Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Nov 1.
Published in final edited form as: J Psychopathol Clin Sci. 2023 Nov;132(8):1043–1050. doi: 10.1037/abn0000861

Prospective Reciprocal Relations between Social Support and Eating Disorder Symptoms

May Stern 1,*, Laura Rubino 1,*, Chris Desjardins 2, Eric Stice 1
PMCID: PMC10683857  NIHMSID: NIHMS1936370  PMID: 38010772

Abstract

Prospective studies have found inconsistent relations between social support deficits and future increases in eating disorder symptoms. Further, no prospective study has tested whether elevated eating disorder symptoms predict a future erosion of social support. Accordingly, the current study investigated the prospective reciprocal relations between perceived social support from both parents and peers and eating disorder symptoms in adolescent girls. In this study, 496 adolescent girls reported perceived social support and completed an eating disorder diagnostic interview annually for 7 years. Deficits in perceived peer, but not parental, support predicted future increases in eating disorder symptoms (p = .019, partial r = −0.10). Further, initial eating disorder symptoms predicted future reductions in perceived peer support (p = .016, partial r = −0.11) but not parental support. Interestingly, these relations became non-significant when we controlled for negative affect and Body Mass Index (BMI), suggesting that comorbid mood disorders and elevated body weight might partially drive these relations. Although both relations were small in magnitude, these results suggest low perceived peer support is a risk factor for future escalations in eating disorder symptoms and that elevated symptoms may contribute to a further erosion of peer support, creating a cyclical relation that maintains eating pathology. Conversely, high levels of perceived peer support could serve as a protective factor against future increases in eating pathology. These findings should advance interpersonal theories of eating pathology and inform the design of more effective prevention programs.

General Scientific Summary

Prospective data from a large sample suggest there are prospective reciprocal relations between perceived peer support and eating disorder symptoms in adolescent females. Findings provide evidence that low peer support, but not low parental support, is a risk factor for eating pathology and that elevated eating disorder symptoms predicted a future erosion of peer support, but not parental support. Results also suggest that future research should investigate whether negative affect is a mediator between social support and eating pathology. These novel findings should advance current intervention efforts by attempting to foster and maintain peer networks in adolescents.

Keywords: social support, eating disorders, risk factors, prospective, prevention, protective factor

Introduction

Eating disorders are serious, life-threatening illnesses associated with functional impairment, substance abuse, suicide, and lifelong medical complications (Casiero & Frishman, 2006; Keski-Rahkonen & Mustelin, 2016). One eating disorder, Anorexia Nervosa (AN), has the second highest mortality rate of any mental illness (Chesney et al., 2014). Regrettably, 80% of individuals with an eating disorder do not receive treatment. Further, treatment for the full spectrum of eating disorders produce lasting symptom remission for fewer than 35% of individuals (Fairburn et al., 2015). Given the severity of these illnesses and the limited efficacy of current treatments it is imperative to advance knowledge on the risk and maintenance factors of eating disorders to better inform intervention efforts.

Social support has frequently been considered when investigating risk and protective factors for eating disorders and provides a promising future focus for prevention and treatment interventions. Generally, higher social support is an established protective factor for psychopathology (Scardera et al., 2020) but to our knowledge, only a few studies have used prospective data to analyze the relation between social support and future change in eating disorder symptoms and these studies have reported mixed findings. Linville et al. (2012) did not find peer and parental support predicted changes in bulimic behaviors during later adolescence. However, Stice et al. (2002) found low social support from peers, but not from parents, predicted onset of binge eating 20 months later during early adolescence. Contrarily, Kirsch et al. (2016) found parental but not peer support predicted future increases in disordered eating symptoms.

Prospective studies which did not distinguish types of support found lower levels of overall social support predicted increased bulimic symptoms eight weeks later (Bodell et al., 2011) and uniquely predicted eating disturbances six weeks later (Jackson et al., 2005). These findings suggest it may be important to distinguish between different types of social support. Depression research suggests different types of support can have distinct impacts on psychopathology and psychopathology can differentially predict future change in types of support. Stice and colleagues (2004) found deficits in parental, but not peer, support predicted future increases in depressive symptoms, and depression symptoms predicted future reductions in peer but not parental support. Consequently, it seems useful to parse the impact of varying types of support.

It also seems important to investigate whether high eating disorder symptoms contribute to a future erosion in social support. A cross-sectional study found more frequent bulimic episodes correlated with a lower intent to continue residing with that individual by their college roommate (Kwan et al., 2017). Furthermore, caregivers of those suffering from eating disorders experience clinically significant levels of distress (Treasure et al., 2001). Additionally, anecdotal evidence from participants in Body Project groups (Stice), provide support for the sentiment that people often do not appreciate peers who consistently seek reassurance about their appearance and fixate on their appearance and weight, potentially resulting in erosion of social support. Although cross-sectional studies have documented that social support deficits correlate with eating disorder symptoms (e.g., Wonderlich-Tierney & Vander Wal, 2010), these studies cannot determine the direction of the relationship.

Current Study

This prospective study addresses a gap in the literature by specifically investigating the relation of both peer and parental support separately to future changes in eating disorder symptoms, as well as whether baseline eating disorder symptoms predict future changes in peer and parental support. We addressed these questions with a larger sample of adolescent girls followed over a 7-year period and used masked diagnostic interviews which is the gold-standard for assessing eating disorder symptoms. We hypothesize that deficits in both peer and parental support will predict future increases in eating disorder symptoms and eating disorder symptoms will predict an erosion in peer, but not parental, support.

Methods

Participants and Procedures

The Austin Adolescent Development Study (completed prior to ClinicalTrials.gov) collected data from 496 adolescent girls (M age = 13.02) who completed diagnostic interviews and surveys at baseline and 7 annual follow-up assessments. Participants in this sample identified as White (68%), Black (7%), Asian (2%), Native American (1%), Pacific Islander (1%), Hispanic (18%) or Other (4%). The University of Texas at Austin IRB approved this study (study title: Etiology of Bulimic Pathology: Multimethod Investigation). Participants were recruited from four public and four private middle schools in (Austin, Texas). An informed consent letter and self-addressed return envelope were sent to parents of girls at the schools. The study was described as an investigation of physical and mental health among adolescent girls. Between 2% and 10% of participants did not provide data at one or more of the follow-up assessments. However, 99% of participants provided data for at least two assessments (Stice et al., 2013).

Perceived Social Support

Social support was measured with 12 items from the Network of Relationships Inventory (Furman & Buhrmester, 1985) that assessed perceived companionship, guidance, intimacy, affection, admiration, and reliable alliance from parents/guardians and peers, which were averaged to form parent and peer support scales. These scales have shown internal consistency (M α = .88) and 1 month test–retest reliability (M r = .69), and predictive validity (Bernasco et al., 2021; Presnell et al., 2004; Stice et al., 1998, 2002, 2004; Stice & Bearman, 2001).

Eating pathology

The semi-structured Eating Disorder Diagnostic Interview (EDDI; Stice et al., 2017) assessed eating disorder symptoms over the past 3 months at baseline and since the past interview at follow-ups on a month-by-month basis using time-line follow-back. The following symptoms were assessed on a month-by-month basis: frequency of binge eating, vomiting, laxative/diuretic use, fasting, and excessive exercise; overvaluation of weight/shape, frequency of feeling fat, and fear of weight gain were rated on 0–6-point scales. Participants who endorsed binge eating were asked about distress regarding binge eating, rapid eating, eating until uncomfortably full, eating large quantities of food when not hungry, eating alone because of embarrassment, feeling disgusted, depressed, or guilty after overeating, using yes or no ratings. Fasting reflected the number of times in the past three months two meals in a row were skipped “as a means of controlling your shape or weight,” and excessive exercise reflected the number of times in the past 3 months of vigorous exercise was used “to compensate for overeating.” We used these items to create an overall eating disorder symptom composite score at each assessment. The lowest a composite score could be is 0, indicating that the person does not engage in any behavioral or cognitive symptoms of an eating disorder, and theoretically there is no upper limit if a person engages in extreme behavioral symptoms. Assessors attended 24 hours of training and demonstrated inter-rater agreement (k>.80) with supervisors on 12 tape-recorded interviews prior to collecting data. This composite has been shown internal consistency (α = .92), interrater agree- ment (ICC r .93), 1-week test–retest reliability (ICC r = .95), predictive validity, and sensitivity to detecting intervention effects (Burton & Stice, 2006; Stice et al., 2009). The symptom composite showed internal consistency (mean α = .70), interrater agreement (ICC = .96; n = 116), and 1-week test–retest reliability (ICC = .96; n = 109).

During post hoc analyses, we created subscales assessing behavioral and cognitive symptoms separately. For behavioral symptoms, items assessing symptoms in the past 3 months were summed. For cognitive symptoms, scores were averaged over the past 3 months. The average composite reliability was 0.43 (SD = 0.06) for the behavioral scale and 0.77 (SD = 0.03) for the cognitive scale.

Transparency and openness

Please contact the corresponding author for the data and study material (e.g., copies of assessment measures) or the third author (cddesjardins@gmail.com) for analysis code.

Statistical Methods

Missingness

Prior to statistical modeling, multiple imputation was used to impute 50 data sets (Rubin, 1974). Missingness for study variables ranged from 0% to 10.7% (peer support in Year 7) and 399 of the participants (80.4%) had complete data. In addition to the variables used in the analyses, the multiple imputation model included negative affect, body mass index (BMI), dietary restraint, body dissatisfaction, current dieting, positive expectancies regarding thinness, parental perceived pressure for thinness, and peer modeling of eating pathology; these measures and their psychometrics are described elsewhere (Yamamiya & Stice, 2022). Imputation was performed using the mice package (van Buuren & Groothuis-Oudshoorn, 2011) and all imputation controlled for the multilevel nature of the data using a two-level normal model with heterogenous group variances for imputation. Imputed variables were post-processed to ensure they corresponded to plausible values on the scales (i.e., out of bounds values were recoded as the minimum and maximum, respectively).

Growth Curve Modeling

We began by fitting separate linear growth curve models for perceived parental support, perceived peer support, and eating disorder symptoms. These models enabled us to control for the lack of independence (i.e., the repeated measures on each participant) and to model change over the 7-year study. All models included random intercepts and slopes to account for inter-participant variation at baseline and change over follow-up. Higher order polynomial growth curve models were also fit to examine the plausibility of nonlinear change. Models were fit to the 50 imputed datasets and pooled using Rubin’s rules (Rubin, 2004).

Reciprocal Effects Modeling

Next, we extracted the random intercept and slopes for each of the 50 models for each of our three outcomes and pooled the random effects. Using these pooled random effects, we fit four reciprocal effects regression models. The four models involved: (1) regression of the random slope of perceived parental support onto the random intercept of perceived parental support and the random intercept of eating disorder symptoms; (2) regression of the random slope of eating disorder symptoms onto the random intercept of eating disorder symptoms and the random intercept of perceived parental support; (3) regression of the random slope of perceived peer support onto the random intercept of perceived peer support and the random intercept of eating disorder symptoms; and (4) regression of the random slope of eating disorder symptoms onto the random intercept of eating disorder symptoms and the random intercept of perceived peer support.

Results

Summary statistics

Summary statistics for the three outcome measures are shown in Table 1 and mean change over the 7-year follow-up are shown in Figure 1. In general, perceived parental support and eating disorder symptoms increased during adolescence, while peer support decreased. Change across time was best explained by a linear growth curve, except for perceived parental support which was best modeled with a quadratic growth curve. However, because results did not change when perceived parental support was modeled with a linear growth curve, we present the simpler linear growth curve model for ease of interpretation.

Table 1.

Summary Statistics by Measure Across Time

Outcome Measures
Time (in years) Parental Support Peer Support Eating Disorder Symptoms
1 3.90 (0.92) 4.35 (0.74) 0.41 (0.65)
2 3.88 (0.92) 4.36 (0.76) 0.53 (0.62)
3 3.85 (0.93) 4.24 (0.82) 0.53 (0.53)
4 3.93 (0.91) 4.25 (0.77) 0.53 (0.58)
5 4.04 (0.86) 4.23 (0.74) 0.61 (0.73)
6 4.10 (0.89) 4.16 (0.83) 0.69 (0.68)
7 4.21 (0.83) 4.18 (0.81) 0.69 (0.69)
Open in a new tab

Note: Means and Standard Deviations are reported as raw scores; Means (Standard Deviation)

Figure 1.

Figure 1.

Open in a new tab

Connected means plot for parental support, peer support, and log-transformed eating disorder symptoms from Year 1 to Year 7.

Correlations of random effects

Table 2 shows the correlations between the random effects. For each measure, initial values (random intercepts) were moderately correlated with future change (random slopes) in that measure. This was strongest for perceived parental support (r = −.62). The strongest correlation for the random intercepts was between initial eating disorder symptoms and perceived parental support (r = −0.22).

Table 2.

Correlations of random effects used in multiple linear regression models.

1. 2. 3. 4. 5. Mean SD
1. Parental support intercept 1 3.73 0.73
2. Parental support slope −.62 1 0.05 0.07
3. Peer support intercept .17 −.06 1 4.42 0.42
4. Peer support slope .09 .16 −.37 1 −0.04 0.05
5. Eating disorder symptoms intercept −.22 .14 −.17 −.04 1 0.26 0.24
6. Eating disorder symptoms slope .08 −.09 .00 −.03 −.53 0.03 0.03
Open in a new tab

Note: Mean is the sum of the fixed and random effect.

Testing reciprocal effects of parental support and eating disorder symptoms

Results from the regression models are shown in Table 3. Initial perceived parental support (p < .001, partial r = −.61), but not initial eating disorder symptoms (p = .842, partial r = .01) predicted change in perceived parental support over the 7-year follow-up. Similarly, initial eating disorder symptoms (p < .001, partial r = −.53), but not initial perceived parental support (p = .417, partial r = −04) predicted change in eating disorder symptoms over the 7-year follow-up. These findings suggest no prospective reciprocal effects of perceived parental support and eating disorder symptoms. There was a statistically significant difference in the magnitude of the slopes (x2 = 5.91, df = 1, p = .015), wherein the slope of perceived parental versus perceived peer support was significantly steeper. When models were fit with initial values of eating symptoms, the significant difference of the slopes’ magnitude persisted (x2 = 4.17, df = 1, p = .041).

Table 3.

Estimated parameters from the reciprocal effects multiple regression models.

Outcome Predictor Estimate SE t-value p-value partial r
(95% CI)
Change in Parental Support Initial Parental Support −0.056 0.003 −17.026 < .001 −.61
(−.66, −.55)
Initial ED Symptoms 0.003 0.010 0.255 .799 .01
(−.08, .10)
Change in ED Symptoms Initial Parental Support −0.001 0.002 −0.890 .374 −.04
(−.13, .05)
Initial ED Symptoms −0.065 0.005 −13.257 < .001 −.51
(−.57, −.44)
Change in Peer Support Initial Peer Support −0.044 0.005 −9.640 <.001 −.40
(−.47, −.32)
Initial ED Symptoms −0.020 0.008 −2.408 .016 −.11
(−.19, −.02)
Change in ED Symptoms Initial Peer Support −0.006 0.003 −2.346 .019 −.10
(−.19, −.02)
Initial ED Symptoms −0.066 0.005 −13.654 <.001 −.52
(−.58, −.46)
Open in a new tab

Note: ED = Eating Disorder; SE = standard error; Initial refers to random intercept; and Change refers to random slope.

Testing reciprocal effects of peer support and eating disorder symptoms

In contrast, we found evidence of prospective reciprocal relations between perceived peer support and eating disorder symptoms. Initial perceived peer support (p < .001, partial r = −.38) and eating disorder symptoms (p = .017, partial r = −.11) predicted change in peer support over the 7-year follow-up. Similarly, initial eating disorder symptoms (p < .001, partial r = −.54) and initial perceived peer support (p = .022, partial r = −.10) predicted change in eating disorder symptoms over the 7-year follow-up. These findings indicate that deficits in perceived peer support predicts furture increases in eating disorder symptoms and that initial elevations in eating disorder symptoms predicts future decreases in perceived peer support. These results suggest there are reciprocal effects between perceived peer support and eating disorder symptoms.

Sensitivity analysis

To assess the robustness of our findings, we performed four sensitivity analyses. First, we examined whether initial perceived peer support continued to predict change in eating disorder symptoms while controlling for perceived parental support and whether there was an interactive effect of these two supports on change in eating disorder symptoms. After controlling for initial perceived parental support, initial perceived peer support continued to predict future change in eating disorder symptoms (p = .028); there was no interaction between perceived parental and peer support (p = .523).

Second, we examined whether BMI at baseline predicted change in eating disorder symptoms controlling for initial eating disorder symptoms. There was no evidence of this (p = .838).

Third, we separated the eating disorder symptom composite into a cognitive and a behavioral scale. The results were qualitatively the same. That is there was no reciprocal effect for perceived parental support, but there was a reciprocal effect for perceived peer support. For the behavioral scale, both reciprocal effects were significant (p = .022 for initial perceived peer support predicting change in behavioral symptoms and p = 0.017 for initial behavioral symptoms predicting change in perceived peer support). For the cognitive scale, initial cognitive symptoms was a marginally significant predictor of change in perceived peer support (p = .069), while initial perceived peer support was a significant predictor of change in cognitive symptoms (p = 0.049). It should be noted that compared to the original scale, these two composites had much lower internal consistency (0.43 for the behavioral symptom scale and 0.77 for the cognitive symptom scale).

Finally, we assessed whether reciprocal effects for peer symptom and overall eating disorder symptoms persisted after controlling for baseline BMI and negative affect. For the effect of baseline eating disorder symptoms predicting change in perceived peer support, this effect becomes non-significant when controlling for just BMI (p = .073) and remains non-significant when controlling for BMI and negative affect (p=.178). In contrast, the effect of baseline perceived peer support predicting change in eating disorder symptoms remains significant when controlling for just BMI (p = .028), but became non-significant once negative affect was also controlled for (p=.086). In both cases, the effects become non-significant when just controlling for negative affect alone.

Discussion

Social support is commonly investigated as both a protective and risk factor for various psychopathologies. In the current study we investigated the prospective reciprocal relations between both perceived peer and parental support and eating disorder symptoms. Our hypotheses were partially supported. We found that deficits in perceived peer, but not perceived parental, support predicted future increases in eating disorder symptoms. Second, supporting our hypothesis, initial eating disorder symptoms predicted a reduction in perceived peer, but not perceived parental, support. Further, in exploratory analysis, we found the autoregressive relation of initial levels to future change in support were larger for perceived parental than for perceived peer support. Findings suggest that although perceived parental support remains stable during adolescence, perceived peer support decreases during this developmental period. Together, these findings provide insight into low perceived peer support as a potential risk factor for the development and maintenance of eating pathology and will help inform intervention efforts.

One reason we might have seen a relationship between perceived peer, but not perceived parental, support and eating disorder symptoms is that parental support may be more stable throughout this developmental period. Consistent with this idea, the auto-regressive effects of initial levels and change in support were larger for parental than for peer support. That is, the temporal stability of parental support was about 50% greater than for peer support. This converges with evidence that sociometric measures (i.e., being popular, accepted, neglected, or rejected) by peers are unstable in both the short-term and long-term (Jiang & Cillessen, 2005). Another study found that initial depressive symptoms predict future decreases in peer support but not parental support (Stice et al., 2004). Thus, in both studies initial adolescent psychopathology predicted reductions in peer but not parental support. Theoretically, parental support may be less susceptible to change. Familial ties might make parents more willing than peers to maintain consistent support when an adolescent displays interpersonal difficulties. Interestingly, we found that while both eating disorder symptoms and perceived parental support increased during adolescence, perceived peer support decreased during this developmental period. This suggests that declining levels of peer support play a significant role in the timing of the emergence of eating pathology. As adolescence is also a period during which eating disorder symptoms increase (Stice et al., 2013), lower levels of peer support may be a contributing factor.

Relatedly, initial deficits in perceived peer social support predicted future increases in eating disorder symptoms. Research has examined the importance of peer social support on adolescent development (Brown & Larson, 2009). Adolescents almost double the time they spend with their peers (Collins & Steinberg, 2006) and increasingly rely on peers to build selfesteem and test their sense of identity (Albarello et al., 2018; Tahir et al., 2015). Research has found that high levels of peer support is a protective factor against psychopathology (Scardera et al., 2020), suggesting adolescents may benefit from higher levels of peer support during this developmental period making a decrease in peer support even more detrimental to their eating pathology. Moreover, lower peer support during adolescence has been linked to increases in various psychopathologies (Czyz et al., 2012; Jacobson & Newman, 2016), and specifically has predicted higher levels of body dissatisfaction (Kirsch et al., 2016). Research has found that psychosocial impairment with family, friends, and schoolmates increase the likelihood that adolescent girls will pursue the appearance ideal (Stice & Van Ryzin, 2018), potentially to gain greater social acceptance.

Pathological behaviors may increase the probability of further erosion of peer support. In the current analyses, higher levels of eating disorder symptoms at year 1 predicted future decreases in perceived social support over the 7-year follow-up. This provides support for the anecdotal evidence from patients in Body Project groups (Stice), which suggests that youth can be put off by peers who are constantly seeking reassurance about their appearance and are overly focused on weight and dieting. These behaviors may cause an erosion of social support. This converges with work done by Rayner and colleagues (2013), which found that adolescents with lower levels of dieting had the most friendship nominations, suggesting that individuals who are overly focused on their appearance and weight may alienate themselves from their sources of peer support (Rayner et al., 2013). This may create a cyclical maintaining relationship in which greater eating disorder symptoms contribute to a reduction in peer support which in turn increases risk for a future escalation in eating disorder symptoms. The evidence that elevated eating disorder symptoms predict future decreases in social support dovetails with evidence from a recent randomized trial comparing two transdiagnostic treatments, which found that the intervention that produced significantly larger reductions in eating disorder symptoms (Body Project Treatment) also produced significantly greater improvements in social functioning than the alternative treatment (Interpersonal Psychotherapy; Stice et al., 2023). However, these results are unexpected given that Interpersonal Psychotherapy, which focuses on improving social functioning, produced significantly weeker reductions in both eating disorder symptoms and weaker improvements in social functioning. Although the current prospective effects were relatively small, results suggest it might be useful for treatment to simultaneously focus on improving peer support and reducing eating disorder symptoms.

The current results suggest peer support might act as a mediator between interpersonal difficulties and eating pathology, potentially providing an explanation as to why interpersonal difficulties increase risk for future onset of anorexia nervosa, bulimia nervosa, binge eating disorder, and purging disorder (Stice et al., 2017). These findings are consistent with the Interpersonal Theory of eating disorders and provide support for interventions focused on mechanisms that foster and maintain peer social relationships for adolescents with eating disorders symptoms. For example, Dialectical Behavioral Therapy (DBT) which integrates cognitive behavioral and mindfulness-based strategies to increase interpersonal effectiveness, emotion regulation, and distress tolerance to help adolescents, in part, make and maintain relationships (Lenz et al., 2016; Rathus & Miller, 2014; Vogel et al., 2021) has shown promise in the treatment of bulimia nervosa (Safer et al., 2001) and binge eating disorder (Telch et al., 2001). The current findings support calls (Reilly et al., 2020) for further research into the effectiveness of DBT for treatment of eating disorders.

A sensitivity analysis indicated that when baseline BMI and negative affect were included as covariates, the reciprocal effects of eating disorder symptoms and peer support became non-significant. Further sensitivity analyses were conducted to test the effects of controlling for BMI and negative affect individually. Interestingly, when BMI and negative affect were controlled for separately, eating disorder symptoms no longer significantly predicted change in perceived peer support. In contrast, when controlling for BMI alone perceived peer support continued to predict change in eating disorder symptoms but when negative affect alone was controlled for these results became non-significant. Collectively, these sensitivity analyses may suggest that the comorbidity between eating pathology and affective disturbances (Kitsantas et al., 2003) may have partially driven the reported prospective reciprocal relations between perceived peer support and eating disorder symptoms.

In the current analyses, we found that baseline eating disorder symptoms predicts a future decrease in perceived peer support. We also found that this effect dropped to a marginal effect after controlling for BMI, suggesting that adolescents with a higher BMI are potentially being rejected by their peers and that the predictive relations disappeared because eating pathology is comorbid with higher body weights (e.g., Albano et al., 2019; Gunnarsdottir et al., 2012; Strauss & Pollack, 2003).

Another interpretation of these results is that negative affect acts as a mediator between social support and eating pathology. While this could not be tested in the current study, it merits investigation in future research. In our sample, increased negative affect predicted lower peer support and increased eating disorder symptoms. This aligns with previous findings that low social support leads to negative affect, which is known to increase risk for onset of eating pathology (Stice et al., 2011; Pössel et a., 2018; Santos et al., 2007). Along similar lines, eating pathology (e.g., binge eating) may contribute to increases in BMI and, thus, rejection of social support, potentially producing the reciprocal relationship between social support and eating disorder symptoms found in our initial analyses.

Our results suggest that there may be a relationship between increased eating disorder symptoms and erosion of peer support, creating a maintenance cycle. Future research should investigate which aspects of peer relationships are most impactful on disordered eating so that interventions can be designed to target these relational gaps to help break this maladaptive cycle. Additionally, research should investigate which mechanisms are important for the development and maintenance of secure peer relationships. Prevention efforts designed to provide tools that cultivate and maintain stronger connectivity with and social support from peers, might be more effective than prevention programs that do not.

This study had several limitations. First, while we differentiated between perceived peer and parental support, we did not differentiate between maternal and paternal support. Given that distinct types of parental support are sometimes found to have differential impact on adolescent psychopathology (Pan et al., 2016), this will be an important distinction for future research. Second, our sample consisted only of female adolescents and the majority were white, which limits generalizability. Third, it is not possible to rule out the possibility that some omitted third variable explains the observed prospective relations. Future research should experimentally manipulate social support so that firmer inferences can be drawn, as recently done by Stice and colleagues (Stice et al., 2023). However, this study also had strengths. First, we used a diagnostic interview to assess eating disorder symptoms instead of a self-report measure. Second, we collected data over seven years which allowed us to capture most of adolescent development. Third, we parsed out the differences between peer and parental support on eating disorder symptoms. Finally, we used measures with established reliability and validity.

Conclusion

The current study provides support for a potential prospective cyclical relationship between levels of perceived peer support and eating disorder symptoms. These results have implications for eating disorder intervention efforts. Prevention and treatment options focusing on mechanisms that help foster connection and support from peers may be more effective than those that do not. Future research should investigate potential mechanisms between levels of peer support and eating disorder symptoms to better inform interventions. Future research should also investigate the importance of different types of parental support on eating disorder symptoms. It might also be helpful to parse the elements of familial support that are most important in protecting against disordered eating, such as reliable alliance versus instrumental support. Additionally, to provide an experimental test of the relation between peer social support and future change in eating disorder symptoms, future randomized trials should evaluate an eating disorder prevention program designed to improve mechanisms that build social connectivity with peers and others. The current study adds to the body of literature on the impact of perceived social support on adolescent well-being and fills a gap by prospectively exploring the specific relationship between social support and eating disorder symptoms.

Funding Statement:

This research was supported by NIH grants (grant numbers MH01708 and MH64560).

Footnotes

We have no known conflict of interest to disclose.

References

  1. Albarello F, Crocetti E, & Rubini M (2018). I and Us: A Longitudinal Study on the Interplay of Personal and Social Identity in Adolescence. Journal of youth and adolescence, 47(4), 689–702. 10.1007/s10964-017-0791-4 [DOI] [PubMed] [Google Scholar]
  2. Albano G, Rowlands K, Baciadonna L, Coco GL, & Cardi V (2019). Interpersonal difficulties in obesity: A systematic review and meta-analysis to inform a rejection sensitivity-based model. Neuroscience & Biobehavioral Reviews, 107, 846–861. [DOI] [PubMed] [Google Scholar]
  3. Bernasco EL, Nelemans SA, van der Graaff J, & Branje S (2021). Friend support and internalizing symptoms in early adolescence during COVID-19. Journal of Research on Adolescence, 31(3), 692–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bodell LP, Smith AR, Holm-Denoma JM, Gordon KH, & Joiner TE (2011). The impact of perceived social support and negative life events on bulimic symptoms. Eating behaviors, 12(1), 44–48. 10.1016/j.eatbeh.2010.11.002 [DOI] [PubMed] [Google Scholar]
  5. Brown BB, & Larson J (2009). Peer relationships in adolescence. In Lerner RM & Steinberg L (Eds.), Handbook of adolescent psychology: Contextual influences on adolescent development (pp. 74–103). John Wiley & Sons, Inc.. 10.1002/9780470479193.adlpsy002004 [DOI] [Google Scholar]
  6. Burton E, & Stice E (2006). Evaluation of a healthy-weight treatment program for bulimia nervosa: a preliminary randomized trial. Behaviour research and therapy, 44(12), 1727–1738. 10.1016/j.brat.2005.12.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Casiero D, & Frishman WH (2006). Cardiovascular complications of eating disorders. Cardiology in review, 14(5), 227–231. 10.1097/01.crd.0000216745.96062.7c [DOI] [PubMed] [Google Scholar]
  8. Chesney E, Goodwin GM, & Fazel S (2014). Risks of all-cause and suicide mortality in mental disorders: a meta-review. World psychiatry : official journal of the World Psychiatric Association (WPA), 13(2), 153–160. 10.1002/wps.20128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Collins WA, & Steinberg L (2006). Adolescent Development in Interpersonal Context. In Eisenberg N, Damon W, & Lerner RM (Eds.), Handbook of child psychology: Social, emotional, and personality development (pp. 1003–1067). John Wiley & Sons, Inc.. [Google Scholar]
  10. Czyz EK, Liu Z, & King CA (2012). Social connectedness and one-year trajectories among suicidal adolescents following psychiatric hospitalization. Journal of Clinical Child & Adolescent Psychology, 41(2), 214–226. 10.1080/15374416.2012.651998 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fairburn CG, Bailey-Straebler S, Basden S, Doll HA, Jones R, Murphy R, O'Connor ME, & Cooper Z (2015). A transdiagnostic comparison of enhanced cognitive behaviour therapy (CBT-E) and interpersonal psychotherapy in the treatment of eating disorders. Behaviour research and therapy, 70, 64–71. 10.1016/j.brat.2015.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Furman W, & Buhrmester D (1985). Children's perceptions of the personal relationships in their social networks. Developmental psychology, 21(6), 1016. 10.1037/0012-1649.21.6.1016 [DOI] [Google Scholar]
  13. Gunnarsdottir T, Njardvik U, Olafsdottir AS, Craighead LW, & Bjarnason R (2012). Teasing and social rejection among obese children enrolling in family-based behavioural treatment: effects on psychological adjustment and academic competencies. International Journal of Obesity, 36(1), 35–44. [DOI] [PubMed] [Google Scholar]
  14. Jackson T, Weiss KE, Lunquist JJ, & Soderlind A (2005). Sociotropy and perceptions of interpersonal relationships as predictors of eating disturbances among college women: Two prospective studies. The Journal of genetic psychology, 166(3), 346–360. 10.3200/GNTP.166.3.346-360 [DOI] [PubMed] [Google Scholar]
  15. Jacobson NC, & Newman MG (2016). Perceptions of close and group relationships mediate the relationship between anxiety and depression over a decade later. Depression and anxiety, 33(1), 66–74. 10.1002/da.22402 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jiang XL, & Cillessen AH (2005). Stability of continuous measures of sociometric status: A meta-analysis. Developmental Review, 25(1), 1–25. 10.1016/j.dr.2004.08.008 [DOI] [Google Scholar]
  17. Keski-Rahkonen A, & Mustelin L (2016). Epidemiology of eating disorders in Europe: prevalence, incidence, comorbidity, course, consequences, and risk factors. Current opinion in psychiatry, 29(6), 340–345. 10.1097/YCO.0000000000000278 [DOI] [PubMed] [Google Scholar]
  18. Kirsch AC, Shapiro JB, Conley CS, & Heinrichs G (2016). Explaining the pathway from familial and peer social support to disordered eating: Is body dissatisfaction the link for male and female adolescents?. Eating behaviors, 22, 175–181. 10.1016/j.eatbeh.2016.06.018 [DOI] [PubMed] [Google Scholar]
  19. Kitsantas A, Gilligan TD, & Kamata A (2003). College women with eating disorders: Self-regulation, life satisfaction, and positive/negative affect. The Journal of Psychology, 137(4), 381–395. [DOI] [PubMed] [Google Scholar]
  20. Kwan MY, Minnich AM, Douglas V, Gordon KH, & Castro Y (2017). Bulimic symptoms and interpersonal functioning among college students. Psychiatry research, 257, 406–411. 10.1016/j.psychres.2017.08.021 [DOI] [PubMed] [Google Scholar]
  21. Lenz AS, Del Conte G, Hollenbaugh KM, & Callendar K (2016). Emotional regulation and interpersonal effectiveness as mechanisms of change for treatment outcomes within a DBT program for adolescents. Counseling Outcome Research and Evaluation, 7(2), 73–85. [Google Scholar]
  22. Linville D, Brown T, Sturm K, & McDougal T (2012). Eating disorders and social support: perspectives of recovered individuals. Eating disorders, 20(3), 216–231. 10.1080/10640266.2012.668480 [DOI] [PubMed] [Google Scholar]
  23. Pan Y, Zhang D, Liu Y et al. Different Effects of Paternal and Maternal Attachment on Psychological Health Among Chinese Secondary School Students. J Child Fam Stud 25, 2998–3008 (2016). 10.1007/s10826-016-0463-0 [DOI] [Google Scholar]
  24. Pössel P, Burton SM, Cauley B et al. Associations between Social Support from Family, Friends, and Teachers and depressive Symptoms in Adolescents. J Youth Adolescence 47, 398–412 (2018). 10.1007/s10964-017-0712-6 [DOI] [PubMed] [Google Scholar]
  25. Presnell K, Bearman SK, & Stice E (2004). Risk factors for body dissatisfaction in adolescent boys and girls: A prospective study. International Journal of eating disorders, 36(4), 389–401. [DOI] [PubMed] [Google Scholar]
  26. Rathus JH, & Miller AL (2014). DBT skills manual for adolescents. Guilford Publications. [Google Scholar]
  27. Rayner KE, Schniering CA, Rapee RM, Taylor A, & Hutchinson DM (2013). Adolescent girls' friendship networks, body dissatisfaction, and disordered eating: examining selection and socialization processes. Journal of abnormal psychology, 122(1), 93. [DOI] [PubMed] [Google Scholar]
  28. Reilly EE, Orloff NC, Luo T, Berner LA, Brown TA, Claudat K, … & Anderson LK (2020). Dialectical behavioral therapy for the treatment of adolescent eating disorders: a review of existing work and proposed future directions. Eating disorders, 28(2), 122–141. [DOI] [PubMed] [Google Scholar]
  29. Rubin DB (1974). Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology, 66(5), 688–701. 10.1037/h0037350 [DOI] [Google Scholar]
  30. Rubin DB (2004). Multiple imputation for nonresponse in surveys (Vol. 81). John Wiley & Sons. [Google Scholar]
  31. Safer DL, Telch CF, & Agras WS (2001). Dialectical Behavior Therapy for Bulimia Nervosa. The American Journal of Psychiatry. 10.1176/appi.ajp.158.4.632 [DOI] [PubMed] [Google Scholar]
  32. Santos M, Richards CS, & Bleckley MK (2007). Comorbidity between depression and disordered eating in adolescents. Eating behaviors, 8(4), 440–449. [DOI] [PubMed] [Google Scholar]
  33. Scardera S, Perret LC, Ouellet-Morin I, Gariépy G, Juster RP, Boivin M, Turecki G, Tremblay RE, Côté S, & Geoffroy MC (2020). Association of Social Support During Adolescence With Depression, Anxiety, and Suicidal Ideation in Young Adults. JAMA network open, 3(12), e2027491. 10.1001/jamanetworkopen.2020.27491 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Stice E, Barrera M Jr., & Chassin L (1998). Prospective differential prediction of adolescent alcohol use and problem use: Examining the mechanisms of effect. Journal of Abnormal Psychology, 107(4), 616–628. 10.1037/0021-843X.107.4.616 [DOI] [PubMed] [Google Scholar]
  35. Stice E, & Bearman SK (2001). Body-image and eating disturbances prospectively predict increases in depressive symptoms in adolescent girls: a growth curve analysis. Developmental psychology, 37(5), 597. [DOI] [PubMed] [Google Scholar]
  36. Stice E, Gau JM, Rohde P, & Shaw H (2017). Risk factors that predict future onset of each DSM-5 eating disorder: Predictive specificity in high-risk adolescent females. Journal of abnormal psychology, 126(1), 38–51. 10.1037/abn0000219 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Stice E, Marti N, & Rohde P (2013). Prevalence, incidence, impairment, and course of the proposed DSM-5 eating disorder diagnoses in an 8-year prospective community study of young women. Journal of Abnormal Psychology, 122, 445–457. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Stice E, Marti CN, & Durant S (2011). Risk factors for onset of eating disorders: Evidence of multiple risk pathways from an 8-year prospective study. Behaviour research and therapy, 49(10), 622–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Stice E, Presnell K, & Spangler D (2002). Risk factors for binge eating onset in adolescent girls: a 2-year prospective investigation. Health psychology, 21(2), 131. https://doiorg.laneproxy.stanford.edu/10.1037/0278-6133.21.2.131 [PubMed] [Google Scholar]
  40. Stice E, Ragan J, & Randall P (2004). Prospective Relations Between Social Support and Depression: Differential Direction of Effects for Parent and Peer Support? Journal of Abnormal Psychology, 113(1), 155–159. 10.1037/0021-843X.113.1.155 [DOI] [PubMed] [Google Scholar]
  41. Stice E, Rohde P, Gau J, & Shaw H (2009). An effectiveness trial of a dissonance-based eating disorder prevention program for high-risk adolescent girls. Journal of Consulting and Clinical Psychology, 77, 825–834. 10.1037/a0016132 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Stice E, Rohde P, Yokum S, Gau J, Bohon C, & Shaw H (2023). A randomized trial of two group-delivered transdiagnostic eating disorder treatments: Dissonance-based treatment versus interpersonal psychotherapy. Under review. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Stice E, & Van Ryzin MJ (2019). A prospective test of the temporal sequencing of risk factor emergence in the dual pathway model of eating disorders. Journal of abnormal psychology, 128(2), 119–128. 10.1037/abn0000400 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Strauss RS, & Pollack HA (2003). Social marginalization of overweight children. Archives of pediatrics & adolescent medicine, 157(8), 746–752. [DOI] [PubMed] [Google Scholar]
  45. Tahir WBE, Inam A, & Raana T (2015). Relationship between social support and selfesteem of adolescent girls. IOSR Journal of Humanities And Social Science (IOSR-JHSS), 20(2), 42–46. 10.9790/0837-20254246 [DOI] [Google Scholar]
  46. Telch CF, Agras WS, & Linehan MM (2001). Dialectical behavior therapy for binge eating disorder. Journal of consulting and clinical psychology, 69(6), 1061. [DOI] [PubMed] [Google Scholar]
  47. Treasure J, Murphy T, Szmukler T, Todd G, Gavan K, & Joyce J (2001). The experience of caregiving for severe mental illness: a comparison between anorexia nervosa and psychosis. Social Psychiatry and Psychiatric Epidemiology, 36(7), 343–347. 10.1007/s001270170039 [DOI] [PubMed] [Google Scholar]
  48. Van Buuren S, & Groothuis-Oudshoorn K (2011). mice: Multivariate imputation by chained equations in R. Journal of statistical software, 45, 1–67. 10.18637/jss.v045.i03 [DOI] [Google Scholar]
  49. Vogel EN, Singh S, & Accurso EC (2021). A systematic review of cognitive behavior therapy and dialectical behavior therapy for adolescent eating disorders. Journal of Eating Disorders, 9(1), 1–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wonderlich-Tierney AL, & Vander Wal JS (2010). The effects of social support and coping on the relationship between social anxiety and eating disorders. Eating behaviors, 11(2), 85–91. 10.1016/j.eatbeh.2009.10.002 [DOI] [PubMed] [Google Scholar]
  51. Yamamiya Y, & Stice E (2022). Risk factors that predict future onset of anorexia nervosa, bulimia nervosa, binge eating disorder, and purging disorder in adolescent girls. Under review. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES