A proof-of-concept study applying machine learning methods to putative risk factors for eating disorders: results from the multi-centre European project on healthy eating.

BACKGROUND: Despite a wide range of proposed risk factors and theoretical models, prediction of eating disorder (ED) onset remains poor. This study undertook the first comparison of two machine learning (ML) approaches [penalised logistic regression (LASSO), and prediction rule ensembles (PREs)] to conventional logistic regression (LR) models to enhance prediction of ED onset and differential ED diagnoses from a range of putative risk factors. METHOD: Data were part of a European Project and comprised 1402 participants, 642 ED patients [52% with anorexia nervosa (AN) and 40% with bulimia nervosa (BN)] and 760 controls. The Cross-Cultural Risk Factor Questionnaire, which assesses retrospectively a range of sociocultural and psychological ED risk factors occurring before the age of 12 years (46 predictors in total), was used. RESULTS: All three statistical approaches had satisfactory model accuracy, with an average area under the curve (AUC) of 86% for predicting ED onset and 70% for predicting AN v. BN. Predictive performance was greatest for the two regression methods (LR and LASSO), although the PRE technique relied on fewer predictors with comparable accuracy. The individual risk factors differed depending on the outcome classification (EDs v. non-EDs and AN v. BN). CONCLUSIONS: Even though the conventional LR performed comparably to the ML approaches in terms of predictive accuracy, the ML methods produced more parsimonious predictive models. ML approaches offer a viable way to modify screening practices for ED risk that balance accuracy against participant burden.

[Child and adolescent psychiatric care 2019 in Austria-steps of care, current state and lookout]. / Kinder- und jugendpsychiatrische Versorgung 2019 in Österreich ­ Stufen der Versorgung, Ist-Stand und Ausblick.

BACKGROUND: Child and adolescent psychiatric disorders are prevalent, their care in Austria is still in progress. METHODS: We investigated whether the child and adolescent psychiatric care made progress dsince the establishment of the medical special discipline. RESULTS: There remains massive heterogeneity between the Austrian provinces. The "Mangelfachregelung" in ist current version is insufficient in order to consolidate our medical special discipline and to guarantee full carewithin the next years. CONCLUSION: Hints are given how the situation of care in our discipline can be strengthened.

A genome-wide association study of anorexia nervosa.

Anorexia nervosa (AN) is a complex and heritable eating disorder characterized by dangerously low body weight. Neither candidate gene studies nor an initial genome-wide association study (GWAS) have yielded significant and replicated results. We performed a GWAS in 2907 cases with AN from 14 countries (15 sites) and 14 860 ancestrally matched controls as part of the Genetic Consortium for AN (GCAN) and the Wellcome Trust Case Control Consortium 3 (WTCCC3). Individual association analyses were conducted in each stratum and meta-analyzed across all 15 discovery data sets. Seventy-six (72 independent) single nucleotide polymorphisms were taken forward for in silico (two data sets) or de novo (13 data sets) replication genotyping in 2677 independent AN cases and 8629 European ancestry controls along with 458 AN cases and 421 controls from Japan. The final global meta-analysis across discovery and replication data sets comprised 5551 AN cases and 21 080 controls. AN subtype analyses (1606 AN restricting; 1445 AN binge-purge) were performed. No findings reached genome-wide significance. Two intronic variants were suggestively associated: rs9839776 (P=3.01 × 10(-7)) in SOX2OT and rs17030795 (P=5.84 × 10(-6)) in PPP3CA. Two additional signals were specific to Europeans: rs1523921 (P=5.76 × 10(-)(6)) between CUL3 and FAM124B and rs1886797 (P=8.05 × 10(-)(6)) near SPATA13. Comparing discovery with replication results, 76% of the effects were in the same direction, an observation highly unlikely to be due to chance (P=4 × 10(-6)), strongly suggesting that true findings exist but our sample, the largest yet reported, was underpowered for their detection. The accrual of large genotyped AN case-control samples should be an immediate priority for the field.

Eating-related environmental factors in underweight eating disorders and obesity: are there common vulnerabilities during childhood and early adolescence?

OBJECTIVE: This study aimed to examine whether there is an association between individual, social and family influences and dysfunctional eating patterns early in life and the likelihood of developing a subsequent underweight eating disorder (ED) or obesity. METHOD: The total sample comprised 152 individuals (underweight ED, n = 45; obese patients, n = 65; healthy controls; n = 42) from Barcelona, Spain. The Cross-Cultural Questionnaire (CCQ) was used to assess early eating influences as well as individual and family eating patterns and attitudes towards food. RESULTS: Even though a few shared eating influences emerged for both groups, unique factors were also observed. Whereas relationship with friends, teasing about eating habits by family members and the mass media were of specific relevance to the underweight ED group, the patient's own physical appearance, body dissatisfaction, teasing about eating habits by friends, teasing about body shape by family members and dysfunctional eating patterns were unique to obesity. CONCLUSIONS: Overlapping environmental risk factors provide evidence for integral prevention and intervention approaches that simultaneously tackle a range of weight-related problems. The unique factors might be important for targeting high-risk individuals.

Family-based association analysis of functional VNTR polymorphisms in the dopamine transporter gene in migraine with and without aura.

Because of the role of dopamine in triggering migraine attacks, genes of the dopamine system are candidates for involvement in migraine. We examined three VNTR polymorphisms in the dopamine transporter, the 5'UTR VNTR, the intron 8 VNTR and the intron 14 VNTR, in a sample of 205 family trios. We used the transmission disequilibirium test (TDT) to examine the transmission of these three markers and their haplotypes to offspring affected by migraine. We found no significant transmission distortion of any marker. Likewise haplotypes of the three markers did not show significant overall or individual association with migraine. Finally we examined migraine with and without aura, and likewise found no association between dopamine transporter VNTRs or their haplotypes and either classification of the disease. We conclude that functional genetic variation in the dopamine transporter does not act as a significant risk factor for migraine.

Family-based analysis of serotonin transporter gene polymorphisms in migraine with and without aura.

Genetic epidemiological twin studies have demonstrated a significant heritability for migraine, with > 60% of liability to migraine either with or without aura coming from additive genetic factors. Because of the essential role of serotonin in the pathophysiology and treatment of migraine, genes of the serotonin system are candidates for involvement in migraine. Consequently, we examined two functional VNTR polymorphisms in the serotonin transporter gene, the 5-HTTLPR and the intron 2 VNTR, in a sample of 212 family trios each with a proband with childhood migraine, 153 with migraine without aura (MoA) and 59 with migraine with aura (MA). For the first time, we used transmission disequilibrium test analysis with the program TDTPHASE to examine the transmission of these two markers and their haplotypes to offspring affected by migraine. We found no significant transmission distortion of any marker, with the common L allele of the 5-HTTLPR transmitted 170 times and not transmitted 178 times, and the S allele 130 vs. 122 times. Likewise, the common 12 allele of the intron 2 VNTR was transmitted 201 times and not transmitted 188 times, and the 10 allele 107 vs. 120 times. The markers were not associated with MoA and MA and none of the haplotypes was associated with overall migraine, MoA or MA. The 5-HTTLPR and the intron 2 VNTRs do not play a major role in susceptibility to migraine.