Abstract
Objective:
The purpose of this study is to understand the self-reported advantages and disadvantages of socially networking about body image/eating disorders (EDs); and, to examine the openness of these participants to online outreach and support for ED symptoms.
Method:
A cross-sectional online survey was conducted with a sample of N=598. Eligible participants were ≥15 years old, English-speaking, and U.S. residents who endorsed posting or following thin-ideal/body image content on social media. Quantitative measures were used to assess online peer support and online interaction preferences, and to identify ED symptoms. Deductive and inductive qualitative approaches were used to analyze open-ended items about the advantages and disadvantages of social networking about thin-ideal content on social media platforms (SMPs).
Results:
Among those who posted about the thin-ideal on social media, 70% felt the peer responses were positive and supportive. Participants generally favored online interaction, and a third stated they would accept support from someone they did not know online (38%). The most common advantages noted for posting/following thin-ideal content on SMPs were motivation/encouragement to engage in a certain behavior, socializing, and information giving/seeking. The most common disadvantages mentioned for posting/following thin-ideal content on SMPs were that the content elicits negative/bad feelings, having to deal with the negative consequences/reactions of others when socially networking about this topic, and that it triggers a desire to engage in ED behaviors.
Discussion:
With these findings, researchers, health practitioners, and social media administrators can devise ways to reduce harmful consequences of posting/following body image/ED content on social media.
Keywords: eating disorders, social media, body image
Introduction
Social media platforms (SMPs) are online communities where people can network with one another and create, promote, and trade various information and ideas (Kietzmann, Hermkens, McCarthy, & Silvestre, 2011). SMPs, including Instagram and Facebook, have prominent visual and linguistic features that can serve to promote and normalize potentially harmful thin-ideals about body image (Mingoia, Hutchinson, Wilson, & Gleaves, 2017). Use of SMPs has been associated with negative body image and disordered eating, with specific activities such as viewing and uploading photos and seeking feedback via status updates being particularly problematic (Holland & Tiggemann, 2016). Existing studies signal the potential for SMPs to perpetuate an unrealistic thin-ideal through images/text that glamorize and encourage disordered eating behaviors, accompanied by harmful advice within online forums on how to become and remain thin (i.e. “thinspiration”, “thinspo”, “pro-anorexia”.; e.g., Borzekowski, Schenk, Wilson, & Peebles, 2010). Users may then be more likely to adopt thin-ideal messages due to both repeated exposure to images/text embodying the thin-ideal and associated social rewards in the form of praise of these images/text by peers (Borzekowski et al., 2010). On the other hand, body image/ED-focused online communities can serve as refuge for those discussing this sensitive and often stigmatized topic, by helping these individuals establish a support system (e.g., Juarascio, Shoaib, & Timko, 2010).
In order to mitigate risks among individuals who are on SMPs and who may already have a number of risk factors for EDs or some symptoms of disordered eating, an improved understanding of the potential harm and benefits of socially networking about body image/EDs is needed. In previous studies, the most frequently reported reasons to seek out pro-ED websites were weight loss tips, motivation for weight loss, curiosity, and meeting people (Peebles et al., 2012), while reported motivations for using pro-ED websites included a sense of belonging (77%), social support (75%) and to support to continue their ED-related behaviors (52%) (Ransom, La Guardia, Woody, & Boyd, 2010). However, to date, no known studies have systematically evaluated the self-reported positive and negative effects of networking specifically on SMPs about body image/EDs among those who engage in these online activities. As a result, the primary objective of the current study is to explore the perspectives of individuals who are identified on social media as posting or following body image and/or ED-related content on SMPs, and to compare the advantages/disadvantages of social networking reported across relative ED risk groups. We hypothesize that teens with a clinical/subclinical ED will more often socially network about thin-ideal content and will endorse different advantages and disadvantages related to this social networking than teens without an ED. Additionally, given that individuals with EDs tend to be stigmatized or hard-to-reach, a secondary objective of this study is to explore participant interest in online outreach and support related to ED symptoms, a significantly understudied topic.
Methods
Participants and recruitment
Participants were recruited during March - June 2017. We used paid advertisements on Instagram, Facebook, and Twitter to target individuals in the U.S. who spoke English and posted and/or followed ED-related content and/or accounts; total cost for the advertisements was approximately $500. Targeted advertisements on social media have been previously used in mental health-related research and have been described as an efficient way to reach a large audience (e.g., Carter-Harris, Bartlett Ellis, Warrick, & Rawl, 2016). SMPs utilize proprietary algorithms to target online advertisement campaigns at individuals who are socially networking about the topic of interest as indicated by their use of certain keywords and hashtags. We selected the following keywords and hashtags to identify potential participants: body image, body shape, dieting, female body shape, Eating Disorders Anonymous, National Eating Disorders Association, and recovering from an EDs. To recruit on Reddit, we gained moderator approval versus advertisements to post about our study on two forums focused on maintaining ED-related behaviors.
The advertisements and posts included a link to our study website. When a potential participant clicked the link to our website, another link took them to the online eligibility survey. Eligible participants were ≥15 years old, U.S. residents, who endorsed either having posted on social media about eating/weight/body image, emphasizing that being thin is important or attractive, or following social media accounts with this emphasis (“Do you ever post about your body-image, weight or relationship with food on social media?” or /and “Do you follow/subscribe to any social media accounts/groups that post about body-image, weight, or relationship with food?”). Eligible participants were then provided an online consent document and were forwarded to the full survey after consenting. Washington University IRB waived parental consent due to the low risks associated with this study and provided study approval. The survey was distributed using Qualtrics and completed via computer or mobile device. Compensation was a $10 Amazon.com gift card.
Among 1,062 eligible participants, 1,055 consented. To maintain high quality responses, we used measures to help prevent machine responses and repeat survey takers (i.e., Captcha and the “Prevent Ballot Box Stuffing” feature in Qualtrics, which uses cookies to prevent participants from taking the survey again from the same web browser). We also used several data cleaning steps to remove low quality responses. Individuals who did not progress at least 50% of the way through the survey were removed (n=334), as this would have precluded the identification of ED symptoms for diagnostic screening. Short completion time is also known to be a good indicator of meaningless data (Leiner, 2013); therefore, we removed 68 participants in the lowest 10th percentile of survey completion time (8 minutes). We also removed 37 duplicate responses and 18 responses with inconsistent response patterns (e.g., differing responses on items asking about the same construct). This left 598 participants for analysis (median survey completion time 22 minutes, inter-quartile range [IQR] 17-30 minutes).
Quantitative survey measures
To examine exposure to thin-ideal content on social media, the survey queried how often in the past month (i.e., Never, Once, 2-5 times, 6-10 times, >10 times), individuals had seen a social media post from a peer (“In the past month, how often did you see a post from a peer about eating/weight/body image emphasizing that being thin is important or attractive on social media sites?”) or posted on social media about their own thoughts about eating/weight/body image emphasizing being thin is important or attractive (“In the past month, how often have you posted your thoughts about eating/weight/body image emphasizing that being thin is important or attractive on social media sites?”). The survey also queried whether participants followed accounts on SMPs that posted about this type of content (yes/no). SMPs included Twitter, Facebook, Snapchat, Instagram, Tumblr, YouTube, Pinterest, Google+, LinkedIn, Reddit, Yik Yak, Whisper, and participants were asked to specify any others that were not listed. Those who had posted this content in the past month were queried about the type of response they received from others (five categories of potential responses ranging from “People are mostly positive. They almost always support what I post.” to “People are mostly negative. They almost never support what I post”).
To examine participants’ perceived support from their online network, we modified the Social Support Questionnaire (Sarason, Levine, Basham, & Sarason, 1983), which has been validated with individuals with eating and body image concerns (e.g., Pace, D'Urso, & Zappulla, 2018) and asks about support from friends, to focus on support from “online friends.” These six items were statements of various types of support participants may feel they receive from their “online friends” (e.g., are dependable when participant needs help, are accepting of the participant including worst and best points, make participant feel better when feeling down), with a five-point Likert scale responses ranging from 1=Very untrue to 5=Very true. Internal consistency within our data was excellent (Cronbach’s alpha=0.95); thus, we calculated the average across all six items as an indicator of online social support. We also assessed participants’ preference for online social interaction (POSI) with three items: “I prefer communicating with other people online rather than face-to-face,” “I feel like I have more control over conversations online than I do in face-to-face conversations,” and “Meeting and talking with people is better when done online rather than in face-to-face situations.” (Caplan, 2005). Participants rated their agreement on a Likert scale ranging from 1=Strongly disagree to 5=Strongly agree. Internal consistency was good (Cronbach’s alpha=0.82), and we averaged each participant’s responses across these three items.
We also queried participants’ likelihood of accepting emotional support (e.g., advice or a listening ear) from several types of individuals if they reached out to them on social media: 1) someone they know (online or in-person), 2) someone they did not know (online or in-person), 3) their own health professional (e.g., counselor/doctor), 4) a health professional they did not know. Potential responses ranged from 1=Extremely unlikely to 7=Extremely likely. We dichotomized responses as extremely/moderately likely vs all others.
The Stanford-Washington ED Screen (SWED) was used to examine participant’s risk of having an ED based on DSM-5 criteria (i.e., possible anorexia nervosa (AN), bulimia nervosa (BN), binge eating disorder (BED), subclinical BN, subclinical BED, purging disorder, unspecified feeding or eating disorder (UFED, and no ED)). The SWED is an online self-report tool with good sensitivity and specificity for most diagnoses (Graham et al., 2019). Participants were grouped into three ED screening groups based on their results; clinical ED (i.e., AN, BN, BED, purging disorder), subclinical ED (i.e., subclinical BN, subclinical BED, UFED), and no ED (i.e., high risk for an ED and not currently at risk for an ED). Participants also reported whether they had received treatment for their eating related problems in the last six months.
Demographics assessed included age, gender, race/ethnicity, school enrollment, employment status, and household income. Gender categories included male, female, transgender, non-binary, and a text box for participants to self-identify their gender.
Qualitative survey measures
Participants were asked two open-ended items about the pros and cons of networking about thin-ideal content on SMPs: 1) “What are some advantages that you experience from posting this type of content/following social media sites that encourage this type of content?”; and 2) “What are some disadvantages that you experience from posting this type of content/following social media sites that encourage this type of content?”. To analyze responses to open-ended items, we used both deductive and inductive content qualitative analysis to test existing theories and generate new themes from the current data, respectively (Elo & Kyngäs, 2008).
Advantages to posting/following
Previous research has explored why individuals engage in online communities using the uses and gratifications theory as a framework (Alhabash & Ma, 2017). In a factor analysis conducted by Park et al. (2009), four main reasons for participation in Facebook groups were identified, in line with this theory: 1) Socializing; 2) Entertainment; 3) Self-status seeking; 4) Information seeking (Park, Kee, & Valenzuela, 2009). We similarly used these four themes as the foundation of the codebook that was created for the advantages responses (Table 1).
Table 1.
Perceived advantages of posting about/following thin-ideal content on social media (n = 566 participants responded to these open-ended items in the survey)
| Total (%) | Clinical ED (n = 138) |
Subclinical ED (n = 338) |
No ED (n = 90) |
p-value | Pairwise comparisons |
|
|---|---|---|---|---|---|---|
| n (%) | ||||||
| Motivation/Encouragement | 345 (60.95) | 76 (55.07) | 224 (66.27) | 45 (50.00) | 0.005 | Subclinical ED > Clinical ED, No ED |
| • “This allows me to stay motivated and remind myself of my goals.” • “Some people have the same views and are incredibly supportive/motivating.” • “It makes you feel motivated to do something about yourself and change things.” | ||||||
| Motivation to… | ||||||
| …be thin | 129 (37.39) | 33 (43.42) | 87 (38.84) | 9 (20.08) | 0.027 | Clinical ED, Subclinical ED > No ED |
| • “It gives me motivation to achieve my thinness related goals.” • “I use it to motivate myself to get skinnier.” • “It’s almost like I’m persuading myself to stick with losing more and more weight; it keeps me motivated.” | ||||||
| …get healthy | 82 (23.77) | 9 (11.84) | 51 (22.77) | 22 (48.89) | <0.001 | No ED > Subclinical ED > Clinical ED |
| • “It encourages me to keep up with healthy habits (e.g., exercise, nutritious eating, etc.) and reminds me of what is possible.” • “They can help motivate people to get healthy, when the activity or results being encouraged are healthy.” • “Motivation to have a more active and healthy lifestyle.” | ||||||
| …work out | 79 (22.90) | 9 (11.84) | 45 (20.09) | 25 (55.56) | <0.001 | No ED > Subclinical ED, Clinical ED |
| • It makes me want to work out in order to look more like the models or people in the photos/videos.” • “It gives me motivation to eat better and work out more. Working out can be fun if you do activities you enjoy.” • “I feel focused on my fitness goals and stops me from slacking off.” | ||||||
| …restrict, not eat, and/or diet | 38 (11.01) | 13 (17.11) | 25 (11.16) | 0 | 0.015 | Clinical ED, Subclinical ED > No ED |
| • “It helps me to remind myself that I should continue dieting.” • “when I get really hungry, I go in these sites to get a little extra motivation to not eat for just a bit longer” • “It makes it easier to not eat if I remind myself what I want my body to look like, and how I want to control my eating behaviors” | ||||||
| …recover/seek treatment/not engage in ED behavior | 10 (2.90) | 5 (6.58) | 5 (2.23) | 0 | 0.069 | -- |
| • “It helps motivate me to combat my binge eating disorder and makes me feel more comfortable and less alone with some harmful thoughts I have” • “Hearing others' ED struggles and recovery victories makes me believe that I might be able to recover too. Finding these online communities was the main push that actually got me to seek out help from a therapist and nutritionist.” • “Sometimes the bodies look really scary so it's like a warning to not go too far and it's a reminder that eating disorders aren't pretty.” |
||||||
| Socializing | 211 (37.28) | 88 (63.77) | 111 (32.84) | 12 (13.33) | <0.001 | Clinical ED> Subclinical ED > No ED |
| • “Getting support from others who are struggling with similar issues, feeling less alone and isolated” • “Eating disorders and body dysmorphia can be isolating, being able to talk about things in a somewhat confidential environment helps me feel not as alone.” • “The advantages are that I can discuss the things I'm experiencing with others who relate. A lot of the discussions help me learn more about myself and understand myself better.” | ||||||
| Information giving/seeking | 138 (24.38) | 32 (23.19) | 79 (23.37) | 27 (30.00) | 0.400 | -- |
| • “Sometimes I feel like I am influencing people positively, to make better lifestyle choices.” • “I’ve learned a lot about calories and nutrition from ED communities, and despite my ED I feel like I have a lot of knowledge on those subjects now.” • “They give tips on how to become skinny.” |
||||||
| Emotional outlet | 113 (19.96) | 46 (33.33) | 60 (17.75) | 7 (7.78) | <0.001 | Clinical ED> Subclinical ED > No ED |
| • I use it as a former of venting…I have high and low days regarding my self esteem, and on my low days it's nice to have somewhere I can let out how I feel and not hold it in.” • “It gets my feelings out and it helps me connect with others who feel the same way and we can help each other. It makes me feel less alone and it helps me to motivate myself to become prettier” • “It lets me speak my mind and get things off of my chest that I otherwise wouldn't be able to, as it's not something I can really discuss with friends or family.” | ||||||
| Self-status seeking | 91 (16.08) | 22 (15.94) | 52 (15.38) | 17 (18.89) | 0.723 | -- |
| • “It gives me attention thus temporarily making me feel better about my appearance.” • “It makes me feel better about my appearance when people validate me and my body type” • “I get more likes on my photos when I post a picture that accentuates a certain part of [my] face or body… I sometimes post about how unhappy I am with my appearance and people post nice things about me. When I post that I lost weight, people comment “yay!” with celebration emojis.” |
||||||
| Entertainment | 32 (5.65) | 7 (5.07) | 19 (5.62) | 6 (6.67) | 0.878 | -- |
| • “Usually the posts are humorous and my friends can relate.” • “I just think it's pretty but there aren't really any advantages besides being something nice to look at when I'm bored looking at social media.” • “They're entertaining and 'trendy'.” |
||||||
| No benefit | 36 (6.36) | 11 (7.97) | 21 (6.21) | 4 (4.44) | 0.559 | -- |
| • “I don't experience any advantages or positive feelings from posting about this content” • “there's really no rational advantage to following the accounts” • “There are no real advantages. As someone who has been hospitalized for my eating disorder… there is logically nothing positive about being encouraged or encouraging someone to engage in disordered behavior.” |
||||||
Note: Pairwise comparisons listed were significant at least at p < 0.05.
In addition to these four themes, our team inductively identified additional themes based on a review of approximately 100 responses to the advantages questions. It was noted if 1) participants were motivated/encouraged to do something or engage in a behavior and/or if 2) they expressed posting/following accounts as way to vent or as an emotional outlet. Responses indicating motivation/encouragement as an advantage were further coded if one of the following was specifically mentioned: 1) Work out or be active; 2) Get healthy (e.g., eat better, maintain healthy weight); 3) Not eat, diet, or restrict food intake; 4) Be thin (e.g., aspire for thin-ideal, “thinspiration”) or lose weight, 5) Seek recovery/treatment, not engage in ED behavior. If participants indicated that there were no advantages, this was also recorded. These themes and examples are also included in Table 1.
Disadvantages to posting/following
Disordered eating is influenced by a range of sociocultural factors (Weissman, 2019). Sociocultural theory explains the process of how women can internalize the thin-ideal and unrealistic beauty standards portrayed in the media (e.g., Holland & Tiggemann, 2016). Consequently, the themes used to code the responses to the disadvantages questions in this project were based on sociocultural factors that have been shown to contribute to the development and maintenance of body dissatisfaction (Fitzsimmons-Craft et al., 2014) (Table 2).
Table 2.
Perceived disadvantages of posting about/following thin-ideal content on social media (n=565 participants responded to these open-ended items in the survey).
| Total (%) | Clinical ED (n = 138) |
Subclinical ED (n = 338) |
No ED (n = 90) |
p-value | Pairwise comparisons |
|
|---|---|---|---|---|---|---|
| n (%) | ||||||
| Conjures up negative/bad feelings, lowers self-esteem… | 347 (61.31) | 73 (52.90) | 214 (63.31) | 60 (66.67) | 0.173 | -- |
| • “Contributes to low self esteem, reinforces negative thoughts” • “Negative impact on self-esteem, feeling less worthy” • “It makes people feel bad about themselves. | ||||||
| … mentions anxiety or depression | 37 (6.54) | 11 (7.79) | 23 (6.80) | 3 (3.33) | 0.631 | -- |
| • “It can send me into a depressive state” • “Thinspo accounts can be pretty depressing and triggering. They post photos of bone-thin girls and it definitely turns my focus to my body in negative ways.” • “Encouragement. Sadness. Anxiety.” | ||||||
| General outcomes of socially networking about this topic. | 172 (30.39) | 41 (29.71) | 105 (31.07) | 26 (28.89) | 0.896 | -- |
| • “People judge me.” • “People will dm me mean things and it makes me mad” • “No 'likes' or compliments from others” | ||||||
| Triggering ED behaviors | 149 (26.33) | 46 (33.33) | 89 (26.33) | 14 (15.56) | 0.041 | Clinical ED, Subclinical ED > No ED |
| • “It triggers me into restricting and self hatred” • “I usually want to relapse with my eating disorders” • “I will sometimes push myself to stop eating” | ||||||
| Promoting thinness as attractive/pressure to be thin | 129 (22.79) | 29 (21.01) | 83 (24.56) | 17 (18.89) | 0.641 | -- |
| • “Promotes problematic and unrealistic beauty standards” • “It can be unhealthy thinspiration” • “They hinder recovery and enforce the thin ideal. Although they don't always have a substantial effect on me, at times of weakness they can push me into relapse.” | ||||||
| Body comparisons/feeling competitive towards others | 125 (22.08) | 32 (23.19) | 74 (21.89) | 19 (21.11) | 0.899 | -- |
| • “It could make a person feel down on their self because they might not think they look as good as the person or people they follow on social media” • “I obsessively compare my body and my lifestyle to others.” • “It makes me less satisfied with my own body and more jealous/competitive toward other attractive women.” |
|
|||||
| Obsessing over weight/need to be thin | 37 (6.54) | 15 (10.87) | 18 (5.33) | 4 (4.44) | 0.155 | -- |
| • “It leads to me becoming more obsessive about eating” • “Makes me fixate on my weight and body to an unhealthy degree that I probably wouldn't if I weren't consuming this type of content.” • “It fuels my obsession with my weight, food, exercise, etc. I read other's stats who have a lower BMI than me and I get envious. I also get sickeningly proud when I see my BMI is lower than other posters. I compulsively check the proED subreddit which probably isn't a healthy behavior.” | ||||||
| Fear of exposing ED/identification (e.g. doxing) | 23 (4.06) | 11(7.97) | 12 (3.55) | 0 | 0.034 | Clinical ED > Subclinical ED, No ED |
| • “Potential exposure. On more public social media websites like facebook or instagram, I try to mask anything that might indicate an eating disorder so that anyone who doesn't know what to look for would assume that my eating habits, weight, and way that I portray myself are at worst unusual but healthy. On more anonymous social media websites I openly talk about it, but there is of course always a risk of being identified.” • “I'm afraid someone will find out who I am.” • “There's always a small chance someone will recognize me from real life” |
||||||
| No disadvantages | 33 (5.83) | 7 (5.07) | 21(6.21) | 5 (5.56) | 0.899 | -- |
| • “Nothing” • “Nothing, maybe someone doesn't agree but they don't say anything.” • “None.” | ||||||
Note: Pairwise comparisons listed were significant at least at p < 0.05.
In addition, themes were inductively developed based on a review of approximately 100 responses to the disadvantages question. For those participants who mentioned that thin-ideal content leads to negative/bad feelings, it was noted if they specifically mentioned that it caused feelings of anxiety or depression. Responses were additionally coded for general social networking-related consequences of posting/following about a thin-ideal. Responses were also coded for obsessing over weight and/or the need to be thin; as well as, fear of identification and/or exposing ED thoughts and behaviors. Lastly, it was noted when participants mentioned there were no disadvantages of posting/following this content. These themes and examples are illustrated in Table 2.
Coding qualitative responses
The responses to the advantages/disadvantages questions were coded by two team members with a thorough understanding of EDs and mental health (advantages: EFC, SJC; disadvantages: EFC, NA). Responses were coded first in sets of 50 to refine the codebook, then once the codebook was well-defined and reliability was established, larger sets were coded (i.e., 100-216). After each set, the two coders would meet to discuss any discrepancies and the agreed upon codes were recorded in a master database. Inter-coder reliability for each theme across all coded responses was assessed using Krippendorff’s α (range 0 to 1 with higher numbers indicating better agreement, 1= perfect reliability) (Krippendorff, 2004, 2011). Agreement across the two coders for the advantages themes was excellent (median Krippendorff’s α 0.83, minimum 0.73, maximum 0.95). Agreement across the codes for the disadvantages themes was excellent as well (median Krippendorff’s α 0.81, minimum 0.6, maximum 0.93).
Statistical Analysis
Codes from the qualitative analysis were merged with participants’ survey data. We performed descriptive statistics and bivariate analyses to test for differences in sample characteristics across ED screening groups. Chi-square tests were conducted for the categorical variables and ANOVA tests and Tukey-Kramer tests were conducted for the continuous variables. P-values of less than 0.05 were considered statistically significant. All analyses were conducted using Stata MP. Version 16.
Results
Sample characteristics
The self-reported characteristics of our 598 participants are shown in Table 3 by ED screening groups. The mean age of our sample was 19.77 (SD = 5.03). The majority of participants screened positive for a subclinical ED (59.7%), while 23.7% had a clinical ED. Thus, only 16.6% did not have an ED. . Most were female (77.26%), non-Hispanic white (69.06%), and enrolled in school (69.90%). Nearly half (46.32%) were recruited from Instagram, a quarter (24.92%) from Reddit, and fewer from Facebook (17.89%) and Twitter (10.87%). Two thirds (67.4%) had never received treatment for an ED. Most participants had seen thin-ideal content on SMPs in the past month (96.49%) and/or followed accounts that focused on posting this content (94.82%); nearly two-thirds of participants (71.74%) reported posting about their own thoughts about eating/weight/body image that emphasized a thin-ideal. Many reported seeing or following such content on multiple SMPs (M = 4.38, SD =2.17 for seeing content, M = 3.12, SD =1.89 for following thin-ideal focused accounts).
Table 3.
Demographic characteristics, ED characteristics and social networking about thin-ideal of the total sample (N = 598 unless otherwise noted)
| Total | Clinical ED (n= 142) |
Subclinical ED (n= 358) |
No ED (n=98) |
p-value | Pairwise comparisons | |
|---|---|---|---|---|---|---|
| N (%) or Mean [SD] | ||||||
| Demographic characteristics | ||||||
| Gender (n=548) | <0.001 | |||||
| Male | 29 (4.85) | 0 | 18 (5.03) | 11 (11.22) | No ED > Clinical ED | |
| Female | 462 (77.26) | 129 (90.85) | 259 (72.38) | 74 (75.51) | -- | |
| Transgender | 57 (9.53) | 5 (3.52) | 48 (13.41) | 4 (4.08) | Subclinical ED > No ED | |
| Age in years (n=541) | 19.77 [5.03] | 20.95 [4.04] | 19.29 [4.87] | 19.74 [5.47] | 0.006 | Clinical ED > Subclinical ED |
| Race (n=543) | 0.253 | |||||
| Non-Hispanic White | 413 (69.06) | 102 (71.83) | 242 (67.60) | 69 (70.41) | -- | |
| Non-Hispanic Black | 35 (5.85) | 4 (2.82) | 28 (7.82) | 3 (3.06) | -- | |
| Non-Hispanic Asian | 35 (5.85) | 7 (4.93) | 22 (6.15) | 6 (6.12) | -- | |
| Other | 60 (10.03) | 19 (13.38) | 30 (8.36) | 11 11.22) | -- | |
| Enrolled as a student (n=548) | 418 (69.90) | 87 (61.27) | 262 (73.18) | 69 (70.41) | 0.005 | Subclinical ED > Clinical ED |
| Employed (n=548) | 272 (45.48) | 77 (54.23) | 158 (44.13) | 37 (37.76) | 0.107 | -- |
| Annual household income (n=532) | 0.021 | |||||
| <$25,000 | 178 (29.77) | 39 (27.46) | 119 (33.34) | 20 (20.41) | Subclinical ED > Clinical ED, No ED | |
| $25,000 to $49, 999 | 119 (19.90) | 38 (26.76) | 67 (18.72) | 14 (14.29) | -- | |
| $50,000 to $74,999 | 96 (16.05) | 22 (15.49) | 56 (15.64) | 18 (18.37) | -- | |
| ≥$75,000 | 139 (23.24) | 33 (23.24) | 74 (20.67) | 32 (32.65) | Subclinical ED > Clinical ED, No ED | |
| Recruited from… | <0.001 | |||||
| 277 (46.32) | 30 (21.13) | 187 (52.23) | 60 (61.22) | No ED, Subclinical ED > Clinical ED | ||
| 149 (24.92) | 62 (43.66) | 77 (21.51) | 10 (10.20) | Clinical ED > Subclinical ED > No ED | ||
| 107 (17.89) | 28 (19.72) | 55 (15.36) | 24 (24.49) | No ED > Subclinical ED | ||
| 65 (10.87) | 22 (15.49) | 39 (10.89) | 4 (4.08) | Clinical ED, Subclinical ED > No ED | ||
| ED characteristics | ||||||
| Treatment status (n=576) | <0.001 | |||||
| Received treatment in the past six months | 74 (12.37) | 32 (22.54) | 36 (10.06) | 6 (6.12) | Clinical ED > Subclinical ED, No ED | |
| Received treatment at some point in the life | 99 (16.56) | 33 (23.24) | 55 (15.36) | 11(11.22) | Clinical ED > No ED | |
| Never received treatment | 403 (67.39) | 77 (54.23) | 246 (68.72) | 80 (81.63) | No ED, Subclinical ED > Clinical ED | |
| Social networking about thin-ideal in past month on any social media platform | ||||||
| Saw this content | 577 (96.49) | 139 (97.89) | 346 (96.65) | 92 (93.88) | 0.244 | -- |
| Posted this content (n=597) | 429 (71.74) | 118 (83.10) | 259 (72.35) | 52 (53.06) | <0.001 | Clinical ED > Subclinical E > No ED |
| Followed accounts posting this content (n=597) | 567 (94.82) | 138 (97.18) | 338 (94.41) | 91 (92.86) | 0.539 | -- |
| Number of social media platforms on which they saw this content | 4.38 [2.17] | 4.68 [2.06] | 4.32 [2.11] | 4.18 [2.52] | 0.155 | -- |
| Number of social media platforms on which they posted this content (n=597) | 1.72 [1.65] | 2.11 [1.67] | 1.72 [1.61] | 1.17 [1.59] | <0.001 | Clinical ED, Subclinical ED > No ED |
| Number of social media platforms on which they followed this content (n=597) | 3.12 [1.89] | 3.50 [1.84] | 3.11 [1.86] | 2.58 [1.98] | 0.001 | Clinical ED > No ED |
Note: Pairwise comparisons listed were significant at least at p <0.05.
As shown in Table 3, the subclinical ED group had a significantly higher proportion of transgender participants compared with the clinical ED group and the no ED group (13.41% vs. 3.52% & 4.08%). As hypothesized, the clinical ED group had a significantly higher proportion of participants who posted thin-ideal content on any SMP relative to the subclinical ED group, which in turn had a significantly higher proportion compared to the no ED group (83.10% vs. 72.35% vs. 53.06). People in the clinical ED and subclinical ED groups posted thin-ideal content on significantly more SMPs compared with people in the no ED group (M = 2.11, SD = 1.67 & M = 1.72, SD = 1.61 vs. M = 1.17, SD = 1.59).
Online social support and outreach
Preferences for online communication and the social support participants felt they received from their online friends are described in Table 4. On a scale from 1 to 5, participants with a clinical ED felt that their online friends were significantly more supportive compared with those in the subclinical ED group and the no ED group (M = 3.74, SD = 1.03 vs. M = 3.49, SD = 1.17 & M = 3.38, SD = 0.98). Among those that posted their thoughts about the thin-ideal on SMPs, most (64.88%), across all diagnosis groups felt that the responses they have received are, at least sometimes, positive and supportive of this thin-ideal content. On a scale from 1 to 5 in rating their preference for online communication, all participants generally swayed toward favoring online interaction (M = 3.30, SD = 1.03), and there was no significant difference across the ED screening groups (p = 0.113). For participants without an ED (67.35%) and those in the subclinical ED group (58.10%), a majority claimed that they would accept emotional support from someone they knew and only few of them would accept emotional support from someone they did not know (26.53%). Similar results were found for participants in the clinical ED group, with a comparable proportion of participants claiming that they would accept emotional support from someone they knew, and they did not know (both 47.18%).
Table 4.
Perceived support, online interactions, and likeliness to accept emotional support (N = 598 unless otherwise noted)
| Total | Clinical ED (n= 142) |
Subclinical ED (n= 358) |
No ED (n=98) |
p-value | Pairwise comparisons | |
|---|---|---|---|---|---|---|
| n (%) or Mean [SD] | ||||||
| Online social support score (range 1 to 5) (n=542) | 3.53 [1.06] | 3.74 [1.03] | 3.49 [1.17] | 3.38 [0.98] | 0.023 | Clinical ED > Subclinical ED, No ED |
| Preference for online social interaction score (range 1 to 5) (n=556) | 3.30 [1.03] | 3.43 [1.05] | 3.28 [1.03] | 3.15 [1.03] | 0.113 | -- |
| Accepting emotional support from individuals on social media (People may choose more than one) | ||||||
| Someone you know | 341 (57.02) | 67 (47.18) | 208 (58.10) | 66 (67.35) | 0.007 | No ED, Subclinical ED > Clinical ED |
| Someone you didn’t know | 228 (38.13) | 67 (47.18) | 135 (37.71) | 26 (26.53) | 0.005 | Clinical ED, Subclinical ED > No ED |
| Your health professional | 250 (41.81) | 46 (32.39) | 151 (42.18) | 53 (54.08) | 0.009 | No ED > Subclinical ED > Clinical ED |
| A health professional you didn’t know | 159 (26.59) | 36 (25.35) | 91 (25.42) | 32 (32.65) | 0.331 | -- |
| Response type (Among those who posted on social media, n=413) | 0.002 | |||||
| People are mostly positive. | 154 (25.75) | 50 (35.21) | 84 (23.46) | 20 (20.41) | Clinical ED > Subclinical ED | |
| People are sometimes positive. | 136 (22.74) | 34 (23.94) | 86 (24.02) | 16 (16.33) | -- | |
| People are sometimes positive and sometimes negative | 98 (16.39) | 20 (14.08) | 64 (17.88) | 14 (14.29) | -- | |
| People are sometimes negative. | 16 (2.68) | 6 (4.34) | 9 (2.51) | 1 (1.02) | -- | |
| People are mostly negative. | 9 (1.51) | 2 (1.41) | 7 (1.96) | 0 | -- | |
Note: Pairwise comparisons listed were significant at least at p <0.05.
Advantages of socially networking
Among our 598 participants, 566 (94.6%) responded to open-ended items asking about the perceived advantages of posting about or following thin-ideal content on SMPs. Results are shown in Table 1. The most common theme was motivation/encouragement to do something or engage in a behavior (60.95%), specifically motivation to be thin (37.39%), get healthy (23.77%), work out (22.90%), or restrict their eating or diet (11.01%). Motivation to recover or seek treatment was mentioned in far fewer responses (2.90%). The group of participants without an ED had a significantly higher proportion of people that claimed a motivation to get healthy (48.89% vs. 22.77% & 11.84%) and work out (55.56% vs. 20.09% & 11.84%) compared to those in the clinical ED and subclinical ED groups. Conversely, the clinical ED and subclinical ED groups had a significantly higher proportion of people that claimed the motivation to be thin (43.42% & 38.84 % vs. 20.08%) and restrict their eating or diet (17.11% & 11.16 % vs. 0) compared to the group without an ED. Other common advantages mentioned were socializing (37.28%), seeking/giving information (24.38%), as a way to vent or as an emotional outlet (19.96%), for self-status seeking (16.08%) entertainment (5.65%), and 6.36% of participants stated that posting/following had no benefit.
Disadvantages of socially networking
Within our 598 participants, 565 responded to open-ended items asking about the perceived disadvantages of posting about or following thin-ideal content on SMPs. The results are shown in Table 2. The most common disadvantage mentioned was that the content elicits negative/bad feelings and/or lowers self-esteem (61.31%), with some responses specifically mentioning anxiety and/or depression (6.54%). The second most common disadvantage was dealing with the negative consequences/reactions of others when socially networking about this topic (i.e., fear of other people’s reactions, fear of judgment, lack of comments/likes) (30.39%). The third most common disadvantage was that it triggers a desire to engage in ED behaviors (26.33%). Other disadvantages noted were promoting thinness as attractive/increasing the pressure to be thin (22.79%), engaging in body comparisons/feeling competitive towards others (22.08%), obsessing over weight/the need to be thin (6.54%), fear of exposing ED thoughts/being identified (4.06%), and 5.83% of participants stated that posting/following had no disadvantages. For most disadvantages, there were no statistically significant disadvantages in themes across the ED screening groups; meaning, participants’ expressions of a disadvantage theme may not be determined/influenced by their ED status. For the people who claimed disadvantages related to the trigging ED behaviors (33.33%% vs. 26.33% & 15.56%) and the fear of exposing ED/identification (7.97% vs. 3.55% & 0), the clinical ED group has a significantly higher proportion of people compared with the subclinical ED group and the no ED group.
Discussion
The present study examines the self-reported advantages and disadvantages of ED-social networking to improve understanding of the potential motivations and outcomes of these online social networking behaviors from the perspectives of individuals who endorse communicating online in this way. Aligning with our hypotheses, in examining participants’ self-reported advantages of ED social networking, a statistically significant difference was found between participants with a clinical ED, a subclinical ED, or no ED for most of the advantages examined. In particular and of concern, the advantages to ED social networking endorsed by participants with a clinical and/or subclinical ED (versus participants without an ED) included increasing motivation to engage in harmful eating/dieting thoughts and habits. Increasing motivation to engage in healthy eating/dieting habits, such as to “get healthy” and “work out,” tended to be advantages to ED social networking identified by participants without an ED. These findings corroborate prior research about individuals who socially network about pro-ED online content in order to reinforce ED thoughts and behaviors in such a way that this mental illness is exacerbated and/or sustained (Wilson, Peebles, Hardy, & Litt, 2006). Of note, some participants with a clinical and/or subclinical ED in our study cited motivation/encouragement to ED socially network in order to “recover/seek treatment/not engage in ED behavior”; however, this was endorsed to a low degree (fewer than 2% of the entire study population). As such, it seems individuals with EDs were engaging in ED-related social networking behavior more so to fuel their ED rather than to assist with recovery.
Moreover, participants with a clinical and/or subclinical ED disproportionally cited “socializing” and using social networking as “emotional outlet” as advantages to ED-social networking. Related, participants in these subcategories had a higher median “online social support” and “preference for online social interaction” score versus participants without an ED. Furthermore, there was a statistically significant difference found in the amount of posting of thin-ideal content between screen result groups. More participants in the clinical and subclinical groups posted thin-ideal content on social media, and they tended to post on more SMPs than those without an ED. As a whole, the clinical implications of our findings suggest the presence of unique and identifiable social networking behaviors and preferences among individuals with a clinical or subclinical ED versus those without an ED.
Many of our participants expressed their experience of serious negative consequences including potential for triggering ED-related symptomatology, increasing the pressure one feels to become thin, and feeling more pressure to engage in body comparisons. Our findings lend support to surveillance and moderation of pro-ED content online (Tong, Heinemann-Lafave, Jeon, Kolodziej-Smith, & Warshay, 2013). It is also noteworthy that participants with a clinical ED are just as likely to state that posting/following thin-ideal content online “conjures up negative/bad feelings and/or lowers self-esteem” as those without an ED. Therefore, online content that perpetuates the thin-ideal may be harmful to all viewers, not just those with a clinical or subclinical ED. However, in order to manage these negative outcomes, individuals may seek out more engagement with social media, which may inform and fuel ED behaviors, creating a challenging cycle of thin-ideal content exposure and subsequent engagement in ED-related behaviors (Kelly, Zilanawala, Booker, & Sacker, 2018). Further, because individuals with EDs often label their symptoms as in line with their self-image (Gregertsen, Mandy, & Serpell, 2017), such interactions on pro-ED websites can serve to further normalize these beliefs/values, which may in turn decrease awareness of risks, increase social pressure, and decrease motivation to alter these ED behaviors.
When assessing participants’ interest in online support, while 57% would accept support from someone they knew (in person or online) and 42% would accept support from their own health professional, we found that fewer would accept support from someone they did not know (38%) or a health professional they did not know (27%). Future research could explore the utility of training forum moderators to connect individuals to support and peer-delivered interventions. Participant’s preference for online communication along with their interest in online recruitment and support via SMPs outlines the potential to engage persons on SMPs to participate in an online intervention. This is further supported by past findings which demonstrate successful online social media recruitment and connection to interventions (e.g., (Raggatt et al., 2018).
Limitations
With the distribution of percentages from our results, we are able to provide plausible reasons for the statistically significant differences between those with a clinical/subclinical ED and those without an ED. However, our findings alone do not help to provide an explanation for what is driving the statistically significant differences observed between those in the clinical and subclinical ED groups. Additionally, our findings collected from an online self-report questionnaire are subject to social desirability bias. Our sample excludes adolescents under 15 who can be affected by EDs and/or engage in ED-related social networking behavior (Wilksch, O'Shea, Ho, Byrne, & Wade, 2020). We also did not include different response options for transgender males versus transgender females, limiting the interpretation of our results related to this at-risk group (Coelho et al., 2019) and representing an important consideration for more accurate gender identity descriptions in future studies. Additionally, there are limitations associated with our use of targeted advertisements for recruitment, including the changing oSMP algorithms to target individuals that can occur sporadically unbeknownst to the researcher (Kayrouz, Dear, Karin, & Titov, 2016), and the fact that these advertisements did not exclusively target those who objectively posted or followed pro-ED content. There was not a distinction made between motives for posting versus following ED-related content when questioning participants. There are likely different motives and rewards associated with posting versus following content and these should be taken into consideration in future studies. Further, our study sample may not be representative of all ED-websites and online communities as individuals with alternative or more severe diagnoses than were most commonly seen in our study (UFED and subclinical BN) may contribute to these sites but may not have been as inclined to participate in research. Lastly, we do not know the specific impact of these self-reported advantages/disadvantages on the mental health/well-being of our participants. The questionnaire did not assess the direct short or long-term impacts of posting or following body image/ED related content online.
Implications for Future Research
Our findings can be beneficial for those designing future ED studies with social networking components. Our results can also inform the methodology of future research that aims to investigate how online social networks can improve body image, foster positive social support, and/or reduce eating disorder symptoms, perhaps through interventions that aim to increase critical thinking and media literacy skills used when viewing ED-related posts (McLean, Wertheim, Masters, & Paxton, 2017). These findings can help researchers integrate advantageous aspects of socially networking about this topic while minimizing the disadvantages in order to make the online environment more conducive for treatment and/or recovery. For example, researchers and social media administrators can increase language monitoring of tips and tricks for extreme weight loss, increase screening for pro-ED communities, images, and posts, and work to track increasing interest in and ultimately limit thinspiration content. Lastly, these findings can be utilized by practitioners to better understand why patients seek online social support and/or to assess whether the benefits outweigh the potential risks for persons with EDs.
Acknowledgments
Funding: This study was funded by the National Institute of Mental Health [R21 MH112331, R34 MH119170, K08 MH120341] and the National Institutes of Health [K02 DA043657].
Footnotes
There are no conflicts of interest to report.
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- Alhabash S, & Ma M (2017). A tale of four platforms: Motivations and uses of Facebook, Twitter, Instagram, and Snapchat among college students? Social Media+ Society, 3(1), 2056305117691544. [Google Scholar]
- Almenara CA, Machackova H, & Smahel D (2016). Individual Differences Associated with Exposure to “Ana-Mia” Websites: An Examination of Adolescents from 25 European Countries. Cyberpsychology, Behavior, and Social Networking, 19(8), 475–480. doi: 10.1089/cyber.2016.0098 [DOI] [PubMed] [Google Scholar]
- Borzekowski DL, Schenk S, Wilson JL, & Peebles R (2010). e-Ana and e-Mia: A content analysis of pro-eating disorder Web sites. Am J Public Health, 100(8), 1526–1534. doi: 10.2105/ajph.2009.172700 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Branley DB, & Covey J (2017). Pro-ana versus pro-recovery: A content analytic comparison of social media users’ communication about eating disorders on Twitter and Tumblr. Frontiers in psychology, 8, 1356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Caplan SE (2005). A social skill account of problematic Internet use. Journal of communication, 55(4), 721–736. [Google Scholar]
- Carter-Harris L, Bartlett Ellis R, Warrick A, & Rawl S (2016). Beyond Traditional Newspaper Advertisement: Leveraging Facebook-Targeted Advertisement to Recruit Long-Term Smokers for Research. J Med Internet Res, 18(6), e117. doi: 10.2196/jmir.5502 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Coelho JS, Suen J, Clark BA, Marshall SK, Geller J, & Lam P-Y (2019). Eating Disorder Diagnoses and Symptom Presentation in Transgender Youth: a Scoping Review. Current Psychiatry Reports, 21(11), 107. doi: 10.1007/s11920-019-1097-x [DOI] [PubMed] [Google Scholar]
- Elo S, & Kyngäs H (2008). The qualitative content analysis process. Journal of Advanced Nursing, 62(1), 107–115. doi: 10.1111/j.1365-2648.2007.04569.x [DOI] [PubMed] [Google Scholar]
- Fitzsimmons-Craft EE, Bardone-Cone AM, Bulik CM, Wonderlich SA, Crosby RD, & Engel SG (2014). Examining an elaborated sociocultural model of disordered eating among college women: The roles of social comparison and body surveillance. Body Image, 11(4), 488–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Graham AK, Trockel M, Weisman H, Fitzsimmons-Craft EE, Balantekin KN, Wilfley DE, & Taylor CB (2019). A screening tool for detecting eating disorder risk and diagnostic symptoms among college-age women. Journal of American College Health, 67(4), 357–366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gregertsen EC, Mandy W, & Serpell L (2017). The Egosyntonic Nature of Anorexia: An Impediment to Recovery in Anorexia Nervosa Treatment. Frontiers in psychology, 8, 2273–2273. doi: 10.3389/fpsyg.2017.02273 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Holland G, & Tiggemann M (2016). A systematic review of the impact of the use of social networking sites on body image and disordered eating outcomes. Body Image, 17, 100–110. [DOI] [PubMed] [Google Scholar]
- Juarascio AS, Shoaib A, & Timko CA (2010). Pro-eating disorder communities on social networking sites: a content analysis. Eating disorders, 18(5), 393–407. [DOI] [PubMed] [Google Scholar]
- Kayrouz R, Dear BF, Karin E, & Titov N (2016). Facebook as an effective recruitment strategy for mental health research of hard to reach populations. Internet Interventions, 4, 1–10. doi: 10.1016/j.invent.2016.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kelly Y, Zilanawala A, Booker C, & Sacker A (2018). Social Media Use and Adolescent Mental Health: Findings From the UK Millennium Cohort Study. EClinicalMedicine, 6, 59–68. doi: 10.1016/j.eclinm.2018.12.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kietzmann JH, Hermkens K, McCarthy IP, & Silvestre BS (2011). Social media? Get serious! Understanding the functional building blocks of social media. Business Horizons, 54(3), 241–251. [Google Scholar]
- Krippendorff K (2004). Reliability in content analysis: Some common misconceptions and recommendations. Human communication research, 30(3), 411–433. [Google Scholar]
- Krippendorff K (2011). Computing Krippendorff's alpha-reliability. [Google Scholar]
- Leiner DJ (2013). Too fast, too straight, too weird: Post hoc identification of meaningless data in internet surveys. SSRN Electronic Journal. [Google Scholar]
- McLean SA, Wertheim EH, Masters J, & Paxton SJ (2017). A pilot evaluation of a social media literacy intervention to reduce risk factors for eating disorders. International Journal of Eating Disorders, 50(7), 847–851. doi: 10.1002/eat.22708 [DOI] [PubMed] [Google Scholar]
- Mingoia J, Hutchinson AD, Wilson C, & Gleaves DH (2017). The relationship between social networking site use and the internalization of a thin ideal in females: A meta-analytic review. Frontiers in psychology, 8, 1351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pace U, D'Urso G, & Zappulla C (2018). Negative eating attitudes and behaviors among adolescents: The role of parental control and perceived peer support. Appetite, 121, 77–82. doi: 10.1016/j.appet.2017.11.001 [DOI] [PubMed] [Google Scholar]
- Park N, Kee KF, & Valenzuela S (2009). Being immersed in social networking environment: Facebook groups, uses and gratifications, and social outcomes. CyberPsychology & Behavior, 12(6), 729–733. [DOI] [PubMed] [Google Scholar]
- Peebles R, Wilson JL, Litt IF, Hardy KK, Lock JD, Mann JR, & Borzekowski DL (2012). Disordered eating in a digital age: eating behaviors, health, and quality of life in users of websites with pro-eating disorder content. J Med Internet Res, 14(5), e148. doi: 10.2196/jmir.2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Prochaska JO, & Velicer WF (1997). The Transtheoretical Model of Health Behavior Change. American Journal of Health Promotion, 12(1), 38–48. doi: 10.4278/0890-1171-12.1.38 [DOI] [PubMed] [Google Scholar]
- Raggatt M, Wright CJC, Carrotte E, Jenkinson R, Mulgrew K, Prichard I, & Lim MSC (2018). “I aspire to look and feel healthy like the posts convey”: engagement with fitness inspiration on social media and perceptions of its influence on health and wellbeing. BMC Public Health, 18(1), 1002. doi: 10.1186/s12889-018-5930-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ransom DC, La Guardia JG, Woody EZ, & Boyd JL (2010). Interpersonal interactions on online forums addressing eating concerns. Int J Eat Disord, 43(2), 161–170. doi: 10.1002/eat.20629 [DOI] [PubMed] [Google Scholar]
- Sarason IG, Levine HM, Basham RB, & Sarason BR (1983). Assessing social support: The social support questionnaire. Journal of personality and social psychology, 44(1), 127. [Google Scholar]
- Tong ST, Heinemann-Lafave D, Jeon J, Kolodziej-Smith R, & Warshay N (2013). The use of pro-ana blogs for online social support. Eating disorders, 21(5), 408–422. [DOI] [PubMed] [Google Scholar]
- Weissman RS (2019). The role of sociocultural factors in the etiology of eating disorders. Psychiatric Clinics, 42(1), 121–144. [DOI] [PubMed] [Google Scholar]
- Wilksch SM, O'Shea A, Ho P, Byrne S, & Wade TD (2020). The relationship between social media use and disordered eating in young adolescents. International Journal of Eating Disorders, 53(1), 96–106. doi: 10.1002/eat.23198 [DOI] [PubMed] [Google Scholar]
- Wilson JL, Peebles R, Hardy KK, & Litt IF (2006). Surfing for Thinness: A Pilot Study of Pro–Eating Disorder Web Site Usage in Adolescents With Eating Disorders. Pediatrics, 118(6), e1635. doi: 10.1542/peds.2006-1133 [DOI] [PubMed] [Google Scholar]