Introduction

In the last decades, the ideal of thinness has become not only a sign of beauty but also of success and self-control [1]. At the same time, prevalence rates of eating disorders (ED) show a two- to five times increase in the past three decades as [2] described using current data in Spain [3, 4]. Similarly, Yeo and Hughes [5] documented that ED are the third most common pathology in adolescent females after obesity and asthma. EDs can also have severe medical and psychological consequences [2, 6, 7]; for example, anorexia nervosa is associated with the highest mortality rate of any psychiatric disorder [7–9]. Thus, it is both essential and urgent to investigate cognitive and behavioral aspects of ED and develop effective treatments to alleviate its symptoms.

Fairburn holds that EDs are in essence ‘cognitive disorders’ [10]. Some disordered eating-related behaviors are driven by negative self-evaluations, emotions or thoughts [11]. For example, patients with anorexia nervosa and bulimia nervosa often experience negative emotion states [6, 12], and focus repetitively on those experiences and its consequences, which is a maladaptive coping strategy called rumination [13, 14]. This ruminative thinking seems to share characteristics such as ‘inflexible’ and ‘difficult to control behavior’, proposed by Deci and Ryan [15], to define automatic behaviors. Nevertheless, Rawal, Park, and Williams [16] suggested that rumination may be a relevant feature in the information processing in ED, dealing with negative consequences as psychological distress and the maintenance of abnormal eating [17]. In addition, recent studies show the importance of rumination as a vulnerability factor for disordered eating, and in the aggravation of ED, since it is a coping strategy to distress related to negative affect, maladaptive behavior and psychological distress [18, 19].

Mindfulness is known for helping individuals to disengage from automatic thoughts, dysfunctional thinking, habits, and unhealthy behaviors patterns [11, 20]. Mindfulness has been defined as open receptivity to the present and includes acceptance of the current moment, focused attention and awareness, as well as the metacognitive ability to take distance from one’s cognitions [20]. Deci and Ryan [15] explained that automatic or ‘mindless’ behaviors resist changes, particularly when motives or processes remain out of awareness. And, according to Brown and Ryan [20], attention and awareness are crucial notions in their definition of mindfulness, considered both as an ‘inherent state of consciousness’ and as a disposition (trait) that vary both inter- and intra-individuals. Attention and awareness are particularly important to avoid rumination and thus to improve self-regulation, with individual differences in the frequency of mindful states over time being inversely related to rumination behavior [17, 20]. Lattimore, Mead, Irwin, Grice, Carson, and Malinowski [21] also provided evidence that supports the relevancy of interoceptive awareness, in the relationship between dispositional mindfulness and psychological problems associated with development and maintenance of ED. Additionally, Brown and Ryan [20] pointed to rumination as an example of blunting consciousness and its negative consequences for well-being; while mindfulness was related to positive well-being and less cognitive and emotional disturbance.

Several studies have examined the importance of mindfulness training and other metacognitive skills in relation to ED, showing the individual alternative ways of experiencing mental events, particularly taking distance from them, or dis-identification, that involves becoming aware of one’s mental functioning and taking distance from it, which is believed that dysfunctional coping strategies are less likely to occur [11, 17, 22, 23]. Elices et al [24] found that individuals with ED, compared to healthy controls, engage to a lesser extent in a direct experience while eating; but they did not include potential psychopathological confounders as rumination or coping in their analyses. Additionally, the study conducted by Masuda, Price, Anderson, and Wendell [25] showed the importance of how a person responds to events, and its role in the development and maintenance of negative psychological outcomes, supporting the premises of mindfulness-based interventions. Alberts et al., [11] suggested that mindfulness could help to reduce food cravings, body image concern, dichotomous thinking or avoidance-based coping strategies. For these reasons, mindfulness training could be a useful intervention to reduce rumination, and increase effective coping skills, though first, the role of trait mindfulness needs to be better understood.

Based on previous findings, we aimed to investigate in a clinical context if rumination, effective coping and mindfulness differ among young women with an ED diagnose and healthy controls. Furthermore, we tested the hypothesis that mindfulness, rumination and effective coping are significant and independent predictors of ED and significantly differ between those subsamples.

Materials and methods

This research has been evaluated and approved by the "Comité Ético de Investigación Clínica de Aragón" (CEICA, C.I. PI17/0364). Every individual entering the study received information about privacy data norms and regulations (Law 15/1999) [26]. In addition, this study fulfilled Spanish national regulations regarding patients’ autonomy and rights and biomedical research, including the possibility to revoke their consent and to drop out the study when any participant would desire (Law 41/2002 and Law 14/2007) [27, 28].

Finally, subjects were asked to sign two copies of the consent as they accept to provide information and complete the measures. Participants were provided with instructions before assessment, which included information about the instructions of both tests and how they must be completed.

Results and discussion

Our original sample was composed of 50 young women. A detailed review of every questionnaire was carried out prior to data analyses. One patient was excluded from final analyses because as the Field Study Record confirmed, she filled the questionnaires after a long evaluation at the ED Unit, so she was most probably cognitively exhausted. Information regarding scores on the scales refers to the whole sample, and indicate descriptive statistics for mindfulness (mean = 35.43; SD = 5.76), rumination (mean = 52.12; SD = 9.75) and effective coping (mean = 45.12; SD = 10.41). The study sample was composed of two subsamples:

In the clinical subsample the mean age was 16.6 (SD = 2.24); all were students of bachelor’s degree (24%) or secondary education (76%), and according to DSM-IV-TR criteria [31] they were diagnosed as having: 307.5 EDNOS (44%), 307.1 AN-R (52%) and 307.51 BN (4%).

Healthy controls showed a mean age of 19.08 years (SD = 0.64, range = 18–21). All of the participants are students of bachelor’s degree (92%) or secondary education (8%).

Comparative statistics and Student’s t-tests results and descriptive statistics are shown in Table 1. The descriptive analyses show the mean and the standard deviation of both groups regarding the three variables. Student’s t-tests were conducted to compare means of both groups across the three variables representing metacognitive processes: mindfulness ability, rumination and effective coping skills. Results showed that mindfulness and effective coping means were significantly lower in the clinical subsample (Student´s t = .015 and .007, respectively), while rumination mean was marginally higher in that group.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Comparison of means and independent samples t-test in mindfulness, rumination and effective coping between patients and healthy women groups.

https://doi.org/10.1371/journal.pone.0213985.t001

According to data shown in Table 1, the clinical subsample mean ( = 33.42) was statistically significantly lower in mindfulness, when compared to the healthy subsample ( = 37.36). Moreover, the clinical subsample mean in effective coping ( = 41.12) was also significantly lower when compared to the healthy controls ( = 48.95). In contrast, the average score in rumination of the clinical subsample was higher than healthy subsample, although this difference was marginally significant.

In addition, bivariate Binary Logistic Regression (LR) was conducted for ED risk estimation through variables based on the four predictor variables: age, mindfulness, rumination and effective coping.

Before a multivariate binary logistic regression study would be conducted, we tested if age fulfilled appropriate statistical assumptions, because OR was suspiciously high and age only ranged from 13 to 21. For instance, normality test (using Kolmogorov-Smirnov test and Shapiro-Wilk) were carried out for age, showing that this variable did not follow a normal distribution (p<0.001 in both normality tests). In addition, Levene’s test showed that age did not share the same variance in the two groups of ED (F = 44.125; p<0.001). Thus, we considered that age was skewing these specific results, so this variable was excluded from further LR analyses. Due to evident and model-guided theoretical reasons, we additionally hypothesized that the three factors mindfulness, rumination, and effective coping were high related to each other, most probably due to sharing a common cognitive construct. Thus, a Correlation analysis was conducted.

On one hand, results showed medium-high correlations between mindfulness and effective coping (r = -0.38), while on the other hand, those variables showed a low correlation with rumination (mindfulness: r = 0.10; effective coping: r = -0.01). To avoid confounding between these factors, we conducted multivariate analysis in two separate models in order to exclude both biases (age abnormal distribution and mindfulness and effective coping high correlation). Hierarchical multivariate LR analysis results are shown in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Hierarchical multivariate logistic regression of metacognitive skills by the presence of ED.

https://doi.org/10.1371/journal.pone.0213985.t002

As shown in Table 2, multivariate LR models showed that mindfulness and effective coping independently predicted the presence or absence of ED in the study sample. LR analyses indicate that every increased score in mindfulness (FMI has a maximum score of 30) increases 12% the risk of ED and 8% in the case of rumination.

Conclusions

To our knowledge, his is the first study that compares mindfulness, effective coping and rumination in ED female patients compared to healthy women. Our findings demonstrate that young patients diagnosed with ED show significantly lower levels of mindfulness (p = 0.015) and effective coping (p = 0.007) when compared to a group of healthy women. Specifically, our study supports the hypothesis that patients diagnosed with ED are more likely to engage in dysfunctional coping strategies, which has been previously associated with disordered eating behaviors [13, 16, 22]. Consistent with previous findings [11, 15, 17, 21], our study provides new data supporting the fact that individuals who exhibit higher levels of awareness and attention to one’s thoughts, may be less likely to engage in a disordered eating-related behavior, highlighting the role of mindfulness in self-regulatory and metacognitive processes. This position concurs with Brown and Ryan [20], who found that mindfulness, involving awareness of one’s behavior, was related to and predicted more positive well-being and less cognitive disturbance.

In contrast, our data do not reach strictly significant differences in rumination between patients with ED and healthy participants (p = 0.052). This data in relation to rumination doesn’t allow us to make certain conclusions about whether rumination is clearly related to ED or not. However, past research does support this association [13, 17, 18, 20]. This lack of association may be due to are small sample size, and larger studies are needed to confirm whether rumination plays a role.

Bearing in mind both statements, a mindfulness and effective coping focused interventional approach, would be potentially more effective in treating ED, than directing main intervention aims at rumination. In sum, mindfulness practice could be a helpful resource to create new mental landscape and behaviors to deal with different situations (emotions, reactions…) in these patients, bringing up alternative and more effective coping responses [11, 13, 18, 21, 22]. Furthermore, clinical research innovations are suggested by our study data, as including new questionnaires related to disordered eating-related behavior (as EAT-40 for example), in this sense, we could define better clinical and healthy subsamples. Likewise, clinical subsample could be differentiated by specific diagnosis (AN, BN, EDNOS) in further studies. Moreover, those studies should also include other questionnaires as the Spanish version of MAAS (Mindful Attention Awareness Scale; Soler et al) [34] or the FFMQ (Five Facets of Mindfulness Questionnaire; Cebolla et al) [35]; and finally, comparison with men subsamples could contribute with more data about self-regulatory processes in general [23] and its relationship with ED development, course, treatment and prognosis.

There were also limitations in the present study; firstly, questionnaires were completed in exams period (June). Considering that our sample were college students, factors related to this situation (anxiety, stress, fatigue…) could skew our results; thus, administrate questionnaires in another period of the academic year could be an adequate alternative. Secondly, there is a difference of 2.48 years between the mean age of each subsample (clinical and healthy subjects). Despite it is relatively frequent to see both adolescents and young women included together in ED-related articles [24], the potential bias produced by the age difference in our results has to be considered.

Another limitation of our study is related to the small sample size, but the study subject’s characteristics and inclusion/exclusion criteria are very specific (only young women aged 14-21years, clinical subsample with an ED diagnose), and recent and similar studies in which clinical ED patients are compared to healthy controls, such as Elices et al [24], also use similar sample sizes. Moreover, our study is based on a cross-sectional design, which implies that causal inferences cannot be made. Further studies with a higher sample size, comparable age groups in both subsamples and a prospective longitudinal approach, could permit test new hypotheses concerning the role of these factors over time. This would allow us to better understand the role of rumination include other protective or risk factors to develop problematic eating behavior or ED [17, 22].

Despite these limitations, our study contributes towards a promising emerging clinical perspective exploring the investigation of mindfulness in eating behavior domain [11]. In this line, recent studies have appeared, focusing on different but related characteristics as impulsivity and interoceptive awareness [21]; or ruminative thinking although by using only clinical or non-clinical samples [19, 18]. Moreover, our findings suggesting that mindfulness is associated to a lesser presence of an EDs diagnosis, point out that mindfulness intervention studies are to be developed in order to test the mid- to long-term efficacy of mindfulness training on ED outcomes [16, 19, 22, 23]. If those potential trials would show positive results, the quality of life and prognosis of individuals with disordered eating-related cognitions, behaviors or disorders, could be substantially improved.

Supporting information

Acknowledgments

Ciberobn is an initiate of the ISCIII.

JGC appreciates the kind support provided by the Prevention and Promotion of Health Network in Primary Care (REDIAPP) of the Carlos III Health Institute of the Ministry of Economy and Competitiveness of Spain (RD16/0007/0005).