Authors: Bar Zissu, Helene Sher, Ortal Slobodin
Categories: Research, Eating disorder, Parents, Sociocultural attitudes, Transgenerational transmission
Source: Journal of Eating Disorders
Authors: Bar Zissu, Helene Sher, Ortal Slobodin
Eating disorders may be transmitted from one generation to the others through various trajectories, including genetic and epigenetic factors, parent–child relationships, and behavioral factors. In the current study, we examined whether parents' eating disorder symptoms and sociocultural attitudes toward appearance are associated with the diagnosis of an eating disorder or the level of eating disorder symptoms among female adolescents and young adults, and the nature of these associations. We also examined whether fathers' and mothers' effects on offsprings' eating disorder symptoms interact.
The study included 65 triads of mothers, fathers, and their female offspring aged 11–22 (N = 195). Thirty-two offspring were diagnosed with an eating disorder and recruited from the in-patient eating disorder unit in a public hospital. The remaining 33 offspring, who formed the control group, were recruited via social media platforms.
In contrast to our expectations and previous findings, mothers' and fathers' levels of eating disorder symptoms and sociocultural attitudes toward appearance were negatively associated with their offspring's eating disorder symptoms. Also, inconsistent with mental health studies that showed that the behaviors and attitudes of one parent are exacerbated or attenuated by the other parent, we failed to find an interaction effect between mothers' and fathers' variables on their offspring's level of eating disorder symptoms.
The current study adds to the limited number of studies that focused on fathers in the transgenerational transmission of eating disorders and encourages further research on the effects of each parent and the combined effects of both in the development and maintenance of eating disorders in their offspring.
Eating disorders are serious, life-threatening, psychiatric illnesses that are associated with increased mortality and high treatment costs [1]. Eating disorders are considered heterogeneous disorders with a complex multifactor etiology that involves biological (e.g., structural and functional brain abnormalities, pubertal status, excess body fat mass), psychological (e.g., perfectionism, negative affectivity, body dissatisfaction), and socioenvironmental factors (familial pressure, media body ideals, cultural values) [2, 3]. Research focusing on the transgenerational transmission of eating disorders identified several trajectories through which parents may affect their children's risk for developing the disorders, including genetics and epigenetics factors [4], parent–child interactions (e.g., attachment style, [5], and behavioral factors, such as eating habits [6].
In the current study, we examined whether parents'eating disorder symptoms and sociocultural attitudes toward appearance are associated with the diagnosis of an eating disorder or the level of eating disorder symptoms among female adolescents and young adults, and the nature of these associations. We also examined whether fathers'and mothers'effects on offsprings'eating disorder symptoms interact. Given the importance of both parents in the development, maintenance, and treatment of eating disorders [7], we examined our research questions in triads of mothers, fathers, and offspring.
Transgenerational transmission models of psychiatric disorders were developed and advanced to explain why children of parents with mental health disorders are at a higher risk of developing such disorders during their life course [8, 9]. Studies on mental health disorders suggested that both transgenerational concordance (specific parental mental disorders increase children’s risk for certain disorders) and multifinality (parental mental disorders increase children’s risk for mental disorders in general) are used to define the scope of the child’s diagnostic outcomes of a particular parental disorder [10]. Transgenerational transmission models of mental disorders traditionally identify five mechanisms through which parental mental health may affect children's genetics, prenatal factors, parent–child interaction, family, and social factors [11].
In the case of eating disorders, candidate gene studies identified genes involved in weight regulation, eating behavior, neuropsychological profiles, mood, neurodevelopment, and stress responsivity. However, the inherited risk involves many genes of small effect sizes, suggesting that gene-environment interactions are likely to be involved [4, 12]. Another strand of literature focused on the role of parent–child interaction in the development of eating disorders. Studies focusing on the role of attachment style showed that, overall, patients with eating disorders as well as their parents were more likely to have insecure attachment styles compared with families of controls [13]. Attili et al. [14], for example, showed that patients with eating disorders and their mothers showed a higher prevalence of ambivalent and avoidant attachment styles compared with controls, and fathers showed a higher prevalence of avoidant attachment styles. Furthermore, attachment style differences between parents or between patients and mothers were larger in families of eating disorder patients [14].
Parent–child communication, especially related to the child's weight, body shape, and restrictive diet, was also studied as a risk factor for eating disorders [6]. Importantly, excessive body talk—even when positive—can reinforce the idea that appearance is central to self-worth, a sociocultural message identified as a risk factor in the development of eating disorders [15–18].
Transgenerational transmission of eating disorders is also associated with behavioral factors, such as modeling and family eating habits, and their interactions with genetic and epigenetic factors [19, 20]. Previous studies showed that children of mothers with eating disorders manifested disordered eating habits and attitudes compared with controls [21, 22]. Mothers with disordered eating habits may restrict their children’s food intake, be anxious about handling food, and experience difficulties modeling healthy eating habits for their children [23, 24].
Sociocultural factors received major attention in eating disorder research [25]. According to the tripartite influence model (TIM), individuals are pressured by peers, family, and media to adhere to culturally desirable appearance ideals [26, 27]. Studies that focused on sociocultural factors in the development of eating disorders showed that the internalization of appearance ideals and appearance pressures were risk factors for body image disturbance and disordered eating [28]. Traditionally, the thin ideal has been the dominant standard of female beauty but in recent years, the ‘fit’ ideal (which is characterized by a more toned and athletic figure) emerged as a competing alternative [29]. However, research suggests that fit ideal internalization, although meant to inspire a fit, healthy, and active lifestyle may be similarly [30] or even more detrimental [31] to female body satisfaction, disordered eating, and compulsive exercise. Recently, the “slim-thick” body type emerged in social media as a new beauty ideal. This ideal is a modified version of the thin ideal that is characterized by having a small waist and flat stomach, but large thighs and butt [32]. While research on the impact of slim-thick media on women’s body image is scarce, there is evidence to suggest that exposure to all three body ideals (thin, fit, and slim-thick) resulted in increased weight and appearance dissatisfaction and lower overall body satisfaction [12].
Several decades of research have demonstrated links between parental attitudes, perceptions, and behaviors related to body image and their offsprings'development of eating disorders. It has been shown that when parents believe that thin body silhouettes of women are more appealing than larger ones, their children are more likely to develop eating disorders to achieve their parents'ideal preferred body silhouette [27, 33]. Parents'perceptions of the ideal body may influence their children directly and indirectly [34]. Examples of direct influence are parents’ comments about body shape and/or the need for weight control by children [35, 36], while an example of indirect influence is the parents’ behaviors toward their own body [37]. Both types of influence may convey several messages, including the importance of the functionality of the body, care through a healthy lifestyle, messages of affection and respect [38], or, in contrast, body dissatisfaction [39]. Less is known, however, about the associations between mothers'and fathers'perceptions of eating, body, and appearance and children's eating disorder symptoms during adolescence and early adulthood [35, 40] and whether parents'variables interact.
The current study had two research objectives. First, we wished to examine whether parents'eating disorder symptoms and sociocultural attitudes toward appearance are associated with the diagnosis of an eating disorder or the level of eating disorder symptoms among female adolescents and young adults, and the nature of these associations. We also examined whether mothers'and fathers'effects on their children's eating disorder symptoms interact. To date, most research on the transgenerational transmission of eating disorders has predominantly focused on mothers rather than fathers [7]. There are several reasons why transgenerational transmission of eating disorders should include fathers. First, evidence suggests that while both parents are greatly affected by a child's eating disorder, fathers and mothers may feel and react differently [41]. Second, including both parents in eating disorder studies allows us to examine whether the association between mother and child's eating behavior and attitudes differs from the father–child association. Including both parents in eating disorders studies will also allow us to examine whether the behaviors and attitudes of one parent are exacerbated or attenuated by the other parent [42]. For instance, previous studies on children's depression showed that fathers'mental health serves as a protective factor in the relationship between mothers'and children's depressive symptoms [43]. Investigating the main and interaction effects of fathers and mothers may also provide insight into the gendered aspects of transgenerational transmission of eating disorders [44]. Finally, fathers benefit from family-based interventions for eating disorders like mothers, and their involvement in the treatment was associated with better patient outcomes [45, 46].
Given accumulating evidence on the positive relationship between parents'and children's behaviors, attitudes, and perceptions related to eating, weight, and appearance [6, 20], we have hypothesized that offsprings'eating disorder symptoms will be positively correlated with fathers'and mothers'eating disorders (H1) and with parents'sociocultural attitudes toward appearance (H2). We also hypothesized that fathers'and mothers'effects will interact so that the adverse effects of eating disorder symptoms and sociocultural attitudes toward appearance of one parent will be exacerbated by increased levels of eating disorder symptoms and sociocultural attitudes toward appearance of the other parent. In contrast, lower levels of eating disorder symptoms and sociocultural attitudes toward appearance of one parent will attenuate the adverse effects of eating disorder symptoms and sociocultural attitudes toward appearance of the other parent (H3).
The study included 65 female adolescents and young adults (Mean = 16.10, S.D = 2.65) and both of their parents. Of them, 32 participants were diagnosed with an eating disorder and were current or past inpatients of an eating disorder unit in a public hospital. The other 33 participants, without an eating disorder, were recruited using social media platforms, as detailed below. Inclusion criteria for the eating disorder group were (1) past or present DSM-5 [47] diagnosis of an eating disorder, given by a psychiatrist in the inpatient unit (2) No diagnosis of psychotic disorders or severe intellectual deficits. Seven families refused to participate, and two left the questionnaire without completion. Because we were interested only in full triads of offspring-mother–father, we had to remove 32 participants (17 parents and 15 girls) for whom at least one member of the triad was missing. After removing these participants, 32 triads of offspring-mother–father were included in the eating disorder group.
Inclusion criteria for the control group were (1) an absence of an eating disorder (scored below 42 in the Eating Disorder Inventory-2; [48, 49]) (2) no history of major psychiatric disorders or psychiatric medications use. Recruitment approach for the control group included three strategies. First, we used snowball nonprobability convenience sampling [50, 51], by sharing posts on Facebook group pages (with a total number of 74,000 members). For example, we shared posts in Facebook groups for local and national Israeli women, research Facebook groups (including a specific group of mothers interested in research), and municipalities'groups. Posts for this project included a link to the online survey and text with information about the survey. No cost was involved in generating these posts. Second, the first author and another research assistant shared the post on their personal social media sites and WhatsApp groups. Third, we used traditional snowball sampling [52], by asking personal contacts to forward the research opportunity to other participants. Participants accessed an anonymous online survey through a link. While it is impossible to know how many people have seen the invitation to participate, 134 girls and 143 parents accessed the survey via the Qualtrics link. After removing 101 girls and 77 parents for whom at least one member of the triad was missing, 33 full triads remained. Ethical approval was received from the Helsinki Committee of the hospital and the University's IRBs. All participants above 18 years of age signed informed consent. When participants were under 18 years old, one of the parents (or both in the case of divorced parents) signed an informed consent. In these cases, participants signed an agreement to participate form. All participants completed online questionnaires using the Qualtrics platform. In accordance with the standard procedure for minimal-risk online surveys, the informed consent of the participants was implied through survey completion. Children and parents were matched using identifying ID codes. No personal identifying information was collected. No compensation was offered. If participants'responses raised a concern regarding disordered eating (EDI-2 > 42 for offspring, or Eating Attitudes Test > 20 for parents, [53]), an automated response appeared in the survey, suggesting that they consult a professional. Participants were included in the final sample only if both parents completed the study questionnaires.
Offsprings'background variables included age, education, and place of birth. Parents'background variables included age, education level, family status, income, place of birth, number of children, and level of religiosity.
Offspring eating disorders were evaluated using the Eating Disorder Inventory-2 (EDI-2; [49]). This is a 91-item, standardized self-report measure consisting of 11 subscales that assess specific cognitive and behavioral dimensions of eating drive for thinness, bulimia, body dissatisfaction, ineffectiveness, perfectionism, interpersonal distrust, interoceptive awareness, maturity fears, asceticism, impulse regulation, and social insecurity. All scales have high rates of statistical validity and reliability. The EDI-2 is commonly used as a measure of symptom severity [54, 55] and has been found to successfully differentiate between participants with and without eating disorders [49]. All items are rated on a 6-point Likert scale ranging from 0 (never) to 5 (always). A high score on the questionnaire indicates a higher severity or frequency of specific thoughts, feelings, and behaviors associated with eating disorders. In the current study, Cronbach’s alpha was 0.97.
Parents'eating disorder symptoms were evaluated using the Eating Attitudes Test (EAT-26; [53]). It is a valid and reliable self-report questionnaire consisting of 26 items that form three dieting, bulimia and increased preoccupation with food, and oral control. Each item is scored on a 6-point Likert scale and summed for a total score, with answers ranging from'never'to'always'. The questionnaire alone does not yield a specific diagnosis of an eating disorder but a high score on it indicates a higher level of concern or problematic eating attitudes and behaviors, which may indicate a potential risk for eating disorders. In the current study, Cronbach’s alpha was 0.80.
Parents'attitudes toward body and appearance were measured by the Sociocultural Attitudes Towards Appearance Scale (SATAQ-3; [56]). This is a self-report questionnaire that examines the cultural perception of the desired body. It was designed to assess people's attitudes and beliefs regarding appearance and body image in the context of socio-cultural influences. The questionnaire consists of 30 items divided into four general internalization (a nine-item scale describing the degree of internalization of cultural-social standards of body image), sports internalization (a five-item scale measuring the degree of internalization of an athletic physique), pressures (a seven-item scale describing the intensity of pressure a person feels to conform to socio-cultural standards), and information (a nine-item scale describing the frequency with which a person searches for information about body image and socio-cultural standards of appearance). Each item can be answered using a 5-point Likert scale (0—completely disagree, 4—completely agree). A high score in the questionnaire indicates a higher level of internalization of socio-cultural attitudes towards appearance and body image, which indicates that the individual is more influenced by social norms and ideals about how a person should look and the importance of physical appearance. The scale showed good psychometric qualities in different cultural contexts [57, 58]. Cronbach’s alpha in the current study was 0.88.
To compare participants in the eating disorder group to those without an eating disorder in demographic and clinical variables, we used t-tests for independent samples and Chi-square tests. Next, Bivariate Pearson correlations between study variables were calculated. Finally, a conditional process modeling was used to examine the main and interaction effects of fathers'and mothers'variables in predicting offsprings'eating disorder symptoms, as outlined by Hayes [59], using the PROCESS macro. Offspring's age served as a covariate. We used an alpha level of 0.05 for all statistical tests. Data analyses were carried out on SPSS Windows 29.0.
Given that the current study used a cross-sectional design with variables being correlated and continuous, identifying interaction effects might be more complex [60]. A-priori power calculations using G*POWER software revealed that for multiple regression with interaction effects, four predictors, a power of 0.8, and an effect size of 0.15, a minimum of 89 participants was required.
Participants were Jewish, Hebrew-speaking, and residing in Israel at the time of the study. All offspring were Israeli-born. Nine mothers (13.8%) and eight fathers (12.3%) were immigrants, mostly from the Former Soviet Union. All parent couples were heterosexual. Table 1 presents the results of independent t-tests and Chi-square tests comparing the two study groups. In terms of demographic variables, participants with eating disorders were significantly younger than those without eating disorders. Furthermore, the fathers of offsprings without an eating disorder had higher levels of education than fathers of offsprings with an eating disorder (in the control group, 84% of fathers had an academic education compared to 41% in the clinical group). Table 1Group differences between participants with and without eating disordersEating disorder group (n = 32)Control group (n = 33)Group Differences (t)%%χ^2^Mother's education2.13, p =.547Highschool19.412.1Secondary education16.19.1Bachelor degree2927.3Masters'degree and higher35.551.5Father's education15.98, p <.001Highschool21.912.1Secondary education37.53Bachelor degree18.848.5Masters'degree and higher21.936.4Income^1^ < 12, 00023.39.42.82, p =.58912,000–20,0002031.321,000–30,0003028.131,000–40,0002021.940,000 < 6.79.4MeanS.DS.EMeanS.DS.EtOffspring's age14.921.730.3117.393.150.553.94, p <.001Mother's age47.595.270.9349.065.010.871.15, p =.127Father's age516.751.1950.675.480.95−0.22, p =.415Offspring's eating disorder symptoms^2^3.740.650.112.630.820.14−6.06, p <.001Mother's eating disorder symptoms ^3^2.050.490.092.480.440.083.76, p <.001Mother's socio-cultural attitudes of appearance^4^2.500.680.122.960.350.063.38, p <.001Father's eating disorder symptoms ^3^1.850.500.092.430.530.094.45, p <.001Father's socio-cultural attitudes of appearance ^4^2.500.830.152.990.340.063.21, p <.001^1^20,000 NIS is the national average income^2^Offspring eating disorders were evaluated using the Eating Disorder Inventory-2 (EDI-2; [49])^3^Parents'eating disorder symptoms were evaluated using the Eating Attitudes Test (EAT-26; [53])^4^Parents'attitudes toward body and appearance were measured by the Sociocultural Attitudes Towards Appearance Scale (SATAQ-3; [56])
Additionally, offsprings in the eating disorder group reported higher levels of eating disorder symptoms compared to those without an eating disorder. In contrast, fathers and mothers of offspring with a diagnosed eating disorder reported lower levels of eating disorder symptoms and decreased internalization of sociocultural attitudes (i.e., thin ideals) than parents of offsprings without an eating disorder diagnosis.
Table 2 presents Pearson correlation matrix for the study variables. In contrast to the first and second hypotheses, the offsprings'eating disorder symptoms were negatively associated with their mothers'(r =.−43, p < 0.001), and fathers'eating disorder symptoms (r =.−31, p = 0.014). Offsprings'eating disorder symptoms were also negatively associated with mothers'(r = −0.35, p = 0.004) and fathers'sociocultural perceptions toward appearance (r = −0.28, p = 0.023). Finally, we conducted a moderation analysis using PROCESS to examine whether fathers'and mothers'variables interact in predicting the offsprings'eating disorder symptoms. Due to significant age differences between participants with and without an eating disorder diagnosis, the offsprings'age served as a covariate. The first regression equation examined whether fathers'eating disorder symptoms moderated the relationship between mothers'and the offsprings'eating disorder symptoms. Results showed that the model was significant, F (4,57) = 4.81, p = 0.002, explaining 25% of variance in offsprings'eating disorder symptoms. However, as seen in Table 3, no main or interaction effects were observed. Cohen's f^2^ value for the interaction effect was 0.007. The second regression examined whether fathers'sociocultural attitudes toward appearance moderate the relationship between mothers'sociocultural attitudes and offsprings'eating disorder symptoms. Results showed that the model was significant, F (4,60) = 3.41, p = 0.013, explaining 19% of the variance. As seen in Table 3, no main or interaction effects were observed. Cohen's f^2^ value for the interaction effect was 0.01. Table 2Pearson correlations between study variablesOffspring's eating disorder symptomsMother's eating disorder symptomsMother's socio-cultural attitudes of appearanceFather's eating disorder symptomsFather's socio-cultural attitudes of appearanceOffspring's eating disorder symptoms^1^1−0.43***−0.35***−0.31*−0.28Mother's eating disorder symptoms^2^10.360.350.31Mother's socio-cultural attitudes of appearance ^3^10.380.40Father's eating disorder symptoms^2^10.43Father's socio-cultural attitudes of appearance ^3^1p <.05, **p <.01, **p <.001^1^Offspring eating disorders were evaluated using the Eating Disorder Inventory-2 (EDI-2; [49])^2^Parents'eating disorder symptoms were evaluated using the Eating Attitudes Test (EAT-26; [53])^3^Parents'attitudes toward body and appearance were measured by the Sociocultural Attitudes Towards Appearance Scale (SATAQ-3; [56])Table 3Regression predicting offspring's eating disorders symptomsBS.EBetatp*Offspring's age−0.060.04−0.19−1.600.11Mother's eating disorder symptoms−0.070.79−0.36−0.090.93Father's eating disorder symptoms0.400.87−0.130.470.64Mother's * Father's eating disorder symptoms−0.270.36−0.09−0.740.46Offspring's age−0.60.4−0.17−1.450.15Mother's socio-cultural attitudes−1.030.74−0.24−1.380.171Father's socio-cultural attitudes−0.770.74−0.10−1.050.29Mothers * Father's socio-cultural attitudes0.240.280.100.840.40
The current study examined the role of transgenerational transmission in the development of eating disorders. Specifically, we examined the main effects and interactions between fathers'and mothers'eating disorder symptoms and sociocultural attitudes toward appearance on offsprings'eating disorder symptoms. In contrast to our expectations and to previous findings [13, 61], both fathers'and mothers'levels of eating disorder symptoms and sociocultural attitudes were negatively associated with the offsprings'eating disorder symptoms. Also, in contrast to previous studies in mental health that showed that the behaviors and attitudes of one parent are exacerbated or attenuated by the other parent [42, 43], we failed to find an interaction effect between mothers'and fathers'variables on offsprings'level of eating disorder symptoms.
Several explanations may shed light on our findings. First, while the current study focused on parents'perceptions of eating and appearance, it overlooked how these perceptions were communicated to adolescents and young women. Previous studies suggested that parents with eating disorder symptoms transmit their perceptions through verbal and non-verbal communication during eating-related and non-eating interactions [62, 63]. These communication styles may include"fat talk", teasing about weight, and preoccupation with weight and diet [40]. Our findings suggest that parents'attitudes alone may not be enough to explain the risk of eating disorders, and that further investigation should focus on how such attitudes are transmitted. For example, Cimino et al. [64], who compared young children of parents with and without eating disorders, found that parents'eating disorders had no main effect on children’s psychopathology but influenced children’s behavioral and mood difficulties through parent–child interactions.
Another possible explanation for the negative association between parents'and offsprings'eating disorder symptoms may be related to intentional or unintentional reporting bias. Parents of children with eating disorders may be more affected by aspects of social desirability than parents of children without an eating disorder or more aware of the negative effects of their own disordered eating on their children [65]. Therefore, they might underreport symptoms or attitudes that might be associated with eating disorder pathology [66]. Another possible explanation for our findings may be related to the history and chronicity of parents'and children's eating disorder symptoms [62]. Although a history of eating disorder symptoms may increase transmission risk from parents to children [11], we did not control for such variables in our study. If a substantial proportion of parents in the eating disorder group were experiencing a decrease in their disordered eating due to their children's diagnosis and treatment, it could have impacted the results. This was a limitation noted by many authors and may be particularly influential when the parenting constructs under assessment are in eating disorder-specific domains [62]. Our surprising results may also be explained by the sample characteristics. Unlike previous studies on young children [62], the current study focused on adolescents and young adults. It is possible that children at these ages become more aware of their parents'disordered eating, and may develop a better insight that could serve as a protective factor. However, future research is needed to further explore this possibility. Finally, it is important to point out that parental psychopathology does not always negatively impact child development. This might be due to numerous factors, including a protective family network, child resilience, early access to mental health services, and parenting abilities [67]. In the context of eating disorders, de Barse et al. [68] found that mothers with a history of eating disorders reported less pressuring feeding than mothers without such a history because they were much more concerned with the stress related to eating disorder and more hesitant to interfere with their children's eating habits. In the same vein, Doersam et al. [65] found that mothers with an eating disorder history were more responsive to infant's signals during feeding than mother without such history (e.g., permitting self-feeding, avoiding negative statements when child rejected food, adopting the pace of feeding to the child). Based on these findings, it might be hypothesized that parents with increased awareness of the detrimental effects of their own disordered eating may attenuate and monitor their behavior in a way that reduces their children's risk [69]. Increased awareness and knowledge about eating disorders may also encourage parents to adopt healthier approach towards eating, weight, and appearance [70]. Therefore, interventions focused on improving parental knowledge, self-awareness, and self-efficacy in the context of eating disorders may offer a valuable venue for intervention [71, 72]. For example, studying the efficacy of parent intervention, Bryson et al. [73] found that parents'increased knowledge of eating disorders may improve self-efficacy, which plays a critical role in parents’ ability to adopt eating disorder treatments.
One notable strength of our study is that in contrast to many previous studies that relied on single source data (i.e., by asking parents to report about their own and their child's perceptions and behaviors, or by asking the children to report about their own and their parents'perceptions and behaviors), the current study used multisource data that included the mother, the father, and their offspring. This approach enabled us to overcome some of the limitations of shared method variance when parents'or children's reports are subject to their own perceptions [68]. Additionally, the current study adds to the limited number of studies that focused on fathers in the transgenerational transmission of eating disorders [7]. More research is needed to determine the main and combined effects of both parents in the development and maintenance of eating disorders, as well as the role of marital relationships. An interesting finding of the current study was the association between fathers'and mothers'disordered eating and socio-cultural perceptions of appearance. In other words, we found that when one parent was preoccupied with the body and eating, the other parent was more likely to hold similar perceptions and behaviors. These findings provide further support for the spousal concordance in lifestyle and eating [74], calling for future studies on the role of family dynamic in eating disorder [23, 75].
Several limitations should be considered when interpreting the study results. First, the small sample size limited our ability to generalize our findings and did not allow us to account for potential clinical and background confounders, such as type of diagnosis, duration of eating disorder, and psychiatric co-morbidities. This small sample size also limits the power needed to detect interaction effects in a cross-sectional study with correlated variables [60].
Another limitation is the cross-sectional design of the study, which did not allow us to track time-related changes in the associations between parents'and children's eating attitudes and behaviors. Previous studies suggested that parents perceive the effects of their eating disorder to differ for children of different ages and that concerns about eating disorder development appeared to be heightened for mothers who had children approaching adolescence [76–78]. Due to our cross-sectional design, findings could also indicate increased awareness or sensitivity to eating disorder behaviors rather than reflect those behaviors [61]. Longitudinal research and controlled trials are required to examine the family dynamic of eating disorders. Additionally, the current study focused only on attitudes and behaviors, overlooking other transgenerational trajectories of eating disorders, such as actual parent–child interactions [5], and genetics and epigenetics factors [4] and their interactions. More research is needed to explore how these multiple risk factors interact in the development of eating disorders. Finally, the ability to generalize our findings is limited due to the self-selection method of the control group. While internet-based surveys are a common recruitment strategy, which may reduce measurement bias related to answers on or stigmatizing topics, it relies on self-selection surveys that are not based on probability sampling [79]. Therefore, online self-selection surveys are subject to coverage and selection bias, limiting the external validity and the generalization of findings [80]. In the current study, participation in the control group was not only limited to people with internet access and members of social media networks but may also represent people with a disproportionate preoccupation with eating and appearance.
Findings regarding the relationship between children's and parents'eating disorders are mixed. While many studies supported the association between parents'and offsprings'eating disorder attitudes and behaviors [13, 62], other studies [68, 81, 82] did not observe such an association. The current study suggests that while fathers'and mothers'levels of eating disorder symptoms and sociocultural attitudes toward appearance were related to each other, they were reversibly associated with their offspring's eating disorder symptoms. Further research should focus on the risk and resilience of maternal and paternal factors in the development and maintenance of eating disorders in different life stages.