outcomes.

The results obtained in the present meta-analysis extend previous findings by showing that emotion dysregulation, depressive symptoms, body dissatisfaction, neuroticism, perfectionism, mindfulness and social comparison could be essential psychological mechanisms for explaining the pathways through which insecure attachment may increase the vulnerability to eating symptoms. However, since the application of mediation analysis in disordered eating research is still in its early stages, more studies are needed to corroborate our results. Specifically, longitudinal studies are required to clarify the interplay of the mediators between insecure attachment and ED symptoms.




Leave a Reply

Your email address will not be published.