"Can you see me?" Videoconferencing and eating disorder risk during COVID-19: Anxiety, impairment, and mediators.

OBJECTIVE: The use of videoconferencing has increased during the pandemic, creating prolonged exposure to self-image. This research aimed to investigate whether eating disorder (ED) risk was associated with videoconferencing performance for work or study and to explore whether the use of safety behaviors and self-focused attention mediated the relationship between ED risk and perceived control over performance anxiety, impaired engagement, or avoidance of videoconferencing for work or study. METHOD: In 2020, an online survey was distributed within Australia to those aged over 18 years via academic and social networks, measuring: use of videoconferencing for work/study, demographics, ED risk, safety behaviors for appearance concerns, self-focused attention, perceived control over performance anxiety, perceived engagement impairment, and avoidance of videoconferencing. A total of 640 participants (77.3% female, Mage  = 26.2 years) returned complete data and were included in analyses. RESULTS: 245 participants (38.7%) were considered at-risk for EDs (SCOFF > 2). Those at-risk reported significantly more safety behaviors, self-focused attention, impaired engagement, and avoidance, plus lower perceived control over performance anxiety than those not at-risk. Multiple mediation models found the effects of ED risk on control over performance anxiety, impaired engagement, and avoidance were partially mediated by safety behaviors and self-focused attention. DISCUSSION: Our cross-sectional findings suggest videoconferencing for work/study-related purposes is associated with performance anxiety, impaired engagement, and avoidance among individuals at-risk for EDs. Poorer videoconferencing outcomes appear more strongly related to social anxiety variables than ED status. Clinicians and educators may need to provide extra support for those using videoconferencing. PUBLIC SIGNIFICANCE: Because videoconferencing often involves seeing your own image (via self-view) we wondered whether the appearance concerns experienced by those with eating disorders (EDs) might interfere with the ability to focus on or to contribute to work/study videoconferencing meetings. We found that although those with EDs experience more impairments in their videoconferencing engagement/contribution, these were linked just as strongly to social anxiety as they were to appearance concerns.

The meditating role of sleep in the longitudinal associations between peer victimization and internalizing symptoms: A cross-lagged panel analysis.

Adolescence is a time of heightened vulnerability for both peer victimization (PV) and internalizing symptoms. While the positive association between them is well established, there is little understanding of the mechanisms underpinning this relationship. To address this gap, the current study aimed to investigate sleep hygiene and school night sleep duration as individual and sequential mediators of the relationship between PV and both depressive and social anxiety symptoms during pre- to mid-adolescence. The study drew upon a community sample of 528 Australian youth aged 10-12 years at baseline (M age = 11.19, SD = .55; 51.1% boys) and data were collected over five annual measurement occasions. Direct and indirect longitudinal and bidirectional associations were examined using cross-lagged panel analysis. There was no evidence of sequential mediation through both sleep hygiene and sleep duration to depression and social anxiety. Instead, the findings show that sleep hygiene mediated the prospective association between PV and both depressive and social anxiety symptoms, and between PV and sleep duration. Overall, sleep hygiene represents a modifiable transdiagnostic factor that can be targeted to break the cycle of PV, inadequate sleep, and internalizing symptoms.

Testing the Identity Disruption Model among Adolescents: Pathways Connecting Adverse Childhood Experiences to Body Dissatisfaction.

The Identity Disruption Model posits that early adversity is associated with lower self-concept clarity, which in turn increases vulnerability to sociocultural appearance factors and body dissatisfaction, but this model has not previously been tested among adolescents. Testing the model during adolescence is critical because this is a key point of development of both identity and body dissatisfaction. This paper presents two studies with adolescents recruited through social media (Study 1: n = 213; 78% female; mean age = 15.7 years, SD = 1.14) and from high schools (Study 2; n = 228; 43% female; mean age = 13.8 years, SD = 1.15). In both studies, self-reported early adversity was associated with lower self-concept clarity; lower self-concept clarity was associated with greater internalization of appearance ideals and more frequent appearance comparisons; and internalization and appearance comparisons were associated with greater body dissatisfaction. This research builds on previous sociocultural models of body dissatisfaction by pointing to processes that occur early in life that could be potential targets of intervention and prevention efforts.

Examining the Prospective Bidirectional Associations between Subjective and Objective Attractiveness and Adolescent Internalizing Symptoms and Life Satisfaction.

Research has consistently shown that more physically attractive individuals are perceived by others to be happier and better psychologically adjusted than those perceived as less attractive. However, due to the lack of longitudinal research in adolescents, it is still unclear whether poor mental health predicts or is predicted by either objective or subjective attractiveness during this critical developmental period. The purpose of the current study was to examine prospective bidirectional associations between both subjective and objective ratings of attractiveness, life satisfaction and symptoms of social anxiety, depression and eating disorders (i.e., internalizing symptoms) from early to mid-adolescence. Participants (T1: N = 528, 49.9% girls; Mage = 11.19; SD = 0.55) were followed annually over four time points. The cross-lagged panel model results revealed evidence of prospective associations between both forms of attractiveness and life satisfaction and internalizing symptoms, which were driven more by changes in the mental health outcomes than by changes in the subjective and objective attractiveness ratings. The results also indicated that the pattern, strength, and direction of the associations tested were robust across boys and girls, and white and non-white ethnic groups. Overall, the findings suggest that it is important to find effective ways of educating adolescents who are unhappy with their appearance that making changes to improve their mental health, rather than focusing on their physical appearance, will have benefits not only for how they perceive themselves but also for how they are perceived by others.

Zoomers: videoconferencing, appearance concerns, and potential effects on adolescents.

PURPOSE OF REVIEW: The cessation of in-person teaching to reduce the spread of COVID-19 dramatically increased the use of videoconferencing for home learning among adolescents. RECENT FINDINGS: A consistent finding across studies assessing the relationship between videoconferencing and appearance concerns was that time spent focused on self-view was related to greater appearance concerns. Videoconferencing was associated with an increase in desire for cosmetic surgery and other nonsurgical treatments. Among those at risk of eating disorders, videoconferencing was associated with more appearance-management behaviours and less engagement in the video call. Research to date has been correlational and predominantly involved White, adult women in the USA and in Australia, leaving important gaps in our knowledge, especially around impacts on adolescents. SUMMARY: Given their vulnerability to appearance concerns, we encourage practitioners to speak to adolescents about their use of videoconferencing and how they can migitate the potential negative impacts on body image.

The Effect of Spirituality on Mood: Mediation by Self-Esteem, Social Support, and Meaning in Life.

Negative correlations between religiosity or spirituality and depression symptoms have been frequently reported, but relatively few empirical studies have investigated the processes that mediate the relationships. This study investigated four theorized mediators in a single model to assess the unique contributions of self-esteem, social support, meaning in life, and positive religious coping to the relationship between religiosity, spirituality, and two markers of depression, positive affect and negative affect. Path analysis was employed to investigate multiple mediation models in a sample of 352 undergraduates. Non-significant paths were removed, and a second independent sample of 316 undergraduates was used to validate the trimmed models. Results indicated that self-esteem mediated the relationships between spirituality and both positive and negative affect. Meaning in life also mediated the relationship between spirituality and positive affect, while social support mediated the relationship between spirituality and negative affect. Positive religious coping did not mediate either relationship, possibly because the samples were not drawn from populations under stress. The relationship between intrinsic religiosity and positive affect was similarly mediated by self-esteem and meaning in life. Religiosity and negative affect were related through the mediators self-esteem and an unexpectedly adverse factor captured by positive religious coping in the models used.

The role of parenting behaviors in the bidirectional and intergenerational transmission of depression and anxiety between parents and early adolescent youth.

BACKGROUND: Parenting is a modifiable factor proposed to underpin the transmission of anxiety and depression from parents to children. This study examined the role of parenting in the intergenerational transmission of anxiety and depression across pre- and early adolescence. METHOD: Participants were 531 youth (Mage = 11.18, SD = 0.56; 50.85% boys) and their parent. Child and parent anxiety and depression, and parental rejection, warmth and overprotection were assessed annually over 3 years. Bidirectional relationships between parent and child anxiety and depression, and the mediating role of parenting behaviors, were examined using cross-lagged panel models. RESULTS: Results suggest bidirectional associations over time between parent and child depression, and parental rejection and child depression. Parental rejection and low warmth were associated with increases in child depression, but did not mediate depression transmission. Parental anxiety was associated with increases in child anxiety and depression, but there was no bidirectional association from child psychopathology to parental anxiety. There was little evidence that parenting predicted changes in child anxiety over time. Child anxiety and depression were associated with subsequent increases in parental depression. CONCLUSION: Parental depression, rejection and low warmth are independent risk factors for child depression. Parental rejection may also be a consequence of parenting a depressed youth. Parenting did not account for the apparent transmission of parental anxiety to increased child anxiety and depression. Child psychopathology increases risk of parental depression. Parental rejection may be an important modifiable risk factor for youth depression in early adolescence, and may also reduce later risk of parent depression.

Risk and Protective Factors for Prospective Changes in Adolescent Mental Health during the COVID-19 Pandemic.

The restrictions put in place to contain the COVID-19 virus have led to widespread social isolation, impacting mental health worldwide. These restrictions may be particularly difficult for adolescents, who rely heavily on their peer connections for emotional support. However, there has been no longitudinal research examining the psychological impact of the COVID-19 pandemic among adolescents. This study addresses this gap by investigating the impact of the COVID-19 pandemic on adolescents' mental health, and moderators of change, as well as assessing the factors perceived as causing the most distress. Two hundred and forty eight adolescents (Mage = 14.4; 51% girls; 81.8% Caucasian) were surveyed over two time points; in the 12 months leading up to the COVID-19 outbreak (T1), and again two months following the implementation of government restrictions and online learning (T2). Online surveys assessed depressive symptoms, anxiety, and life satisfaction at T1 and T2, and participants' schooling, peer and family relationships, social connection, media exposure, COVID-19 related stress, and adherence to government stay-at-home directives at T2 only. In line with predictions, adolescents experienced significant increases in depressive symptoms and anxiety, and a significant decrease in life satisfaction from T1 to T2, which was particularly pronounced among girls. Moderation analyses revealed that COVID-19 related worries, online learning difficulties, and increased conflict with parents predicted increases in mental health problems from T1 to T2, whereas adherence to stay-at-home orders and feeling socially connected during the COVID-19 lockdown protected against poor mental health. This study provides initial longitudinal evidence for the decline of adolescent's mental health during the COVID-19 pandemic. The results suggest that adolescents are more concerned about the government restrictions designed to contain the spread of the virus, than the virus itself, and that those concerns are associated with increased anxiety and depressive symptoms, and decreased life satisfaction.

Longitudinal Associations between Coping Strategies and Psychopathology in Pre-adolescence.

Much of the literature investigating the association between coping and psychopathology is cross-sectional, or associations have been investigated in a unidirectional manner; hence, bidirectionality between coping and psychopathology remains largely untested. To address this gap, this study investigated bidirectional relations between coping and psychopathology during pre-adolescence. Participants (N = 532, 51% male) and their primary caregiver both completed questionnaires assessing pre-adolescents' coping (i.e., avoidant, problem solving, social support seeking) and symptoms of psychopathology (i.e., generalized anxiety, social anxiety, depression, eating pathology) in Wave 1 (Mage = 11.18 years, SD = 0.56, range = 10-12) and Wave 2 (Mage = 12.18 years, SD = 0.53, range = 11-13, 52% male), one year later. Cross-lagged panel models showed child-reported avoidant coping predicted increases in symptoms of generalized and social anxiety, and eating pathology. In separate child and parent models, symptoms of depression predicted increases in avoidant coping. Greater parent-reported child depressive symptoms also predicted decreases in problem solving coping. Taken together, results suggest unique longitudinal associations between coping and psychopathology in pre-adolescence, with avoidant coping preceding increases in symptoms of anxiety and eating pathology, and depressive symptoms predicting later increases in maladaptive coping.

Protect me from my selfie: Examining the association between photo-based social media behaviors and self-reported eating disorders in adolescence.

OBJECTIVE: This study examined whether social media behaviors were associated with higher odds of meeting criteria for an eating disorder and whether gender moderated these relationships. METHOD: Australian adolescents (N = 4,209; 53.15% girls) completed the self-report photo investment and manipulation scales. Additional self-report items assessed avoidance of posting selfies and investment in others' selfies. Eating disorders were identified by the Eating Disorder Examination Questionnaire and other self-report measures aligned with diagnostic and statistical manual for mental disorders-5 criteria. RESULTS: A multinomial logistic regression examining the effect of avoidance of posting selfies, photo investment, photo manipulation, and investment in others' selfies on the likelihood of meeting criteria for an eating disorder, compared to no disorder, was significant (χ2 [42] = 1,128.93, p < .001). Greater avoidance was associated with higher odds of meeting criteria for all disorders except clinical/subclinical binge-eating disorder and purging disorder. Increased photo investment was related to greater odds of meeting criteria for all disorders. A similar relationship emerged for photo manipulation, with the exception of clinical/subclinical binge-eating disorder, and unspecified feeding and eating disorder. Investment in others' selfies was associated with higher odds of meeting criteria for all disorders except clinical/subclinical anorexia nervosa and purging disorder. There was a significant interaction between gender and avoidance (χ2 [1] = 5.23, p = .025, OR = 1.74), whereby boys were more likely to meet criteria for clinical/subclinical anorexia nervosa in the context of greater avoidance of posting selfies. DISCUSSION: Appearance-related social media behaviors may be indicative of eating disorder risk. Implications for clinicians and advancements for social media screening tools are discussed.

The use of social media by Australian preadolescents and its links with mental health.

OBJECTIVES: Preadolescent social media use is normative and could influence mental health. This study investigated: (a) Differences between preadolescent users and non-users of various social media platforms on mental health, (b) unique links between time spent on those platforms, appearance-based activities on social media, and mental health, and (c) the moderating role of biological sex on those relationships. METHOD: Preadolescent youth (N = 528; 50.9% male) completed online surveys. RESULTS: Users of YouTube, Instagram, and Snapchat reported more body image concerns and eating pathology than non-users, but did not differ on depressive symptoms or social anxiety. Appearance investment uniquely predicted depressive symptoms. Appearance comparisons uniquely predicted all aspects of mental health, with some associations stronger for females than males. CONCLUSIONS: Preadolescents could be encouraged to reduce their opportunities to make appearance comparisons and to invest less in their appearance on social media. Preadolescents may benefit from social media intervention programs.

The Preteen Perfectionist: An Evaluation of the Perfectionism Social Disconnection Model.

This research investigated associations between socially prescribed and self-orientated perfectionism, and the social functioning of 510 preteens (Mage = 11.2). The study focused on predictions from the Perfectionism Social Disconnection Model (PSDM) by determining whether rejection sensitivity and social isolation, in that sequence, mediated the associations between both perfectionism types and mental health outcomes. Employing both survey and experimental methods, findings indicated that both types of perfectionism in preadolescence were associated with increased interpersonal difficulty, rejection sensitivity and feelings of social isolation, as well as higher levels of eating disorder symptoms, depression, and anxiety. Results from serial mediation analyses found general support for the theoretical predictions of the PSDM for socially prescribed perfectionism, and extending upon previous research, for self-orientated perfectionism. The age of the sample suggests that both forms of perfectionism may be important targets in programs to prevent the development of mental health problems.

Parental Control of the Time Preadolescents Spend on Social Media: Links with Preadolescents' Social Media Appearance Comparisons and Mental Health.

Time spent on social media and making online comparisons with others may influence users' mental health. This study examined links between parental control over the time their child spends on social media, preadolescents' time spent browsing social media, preadolescents' appearance comparisons on social media, and preadolescents' appearance satisfaction, depressive symptoms, and life satisfaction. Preadolescent social media users (N = 284, 49.1% female; aged 10-12) and one of their parents completed online surveys. Preadolescents, whose parents reported greater control over their child's time on social media, reported better mental health. This relationship was mediated by preadolescents spending less time browsing and making fewer appearance comparisons on social media. Parental control over time spent on social media may be associated with benefits for mental health among preadolescents.

Does medium matter? Investigating the impact of viewing ideal image or short-form video content on young women's body image, mood, and self-objectification.

There is a rising prevalence of short-form videos on social media, particularly since the advent of TikTok. Viewing appearance-ideal images has harmful effects on young women's body image. However, the impacts of viewing appearance-ideal short-form videos on body image are largely unknown. This study investigated the impact of viewing appearance-ideal short-form social media video content on young women's (Mage = 19.19, SD = 1.80) state appearance satisfaction, negative mood, self-objectification, and related constructs, compared to viewing appearance-ideal image content and appearance-neutral content. Young women (N = 211) were shown either: (1) appearance-ideal images, (2) appearance-ideal videos, (3) appearance-neutral images, or (4) appearance-neutral videos. Viewing appearance-ideal content regardless of the medium led to decreased appearance satisfaction, and increased negative mood, and self-objectification, and more state internalisation of appearance ideals compared to viewing appearance-neutral content. Further, if women perceived the appearance-ideal content they viewed to be unedited or unenhanced, they reported less appearance satisfaction after viewing video than image content. Thus, the impact of viewing ideal video and image content taken from social media may have similar effects on young women. However, when ideal content is low in perceived enhancement, viewing videos may be more harmful for appearance satisfaction than viewing images.

Digital social evaluation: Relationships between receiving likes, comments, and follows on social media and adolescents' body image concerns.

A defining feature of social media is its social interactivity, but limited research has investigated how various aspects of digital social evaluation (i.e., likes, follows, comments) may be related with users' body image concerns. This study investigated the relationships between aspects of digital social evaluation and body image concerns in adolescents. Australian adolescents (N = 245, 58.4 % male, Mage = 13.79, SD = 1.11) completed online surveys measuring digital social evaluation frequency (likes, follows, positive and negative comments about appearance and non-appearance), perceived importance (likes, number of followers), body dissatisfaction, restrictive eating attitudes, and drive for muscularity. Adolescents reported frequently receiving digital social evaluation, with females receiving more positive, but not negative, comments than males. All aspects of digital social evaluation were positively correlated with at least one aspect of body image concern. Controlling for other factors, all three aspects of body image were associated with receiving more frequent positive appearance-based comments, greater restrictive eating attitudes and body dissatisfaction were also associated with greater importance of likes, whilst greater drive for muscularity was associated with less frequent likes and comments. Digital social evaluation should be considered in future longitudinal or experimental research, and media literacy interventions.

Can following body positive or appearance neutral Facebook pages improve young women's body image and mood? Testing novel social media micro-interventions.

Small changes to social media use could have a large impact across the population. The present study tested novel social media micro-interventions (i.e., brief content delivered in everyday life) in which young women (N = 159) were instructed to either (1) follow a body positive Facebook group, (2) follow an appearance neutral Facebook group, or (3) use Facebook as usual. Relevant content was posted to the Facebook groups three times per day for two weeks. Primary outcomes were trait body image (body dissatisfaction and appreciation) and mood, and secondary outcomes were trait self-objectification, appearance comparison tendency, and body activism. Outcomes were assessed across three timepoints: pre-test (T1), post-test (T2) after the 14-day intervention period, and follow-up (T3) 4-weeks after T2. Participants in both the body positive and appearance neutral conditions reported decreased body dissatisfaction from T1 to T2 (small-medium effects) and participants in the body positive condition reported decreased appearance comparisons from T1 to T2 (medium effect). There were no changes for those variables from T2 to T3. No other significant differences were found. Viewing a small number of body positive or appearance neutral posts on social media may be an effective inexpensive micro-intervention for improving young women's body image.

Exploring the tripartite influence model of body image and disordered eating among adolescent girls living in Australia, China, India, and Iran.

The aim of this study was to evaluate the tripartite influence model of body image and disordered eating among 12-18-year-old girls (N = 900) living in Australia (n = 184), China (n = 293), India (n = 223), and Iran (n = 200). Participants completed measures about appearance pressures from family, peers, and media, thin-ideal internalization, appearance comparisons, appearance satisfaction, and disordered eating. Results indicated that media- pressures were most salient in Australia, whereas Iranians perceived family pressures to be the strongest. Indians felt the most pressure to look attractive from their family and peers, whereas Chinese felt similar levels of pressure from family, peers, and media. Path analyses suggested direct associations from one or more sociocultural factors to appearance satisfaction or disordered eating, with indirect paths via thin-ideal internalization and appearance comparisons emerging in all countries. Certain sociocultural factors contributed more strongly to the internalization of ideals and appearance comparisons within some countries. Multigroup path analyses demonstrated some cross-cultural variations in the strength of the pathways. Findings extend cross-cultural support for a modified tripartite influence model emphasizing culture-based specificities that can further guide preventative efforts to improve body image and eating attitudes among young girls in these four countries.

A longitudinal investigation of sleep hygiene as a mediator linking parental warmth with adolescent sleep.

STUDY OBJECTIVES: Parental warmth in adolescence protects sleep in early adulthood, yet the nature, directions, and mechanisms of this association across adolescence are unknown. This study examined parental warmth, adolescent sleep hygiene and sleep outcomes (morning/eveningness, school night sleep duration, and daytime sleepiness) across five annual waves, spanning four years, using a cross-lagged panel design. METHODS: Adolescents and one primary caregiver (96% mothers) completed questionnaires assessing parental warmth (child- and parent-report) and adolescent sleep hygiene and sleep (child-report), across five annual waves: Wave 1 (N = 531, Mage = 11.18, SD = 0.56, 51% male), Wave 2 (N = 504, Mage = 12.19, SD = 0.53, 52% male), Wave 3 (N = 478, Mage = 13.19, SD = 0.53, 52% male), Wave 4 (N = 440, Mage = 14.76, SD = 0.47, 51% male), and Wave 5 (N = 422, Mage = 15.75, SD = 0.49, 51% male). RESULTS: Greater child-reported parental warmth was indirectly associated with better adolescent sleep (greater morningness, longer school night sleep duration, less sleepiness) through healthier sleep hygiene. The inverse was also often observed. Warmth had a direct relationship with sleep duration and sleepiness, independent of sleep hygiene. Parent-reported parental warmth did not predict, nor was predicted by child-reported adolescent sleep. CONCLUSIONS: Parental warmth may protect against developmental changes in adolescent sleep, partially by improving sleep hygiene practices. Similarly, inadequate adolescent sleep may negatively impact parental warmth via deteriorating sleep hygiene. Sleep hygiene emerged as a key mechanism for protecting adolescent sleep and parent-child relationships.

Zooming in: The relationship between appearance concerns and perceived performance whilst videoconferencing among Australian adults.

Videoconferencing for work/study purposes has increased rapidly due to the COVID-19 global pandemic. Given this practice often involves viewing one's own video image, higher appearance concerns whilst videoconferencing may be linked to poorer performance whereby individuals may not feel they are able to engage or have control during a work/study meeting. The present study cross-sectionally examined whether both facial appearance concerns and fear of negative evaluation during videoconferencing for work/study purposes were directly and indirectly related to performance control and engagement, through self-focused attention and appearance distraction. Adult participants (N = 534, Mage = 26.32, SD = 11.08; 78% female) completed an online survey during the COVID-19 pandemic in Australia. Path analysis revealed direct and indirect effects, suggesting that heightened appearance self-consciousness was associated with greater self- and appearance-focused attention. In turn, this was related to impaired performance during videoconferencing. Multigroup analyses demonstrated that the magnitude of the proposed correlational effects were comparable across men and women. Given the ongoing reliance on videoconferencing, these findings have important theoretical and practical implications.

Potential benefits and limitations of machine learning in the field of eating disorders: current research and future directions.

Advances in machine learning and digital data provide vast potential for mental health predictions. However, research using machine learning in the field of eating disorders is just beginning to emerge. This paper provides a narrative review of existing research and explores potential benefits, limitations, and ethical considerations of using machine learning to aid in the detection, prevention, and treatment of eating disorders. Current research primarily uses machine learning to predict eating disorder status from females' responses to validated surveys, social media posts, or neuroimaging data often with relatively high levels of accuracy. This early work provides evidence for the potential of machine learning to improve current eating disorder screening methods. However, the ability of these algorithms to generalise to other samples or be used on a mass scale is only beginning to be explored. One key benefit of machine learning over traditional statistical methods is the ability of machine learning to simultaneously examine large numbers (100s to 1000s) of multimodal predictors and their complex non-linear interactions, but few studies have explored this potential in the field of eating disorders. Machine learning is also being used to develop chatbots to provide psychoeducation and coping skills training around body image and eating disorders, with implications for early intervention. The use of machine learning to personalise treatment options, provide ecological momentary interventions, and aid the work of clinicians is also discussed. Machine learning provides vast potential for the accurate, rapid, and cost-effective detection, prevention, and treatment of eating disorders. More research is needed with large samples of diverse participants to ensure that machine learning models are accurate, unbiased, and generalisable to all people with eating disorders. There are important limitations and ethical considerations with utilising machine learning methods in practice. Thus, rather than a magical solution, machine learning should be seen as an important tool to aid the work of researchers, and eventually clinicians, in the early identification, prevention, and treatment of eating disorders.

Machine learning models are computer algorithms that learn from data to reach an optimal solution for a problem. These algorithms provide exciting potential for the accurate, accessible, and cost-effective early identification, prevention, and treatment of eating disorders, but this potential is just beginning to be explored. Research to date has mainly used machine learning to predict women's eating disorder status with relatively high levels of accuracy from responses to validated surveys, social media posts, or neuroimaging data. These studies show potential for the use of machine learning in the field, but we are far from using these methods in practice. Useful avenues for future research include the use of machine learning to personalise prevention and treatment options, provide ecological momentary interventions via smartphones, and to aid clinicians with their treatment fidelity and effectiveness. More research is needed with large samples of diverse participants to ensure that machine learning models are accurate, unbiased, and generalisable to all people with eating disorders. There are limitations and ethical considerations with using these methods in practice. If accurate and generalisable machine learning models can be created in the field of eating disorders, it could improve the way we identify, prevent, and treat these debilitating disorders.