Food consumption and eating disorders in Latin America and the Caribbean: a scoping review

Abstract

Background

The relationship between food consumption and eating disorders (ED) is pivotal, given that eating disorders are characterized by abnormal eating habits which can result in significant health issues. However, research on this topic in non-western, economically underdeveloped countries is limited, particularly in Latin America and the Caribbean.

Objective

To gather and examine the available evidence on food consumption and eating patterns in people diagnosed with eating disorders (ED), in studies conducted in Latin American and Caribbean countries, with particular interest in identifying food consumption assessment tools used in this population.

Methods

The scoping review was carried out in three bibliographic databases: Medline (PubMed), Cochrane Library, and LILACS, with no restrictions, up to November 2025. PRISMA guidelines were used for reporting this review.

Results

Four studies were identified that met the eligibility criteria, all conducted in Brazil. The studies included adults diagnosed with binge eating disorder (BED) or bulimia nervosa (BN). No participants with Anorexia Nervosa (AN) were included in these studies. High energy variability was observed in the diets; when considering only binge eating episodes, the highest intake was found in participants diagnosed with BED. Among the most common macronutrients consumed are carbohydrates, lipids, and finally, proteins; there is no greater variation when considering only binge eating episodes. Micronutrient intake is below the RDI levels, with deficiencies of vitamin E, folic acid, magnesium, and iron. Among the most frequently consumed foods are highly palatable foods, as well as beef, rice, and beans. The most common methods for assessing food consumption were food diary, food frecuency, and 24-hour recall, among others.

Conclusion

Common foods in Brasil, such as rice and beans were frequently consumed in the studies selected. Consumption in patients with BN varies, ranging from 19,257 to 158 kcal/day. More studies are needed to validate and culturally adapt food consumption and eating patterns assessment tools in this population.

Plain language summary

Few studies have been published that address food consumption and Eating Disorders in Latin America and the Caribbean with the use of instruments that approximate actual food consumption in this population. The four selected studies belong to Brazil and have a 20-year difference between the first one and the other two more recent ones. The studies addressed BED and BN as diagnosis. The summary findings show a high variability in the diets in terms of calories. Important micronutrients, such as iron, magnesium and folic acid, are below recommended intakes. Beef, rice, and beans were the most consumed foods. Further research should be conducted on the foods most commonly consumed by these patients in other countries in Latin America and the Caribbean to improve the recommendations.

Introduction

Eating disorders (EDs) are a group of serious illnesses characterized by significant and persistent alterations in food consumption and distressing thoughts and emotions associated with a concern for body weight and shape [1–3]. The DSM–5-TR [2] includes the following eating disorder diagnostic categories: Anorexia Nervosa (AN), Bulimia Nervosa (BN), Binge Eating Disorder (BED), Avoidant Restrictive Food Intake Disorder (ARFID), PICA, Rumination Disorder (RD), Other Specified Feeding and Eating Disorders (OSFED) (i.e., Night Eating Syndrome, Purging Disorder, Atypical AN, BN and BED of low frequency and/or limited duration), and Unspecified Feeding or Eating Disorders (UFED). In terms of prevalence, OSFED is the most common ED, followed by BED, BN, and AN [4, 5]. However, AN, BN and BED are the most frequently studied in the literature [6]. EDs can have severe physical and psychosocial consequences in those afflicted by them [7–9]; with high comorbidity and mortality prevalence rates [2, 3, 10–12].

Eating disorders (EDs) often begin in early adolescence or childhood, and are more common in women than men [4, 13, 14]. There is ample literature on EDs across all ages, sexes, and cultures [15–17]. Factors linked to EDs include occupation, marital status, motherhood, income, and social environments that emphasize thinness [3, 18–21].

There is ample evidence in the literature on EDs in Western and economically developed countries, with an emphasis on women [22, 23]. However, in recent years, authors have called for more studies in non-Western and developing countries due to the rising prevalence of EDs in these regions [14, 24–28]. Some studies have argued that the risk of developing an ED has increased in low and middle-income countries due to social media and internet access among youth worldwide [29, 30].

Furthermore, research has highlighted significant disparities between Western and non-Western nations, regarding prevalence data. In a systematic review of the literature (2000 to 2018) on the prevalence of eating disorders worldwide, Galmiche et al. reported differences between Western and non-Western countries, which were aggregated by continent [4]. Data reported on the weighted means (ranges) of point prevalence rates were highest for the Americas (4.6%, 95% CI: 2.0-13.5%), followed by Asia (3.5%, 95% CI: 0.6–7.7%) and Europe, with the lowest rate (2.2%, 95% CI: 0.2–13.1%).

Focusing on the prevalence of EDs in Latin America, Kolar and Mebarak conducted a systematic review of epidemiological data from 2020 to 2022 [31]. The authors synthesized findings on full threshold AN, BN, and BED from four studies in Latin America and the Caribbean (three from Brazil). One of these studies, the Global Burden of Disease (GBD) study [32], reported age-standardized overall ED prevalence estimates of 0.23% (95% CI: 0.17–0.30%) for the region. Complementing this, a previous study highlighted how economic development influences the rates of EDs. Upper-middle-income countries, such as Brazil, reported higher levels of BN (2.0%) and BED (4.7%), compared to lower-middle-income countries like Colombia (0.4% for BN; 0.9% for BED). These findings underscore internal disparities among Latin American nations.

To better understand the complex relationship between food consumption and eating habits in people with EDs, it is necessary to understand the cultural and contextual differences regarding eating practices and food access across different countries and regions [33]. Prevalence data for eating disorders among university students in Brazil showed the following prevalence rates: AN (1.4%), BN (0.7%) and recurrent binge eating (RBE) (6.2%) [34]. Data on binge eating disorder (BED) were compared with bulimia nervosa (BN) using WHO mental health surveys in 14 countries. With a sample of 24,124 adults, it was found that the prevalence of BED consistently exceeds that of BN. Both disorders begin in late adolescence or early adulthood, with greater persistence in BN and in women [35]. Sproesser et al. [36] conducted a systematic literature review of food cultures, including traditional and modern eating practices across ten countries worldwide. The review examined 86 aspects of food cultures, including cooking methods and flavor principles. The analyses identified nine distinct food culture clusters, including: Brazilian, Chinese, Ghanaian, Indian, Japanese, Mexican, Turkish, African and Latin US American samples, and European-descendent (France, Germany and the U.S. European-descendent samples).The authors reported similarities in food culture between Mexico and Brazil -two Latin American countries- while showing significant differences with other, such as those from the European-descendant cluster.

There is some available literature regarding the association between EDs and food consumption and eating patterns. For instance, Moulrihe et al. [37] conducted a systematic review and meta-analysis of studies examining binge eating characteristics in individuals with EDs. The authors analyzed data from 42 studies that examined binge eating episode size in terms of duration and total energy content in people with EDs, specifically AN, BN, and BED. No significant differences were found in mean caloric intake between BN and BED participants in clinical studies. However, in laboratory settings, participants with BN exhibited higher caloric intake during binge eating episodes. It is important to note the differences based on the assessment tools used. For instance, dietary recall methods revealed significantly higher calorie intake than assessments using food diaries. Additionally, the authors mentioned that 90% of the studies included in their systematic review were from European countries or the U.S., suggesting that most studies examining food consumption in individuals with EDs are conducted in Western nations.

In an earlier study, the authors reported that participants with AN and restrained eating behavior (REB) had lower energy, fat, and micronutrient (Na, K, Cu, Zn, Fe, Se, B complex, vitamins D and E) intake [38]. Additionally, this group consumed more caffeine, fiber, fruits, vegetables, and legumes, and less fast food, sweets, and high-fat/high-carbohydrate snacks than controls or healthy adolescents. Individuals with REB had a higher protein intake from meat, poultry, fish, and eggs.

Similarly, Forbush and Hunt reported that girls with AN had lower than average levels of total calories and fat intake, but higher protein intake compared to controls [39]. Furthermore, they also had significantly higher intake of vitamins A, D, and K than the healthy individuals in the population. For individuals with BN or BED, the caloric intake during binge-eating episodes was higher than that of normal-weight or obese control subjects during eating episodes. Among individuals with BED, meal skipping and evening snacking were associated with an increased frequency of binge episodes. However, among individuals with BN, these behaviors were not associated with the frequency of binge eating or purging. The authors concluded that future research should aim to identify differences in eating patterns among different demographic groups of individuals with EDS, and, additionally, identify objective and valid assessment tools that can provide reliable data results.

As identified previously, the assessment of food consumption and eating habits in people diagnosed with EDs is instrumental in understanding the needs of this population, which can enable the development of more effective therapeutic and preventive approaches. Also, of importance is taking into consideration cultural differences when applying frequently used tools to assess food consumption and eating habits from Western cultures. Because of the limited information available, there is particular interest in identifying any existing gaps in knowledge regarding food consumption and eating patterns among people diagnosed with eating disorders (EDs) in Latin America and the Caribbean.

Therefore, the aim of this this scoping review is to gather and examine the available evidence on food consumption and eating patterns in people diagnosed with ED (specifically AN, BN, BED), in studies conducted in the Latin American and Caribbean countries, with particular interest in identifying food consumption assessment tools in this population. Specific questions guiding this scoping review were: (1) What is the extent of the existing literature on the association between food consumption and eating patterns and AN, BN, and BEDS in Latin American and Caribbean populations of all ages? (2) Is there research indicating differences between different countries (Latin America and the Caribbean) and/or contexts regarding food consumption patterns and eating behaviors in people diagnosed with EDs (AN, BN, and BEDS)? (3) What assessment tools were used to measure food consumption and eating patterns in these studies? (4) How were the participants in these studies diagnosed with ED? (5) What was the composition of the research team conducting these studies? This scoping review concludes with an overview of the findings, strengths, and limitations of the study and directions for future research.

Methods

We conducted a scoping review to describe the association between food consumption and eating patterns of people diagnosed with eating disorders, specifically AN, BN and BED, in studies conducted in the Latin American and the Caribbean countries. This scoping review was reported according to the recommendations of the Preferred Reporting Items for Systematic Review and Meta Analysis (PRISMA) Extension for Scoping Reviews [40]. The protocol has been developed and registered at Norbert Wiener University in Peru, with approval code Exp. No. 856–2021.

Eligibility criteria

The research question and eligibility criteria responded to a PCC (Population or participants/Concept/Context) framework (Table 1). Experimental or observational studies were included. Literature reviews, validations of questionnaires or instruments, books, letters to the Editor, and reports without diagnosis terms were excluded from the study.

Table 1.

Eligibility criteria for the selected studies (PCC Framework)

Criterion	Participants	Concept	Context
Inclusion	Participants diagnosed with EDs in the Latin American population and the Caribbean region, of both sexes, all ages, ethnicities and races. Includes information about their eating patterns or food consumption. Diagnostic definition of DSM IV, V or ICD-10 or ICD-11.	Anorexia nervosa is characterized by weight loss (low weight for height and age) leading to starvation, with a high mortality rate. The BMI is usually less than 18.5 in an adult with anorexia nervosa. Dietary behavior in patients is frequently rooted in fear of gaining weight or becoming overweight. Bulimia nervosa, is characterized by periods of strict dieting or eating only low-calorie “safe” foods alternate with binges on high-calorie “forbidden” foods. Binges involve eating large amounts quickly, with a sense of loss of control, often in secret and accompanied by shame. Eating continues past fullness, sometimes causing nausea or discomfort. Binge eating disorder is characterized by episodes of eating large amounts of food in a short time, accompanied by a loss of control and distress. Unlike other EDs, it does not involve regular compensatory behaviors like vomiting, fasting, excessive exercise or abusing laxatives. -Eating Patterns: The term is a general construct used to describe people’s eating patterns at the level of a “meal”, such as one of three main meals or a snack. Meals can be described based on three concepts: Pattern (spacing, frequency, regularity, timing), Format (type of food combinations, nutrient profile) or context (eating alone or with company, with television). For this study we are considering the concept related to format in relation to the type of food combinations or the nutrient profile [41, 42] -Food consumption: refers to the selection and intake of food and drinks by individuals, influenced by a complex array of factors including psychological, social, cultural, biological, and economic determinants.	-Population-based or clinical populations
Exclusion	Latinos living in countries other than Latin America and the Caribbean (i.e., Hispanic Americans or Latino Inmigrants). ED patients diagnosed with other comorbidities and other pathologies (ex., COVID-19, periodontitis, diabetes, cancer, cardiovascular diseases, among others). Participants not formally diagnosed with EDs, for example those participants identified with risky eating behaviors or those studies using ED screening instruments to select participants (no appropriate EDs clinic diagnostic procedures applied)	-Studies that included other EDs such as Pica, Rumination Disorder, Avoidant/Restrictive Food Intake Disorder (ARFID), and Other Specified Feeding and Eating Disorders (OSFED)

Search strategy

The searches were carried out until November 4, 2025. Three databases were searched: Medline (PubMe), Cochrane Library, and LILACS, to identify peer-reviewed literature, without language restriction.

(“eating disorder*“[Title/Abstract] OR “anorexi*“[Title/Abstract] OR “bulimi*“[Title/Abstract] OR “disordered eat*“[Title/Abstract] OR (“Feeding and Eating Disorders“[MeSH Terms] OR “Anorexia Nervosa“[MeSH Terms] OR “Bulimia“[MeSH Terms] OR “Bulimia Nervosa“[MeSH Terms] OR “binge eating disorder“[Title/Abstract])) AND (“Feeding Behavior“[MeSH Terms] OR (“Eating behaviors“[Title/Abstract] OR “Meal pattern“[Title/Abstract] OR “eating habits“[Title/Abstract] OR “dietary habits“[Title/Abstract] OR “diet habits“[Title/Abstract] OR “food intake“[Title/Abstract] OR “feeding behaviour“[Title/Abstract] OR “dietary patterns“[Title/Abstract])) AND (“Latin America“[MeSH Terms] OR “South America“[MeSH Terms] OR “Mexico“[MeSH Terms] OR “Caribbean Region“[MeSH Terms] OR “Guatemala“[MeSH Terms] OR “El Salvador“[MeSH Terms] OR “Honduras“[MeSH Terms] OR “Nicaragua“[MeSH Terms] OR “Costa Rica“[MeSH Terms] OR “Panama“[MeSH Terms] OR “Cuba“[MeSH Terms] OR “Argentina“[MeSH Terms] OR “Bolivia“[MeSH Terms] OR “Chile“[MeSH Terms] OR “Brazil“[MeSH Terms] OR “Colombia“[MeSH Terms] OR “Ecuador“[MeSH Terms] OR “Paraguay“[MeSH Terms] OR “Peru“[MeSH Terms] OR “Uruguay“[MeSH Terms] OR “Venezuela“[MeSH Terms].

Study selection

The selection of the studies was independently performed by four researchers: CB-CK and JDR-MLU. CB and MLU are nutritionists, CK and JDR are clinical psychologists. The Rayyan software, which is free of access, was used to screen the titles and abstracts of the selected articles in this phase. Next, all researchers read independently the full text of shortlisted articles.

Data collection process

All researchers carried out the data collection of each selected study in duplicate. For this, previously designed forms were used that will contain general information on the study (author, year), data of the included studies (country, age, sex, ethnicity, sample size, setting, professions among the research team), main diet-related results of the studies (eating disorder, tools, energy consumption, nutritional contempt, micronutrients, dietary patterns, favorite foods). Discrepancies were resolved through consensus.

Synthesis of results

The studies that met the eligibility criteria were qualitatively synthesized, describing their main findings.

Results

Study selection

The literature review identified 698 eligible studies, after removing duplicate studies, 673 unique titles were identified. The review of titles and abstracts resulted in the exclusion of 632 articles, leaving 41 full texts for review. After the full text review, four articles were included in the scoping review, including one added for other sources (see Fig. 1).

Fig. 1.

Flowchart of identification and selection of studiessss

The four studies included in the analysis were from Brazil, one in Sao Paulo [43] and three in Rio de Janeiro [44–46]. The first study was published in 2003, and the other three followed two decades later. It should be noted that the articles of Ferreira de Moraes et al. [45, 46], are part of the Binge Eating in Rio Study, which consisted of a household survey developed to assess the prevalence of BED and its correlates in a representative sample of Rio de Janeiro residents [33].

The results presented below begin by characterizing the ages and backgrounds of the participants of the studies, then present data on their food consumption and dietary patterns. Later, information is presented on the tools used to measure food consumption and dietary patterns, the diagnostic procedures, and finally, information on the composition of the research teams.

Food consumption and dietary patterns in Latin American and Caribbean populations diagnosed with ED of all ages.

The present study does not have enough data to present food consumption and dietary patterns according to age groups. In the included studies, the sample consisted mostly of adults in the early adulthood period, except for Mourilhe et al. [44], which included adolescents starting from 16 years. No studies were found with child populations that met the inclusion and exclusion criteria defined by the authors.

In the sample of Alvarenga et al. the mean age was 27.23 (SE = 9.6) [43]; in the case of Ferreira de Moraes et al. (2023a) and Ferreira de Moraes et al. (2023b) the same clinical sample was included [45, 46], the mean age being 40.3 (SE = 3.3) for the subsample with presence of BED, 31.9 (SE = 3.7) for the subsample with BN, and 34.7 (SE = 1.4) for the subsample with presence of Recurrent Binge Eating (RBE). RBE was defined as “episodes of binge eating occurring at least on a weekly basis, but not fulfilling criteria for a full ED diagnosis” [46]. Additionally, Ferreira de Moraes et al. [45] included a subsample without a diagnosis of ED, with a mean age of 38.3 (SE = 0.4). In the study of Mourilhe et al. [44], the mean age of the whole sample was 30.3 (SE = 9.9), for the subsample with BED 36.4 (SE = 13.1), and for the BN 33.4 (SE = 12).

Food consumption and dietary patterns in the Latin American and Caribbean population with a diagnosis of ED according to country or context.

The present study does not have enough data to present food consumption and dietary patterns according to country, because no studies were found that met the inclusion and exclusion criteria established for the scoping review, in other Latin American and Caribbean countries, except Brazil. There are differences in food consumption and dietary patterns depending on the context in which the studies were conducted, but this information will be developed in a later section.

According to the context in which the studies were conducted, it was observed that Alvarenga et al. included 30 patients who attended a cognitive behavioral therapy group for the treatment of BN, in a hospital belonging to the tertiary sector of the city of Sao Paulo [43]. Ferreira de Moraes et al. (2023a); Ferreira de Moraes et al. (2023b), conducted the study from a representative sample of the city of Rio de Janeiro [45, 46]. Ferreira de Moraes et al. (2023a) included a sample of 2,297 adults, divided into 2,161 participants without ED and 136 in the binge eating spectrum disorder (BESD) category, which included BN (n = 17), BED (n = 29), and RBE (n = 90). Furthermore, Ferreira de Moraes et al. (2023b) included only the subsample of 136 participants with BESD. The sample from the study of Mourilhe et al., was composed of 114 participants, 57 with a diagnosis of BN, and the other 57 with BED [44]. The study was carried out in an outpatient ED unit belonging to the Universidade Federal do Rio de Janeiro (UFRJ). The number of participants included in the recent studies was larger compared to Alvarenga et al., and with greater diversification in terms of sex and ethnicity.

The data presented below on food consumption and dietary patterns include samples from two of the most populated cities in Brazil, specifically Sao Paulo and Rio de Janeiro. Each of the samples presents differences in terms of the number of participants, contexts, and sociodemographic characteristics. The participants of the studies had a diagnosis of BN, BED, and RBE. It should be noted that no studies were found that included participants with a diagnosis of AN. The characteristics of the studies are presented in Table 2.

Table 2.

Summary of data of the included studies

Author/year	City/country	Sample size	Age	Sex	Ethnicity	Setting	Professions among the research team
Alvarenga et al. (2003)	Sao Paulo, Brazil	30 participants	Mean age = 27.23 ± 9.6	Women	White	Trial CBT for BN, at the biggest tertiary-care hospital.	Dietitians and psychiatrists.
Moraes et al. (2023a)	Rio de Janeiro, Brazil	2297 participants (136 participants diagnosed with BESD (BED = 29; BN = 17; RBE = 90), and 2161 participants without ED).	BESD: BED= mean age of 40.3 ± 3.3; BN= mean age of 31.9 ± 3.7; RBE= mean age of 34.7 ± 1.4. Without ED = 38.3 ± 0.4.	Men, women	White, black, mixed	An in-person, population-based, household survey among a representative sample of the residents of Rio de Janeiro.	Mental health departments (dietitians, psychiatrists, and psychologists).
Moraes et al. (2023b)	Rio de Janeiro, Brazil	136 participants diagnosed with BESD (BED = 29; BN = 17; RBE = 90).	BESD: BED= mean age of 40.3 ± 3.3; BN= mean age of 31.9 ± 3.7; RBE= mean age of 34.7 ± 1.4.	Men, women	White, black, mixed	An in-person, population-based, household survey among a representative sample of the residents of Rio de Janeiro.	Mental health departments (dietitians, psychiatrists, and psychologists).
Mourilhe et al. (2024)	Rio de Janeiro, Brazil	114 participants diagnosed with BESD (BED = 57; BN = 57).	BESD: BED= mean age of 36.4 ± 13.1; BN= mean age of 33.4 ± 12.	Men, women	White, black, mixed	Outpatient ED unit of the Universidade Federal do Rio de Janeiro (UFRJ).	Dietitians and psychiatrists.

BESD Binge Eating Spectrum Disorders (including BED, BN, and RBE), ED Eating disorder, CBT Cognitive behavioral therapy

Food consumption

In the reviewed articles, the authors included various aspects related to food consumption in people diagnosed with ED, such as energy intake, macronutrient intake, micronutrient intake, and food preferences. Additionally, Ferreira de Moraes et al. (2023a) included Hyperpalatable food (HPF) consumption during binge eating episodes [45]. Each of the above aspects reveals dietary specificities according to the ED diagnosis.

In the study by Alvarenga et al. participants diagnosed with BN were included [43]; in the studies by Ferreira de Moraes et al. (2003a) and Ferreira de Moraes et al. (2023b), participants diagnosed with BN, BED, and RBE were included [45, 46]; while Mourilhe et al., incorporated participants with BN and BED [44]. Among the articles analyzed, only Alvarenga et al. presented results on all food consumption variables, except for HPF consumption. Ferreira de Moraes et al. (2023a) presented data such as energy intake and HPF consumption during binge and non-binge meals. Ferreira de Moraes et al. (2023b) reported on energy intake, nutritional content, and preferred foods during objective binge eating episodes (OBE). Meanwhile, Mourilhe et al. presented data related to energy consumption during binge eating episodes, as defined according to DSM-5. Only data from the BN and BED groups without a diagnosis of anxiety or depression were taken into consideration in our scoping review.

Energy intake during binge eating episodes was also presented according to the diagnosis received. The highest consumption during objective binge eating episodes was found in the subsample diagnosed with BED, with a mean of 1184 kcal (95% CI: 891;1476 kcal), followed by the subsample diagnosed with BN, where the mean obtained was 1025 kcal (95% CI: 680; 1365 kcal), and finally, the subsample with the presence of RBE, with a mean of 994 kcal (95% CI: 793; 1195 kcal). It should be noted that no statistically significant differences were found between groups. Ferreira de Moraes et al. (2023a) also included data on energy consumption in subjective binge eating episodes and non-binge meals, according to different diagnoses [45] (see Table 3). Nor were there any statistically significant differences. In the study of Mourilhe et al., the highest caloric intake among the groups without a diagnosis of depression or anxiety was observed in the subsample of BED without anxiety, with a mean of 2026.7 kcal ± 1534.0; while the lower consumption was identified among the BED without depression, with a mean of 1546.3 ± 655.7 [44]. Calorie consumption was significantly higher in the group diagnosed with depression. Data related to comorbidities such as depression and anxiety is not included because it is beyond the scope of this study.

Table 3.

Main diet-related results of the studies

Autor	ED	Tools	Energy consumption	Nutritional contempt	Micronutrients	Dietary patterns		Additional information
Alvarenga et al. (2003)	BN	Food diary (during 14 days).	Highly variable energy consumption (19, 257 kcal to 158 kcal/daily).	51% carbohydrates, 36% fats and 14% proteins.	Below RDI levels: intake of vitamin E, folic acid, magnesium, and iron. After vomiting: vitamin B6, calcium and zinc.	PWSV (breakfast, lunch and dinner). PSV (fewer than three meals on 26% of days).		Prior to vomiting episodes: Biscuits, crackers, cakes, and desserts (74%), rice (63%), beef (58%), fruit (50%), chocolate consumption (37%), and foods with above-average amounts of fat (e.g., butter, lard, eggs, french fries) (32%). milk, soda, and juice were frequently consumed.
Moraes et al. (2023a)	BN, BED, RBE	24-hour dietary recall (non-binge meals) and binge eating recall.	OBE: BED (M = 1184 kcal), BN (M = 1023 kcal), RBE (M = 994 kcal). No significant differences were observed. SBE: BED (M = 710 kcal), BN (M = 954 kcal), RBE (M = 639 kcal). No significant differences were observed. Non-binge meals: PWD (M = 1783 kcal), BED (M = 2153 kcal), BN (M = 2219 kcal), RBE (M = 1822 kcal). No significant differences were observed.	No data	No data	No data		Energy from highly palatable foods in the OBE (BED 69%, BN 62%, and RBE 76%). No significant differences were observed. Energy from highly palatable foods in SBE (BED 76%, RBE 72%, BN 50%). No significant differences were observed. Energy from highly palatable foods in NBE (BED 63%, RBE 63%, BN 44%). No significant differences were observed.
Moraes et al. (2023b)	BN, BED, RBE	24-hour dietary recall (non-binge meals) and binge eating recall	OBE: BED (M = 1184 kcal), BN (M = 1023 kcal), RBE (M = 994 kcal). No significant differences were observed.	OBE: BED (59% carbs; 15% prote; 28% lip), BN (56% carbs; 14% prote; 32% lip), RBE (58% carbs; 14% prote; 32% lip). Did not differ significantly.	No data	No data		During OBE: BED = staple foods (25%) (BED vs. BN: p = 0.01), chocolates/sweets/desserts (18.6%), meats (12.2%), fast foods (10.3%), and sugary drinks (10.3%). BN = chocolates/sweets/desserts (23.4%), sugary drinks (18.7%) (BED vs. BN = 0.05), fast foods (15.6%), and meats (10.9%). RBE = chocolates/sweets/desserts (21.9%), staple foods (18.4%) (RBE vs. BN: p.0.04), and sugary drinks (11.3%).
Mourilhe et al. (2024)	BN, BED	Dietary recall of a representative binge eating episode, defined according to DSM-5.	Total-ND (M = 1770.0), BN-ND (M = 1981.3), BED-ND (M = 1546.3) Total-NA (M = 2026.7 BD-NA (M = 1929.9), BED-NA (M = 2177.9)	No data	No data	No data		Participants with BN or BED comorbid with depression had a higher consumption of calories than those without depression (p.0.025).

OBE Objective binge eating episodes, SBE Subjective binge eating episodes, PWD People without diagnosis, Total-ND Participants with BN or BED with no depression, BN-ND BN with no depression, BED-ND BED with no depression, Total-NA Participants with BN or BED with no anxiety, BN-NA BN with no anxiety, BED-NA BED with no anxiety, Carbs Carbohydrates, Prote Proteins, Lip Lipids, RDI Recommended daily intake, PWSV Patients without self-induced vomiting, PSV Patients with self-induced vomiting, NBE Non-binge eating episodes

Regarding nutritional content, Alvarenga et al. indicate that the most common macronutrients in the participants’ diets were carbohydrates (51%), lipids (36%), and proteins (14%). These figures do not vary greatly when referring exclusively to OBE, with the distribution being 50% carbohydrates, 30% lipids, and 20% protein. When comparing diagnoses, no statistically significant differences were found in the macronutrient consumption [46]. Regarding micronutrients, Alvarenga et al. reported that consumption of vitamin E, folic acid, magnesium, and iron is below the Recommended Daily Intake (RDI) levels. If only the content of meals not resulting in vomiting were considered in this analysis, deficiencies in vitamin B6, calcium, and zinc would be observed. The remaining two articles did not present data on the micronutrient variable.

In terms of frequently ingested foods, the presence of biscuits, crackers, cakes, and desserts was observed in 74% of eating episodes with vomiting, followed by the consumption of beef (58%) and rice (63%). To a lesser extent, the consumption of chocolate (37%) and foods with above-average amounts of fat (e.g., butter, lard, eggs, chips) (32%) occurred [43]. HPF were frequently consumed during binge eating episodes (OBE and SBE), although individuals without an ED diagnosis and with BN reported lower consumption, compared to the RBE and BED groups [45]. No significant differences were observed due to the small sample size in the BN group (n = 17). During OBE, foods such as rice and beans, frequently consumed in Brazil and other Latin American countries, were more frequently consumed by individuals diagnosed with BED (25%) or RBE (18.4%), but not by those with BN (BED vs. BN: p = 0.01; RBE vs. BN: p = 0.04). In BN, consumption of chocolates/sweets/desserts (23.4%), fast foods (15.6%), meats (10.9%), and sugary drinks (18.7%) predominated, but no significant differences among groups were identified. Liquids such as milk, soda, and juice were frequently consumed in each of the subsamples, but participants with BN showed a higher consumption of sugary drinks (18.7%) (BED vs. BN: p = 0.05) (Ferreira de Moraes et al., 2023b).

Eating patterns

Participants’ eating patterns varied depending on the presence or absence of self-induced vomiting. Alvarenga et al. showed that participants without self-induced vomiting ate all three main meals, while those with self-induced vomiting ate fewer than three meals on approximately a quarter of the days (26% of days) [43]. The remaining three articles did not present data on the eating pattern variable. The characteristics associated with food consumption and eating patterns are presented in Table 3.

Each article included in the scoping review presents data that characterizes the diet of people diagnosed with an eating disorder, primarily BN and BED. To obtain this data, the authors used tools frequently employed in nutritional studies. The main results obtained are detailed below.

Tools for dietary assessment

Two tools were used to collect information on the diet of people with an eating disorder: food diaries and dietary recall, both of 24-hour and of binge eating episodes. Ferreira de Moraes et al. (2023a) and Ferreira de Moraes et al. (2023b) utilized a 24-hour recall software for participants to report the amount of food consumed, preparation methods, and any added condiments or sugar/sweeteners [45, 46]. In addition to the 24-hour dietary recall, Ferreira de Moraes et al. (2023a) and Ferreira de Moraes et al. (2023b) used the Questionnaire on Eating and Weight Patterns 5 (QEWP-5) to assess binge eating recall. The following question from the QEWP-5 was included: (e.g., “as best you can remember, please list everything you ate and drank during that episode. Please list the foods eaten and liquids consumed during the episode. Be specific—include brand names where possible and amounts or portion sizes as best you can estimate”) [47]. In the study of Mourilhe et al., a dietary recall was also employed, but with the difference that the time between the last binge episode could be as long as a month [44], unlike the three months in the Ferreira de Moraes et al. studies [45, 46]. While Alvarenga et al. asked participants to complete food diaries to record the type and amount of food consumed, including beverages, for 14 days [43].

Methods used to diagnose the participants’ ED

In all included articles, the clinical assessment of participants was described in the methodology section. Alvarenga et al. conducted clinical interviews based on DSM-IV criteria [43]. Ferreira de Moraes et al. (2023a) and Ferreira de Moraes et al. (2023b) conducted screening of the general population using the Questionnaire on Eating and Weight Patterns (QEWP-5) and then the Structured Clinical Interview of DSM-IV (SCID-P), adapted to DSM-5 [45, 46]. Interviews were conducted by telephone with those who tested positive or close to the threshold for the disorder and with a subsample of negative results. Interviewers during this phase were blinded to the results obtained in the previous screening phase. In the study of Mourilhe et al., the SCID-P was also used to confirm the diagnosis of ED [44].

Research team establishment

Because the research team for this scoping review consisted of dietitians and psychologists, it was of interest to learn how common such interdisciplinary teams are in the Latin American studies identified. It was observed that in all the studies, the research team was composed of dietitians and mental health professionals, mainly psychiatrists. Among the dietitians’ responsibilities, they trained participants to complete food diaries and to differentiate between episodes of binge eating, regular meals, and the presence of induced vomiting [43], or to answer dietary recalls [44, 45, 46], while the mental health professionals were responsible for assessing eating disorders [43, 44]. However, in the case of Ferreira de Moraes et al. (2023a) and Ferreira de Moraes et al. (2023b), it is indicated that two PhD students, both nutritionists, conducted clinical interviews and were supervised by a senior psychiatrist.

Discussion

The evidence searches identified four studies that address the association between food consumption and eating patterns with AN, BN and BED in populations living in Latin America and the Caribbean. One point to note is that this study considered studies that gave their patients a diagnosis of ED through clinical interviews; the application of screening tests with positive scores alone was not considered sufficient.

The use of a two-stage design is widely used and recommended in large population-based studies. Due to operational and financial issues, it is impossible to conduct clinical interviews in representative samples of metropolitan cities [48, 49] In the Ferreira et al. studies (n = 2,297), the screening instrument (QEWP-5) was specifically validated in the Binge Eating in Rio survey, demonstrating adequate psychometric properties to identify binge eating spectrum disorders when compared to the clinical interview SCID-P [34]. Conversely, in the Alvarenga et al. study, the authors conducted only the SCID-P as the assessment was performed in a clinical setting and with a convenience sample (n = 34) of individuals sourced from an RCT.

BED affects approximately 4.9% of women and 4% of men throughout their lives. The prevalence of BN is 1.9% in women and 0.6% in men. Both BN and BED are more common in Latinos and African Americans, and BED is more common in racial/ethnic and non-Latino white Americans [50]. Acculturation is also associated with increased eating pathology, and discrimination contributes to emotional eating [51]. African American and/or women of colour have similar or higher rates of binge eating than white women. Most research on BE and BED has focused on white women, yet African Americans and/or women of colour have been underrepresented in clinical trials [52]. BED is higher among racial/ethnic minorities. BN is more prevalent in Latinos and African Americans, and the prevalence of BED is similar in these groups [53]. Very few studies have been conducted on multiracial/multiethnic individuals. Research is needed on racial/ethnic minorities who are at higher risk for eating disorders and the pathways by which they develop these disorders [54]. The Slovenian population has a high prevalence of binge eating, obesity and anxiety. Anxiety is a factor that links negative affect and binge eating. They describe that younger participants showed the highest anxiety [55]. The higher prevalence of BED, BN and RBE among black people can be attributed to several interrelated factors, such as food insecurity and body image dissatisfaction [56]. Data for the USA, in a sample that included 11,962,287 Medicare beneficiaries, of whom 0.15% had a diagnosis of eating disorder [57].

According to Bjorlie, the median consumption during BEDs was 2134.22 kcal, and during restrictive periods 558.53 kcal [58]. Among the foods that contribute a high amount of kilocalories in the BEDs are highly palatable foods, which could indicate that the vast majority of the sample consumes these foods. During restricting episodes, the consumption of these foods was much more variable. During BED episodes, high palatability foods, mainly (fat) (meal-based foods), accounted for the highest percentage of kcals consumed, followed by the carbohydrate group (mainly snacks). The kcals from the fat and sugar group represented the lowest percentage during BED. During restrictive episodes about half of the kcals consumed came from highly palatable foods.

In 2024, author Jenkins described that people with BN showed wide variation in energy intake. Sodium may be overconsumed in BN and BED. Vitamin D was underconsumed in all disorders [59]. According to author Wiklund, the intake of 391 women and 39 men with BED exceeded the recommendations for saturated fats, while omega fatty acids did not meet the recommendations. The consumption of vitamin D, selenium, and salt was low. Women in the control group had low iron and folate intake [60]. The author Latzer evaluated patients with binge eating disorder (BED) with and without night eating syndrome (NES). The study included 59 women (aged 18–60) who sought treatment for eating disorders (ED) and were diagnosed with BED or bulimia nervosa (BN) (BED) with or without NES. The BED-NSSM group reported a significantly higher frequency of binge days and episodes during the week, as well as higher energy and fat intake than the group without BED-NSSM [61].

In order to know how far food intake falls short of a person’s daily intake requirements, it is necessary to assess the type and frequency of food, the quality of the diet and the calories consumed. Although all existing methods are associated with misreporting, there is some evidence suggesting that 24-hour recalls exhibit a better performance when compared to the food diaries, both in the general population [62, 63] and in individuals with ED [64]. However, when considering the ability to assess dietary patterns, diet history seems to be more appropriate. Furthermore, it may be worth addressing that few studies investigated the accuracy of such methods in ED samples as well as the potential strategies to overcome these limitations (e.g., Multiple-pass method – [65]), as conducted in the Moraes et al. studies.

Other findings are associated with the consumption of beverages such as soda during BED episodes, which reduce hunger and facilitate vomiting. Some patients tend to drink more fluids when experiencing hunger or when they do not want to gain weight; conversely, they may consume less to exert control over their eating behavior [66, 67]. Other commonly consumed beverages include coffee with artificial sweeteners [68].

In order to know how far food intake falls short of a person’s daily intake requirements, it is necessary to assess the type and frequency of food, the quality of the diet and the calories consumed, all of which help in the diagnosis and treatment of eating disorders.

The most accurate tool with the lowest risk of bias for assessing dietary intake is the daily food record, compared to 24-hour recalls and food frequency questionnaires [69]. Inaccurate reports can lead to erroneous conclusions about the dietary patterns of individuals with BED or BN, affecting both clinical assessments and public health recommendations [70]. To develop accurate dietary intake measures, it is essential to involve nutritionists, dietitians, or related professionals who are trained and standardized in the use of dietary assessment tools and procedures. Another important aspect is reactivity—the fact that individuals are aware they will be recording their food intake may influence the types of foods they choose and the quantities consumed [71].

The assessment of patients in Alvarenga’s study was conducted by psychiatrists and dietitians, In contrast, in Ferreiras´ studies, the assessment team consisted of psychiatrists and doctoral students who interviewed participants by telephone, under the supervision of a senior psychiatrist with fieldwork experience.

According to Calyniuk et al., regarding the work of an interdisciplinary team, evidence suggests that psychotherapy combined with nutrition results in better therapeutic outcomes than psychotherapy alone. Nutritional education can teach patients the appropriate composition of meals and enhance their knowledge of healthy eating practices [72]. Among those diagnosed with eating disorders, the majority of participants do not consult a psychotherapy specialist (68%). However, dieticians were the professionals most frequently contacted (around 30%) among those who sought treatment for their condition. In Appolinario’s studio several medical conditions, when controlling for body mass index, were significantly more prevalent in people with BED, BN and RBE, and under half had sought treatment [33].

A systematic review shows that barriers to seeking help for BED treatment involve identifying behaviours that indicate an eating disorder [73]. In the study, individuals with compulsive eating behaviors constituted one of the groups with the lowest levels of nutritional knowledge, whereas those undergoing psychological and/or nutritional therapy demonstrated greater knowledge compared to those not receiving any form of treatment.

There is still limited evidence on this topic in the Latin American population, and there should be more evidence on evaluations and interventions for recently recognized but common and harmful ED that can be applied in the Latin American context. Therefore, promoting research in these countries is necessary.

Future research on the topic should emphasize methodological rigor and standardization for measuring food consumption and eating disorders. Future research should consider a transdiagnostic approach to ED, taking into account existing evidence on diagnostic crossover, as well as shared transdiagnostic factors between eating disorders and other psychiatric diagnoses [74].

Few studies complied with the inclusion criteria. Those without a formal ED diagnostic assessment, and those presenting comorbidities, were not included in the review, forming a stronghold. Another limitation is that all of the selected studies were from only one country, namely Brazil. Given that the Latin American and Caribbean region is a culturally, ethnically, and economically diverse region, the findings presented in this scoping review are not generalizable to other countries of the region. Interesting to note is that Brazil is one of the largest economies of the region, thus, information from other, less economically advantaged countries in the region is not represented in this review. Finally, an important limitation is that none of the reviewed studies included individuals diagnosed with AN, leaving the dietary intake of this population unknown. However, differences in food consumption were observed according to the diagnosis of BN and BED in the studies analyzed in this scoping review.

Conclusions

Common foods in Brasil, such as rice and beans were frequently consumed in the studies selected in this scoping review. A thorough understanding of food consumption and eating patterns is key to designing and implementing evidence-based interventions that can be tailored made to the specific needs of the Brasil population with ED.

Alvarenga describes that the diets of BN patients consisted of 51% carbohydrates, 36% fats, and 14% proteins. The intake of vitamin E, folic acid, magnesium, and iron is below survey levels. The author Moraes 2023 (a) does not describe micro or macronutrients; the patients’ energy consumption comes from palatable foods. Patients obtain energy from palatable foods with a diagnosis of BED 69%, RBE 76%, and BN 62%. In the Moraes (b) study, the diets of patients diagnosed with BED consisted of 59% carbohydrates, 15% protein, and 28% lipids; in patients with BN, the composition was 56% carbohydrates, 14% protein, and 32% lipids. In Mourilhe’s study, energy consumption in patients with BN without depression was 1981.3 kcal and in patients with BED without depression was 1546.3 kcal.

More studies are needed to validate and culturally adapt food consumption and eating patterns assessment tools in this population. Future studies may expand the search as the number of publications in the region related to food consumption among individuals with EDs increases.

Author contributions

M.L.U. and C.B. conceived the idea for the study. C.B. conducted the systematic literature search. C.K., J.D.R., C.B., and M.L.U. participated in the peer evaluation of the selected documents. All authors contributed to the first draft and approved the final version of the manuscript.

Funding

This research was developed with the competitive fund of the UPNW grant number 2021” RESOLUCIÓN N° 851- 2021-R-Universidad Norbert Wiener.

Data availability

The data used to support the findings of this study are included within the article.

Declarations

Ethics approval and consent to participate

Not applicable, we worked with a secondary data.

Consent for publication

All authors gave their consent for publication.

Competing interests

The authors declare no competing interests.