Caroline Barry's research while affiliated with Université Paris-Saclay and other places

Objective Previous studies have highlighted the relevance of perfectionism, self‐esteem, and anxio‐depressive symptoms in anorexia nervosa (AN). However, the relationships between these factors and cardinal eating disorders (ED) symptoms remain unclear, particularly in AN subtypes. This study aimed to examine their interconnections using network analysis. Method The sample included n = 338 inpatients with AN who completed the Eating Disorder Examination Questionnaire, Frost Multidimensional Perfectionism Scale, Rosenberg Self‐Esteem Scale, and Hospital Anxiety and Depression Scale. Using network analysis, we estimated three networks: full sample, AN‐restrictive (AN‐R) and AN‐binge/purging (AN‐BP) subtypes. We estimated central and bridge symptoms using expected influence and conducted an exploratory network comparison test to compare AN subtypes. Results Overvaluation of Weight and Shape, Concern over Mistakes, and Personal Standards were consistently central in all networks. The most central bridge symptoms across all networks were Concern over Mistakes and Self‐Esteem. Concern over Mistakes bridged perfectionism and ED symptoms, while Self‐Esteem was highly connected to all symptom clusters. Anxiety was significantly more central in the AN‐R network compared to the AN‐BP network. Conclusions The present study contributes to a growing body of network studies suggesting that nodes related to perfectionism are just as central as cardinal ED symptoms, indicating the relevance of perfectionism in ED pathology. The high bridge centrality of self‐esteem suggests that it may be an important link between perfectionism, mood, and ED symptoms. Future research should investigate the efficacy of targeting multiple psychological factors in the treatment of AN, as well as their potential transdiagnostic relevance.

Background International guidelines often state that general practitioners (GPs) provide early management for most patients with eating disorders (EDs). GP management of EDs has not been studied in France. Depressive disorders are often a comorbidity of EDs. The aims of this study were to describe in France the characteristics of people with all subcategories of EDs (Anorexia Nervosa, Bulimia Nervosa, ED Not Otherwise Specified) managed by their GPs and to study the management temporality between depression and all subcategories of EDs. Methods Retrospective cohort study of patients with EDs visiting French GPs. Data collected from 1994 through 2009 were extracted from the French society of general electronic health record. A descriptive analysis of the population focused on depression, medication such as antidepressants and anxiolytics, and the management temporality between depression and EDs. Results 1310 patients aged 8 years or older were seen at least once for an ED by a GP participating in the database out of 355,848 patients, with a prevalence rate of 0.3%. They had a mean age of 35.19 years, 82.67% were women. 41.6% had anorexia nervosa, 26.4% bulimia nervosa, and 32% an ED not otherwise specified. Overall, 32.3% had been managed at least once for depression, and 18.4% had been prescribed an antidepressant of any type at least once. Benzodiazepines had been prescribed at least once for 73.9% of the patients treated for depression. Patients with an ED seen regularly by their GP (“during” profile) received care for depression more frequently than those with other profiles. 60.9% had a single visit with the participating GP for their ED Treatment and management for depression did not precede care for EDs. Conclusions Data extracted from the French society of general practice were the only one available in France in primary care about EDs and our study was the only one on this topic. The frequency of visits for EDs was very low in our general practice-based sample. Depressive disorders were a frequent comorbidity of EDs. GPs could manage common early signs of depression and EDs, especially if they improved their communication skills and developed collaborative professional management.

Objective: Previous studies have highlighted the relevance of perfectionism, self-esteem and anxio-depressive symptoms in anorexia nervosa (AN). However, the relationships between these factors and cardinal ED symptoms are still unclear, particularly in AN subtypes. The aim of this study was to visualize their interconnections in a sample of inpatients with AN using network analysis.Method: Participants were n = 338 inpatients seeking treatment for AN. Each patient completed the Eating Disorder Examination Questionnaire, Frost Multidimensional Perfectionism Scale, Rosenberg Self-Esteem Scale, and Hospital Anxiety and Depression Scale. Using network analysis, we visualized three networks: full sample, AN-restrictive (AN-R) subtype and AN-binge/purge (AN-BP) subtype. We estimated central and bridge symptoms using expected influence values. An exploratory network comparison test was used to compare AN subtypes. Results:Overvaluation of Weight and Shape, Concern over Mistakes and Personal Standards were consistently central in all networks. The most central bridge symptoms across all networks were Concern over Mistakes and Self-Esteem. Concern over Mistakes bridged perfectionism and ED symptoms, while Self-Esteem was highly connected to all symptom clusters. Anxiety was significantly more central in the AN-R network compared to the AN-BP network.Conclusions:The present study contributes to a growing body of network studies demonstrating that nodes related to perfectionism are just as central as cardinal ED symptoms, indicating the relevance of perfectionism in ED pathology. Future research should investigate the efficacy of targeting multiple psychological factors, such as perfectionism, anxiety and self-esteem, in the treatment of AN, as well as their potential transdiagnostic relevance.

This paper proposes a proof of concept of using natural language processing techniques to categorize valence of family relationships described in free texts written by french teenagers. The proposed study traces the evolution of techniques for word embedding. After decomposing the different texts in our possession into short texts composed of sentences and manual labeling, we tested different word embedding scenarios to train a multi-label classification model where a text can take several labels: labels describing the family link between the teenager and the person mentioned in the text and labels describing the teenager's relationship with them positive/negative/neutral valence). The natural baseline for word vector representation of our texts is to build a TF-IDF and train classical classifiers (Elasticnet logistic regression, gradient boosting, random forest, support vector classifier) after selecting a model by cross validation in each class of machine learning models. We then studied the strengths of word-vectors embeddings by an advanced language representation technique via the CamemBERT transformer model, and, again, used them with classical classifiers to compare their respective performances. The last scenario consisted in augmenting the CamemBERT with output dense layers (perceptron) representing a classifier adapted to the multi-label classification and fine-tuning the CamemBERT original layers. The optimal fine-tuning depth that achieves a bias-variance trade-off was obtained by a cross-validation procedure. The results of the comparison of the three scenarios on a test dataset show a clear improvement of the classification performances of the scenario with fine-tuning beyond the baseline and of a simple vectorization using CamemBERT without fine-tuning. Despite the moderate size of the dataset and the input texts, fine-tuning to an optimal depth remains the best solution to build a classifier.

The purpose of this study is to clarify the relationship between having a chronic condition (CC) and several types of risky sexual behaviour (RSB) among adolescents and young adults. We used data from a multicentre cross-sectional study carried out on 14,431 adolescents from 137 French schools. Logistic regression was used to assess the association between several types of RSB and CCs among the 2680 participants aged 17 years or over who reported sexual intercourse. Survival analysis was conducted to assess the association between CCs and age at first sexual intercourse across the whole sample. Analyses were conducted separately by gender with and without adjustment for the parents’ education level, early menarche and subjective wellbeing (relationship with mother and father, depression, perceived health status and liking school). Among boys, having a CC was associated with a higher risk of RSB in both univariate (OR: 1.58 [95% CI: 1.10–2.27]) and multivariate analyses (aOR: 1.62 [95% CI: 1.11–2.38]). Among girls, the association between chronic condition and RSB in univariate analysis was non-significant (OR: 1.30 [95% CI: 0.97–1.76]) and disappeared after adjustment on subjective wellbeing (aOR: 1.08 [95% CI: 0.78–1.49]). There was no association between CC and age at first sexual intercourse. Conclusion: There were major gender differences. Boys with a CC were more prone to engage in RSB independent of their subjective wellbeing, whereas in girls, subjective wellbeing seemed to mediate the relationship between CC and RSB. Clinicians should be aware of those gender differences in order to deliver preventive strategies regarding sexuality that target both genders. What is Known: • Young people with chronic conditions have a higher likelihood of engaging in risky sexual behaviour. • Engaging in such behaviours can be much more costly, as it can weaken their underlying state of health. What is New: • We found major gender differences. Boys were more prone to engage in risky behaviour independent of their subjective wellbeing, whereas in girls, it seemed to play an important role. • By understanding how risky sexual behaviour differs according to gender, clinicians can deliver prevention messages that target both genders.

Objective To explore the impact of methodological choices on the results of meta-analyses (MAs), with acupuncture for smoking cessation as a case study. Study Design After performing an umbrella review (using MEDLINE, the COCHRANE Library, the Wan Fang database, and the Chinese Journal Full-text Database/March 2018) of MAs exploring the use of acupuncture for smoking cessation, we extracted all randomized controlled trials (RCTs). Numerous MAs were performed according to every possible combination of various methodological choices (e.g., characteristics of the intervention and control procedures, outcome, publication status, language) to assess their vibration of effects (VoE); or more precisely the existence of a Janus effect, i.e., whether the 10th and 90th percentiles in the distribution of effect sizes were in opposite directions. Results After including 7 MAs and 39 RCTs, we performed 496,528 MAs. The effect size was negative at the 10st percentile (− 0.1, favoring controls) and positive at the 90th percentile (1.17, favoring acupuncture). In all, 104,491 MAs showed a statistically significant difference in favor of acupuncture, while 392,037 failed to demonstrate the efficacy of acupuncture (including 96 that showed a statistically significant difference in favor of the control). Conclusions The methodological choices made in performing pairwise meta-analyses can result in substantial VoE, occasionally leading to opposite results. Study registration https://osf.io/a7ncu/

Migrant adolescents in therapy for psychological problems are at risk of poor attendance or even of dropping out. Transcultural psychotherapy has been developed in France to take cultural diversity into account in psychological treatment and to deal with the specific difficulties encountered in the psychotherapeutic treatment of this population. This study aims to assess adolescents’ attendance rates to this form of psychotherapy and to explore the association of these rates with demographic, cultural, and clinical variables. We conducted a retrospective clinical cohort study of 148 adolescents aged from 11 to 20 years treated between 2008 and 2018 at two transcultural psychotherapy centers in Paris. Statistical analyses tested demographic, cultural, and clinical hypotheses. The main result was the high attendance rate at transcultural psychotherapy sessions among adolescents (77.8%). Attendance rates were not associated with age, gender, family size, generation of migration, or cultural area of origin, but were significantly linked to support in therapy, specifically, the presence at the first transcultural psychotherapy session of the first-line therapist, an interpreter, or both. Transcultural psychotherapy appears to be an effective method for addressing the complex symptoms experienced by migrant adolescents. Better attendance at sessions is statistically significantly associated with factors favoring a therapeutic alliance, specifically, the presence of the first-line therapist or an interpreter in TPT sessions and the existence of support from a social worker. The holistic approach of transcultural psychotherapy to adolescent care may explain the high attendance rates observed.

Objective: To determine if adolescents and adults diagnosed with anorexia nervosa differ in their levels of cognitive flexibility and attention to detail independently of potential confounds. Method: Sixty-two adolescents and 54 adults were assessed while receiving inpatient treatment and completed the following self-reports: Eating Disorders Results: Comparisons of the adolescents and adults with anorexia nervosa revealed no significant differences for any of the neuropsychological test scores even after adjusting for potential confounding factors. Neither cognitive flexibility nor attention to detail were associated with level of eating disorder symptomatology, depression, anxiety or obsessive-compulsive symptom-atology. Unlike age, illness duration was found weakly associated with perseverative errors Wisconsin Card Sorting Test and with the central coherence index of the Rey Complex Figure recall condition. Conclusions: Set-shifting and central coherence performance were independent of age, clinical symptoms severity and emotional status. Additional studies on the relationship between the duration of anorexia nervosa and neuropsychological difficulties are needed.