Psychological and nutritional correlates of objectively assessed physical activity in patients with anorexia nervosa

Momentary physiological indices related to eating disorders: A systematic and methodological review

Eating disorders (ED) are serious psychiatric illnesses, with no everyday support to intervene on the high rates of relapse. Understanding physiological indices that can be measured by wearable sensor technologies may provide new momentary interventions for individuals with ED. This systematic review, searching large databases, synthesises studies investigating peripheral physiological (PP) indices commonly included in wearable wristbands (heart rate [HR], heart rate variability [HRV], electrodermal activity [EDA], peripheral skin temperature [PST], and acceleration) in ED. Inclusion criteria included: (a) full peer-reviewed empirical articles in English; (b) human participants with active ED; and (c) containing one of five wearable physiological measures. Kmet risk of bias was assessed. Ninety-four studies were included (Anorexia nervosa [AN; N = 4418], bulimia nervosa [BN; N = 916], binge eating disorder [BED; N = 1604], other specified feeding and eating disorders [OSFED; N = 424], and transdiagnostic [N = 47]). Participants with AN displayed lower HR and EDA and higher HRV compared to healthy individuals. Those with BN showed higher HRV, and lower EDA and PST compared to healthy individuals. Other ED and Transdiagnostic samples showed mixed results. PP differences are indicated through various assessments in ED, which may suggest diagnostic associations, although more studies are needed to validate observed patterns. Results suggest important therapeutic potential for PP in ED, and larger studies including diverse participants and diagnostic groups are needed to fully uncover their role in ED.

The Relationship between Patient Self-Reported, Pre-Morbid Physical Activity and Clinical Outcomes of Inpatient Treatment in Youth with Anorexia Nervosa: A Pilot Study

Links between premorbid physical activity (PA) and disease onset/course in patients with anorexia nervosa (AN) remain unclear. The aim was to assess self-reported PA as a predictor of change in percent median BMI (%mBMI) and length of hospital stay (LOS). Five PA domains were assessed via semi-structured interview in adolescents with AN at hospitalization: premorbid PA in school grades 1-6 (PA1-6); PA before AN onset (PA-pre) and after AN onset (PA-post); new, pathological motivation for PA (PA-new); and high intensity PA (PA-high). Eating disorder psychopathology was measured via the Eating Disorder Examination Questionnaire (EDE-Q), and current PA (steps/day) with accelerometry. PA1-6 was also assessed in healthy controls (HCs). Using stepwise backward regression models, predictors of %mBMI change and LOS were examined. Compared with 22 HCs (age = 14.7 ± 1.3 years, %mBMI = 102.4 ± 12.1), 25 patients with AN (age = 15.1 ± 1.7 years, %mBMI = 74.8 ± 6.0) reported significantly higher PA1-6 (median, AN = 115 [interquartile range IQR = 75;200] min vs. HC = 68 [IQR = 29;105] min; p = 0.017). PA-post was 244 ± 323% higher than PA-pre. PA1-6 was directly associated with PA-pre (p = 0.001) but not with PA-post (p = 0.179) or change in PA-pre to PA-post (p = 0.735). Lower %mBMI gain was predicted by lower baseline %mBMI (p = 0.001) and more PA-high (p = 0.004; r2 = 0.604). Longer LOS was predicted by higher PA-pre (p = 0.003, r2 = 0.368). Self-reported PA may identify a subgroup of youth with AN at risk of less weight gain and prolonged LOS during inpatient treatment for AN.

Perceived influence of wearable fitness trackers on eating disorder symptoms in a clinical transdiagnostic binge eating and restrictive eating sample

Wearable fitness trackers are an increasingly popular tool for measuring physical activity (PA) due their accuracy and momentary data collection abilities. Despite the benefits of using wearable fitness trackers, there is limited research in the eating disorder (ED) field using wearable fitness trackers to measure PA in the context of EDs. Wearable fitness trackers are often underused in ED research because there is limited knowledge about whether wearable fitness trackers negatively or positively impact PA engagement and ED symptoms in individuals with EDs. The current study aimed to assess the perceived impact wearable fitness trackers have on PA engagement and ED symptoms over a 12-week CBT treatment for 30 individuals with binge eating and restrictive eating that presented to treatment engaging or not engaging in maladaptive exercise. Participants in the maladaptive exercise group (n = 17) and non-maladaptive exercise group (n = 13) wore a fitness tracker for 12 weeks and completed questionnaires assessing participants' perceptions of the fitness trackers' influence on ED symptoms and PA engagement throughout treatment. Results demonstrated a small percentage of individuals perceived the fitness tracker influenced ED behaviors or PA engagement, and there were mixed results on whether participants positively or negatively perceived the fitness tracker influenced them to engage in ED behaviors or PA engagement. Although preliminary, these results demonstrate the need to continue using objective measurements of PA via wearable fitness trackers to further our understanding of the positive and negative effects of fitness trackers on clinical ED samples.Level of Evidence: Level 1, randomized controlled trial.

Sensor technology in eating disorders research: A systematic review

OBJECTIVE

Sensor technologies offer exciting potential to objectively measure psychopathological correlates of eating pathology and eating disorder (ED) research utilizing sensors has rapidly proliferated in the past several years. The aims of the present review are: (1) characterize the types of sensors that have been utilized in ED research, (2) identify the psychopathological factors relevant to EDs that have been assessed using sensors, (3) describe the data supporting the validity and reliability of these sensors, (4) discuss limitations associated with these sensors, and (5) identify gaps that persist within the ED literature with regard to use of sensor technologies.

METHOD

A systematic search was conducted of PubMed, PsycINFO, Web of Science, ProQuest, and "gray" literature sources. Eligible publications were empirical studies that utilized sensors to measure at least one psychological variable among clinical ED populations.

RESULTS

Sensors have been utilized with ED samples to measure eating behaviors, physical activity, sleep, autonomic nervous system activity, eyeblink startle response, visual attention, and visual-haptic object integration. The reliability and validity of these sensors varies widely and there are a number of significant gaps that remain in the literature with regard to the types of sensors utilized, context in which sensors have been used, and populations studied.

DISCUSSION

The existing literature utilizing sensors within ED research largely support the feasibility and acceptability of these tools. Sensors should continue to be utilized within the field, with a specific focus on examining the reliability and validity of these tools within ED samples and increasing the diversity of samples studied.

PUBLIC SIGNIFICANCE STATEMENT

Sensor technologies, such as those included in modern smartwatches, offer new opportunities to measure factors that may maintain or contribute to symptoms of eating disorders. This article describes the types of sensors that have been used in eating disorders research, challenges that may arise in using these sensors, and discusses new applications of these sensors that may be pursued in future research.

High levels of physical activity in female adolescents with anorexia nervosa: medical and psychopathological correlates

PURPOSE

While overexercise is commonly described in patients who experience anorexia nervosa (AN), it represents a condition still underestimated, especially in the paediatric population.

METHOD

The present study aims at assessing the possible associations between levels of physical activity (PA) and clinical features, endocrinological data and psychopathological traits in a sample of 244 female adolescents hospitalised for AN subdivided into two groups according to PA levels (high PA vs. no/low PA). The two groups were compared through multivariate analyses, while multiple regression analysis was conducted to determine whether physical activity predict specific outcomes.

RESULTS

No significant differences were found between the two groups in terms of last Body Mass Index (BMI) before illness, BMI at admission and disease duration, while a difference emerged in delta BMI(rapidity of weight loss), significantly higher in high-PA group (p = 0.021). Significant differences were observed in Free triiodothyronine- (p < 0.001), Free thyroxine (p = 0.046), Follicle-stimulating hormone (p = 0.019), Luteinising hormone (p = 0.002) levels, with values remarkably lower in high-PA group. Concerning psychopathological scales, the high-PA group showed worst Children's Global Assessment Scale (CGAS) scores (p = 0.035). Regression analyses revealed that higher PA predicts higher delta BMI (p = 0.021), presence of amenorrhea (p = 0.003), lower heart rate (p = 0.012), lower thyroid (Free triiodothyronine p < 0.001, Free thyroxine p = 0.029) and gynaecological hormones' levels (Follicle-stimulating hormone p = 0.023, Luteinising hormone p = 0.003, 17-Beta estradiol p = 0.041). Concerning psychiatric measures, HPA predicts worst scores at CGAS (p = 0.019), and at scales for evaluation of alexithymia (p = 0.028) and depression (p = 0.004).

CONCLUSIONS

Results suggest that high levels of physical activity in acute AN associate with worst clinical conditions at admission, especially in terms of endocrinological and medical features.

LEVEL OF EVIDENCE

Level III.