CoChat
Sign Up
Cited by 380Eating Disorders and BehaviorsNutrition, Health and Food BehaviorBody Image and Dysmorphia Studies

Instagram use is linked to increased symptoms of orthorexia nervosa.

Open access
Turner, Pixie G., Lefevre, Carmen E.·2017·Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity
DOI 10.1007/s40519-017-0364-2PMC PMC5440477
Open this paper in CoChat — free accountOpen PDF

Abstract

PURPOSE: Social media use is ever increasing amongst young adults and has previously been shown to have negative effects on body image, depression, social comparison, and disordered eating. One eating disorder of interest in this context is orthorexia nervosa, an obsession with eating healthily. High orthorexia nervosa prevalence has been found in populations who take an active interest in their health and body and is frequently comorbid with anorexia nervosa. Here, we investigate links between social media use, in particularly Instagram and orthorexia nervosa symptoms. METHODS: We conducted an online survey of social media users (N = 680) following health food accounts. We assessed their social media use, eating behaviours, and orthorexia nervosa symptoms using the ORTO-15 inventory. RESULTS: Higher Instagram use was associated with a greater tendency towards orthorexia nervosa, with no other social media channel having this effect. In exploratory analyses Twitter showed a small positive association with orthorexia symptoms. BMI and age had no association with orthorexia nervosa. The prevalence of orthorexia nervosa among the study population was 49%, which is significantly higher than the general population (<1%). CONCLUSIONS: Our results suggest that the healthy eating community on Instagram has a high prevalence of orthorexia symptoms, with higher Instagram use being linked to increased symptoms. These findings highlight the implications social media can have on psychological wellbeing, and the influence social media 'celebrities' may have over hundreds of thousands of individuals. These results may also have clinical implications for eating disorder development and recovery.

AI snapshot

Written from the paper’s full text.

TL;DR

In an online survey of 680 female social media users, higher Instagram use was associated with greater orthorexia nervosa symptoms as measured by ORTO-15. The sample showed a high prevalence of ON symptoms (49% using the <35 cut-off) compared with <1% in the general population.

Key findings

  • Instagram use was negatively correlated with ORTO-15 score (r = −0.10, p = 0.01), meaning higher Instagram use was linked to greater ON symptoms; in regression, Instagram use (β = −0.12, p = 0.003) and country of residence (β = 0.23, p < 0.001) were significant predictors.
  • ON prevalence in the sample was 49.0% with the <35 cut-off and 90.6% with the <40 cut-off; among Instagram users it was 49.3% and 90.4%, respectively, versus <1% in the general population.
  • No significant relationships were found between ORTO-15 and number of food types eaten (r = −0.01, p = 0.87), number of social media channels used (r < 0.01, p = 0.99), age (r = −0.06, p = 0.13), or BMI (r = −0.03, p = 0.47); diet types also did not differ [F(5,674) = 0.705, p = 0.620].
  • Twitter showed a small positive correlation with ORTO-15 score (r = 0.12, p = 0.04) in exploratory analysis; the paper describes this inconsistently as a positive association with ON symptoms in the abstract and as possibly protective in the discussion.

Methods

  • Online Qualtrics survey; participants were recruited via unpaid advertisements on Instagram, Facebook, and Twitter, plus the “Plantbased Pixie” blog and “Health Bloggers Community” newsletter. Data were collected from 713 participants (23 males, 686 females, 4 other/prefer not to say; mean age 24.96 ± 8.17 years); final analysis included 680 females after excluding non-female participants and missing ORTO-15 data.
  • Measures included social media channels and time spent, dietary choices (19 food types and diet label), ORTO-15, and demographics. ORTO-15 lower scores indicate higher ON symptoms; cut-offs of <40 and <35 were used.
  • Analyses used Spearman correlations for social media use and ORTO-15, Pearson correlations for age/BMI, one-way ANOVA for diet types, and linear regression with Instagram use, age, BMI, ethnicity, and country of residence.

Limitations

  • The paper states convenience non-random sampling, with 83% of participants recruited from the author’s Instagram account; the study title (“Health habits on social media”) may have attracted health-interested individuals; BMI was self-reported; medical diets were not assessed; and causality cannot be established due to the correlational design.
  • The paper also notes possible interactions among multiple social media channels and that the sample was restricted to females because the number of male participants was too low for adequate power.
  • The introduction notes ORTO-15 psychometric limitations, including internal consistency issues, lack of standardisation, cultural variation, and possible false positives from the high cut-off score.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1007/s40519-017-0364-2

Cites

37 of 37 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 12 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    The prevalence of orthorexia nervosa in the study sample was 90.6% using the ORTO-15 cut-off <40.

    “The prevalence of ON in our sample was 90.6% with a cut-off score of <40, or 49.0% with a cut-off score of <35.”
    Quote in the paperNot a substantial claim
  • Findingown data

    The prevalence of orthorexia nervosa in the study sample was 49.0% using the ORTO-15 cut-off <35.

    “The prevalence of ON in our sample was 90.6% with a cut-off score of <40, or 49.0% with a cut-off score of <35.”
    Quote in the paperNot a substantial claim
  • Findingown data

    ORTO-15 scores were not significantly correlated with number of social media channels used (r < 0.01, p = 0.99).

    “No relationship was found between ORTO-15 scores and number of food types eaten (r = −0.01; p = 0.87), number of social media channels used (r < 0.01; p = 0.99), age (r = −0.06; p = 0.13), or BMI (r = −0.03; p = 0.47).”
    Quote in the paperNot a substantial claim
  • Findingown data

    ORTO-15 scores were not significantly correlated with BMI (r = −0.03, p = 0.47).

    “No relationship was found between ORTO-15 scores and number of food types eaten (r = −0.01; p = 0.87), number of social media channels used (r < 0.01; p = 0.99), age (r = −0.06; p = 0.13), or BMI (r = −0.03; p = 0.47).”
    Quote in the paperNot a substantial claim
  • Findingown data

    Twitter use showed a small positive correlation with ORTO-15 score (r = 0.12, p = 0.04).

    “except Twitter which showed a small positive correlation (r = 0.12, p = 0.04).”
    Quote in the paperNot a substantial claim
  • May combine two claims. 6
  • Classified as probably not a substantial claim. 15
  • The quote only partly states this claim. 1

  • May combine two claims.

    Findingown data

    In the study sample of female social media users following health food accounts (N=680), higher Instagram use was associated with greater orthorexia nervosa symptoms, as indicated by a negative correlation between time spent on Instagram and ORTO-15 score (r = −0.10, p = 0.01).

    “Higher Instagram use was associated with a greater tendency towards orthorexia nervosa, with no other social media channel having this effect.”
    • Also stated in: § Inferential Results · plain
      “As predicted, there was a negative correlation between ORTO-15 score and Instagram use (r = −0.10; p = 0.01) (Fig. 1).”
    • Also stated in: § Discussion · plain
      “We found a significant relationship between symptoms of ON as measured by the ORTO-15 test and Instagram use, with higher Instagram use being associated with a greater tendency towards ON.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    No other social media channel (Facebook, Pinterest, Google+, Tumblr, LinkedIn) showed a significant relationship with ORTO-15 score, except Twitter.

    “with no other social media channel having this effect.”
    • Also stated in: § Inferential Results · plain
      “No significant relationship was found between ORTO-15 score and other social media use (Table 3), except Twitter which showed a small positive correlation (r = 0.12, p = 0.04).”
    • Also stated in: § Discussion · hedged
      “No other social media channels were found to have this effect, although Twitter seemed to have a small protective association.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationown data

    Twitter use had a small protective association with orthorexia nervosa symptoms (i.e., higher Twitter use associated with lower ON symptoms).

    “although Twitter seemed to have a small protective association.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The effect of Instagram use on orthorexia symptoms, though small, is meaningful at population level because of the large number of Instagram users.

    “While the effect size was small, the large population of social media users, now over 500 million on Instagram alone, means that this is a meaningful effect at population level.”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Interpretationown data

    Higher Instagram use is linked to increased symptoms of orthorexia nervosa (generalisation beyond the study sample).

    “with higher Instagram use being linked to increased symptoms.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductreasoning

    BMI was calculated from self-report rather than measured, so error and deliberate misreporting cannot be ruled out.

    “Furthermore, BMI was calculated from self-report rather than measured, accordingly, we cannot rule out error and deliberate miss-reporting.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    Participants were recruited through convenient non-random sampling, with the majority from the author's Instagram account (83%), which is not representative of the general population.

    “Participants were recruited through convenient non-random sampling, with the majority coming from the author’s Instagram account (83%), which is not representative of the general population.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Non-female participants were excluded because the number recruited was too low to provide adequate power for statistical analyses.

    “We decided to concentrate on female participants as the N recruited for men was too low to provide adequate power for statistical analyses.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    All participants gave informed consent and the study was approved by the UCL ethics committee.

    “All participants gave their informed consent before starting the questionnaire and the study was approved by the UCL ethics committee.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Two ORTO-15 cut-off scores (<40 and <35) were used due to disagreement in the literature over the appropriate cut-off point.

    “We used two cut-off scores due to disagreement in the literature over the appropriate cut-off point: less than 40 (following [28]), and less than 35 (following [35]).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [28] Arusoğlu G, Kabakci E, Köksal G, Merdol TK (2008) Orthorexia nervosa and adaptaon of ORTO-11 into Turkish. Turk J Psychiatry 19(3):283–291

      Not checked in the free run.

    • cites [35] Missbach B, Dunn TM, König JS (2016) We need new tools to assess Orthorexia Nervosa. A commentary on “Prevalence of Orthorexia Nervosa among College Students Based on Bratman’s Test and Associated Tendencies”. Appetite 108:521 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Social media use has been associated with higher levels of depression in young adults.

    “social media use has been associated with higher levels of depression in young adults [14]”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [14] Sidani JE, Shensa A, Radovic A, Miller E, Colditz JB, Hoffman BL, Giles LM, Primack BA (2016) Association between social media use and depression among US young adults. Depress Anxiety 33(4):323–331 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisenone given

    The relationship between orthorexia nervosa and social media use is currently unclear.

    “Currently the relationship between ON and social media use is unclear, although links between social media use and AN have been reported.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    The research title may have attracted individuals with an interest in health.

    “In addition, the research title (Health habits on social media) may have attracted individuals with an interest in health.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Social media use may contribute to an 'echo-chamber' effect, where users perceive their values and world-views to be more common than they actually are, due to selectively viewing contributions of other, similarly minded people.

    “Previous work around similar disorders suggests that social media use may contribute to an ‘echo-chamber’ effect, where users perceive their values and world-views to be more common than they actually are, due to selectively viewing contributions of other, similarly minded people [5].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] Salathé M, Khandelwal S (2011) Assessing vaccination sentiments with online social media: implications for infectious disease dynamics and control. PLoS Comput Biol 7(10):e1002199 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    High orthorexia nervosa prevalence has been found in populations who take an active interest in their health and body.

    “Higher prevalence has been found in yoga instructors [30], dieticians, [31], nutrition students [32], and exercise science students [33] compared to the general population.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [30] Valera JH, Ruiz PA, Valdespino BR, Visioli F (2014) Prevalence of orthorexia nervosa among ashtanga yoga practitioners: a pilot study. Eat Weight Disord-Stud Anorex Bulim Obes 19(4):469–472 source

      Not checked in the free run.

    • cites [31] Asil E, Sürücüoğlu MS (2015) Orthorexia nervosa in Turkish dietitians. Ecol Food Nutr 54(4):303–313 source

      Not checked in the free run.

    • cites [32] Bo S, Zoccali R, Ponzo V, Soldati L, De Carli L, Benso A, Fea E, Rainoldi A, Durazzo M, Fassino S (2014) University courses, eating problems and muscle dysmorphia: are there any associations? J Transl Med 12(1):1 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Linear regression was conducted to assess the robustness of the correlation between ORTO-15 and Instagram use when controlling for potentially confounding demographic variables.

    “To assess the robustness of the correlation between ORTO-15 and Instagram use when controlling for potentially confounding demographic variables, we additionally conducted linear regression analysis with ORTO-15 score as the outcome measure and Instagram use, age, BMI, ethnicity (dummy coded as white/non-white), and country of residence (dummy coded as UK/non-UK) as the predictor variables.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Social media use is increasing among young adults.

    “Social media use is constantly increasing, with now 90% of UK young adults (aged 16–34 years) accessing social media platforms [1].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [1] Ofcom (2015) Communications Market Report 2015. Ofcom

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    Social media use has been associated with eating disorders and related behaviours.

    “as well as eating disorders and related behaviours [15]”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [15] Carrotte ER, Vella AM, Lim MS (2015) Predictors of “liking” three types of health and fitness-related content on social media: a cross-sectional study. J Med Internet Res 17(8) source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The study did not assess whether some participants followed a strict diet for medical reasons.

    “Finally, we did not assess whether some participants followed a strict diet for medical reasons.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Orthorexia nervosa has some overlap with anorexia nervosa and obsessive compulsive disorder.

    “There is some overlap between ON, AN and obsessive compulsive disorder (OCD) [25].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [25] Koven NS, Abry AW (2015) The clinical basis of orthorexia nervosa: emerging perspectives. Neuropsychiatric Disease & Treatment 11

      Not checked in the free run.

20 of 20

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • A study of 713 participants confirmed orthorexia nervosa prevalence of 49%, far greater than the estimated <1% in the general UK population.

    “A study of 713 participants confirmed orthorexia nervosa prevalence of 49% [70], far greater than the estimated <1% in the general UK population.”

    From The social media diet: A scoping review to investigate the association between social media, body image and eating disorders amongst young people (2023) · cites this paper as [70]

    The Results (Sample characteristics; page unavailable) report 49.0% orthorexia prevalence using an ORTO-15 cutoff of <35. Methods (Participants; page unavailable) say 713 people provided data, but the analysis used 680; the Introduction cites a <1% general-population estimate, while the cited estimate is from a US sample, not the UK. (LLM, conf=0.90)

    Experimental — accuracy not yet measured

  • Disturbingly high prevalence of body image dissatisfaction, disordered eating pathology and comorbid mental health outcomes were reported amongst young social media users in this review.

    “Disturbingly high prevalence of body image dissatisfaction, disordered eating pathology and comorbid mental health outcomes were reported amongst young social media users in this review [57, 66, 70].”

    From The social media diet: A scoping review to investigate the association between social media, body image and eating disorders amongst young people (2023) · cites this paper as [70]

    The Results (page unavailable) report a high prevalence of orthorexia nervosa in a sample of social media users following health-food accounts, supporting the disordered-eating part of the claim. The Introduction discusses links between social media use, body image, and depression, but the paper does not report prevalence of body-image dissatisfaction or comorbid mental-health outcomes in its sample. Its findings also should not be generalized to young social media users overall. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • One cross-sectional and one qualitative study indicated orthorexia nervosa symptomatology amongst participants, ranging from obsessions with 'clean eating' to avoidance of 'demonised' foods and compulsive exercise behaviours.

    “One cross-sectional and one qualitative study indicated orthorexia nervosa symptomatology amongst participants, ranging from obsessions with ‘clean eating’ to avoidance of ‘demonised’ foods and compulsive exercise behaviours [64, 70].”

    From The social media diet: A scoping review to investigate the association between social media, body image and eating disorders amongst young people (2023) · cites this paper as [70]

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • ConsistentF(5, 674) = 0.705reported p = .62 · recomputed p = .620

    “F(5,674) = 0.705, p = .620”

  • ConsistentF(5, 648) = 9.58reported p < .001 · recomputed p < .001

    “F(5,648) = 9.58, p < .001”

2 of 2

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Average number of channels conflicts with channel-use percentages

    Minor

    The seven channel-use percentages in Table 2 imply an average of 3.405 channels per participant, rather than the reported 3.55. This difference is too large to result from rounding the percentages.

  • Twitter association is described in opposite directions

    Minor

    The discussion calls Twitter use positively associated with ON symptoms but also calls its association protective. The reported positive correlation is with ORTO-15 score, for which higher scores indicate fewer symptoms, supporting the protective description.

2 of 2

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Turner, P. G., & Lefevre, C. E.
Affiliations
University College London (GB)
Corresponding author
Lefevre, Carmen E.
ORCID
Lefevre, Carmen E.
Venue
Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity
Publisher
Springer Science and Business Media LLC
Published
2017-03-01
Volume
22
Issue
2
Pages
277-284
ISSN
  • 1124-4909 (ISSN-L)
  • 1590-1262
DOI
10.1007/s40519-017-0364-2
PMC
PMC5440477
OpenAlex
W2593519256
Semantic Scholar
8fa58fca216d463d9400de92c0164d84018f7dc8
Cited by
380
References
37
Citations by year
2026: 39 · 2025: 43 · 2024: 40 · 2023: 52 · 2022: 55 · 2021: 60 · 2020: 43 · 2019: 28 · 2018: 18 · 2017: 1
FWCI
30.45
Type
journal article
Language
en
Open access
Licence
Locations
  • Eating and Weight Disorders - Studies on Anorexia Bulimia and Obesity (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • UCL Discovery (University College London) (submittedVersion)
Topics
  • Social Sciences › Psychology › Clinical Psychology › Eating Disorders and Behaviors
  • Health Sciences › Nursing › Nutrition and Dietetics › Nutrition, Health and Food Behavior
  • Social Sciences › Psychology › Clinical Psychology › Body Image and Dysmorphia Studies
Keywords
orthorexia nervosa, Instagram, social media use, eating disorders, disordered eating, young adults, ORTO-15, Twitter, body image, social media influencers
MeSH terms
Adolescent, Adult, Feeding and Eating Disorders, Body Image, Feeding Behavior, Female, Humans, Health Behavior, Young Adult, Social Media, Symptom Assessment

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-08.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-08.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review