Abstract
Purpose
This study assessed the associations between the consumption of sugar- (SSBs) and artificially sweetened beverages (ASBs) and body weight from childhood through adolescence. Moreover, we examined the relationship between childhood weight status and the joint associations between SSBs and ASBs and weight status on eating disorders (ED) symptoms in early adulthood.
Methods
This study used the Danish National Birth Cohort (DNBC) at the 11- and 18-year follow-up (DNBC-11 and DNBC-18). At age 11, children reported their consumption of SSBs and ASBs. Body weight was assessed using BMI based on information from DNBC-11 and DNBC-18. ED were derived from self-reported symptoms and national health registers. Multivariable logistic regression models were used.
Results
Children with a higher consumption of SSBs and ASBs at age 11 had higher odds of overweight or obesity at age 11 (OR: 1.43, 95%CI 1.24–1.64) and age 18 (OR: 1.20, 95%CI 1.00–1.43). Children with overweight (OR: 1.36, 95%CI 1.08–1.70) and obesity (OR: 2.08, 95%CI 1.09–3.97) at age 11 had higher odds of ED symptoms at age 18. Children with overweight or obesity who consumed SSBs and ASBs below and above the median at age 11 had higher odds of ED symptoms at age 18 than those with underweight or normal weight and a consumption below the median (OR: 1.41, 95%CI 1.04–1.92 and OR: 1.50, 95%CI 1.12–2.00, respectively).
Conclusions
SSBs and ASBs consumption during childhood was associated with overweight and obesity in the short- and long-term. Childhood overweight or obesity was also related to later ED symptoms.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00394-025-03869-6.
Keywords: Sugar-sweetened beverages, Artificially sweetened beverage, Body mass index, Obesity, Eating disorders, Children
Introduction
The growing prevalence of overweight and obesity during childhood represents an escalating public health concern. According to the most recent report from the World Health Organization (WHO) in 2022, approximately one in five children and adolescents aged between 5 and 19 worldwide had overweight or obesity [1], and projections indicate that by 2050, one in three children and adolescents may be living with overweight or obesity [2]. Excess weight in early life can impact health in the short term, including mobility, sleep, and psychological well-being, and later in life, increasing the risk of chronic diseases such as type 2 diabetes, hypertension, cardiovascular disease, and certain types of cancer [3]. Further, the presence of overweight and obesity has been linked to disordered eating behaviors during adolescence [4–6]. In fact, recent estimates suggest that approximately 22% of children and adolescents globally exhibit disordered eating behaviors [7], which may have the potential to progress into clinical eating disorders (ED). While traditionally addressed separately, there is growing recognition of the frequent co-existence of overweight or obesity and ED symptoms, and of the potential for shared risk factors, as well as of possible transitions between these conditions [4, 8]. However, despite this well-established co-occurrence, the underlying mechanisms and directionality remain unsettled, and evidence across studies remains inconsistent [9]. Given that adolescence is a period characterized not only by physical and psychological developments but also by the establishment of lifestyle behaviors that often persist into adulthood [10], early identification of risk factors to enable effective interventions remains crucial.
Among the multiple modifiable lifestyle risk factors [11], dietary habits and, specifically, the consumption of sugar- and artificially sweetened beverages (SSBs and ASBs) has been associated with overweight and obesity, among other cardiometabolic and weight-related outcomes [12, 13]. Although the trend has stabilized in recent years, youth consumption remains high enough to raise public health concern [14]. While SSBs and ASBs consumption has been linked to overweight and obesity in children and adults [15, 16], existing evidence is inconsistent due to study heterogeneity and relatively short follow-up periods. Moreover, most studies have not accounted for weight concerns or disordered eating behaviors, which are common during adolescence [17]. These gaps limit our understanding of how weight status, disordered eating behaviors, and consumption of SSBs and ASBs interrelate during adolescence, and may hinder the development of effective prevention strategies.
Therefore, the main aim of this study was to assess the association between the consumption of SSBs and ASBs at age 11 years and body weight status at both ages 11 and 18 years in a population of Danish children and adolescents. We further examined the association between body weight status at age 11 and the incidence of ED symptoms at age 18. Finally, we assessed the joint association between SSBs and ASBs consumption and weight status at age 11 and ED symptoms at age 18.
Methods
Study design and population
The Danish National Birth Cohort (DNBC) is a longitudinal, nationwide cohort designed to investigate the associations between early life exposures and diseases later in life, aiming to provide insights for disease prevention. Between 1996 and 2002, all pregnant women intending to carry their pregnancy to term and who had sufficient proficiency in Danish to participate in telephone interviews were invited to the DNBC through their general practitioner. This resulted in the enrolment of approximately 100,000 pregnancies and 96,000 liveborn children, representing around 28% of the eligible Danes born during these years [18]. The DNBC data collection encompasses a broad range of information from prenatal life to early adulthood, including 11-year follow-up (DNBC-11) and the most recent 18-year follow-up (DNBC-18), which was initiated in 2016 and finally completed in 2022. Further details of the cohort [18], all the data available, and previous publications can be accessed at https://www.dnbc.dk/. All individuals residing in Denmark are assigned a unique personal identification number [19], which facilitates the linkage of data from the DNBC with register data on additional sociodemographic characteristics of the population and hospital-recorded diagnosed ED.
In the current study, of the 96,822 live-born children in the DNBC cohort, only those invited to both the DNBC-11 and DNBC-18 were included. Among these, 49,956 completed the questionnaires at DNBC-11 and 31,354 at DNBC-18. The population was further restricted to participants with available data on weight status at the 11-year follow-up (n = 40,760), for the cross-sectional associations, and further to those with data on both weight status (n = 24,292) and ED symptoms (n = 23,370) at the 18-year follow-up, for the longitudinal analyses. When ED symptoms were analyzed as the primary outcome, participants without information on this outcome were excluded (n = 374). To account for substantial physical and emotional changes during early adolescence [20], which may potentially influence ED symptoms development [21], children older than 12 years at the time of responding to DNBC-11 (n = 548) were excluded from the analysis, consistent with previous DNBC studies [22, 23]. Data were further restricted to participants with complete information on all covariates included in the analysis. In longitudinal analyses, children with overweight or obesity, or ED symptoms at DNBC-11 were excluded accordingly from the corresponding analysis. The final sample size for weight status analyses was 35,173 at the 11-year follow-up, 21,350 at the 18-year follow-up and 20,697 children-adolescents for ED symptoms analysis at the 18-year follow-up (Supplemental Fig 1).
Sugar-sweetened beverages assessment
In the DNBC-11, children completed up to 25 items specifically designed to evaluate the intake of SSBs and ASBs. Specifically, five items per individual type of sweetened beverages, including sugar-sweetened soda, light soda, lemonade, milkshake (or similar, such as chocolate milk, yogurt drink, etc.), and fruit juice (Supplemental Table 1). The responses to these items were converted into a weekly consumption of servings (200 ml/wk.). Finally, the total consumption of SSBs and ASBs was calculated as the sum of the individual consumption of soda, light soda, and lemonade. Juices and milkshakes consumption were excluded from the total SSBs and ASBs consumption calculation, as their nutritional profiles (e.g., natural sugar content or inclusion of dairy) differ from those of soda, light soda, and lemonade [24]. Total consumption of SSBs and ASBs was categorized into five groups: 0 serv/wk.; > 0 to < 4 serv/wk.; 4 to < 7 serv/wk.; 7 to < 11 serv/wk.; and ≥ 11 serv/wk. While individual consumption of SSBs and ASBs subtypes was categorized into four groups: 0 serv/wk.; > 0 to 2.5 serv/wk.; > 2.5 to 4 serv/wk.; > 4 serv/wk.
Weight status assessment
In the DNBC-11, the parents provided information on the child’s most recent height and weight measurements. In the DNBC-18, the adolescent provided information about their height and weight. Weight status was assessed using the body mass index (BMI), which was calculated by dividing weight in kilograms by the square of height in meters. Weight status at age 11 was categorized in accordance with the extended International Obesity Task Force (IOTF) age- and sex-specific curves [25], and that at age 18 was categorized in accordance with the World Health Organization (WHO) guidelines [1], which aligns with 18-year BMI cutoffs from IOTF [25]. The resulting classification comprised four categories: underweight, normal weight, overweight, and obesity.
Diagnosed eating disorders and self-reported symptoms
A composite outcome was determined by considering two main sources: national registers and self-reported symptoms in DNBC-11 and -18, thereby leveraging the complementarities offered by the availability of both diagnosed ED and self-reported ED symptoms among DNBC participants, as in previously published studies from the same cohort [26]. The Danish National Patient Register (NPR) [27], encompassing all hospital contacts, including private hospitals, and the Danish Psychiatric Central Research Register (PCRR) [28] provided information on diagnosed ED between ages 6–18 years. The International Classification of Diseases, 10th version, was used to define diagnoses of anorexia nervosa (F50.0, F50.1), bulimia nervosa (F50.2, F50.3), and eating disorder not otherwise specified (F50.8, F50.9).
In the DNBC-11 and DNBC-18, information on eating behaviors, weight and shape concerns was collected through web-based questionnaires using items adapted from the Youth Risk Behaviour Surveillance System Survey [29] and the McKnight Risk Factor Survey [30]. Threshold and subthreshold measures of ED symptoms were defined according to the Diagnostic and Statistical Manual of Mental Illnesses, 5th edition, and additional definition criteria stated in previous studies [31–33]. Detailed information about the ED symptoms-related items and possible responses found in the DNBC-11 and DNBC-18 has been described elsewhere [22, 34]. The adopted approach to categorized self-reported ED symptoms in the DNBC can be found in Supplemental Table 2.
In this study, individuals with any self-reported threshold or subthreshold ED symptoms or diagnosed ED were categorized as cases of ED symptoms, reflecting that the majority of cases were ascertained based on self-reported symptoms (95% at DNBC-11 and 85% at DNBC-18).
Covariates
Information on age and sex, parental age at birth, household income, parental educational level, and parental psychiatric disorders at age 11 was obtained from Danish Registers. In the DNBC-11, participants were asked items related to physical activity and screen time, which were subsequently defined according to a previously published DNBC study [35]. Behavioral problems were assessed using the Strengths and Difficulties Questionnaire (SDQ) criteria [36]. Parents reported their height and weight in the DNBC-11, and BMI was calculated accordingly, as previously stated. Dietary intake of fruit, fast food, milk, snacks, ice cream, chips and cake was also collected in the DNBC-11 (Supplemental Table 1). Data on smoking habits were collected during pregnancy, and information on family type in the DNBC-11. If data on these two aforementioned covariates were missing, information from Danish registers was imputed.
Statistical analysis
Differences among categories of total and individual consumption of SSBs and ASBs for general characteristics of the study population were assessed using one-way ANOVA and chi-square tests, as appropriate. The descriptive results were expressed as means and standard deviation (SD) for continuous variables or numbers and percentages for categorical variables.
Logistic regression models were used to estimate the odds ratios (ORs) and 95% confidence intervals (CI) of overweight or obesity at 11 and 18 years of age according to total consumption of SSBs and ASBs at 11 years. The first category (0 serv/wk.) was used as the reference group. Model 1 was adjusted for age (years) and sex (female, male). Model 2 was additionally adjusted for household income (quartiles), parental educational level (low, medium, high), family type at DNBC-11 (living with both parents, living with one parent, living without parents), maternal smoking during pregnancy (non-smoking, stopped during pregnancy, smoking during pregnancy), and maternal and paternal age at birth (years). Model 3 was additionally adjusted for children’s physical activity (inactive, lightly active, moderately active, vigorously active), screen time (< 2 h/d, 2 to < 4 h/d, 4 or > 4 h/d) and dietary variables at DNBC-11, specifically consumption of fruit, milk, fast food, ice-cream, cake, chips and snacks (low, moderate, high). Finally, Model 4 was further adjusted for maternal and paternal BMI (kg/m2) at DNBC-11. A sensitivity analysis was performed by further adjusting model 4 for the individual consumption of milkshakes and juices (0 serv/wk., > 0 to 2.5 serv/wk., > 2.5 to < 4 serv/wk., ≥ 4 serv/wk.) at DNBC-11. Potential effect modification by sex (female, male), physical activity (lightly, moderately, vigorously active), screen time (< 2/2 to < 4/4, > 4 h/d), parental educational level (low, medium, high), household income (quartiles) and maternal BMI (continuous) was tested by including multiplicative interaction terms between these variables and total SSBs and ASBs consumption in the logistic regression models concerning weight status.
Logistic regression models were also performed to investigate the associations between overweight or obesity at age 11 and ED symptoms at age 18. Normal weight was considered as the reference group. Additionally, to explore if SSBs and ASBs consumption could potentially contribute to this process, we assessed the associations between the combination of consumption of SSBs and ASBs and weight status at age 11 and ED symptoms at age 18. A joined independent variable was constructed, incorporating both SSBs and ASBs consumption and body weight status at age 11, which was subsequently categorized into four groups: (1) SSBs and ASBs consumption below the median plus underweight or normal weight; (2) SSBs and ASBs consumption above the median plus underweight or normal weight; (3) SSBs and ASBs consumption below the median plus overweight or obesity; and (4) SSBs and ASBs consumption above the median plus overweight or obesity. The first category was designated as the reference group. All aforementioned analyses on ED symptoms incorporated the same previously mentioned confounding factors in the models, except for Model 1, in which age was not included, and Model 4, in which further adjustment was made for parental diagnosis of psychiatric disorders (yes, no) and SDQ (normal, borderline, abnormal) at age 11.
Various sensitivity analyses were conducted to assess the impact of potential variables on the outcomes. Cross-sectional and longitudinal associations between individual consumption of SSBs and ASBs and weight status were conducted separately. The association between body weight status and anorexia nervosa symptoms (as determined by self-reported threshold, subthreshold, or diagnosis) was also assessed. The joint analysis on ED symptoms was repeated using an alternative cutoff point for SSBs and ASBs consumption (7 serv/wk.) to evaluate the robustness of associations across different definitions. Given that purging disorder shares behavioral and psychological features with other ED [37] and that relevant symptom information was available at DNBC-11 and DNBC-18, an analysis was conducted by additionally considering purging disorder as a part of the definition of ED symptoms, to capture a broader spectrum of disordered eating behaviors.
The management of the data was conducted utilizing SAS version 9.4 (SAS Institute, North Carolina, US), while statistical analyses were performed using Stata/SE software, version 18.0 (StataCorp, College Station, TX). Statistically significant results were defined as two-tailed p-values7 serv/wk < 0.05. In addition, figures were generated using R software v4.4.2; ‘ggplot2’ package 3.5.1.
Results
A total of 35,173 children were included, 51.2% were female, and had a mean (± SD) age of 11.4 ± 0.6 when they completed the DNBC-11. In total, 2,786 (7.9%) children had overweight or obesity at DNBC-11 and 2,936 (8.4%) at DNBC-18. There were 457 (1.3%) and 1,477 (4.2%) children who had ED symptoms at DNBC-11 and DNBC-18, respectively. At DNBC-11, the children presented a mean (± SD) consumption of total SSBs and ASBs of 4.2 ± 5.4 servings/week, with individual consumption of soda being 1.1 ± 2.2 servings/week, light soda 0.9 ± 2.2 servings/week, and lemonade 2.2 ± 4.0 servings/week. Supplemental Fig 2 shows the contribution of each type of beverage to the total consumption of SSBs and ASBs for the total study population and by subgroups of weight status at DNBC-11. Lemonade was the most commonly consumed type of SSBs and ASBs (43.2%), followed by soda (32.5%), and light soda (24.3%). Children with overweight or obesity at age 11 showed a higher percentage of consumption of light soda (37.3%) than soda (23.0%). General characteristics of the study population are shown in Table 1. Children with a higher total consumption of SSBs and ASBs were more likely to be male (58 vs. 43%), to be inactive or lightly active (91 vs. 89%), to spend more than 4 h a day in front of a screen (16 vs. 12%), to have parents with a higher BMI (maternal: 22 vs. 20 kg/m2; paternal: 27 vs. 26 kg/m2), lower income (12 vs. 8%), lower educational level (3 vs. 1%) and a higher prevalence of psychiatric disorders (16 vs. 12%), and to have a mother who smoked during pregnancy (19 vs. 8%). Children who consumed 11 or more servings of SSBs and ASBs per week were more likely to have overweight or obesity at both DNBC-11(12 vs. 6%) and DNBC-18 (10 vs. 7%), report less ED symptoms at DNBC-18 (3 vs. 5%), and show higher scores on the SDQ questionnaire (8 vs. 6.5 points). Additional dietary characteristics at the DNBC-11 across categories of total SSBs and ASBs consumption are shown in Supplemental Table 3. Children with a higher consumption of SSBs and ASBs presented a higher consumption of fast food, ice-cream, cake, chips, and snacks, and a lower consumption of fruit and milk at the 11-year follow-up. In Supplemental Table 4, we provide further general characteristics of the study population across categories of consumption of the different individual subtypes of SSBs and ASBs.
Table 1.
General characteristics of the study population
| Total consumption of SSBs and ASBs | ||||||
|---|---|---|---|---|---|---|
| Total n = 35,173 |
0 serv/wk n = 11,836 |
> 0 to < 4 serv/wk n = 9487 |
4 to < 7 serv/wk n = 6164 |
7 to < 11 serv/wk n = 4027 |
≥ 11 serv/wk n = 3659 |
|
| Sex (% female) | 18,021 (51.2) | 6,784 (57.3) | 4,851 (51.1) | 2,993 (48.6) | 1,840 (45.7) | 1,553 (42.4) |
| Age (years) | 11.4 ± 0.6 | 11.4 ± 0.6 | 11.4 ± 0.6 | 11.4 ± 0.6 | 11.4 ± 0.6 | 11.4 ± 0.6 |
| BMIa (kg/m2) | 17.4 ± 2.4 | 17.2 ± 2.3 | 17.3 ± 2.4 | 17.5 ± 2.5 | 17.5 ± 2.5 | 17.9 ± 2.7 |
| BMIb (kg/m2) | 22.0 ± 3.5 | 21.8 ± 3.3 | 21.9 ± 3.4 | 22.1 ± 3.4 | 22.3 ± 3.7 | 22.7 ± 3.9 |
| Weight status at age 11a | ||||||
| Underweight/Normal weight | 32,387 (92.1) | 11,093 (93.7) | 8,801 (92.8) | 5,641 (91.5) | 3,639 (90.4) | 3,213 (87.8) |
| Overweight/Obesity | 2,786 (7.9) | 743 (6.3) | 686 (7.2) | 523 (8.5) | 388 (9.6) | 446 (12.2) |
| Weight status at age 18b | ||||||
| Underweight/Normal weight | 18,414 (52.4) | 6,760 (57.1) | 5,053 (53.3) | 3,077 (49.9) | 1,929 (47.9) | 1,595 (43.6) |
| Overweight/Obesity | 2,936 (8.4) | 865 (7.3) | 763 (8.0) | 536 (8.7) | 391 (9.7) | 381 (10.4) |
| ED symptoms at age 11a | 457 (1.3) | 159(1.3) | 110 (1.2) | 73 (1.2) | 55 (1.4) | 60 (1.6) |
| ED symptoms at age 18b | 1,477 (4.2) | 573 (4.8) | 378 (4.0) | 266 (4.3) | 141 (3.5) | 119 (3.3) |
| Maternal BMIa (kg/m2) | 21.1 ± 4.2 | 20.5 ± 4.0 | 20.9 ± 4.1 | 21.4 ± 4.2 | 21.7 ± 4.5 | 22.4 ± 4.9 |
| Paternal BMIa (kg/m2) | 25.9 ± 3.5 | 25.6 ± 3.3 | 25.8 ± 3.4 | 26.2 ± 3.6 | 26.3 ± 3.7 | 26.7 ± 3.8 |
| Family incomea | ||||||
| 1 (Lowest quartile) | 3,078 (8.8) | 950 (8.0) | 722 (7.6) | 527 (8.6) | 448 (11.1) | 431 (11.8) |
| 4 (Highest quartile) | 13,543 (38.5) | 5,142 (43.4) | 3,925 (41.4) | 2,252 (36.5) | 1,276 (31.7) | 948 (25.9) |
| Family educational levela | ||||||
| Low | 432 (1.2) | 90 (0.8) | 97 (1.0) | 75 (1.2) | 69 (1.7) | 101 (2.8) |
| Medium | 10,157 (28.9) | 2,419 (20.4) | 2,481 (26.2) | 1,977 (32.1) | 1,521 (37.8) | 1,759 (48.1) |
| High | 24,584 (69.9) | 9,327 (78.8) | 6,909 (72.8) | 4,112 (66.7) | 2,437 (60.5) | 1,799 (49.2) |
| Family typea | ||||||
| Living with both parents | 29,902 (85.0) | 10,101 (85.3) | 8,204 (86.5) | 5,189 (84.2) | 3,332 (82.7) | 3,076 (84.1) |
| Living with one parent | 5,164 (14.7) | 1,699 (14.4) | 1,255 (13.2) | 958 (15.5) | 681 (16.9) | 571 (15.6) |
| Living without parents | 107 (0.3) | 36 (0.3) | 28 (0.3) | 17 (0.3) | 14 (0.4) | 12 (0.3) |
| Maternal smoking during pregnancy | ||||||
| Non-smoking | 28,184 (80.1) | 9,918 (83.8) | 7,742 (81.6) | 4,788 (77.7) | 3,093 (76.8) | 2,643 (72.2) |
| Stopped | 2,999 (8.5) | 983 (8.3) | 785 (8.3) | 578 (9.4) | 317 (7.9) | 336 (9.2) |
| Smoking | 3,990 (11.3) | 935 (7.9) | 960 (10.1) | 798 (13.0) | 617 (15.3) | 680 (18.6) |
| Physical activitya—children | ||||||
| Inactive or lightly active | 31,495 (89.5) | 10,578 (89.4) | 8,437 (88.9) | 5,527 (89.7) | 3,620 (89.89) | 3,333 (91.1) |
| Moderately or vigorously active | 3,678 (10.5) | 1,258 (10.6) | 1,050 (11.1) | 637 (10.3) | 407 (10.11) | 326 (8.9) |
| Screen timea—children | ||||||
| < 2 h/d | 11,781 (33.5) | 5,027 (42.5) | 3,381 (35.6) | 1,743 (28.3) | 974 (24.2) | 656 (17.9) |
| 2–4 h/d | 16,836 (47.9) | 5,267 (44.5) | 4,651 (49.0) | 3,092 (50.2) | 2,034(50.5) | 1,792 (49.0) |
| > 4 h/d | 6,556 (18.6) | 1,542 (13.0) | 1,455 (15.3) | 1,329 (21.6) | 1,019 (25.3) | 1,211 (33.1) |
| Parental psychiatric disordera | 4,454 (12.7) | 1,387 (11.7) | 1,071 (11.3) | 81 3 (13.2) | 587 (14.6) | 596 (16.3) |
| SDQa—children | 6,556 (18.6) | |||||
| Continuous, score | 6.9 ± 4.8 | 6.5 ± 4.7 | 6.6 ± 4.6 | 6.9 ± 4.9 | 7.3 ± 5.0 | 8.0 ± 5.4 |
| Normal | 32,449 (92.3) | 11,028 (93.2) | 8,848 (93.3) | 5,671 (92.0) | 3,675 (91.3) | 3,227 (88.2) |
| Borderline | 1,341 (3.8) | 372 (3.1) | 320 (3.4) | 247 (4.0) | 188 (4.7) | 214 (5.9) |
| Abnormal | 614 (1.8) | 177 (1.5) | 113 (1.2) | 108 (1.8) | 80 (2.0) | 136 (3.7) |
BMI Body Mass Index, ED Eating Disorders, SDQ Strength and Difficulties Questionnaire, SSBs and ASBs Sugar- and Artificially Sweetened Beverages
Total consumption of SSBs and ASBs was considered as the sum of individual consumption of soda, light soda and lemonade. Values are reported as means ± standard deviations for continuous variables and number (%) for categorical variables.
aAt 11-year of follow-up
bAt 18-year of follow-up
Figure 1 depicts the associations between SSBs and ASBs consumption at DNBC-11 and body weight status at DNBC-11 and DNBC-18. The ORs and 95%CI resulting from these associations are tabulated in Supplemental Table 5. In DNBC-11, a statistically significant higher odds of overweight or obesity were observed among children with a higher consumption of SSBs and ASBs after adjusting for sociodemographic, family-related, and lifestyle factors. In the fully adjusted model, children consuming 11 or more servings/week of SSBs and ASBs at age 11 had 43% (95%CI 1.24–1.64) higher odds of overweight or obesity at 11 years old compared to those with no consumption. The findings from the longitudinal analysis also showed a statistically significant positive association between total SSBs and ASBs consumption at DNBC-11 and weight status at age 18 in all adjusted models. In the fully adjusted model, children who had the highest consumption of SSBs and ASBs showed 20% (95%CI 1.00–1.43) higher odds of overweight or obesity at age 18, in comparison to those in the reference category. We also conducted a sensitivity analysis by further adjusting the previous models for the consumption of milkshakes and juices (Supplemental Table 5). Consistent with the results above, total consumption of SSBs and ASBs was positively associated with weight status, with an OR of 1.48 (95%CI 1.28–1.70) at the 11-year follow-up and 1.21 (95%CI 1.02–1.45) at 18-year follow-up when comparing children with the highest SSBs and ASBs consumption to those with no consumption. There was no evidence of statistically significant interactions between total SSBs and ASBs consumption and sex, screen time, or income at age 11 years in relation to weight status at DNBC-11 and DNBC-18. However, we observed significant interactions between total SSBs and ASBs consumption and children physical activity (p = 0.006) and parental educational level (p = 0.020) in relation to weight status at DNBC-11 and maternal BMI (p = 0.040) in relation to DNBC-18. Supplemental Table 6 presents the separate association analyses of SSBs and ASBs consumption with weight status. At DNBC-11, children in the highest ASBs consumption category had 2.03 (95%CI 1.79–2.31) odds of overweight or obesity, while those with the highest SSBs consumption had 0.96 (95%CI 0.87–1.06) odds, compared to the lowest consumption category. Similar to the main analysis, the associations were attenuated at DNBC-18 (ASBs: OR: 1.55, 95%CI 1.31–1.83; SSBs: OR: 0.98, 95%CI 0.87–1.10). Table 2 shows the ORs and 95%CI for ED symptoms incidence at DNBC-18 across weight status categories at the DNBC-11. Children with overweight and obesity at age 11 had significantly higher odds of ED symptoms (OR: 1.36, 95%CI 1.08–1.70; OR: 2.08, 95%CI 1.09–3.97, respectively) at age 18 compared with children with normal weight. Children with underweight presented lower odds of ED symptoms at age 18 (OR: 0.83, 95%CI 0.71–0.97) compared to children with normal weight. These findings were consistent across all adjustment models. However, these associations were no longer observed when analyses were restricted to cases of symptoms of anorexia nervosa only (Supplemental Table 7). The association between the joint assessment of consumption of SSBs and ASBs and weight status at age 11 with ED symptoms at age 18 is provided in Table 3. In the most adjusted model, compared to children with underweight or normal weight and a consumption of SSBs and ASBs below the median at age 11, those with overweight or obesity and a consumption of SSBs and ASBs below and above the median presented a significantly 41% (95%CI 1.04–1.92) and 50% (95%CI 1.12–2.00) higher odds of ED symptoms at age 18, respectively. Results remained similar after further adjusting the model for individual consumption of milkshakes and juices (children with overweight or obesity and a consumption of SSBs and ASBs above the median: OR: 1.39, 95%CI 1.02–1.90; below the median: OR: 1.53, 95%CI 1.15–2.05; compared to reference group). The sensitivity analysis conducted using an alternative cutoff point for SSB and ASB consumption of 7 servings/wk. yielded similar results (Supplemental Table 8). Likewise, the findings remained consistent when purging disorder symptoms were considered part of the ED symptoms definition (Supplemental Tables 9, 10).
Fig. 1.
Odds Ratios and 95%CI for weight status at 11- and 18-year follow-up according to categories of total consumption of SSBs and ASBs at 11-years follow-up. Abbreviations: SSBs and ASBs, Sugar- and Artificially Sweetened Beverages; serv, servings; wk., week. Total consumption of SSBs and ASBs was considered as the sum of individual consumption of soda, light soda and lemonade. Logistic regression models were used to assess the association between overweight or obesity at 11- (at the top) and 18-years follow-up (at the bottom) and categories of total consumption of SSBs and ASBs at 11-year follow-up. The results are shown as Odds Ratios and 95%CI. Model 1 was adjusted for age (years) and sex. Model 2 was additionally adjusted for family income (quartiles), parental educational level (low, medium, high), family type (living with both parents, living with one parent, living without parents), maternal smoking during pregnancy (non-smoking, stopped during pregnancy, smoking during pregnancy), maternal age at birth (years) and paternal age at birth (years). Model 3 was additionally adjusted for physical activity (inactive, lightly active, moderately active, vigorously active), screen time (< 2 h/d, 2 to < 4 h/d, 4 or > 4 h/d) and dietary variables, specifically consumption of fruit, milk, fast food, ice-cream, cake, chips and snacks (low, moderate, high). Model 4 was additionally adjusted for maternal BMI (kg/m2), paternal BMI (kg/m2). In the longitudinal analysis, 1533 children were excluded due to previous prevalence of overweight or obesity at 11 years old
Table 2.
Odds Ratios and 95%CI for ED symptoms incidence at 18-year follow-up according to categories of weight status at 11-year follow-up
| Weight status at 11-year follow-up | ||||
|---|---|---|---|---|
| Underweight | Normal weight | Overweight | Obesity | |
| n = 3499 | n = 15,503 | n = 1322 | n = 104 | |
| BMI at 11 y, kg/m2 | 14.46 ± 0.73 | 17.43 ± 1.50 | 22.41 ± 1.26 | 27.46 ± 1.86 |
| ED symptoms cases at 18 y, n (%) | 210 (6.00) | 961 (6.20) | 105 (7.94) | 12 (11.54) |
| Model 1 | 0.84 (0.72–0.99) | 1 (Ref) | 1.34 (1.09–1.67) | 2.23 (1.19–4.18) |
| Model 2 | 0.85 (0.73–0.99) | 1 (Ref) | 1.34 (1.08–1.66) | 2.15 (1.14–4.05) |
| Model 3 | 0.84 (0.72–0.99) | 1 (Ref) | 1.33 (1.07–1.65) | 2.05 (1.08–3.87) |
| Model 4 | 0.83 (0.71–0.97) | 1 (Ref) | 1.36 (1.08–1.70) | 2.08 (1.09–3.97) |
BMI Body Mass Index, ED Eating Disorders, Ref Reference, y years
Logistic regression models were used to assess the association between ED symptoms incidence at 18-year follow-up and categories of weight status at 11-year follow-up. The results are shown as Odds Ratios and 95%CI. Model 1 was adjusted for age (years) and sex. Model 2 was additionally adjusted for family income (quartiles), parental educational level (low, medium, high), family type (living with both parents, living with one parent, living without parents), maternal smoking during pregnancy (non-smoking, stopped during pregnancy, smoking during pregnancy), maternal age at birth (years) and paternal age at birth (years). Model 3 was additionally adjusted for physical activity (inactive, lightly active, moderately active, vigorously active), screen time (< 2 h/d, 2 to < 4 h/d, 4 or > 4 h/d) and dietary variables, specifically consumption of fruit, milk, fast food, ice-cream, cake, chips and snacks (low, moderate, high). Model 4 was additionally adjusted for maternal BMI (kg/m2), paternal BMI (kg/m2), parental diagnosis of psychiatric disorders (yes/no) and children strength and difficulties questionnaire (normal, borderline, abnormal). In this analysis, 269 children were excluded due to prevalence of ED symptoms at 11 years old. Bold indicates p<0.05
Table 3.
Odds Ratios and 95%CI for ED symptoms incidence at 18-year follow-up according to categories of weight status and SSBs and ASBs at 11-year follow-up
| Weight status and SSBs and ASBs at 11-year follow-up | ||||
|---|---|---|---|---|
| SSBs and ASBs at 11 y | ≤ median | > median | ≤ median | > median |
| Weight status at 11 y | Underweight/Normal weight | Overweight/Obesity | ||
| n = 9994 | n = 9008 | n = 626 | n = 800 | |
| SSBs AND ASBs consumption at 11 y, serv/wk | 0.54 ± 0.84 | 7.50 ± 5.18 | 0.60 ± 0.89 | 8.62 ± 6.20 |
| ED symptoms cases at 18 y, n (%) | 638 (6.38) | 533 (5.92) | 53 (8.47) | 64 (8.00) |
| Model 1 | 1 (Ref) | 1.00 (1.04–1.13) | 1.41 (1.05–1.90) | 1.50 (1.14–1.97) |
| Model 2 | 1 (Ref) | 1.00 (0.89–1.14) | 1.40 (1.04–1.90) | 1.47 (1.11–1.94) |
| Model 3 | 1 (Ref) | 1.03 (0.90–1.17) | 1.40 (1.03–1.90) | 1.47 (1.11–1.96) |
| Model 4 | 1 (Ref) | 1.03 (0.90–1.18) | 1.41 (1.04–1.92) | 1.50 (1.12–2.00) |
| Model 5 (sensitivity analysis) | 1 (Ref) | 1.04 (0.91–1.18) | 1.39 (1.02–1.90) | 1.53 (1.15–2.05) |
BMI Body Mass Index, ED Eating Disorders, Ref, Reference, SSBs and ASBs Sugar- and Artificially Sweetened Beverages; y years
Total consumption of SSBs and ASBs was considered as the sum of individual consumption of soda, light soda and lemonade. Logistic regression models were used to assess the association between ED symptoms incidence at 18-year follow-up and categories of total consumption of SSBs and ASBs at 11-year follow-up. The results are shown as Odds Ratios and 95%CI. Model 1 was adjusted for sex. Model 2 was additionally adjusted for family income (quartiles), parental educational level (low, medium, high), family type (living with both parents, living with one parent, living without parents), maternal smoking during pregnancy (non-smoking, stopped during pregnancy, smoking during pregnancy), maternal age at birth (years) and paternal age at birth (years). Model 3 was additionally adjusted for physical activity (inactive, lightly active, moderately active, vigorously active), screen time (< 2 h/d, 2 to < 4 h/d, 4 or > 4 h/d) and dietary variables, specifically consumption of fruit, milk, fast food, ice-cream, cake, chips and snacks (low, moderate, high). Model 4 was additionally adjusted for maternal BMI (kg/m2), paternal BMI (kg/m2), parental diagnosis of psychiatric disorders (yes/no) and children strength and difficulties questionnaire (normal, borderline, abnormal). Model 5 additionally mutually adjusted for individual subtypes of SSBs and ASBs (milkshake, and juice; 0 serv/wk., > 0 to 2.5 serv/wk., > 2.5 to < 4 serv/wk., ≥ 4 serv/wk.). In this analysis, 269 children were excluded due to prevalence of ED symptoms at 11 years old. Bold indicates p<0.05
Discussion
In this large Danish population-based cohort followed since childhood to early adulthood, higher consumption of SSBs and ASBs at age 11 years was associated with higher odds of overweight or obesity at ages 11 and 18 years. Furthermore, the presence of overweight or obesity during childhood was found to be positively associated with ED symptoms in early adulthood. When examining the joint association of SSBs and ASBs consumption and weight status, children with overweight or obesity and higher beverage consumption at age 11 had slightly higher odds of ED symptoms at age 18 than those with the same weight status but lower consumption. The present findings underscore the public health relevance of reducing SSB and ASB consumption during childhood, given its association with overweight and obesity, and its potential link to later ED symptoms.
The relationship between the consumption of SSBs and ASBs and weight status in younger populations has been widely investigated; however, the evidence remains mixed, with several systematic reviews and meta-analyses (SRMAs) calling for further research [15, 16, 38, 39]. Methodological differences and variations in population characteristics may partially account for these inconsistencies. Nevertheless, the majority of findings suggest that the consumption of these beverages is associated with higher body weight, which is consistent with our results. One of the most recent SRMAs in this field, conducted by Nguyen et al. in 2023 [15], and based on evidence from prospective studies and randomized controlled trials, concluded that SSBs consumption contributes to higher BMI and weight gain in children. Our findings are also in line with results from two other Nordic cohort studies. In a subsample of 283 Danish children from the European Youth Heart Study, SSBs consumption was positively associated with increased BMI and waist circumference from childhood through early adulthood [40]. Similarly, in the Cardiovascular Risk in Young Finns Study, sugar-sweetened soft drink intake from childhood to adulthood was directly associated with higher BMI and overweight in adulthood, although this association was only observed among women [41]. Our study, with a large sample size did not support any interaction by sex, indicating that the association between SSBs and ASBs intake and body weight status might not differ between boys and girls.
Measures such as the taxation of sweetened beverages have already shown positive effects in reducing consumption in several countries [42] and appear to be a promising policy approach. However, educational strategies could also be considered to promote healthier dietary choices, including beverage choices, and to increase awareness of the potential risks associated with excessive consumption of unhealthy foods and drinks [43]. Nutritional education should target not only children but also their families, as previous studies and our findings suggest that parental BMI is closely related to children’s BMI [44, 45]. Furthermore, sociodemographic patterns and behaviors may play an important role in these associations. We observed sociodemographic patterns that have been previously associated with increased consumption of SSBs; this includes lower parental education levels and household income [44, 46]. These findings underscore the need for public health strategies that consider social factors and tailor messages to vulnerable groups.
Some plausible explanations have been proposed for the potential mechanisms underlying the relationship between the consumption of these beverages and body weight. A significant contributing factor is the substantial quantities of added sugars found in SSBs, thereby contributing to the overall energy intake of consumers [47]. In the case of ASBs, previous research has linked their content of non-nutritive sweeteners to appetite regulation [48], reward processing, and gut microbiota composition [49], factors that, in turn, may indirectly impact body weight. Additionally, both SSBs and ASBs have been demonstrated to possess low satiating properties, which may result in impaired regulation of energy intake and, eventually, contribute to a positive energy balance [47]. Moreover, the high palatability of these beverages, irrespective of their caloric content, may serve to reinforce preferences for sweet tastes and, consequently, promote the development of unhealthy long-term dietary habits.
Regarding the relationship between obesity and disordered eating behaviors, it has received increasing attention in recent years, with a growing body of evidence suggesting that these conditions are interconnected and should be studied jointly rather than in isolation [4, 8, 50, 51]. The findings of this study support this perspective, as they reveal a positive association between overweight and obesity at age 11 and ED symptoms at age 18. Additionally, similar results were observed in a previous analysis conducted within the DNBC, which demonstrated that overweight at age 7 was predictive of disordered eating behaviors later at age 11 [23]. It is also noteworthy that in the present study, the observed association was predominantly driven by symptoms of bulimia nervosa and binge eating disorder, as the same trend was not observed in the supplementary analysis restricted solely to symptoms of anorexia nervosa. This distinction, which has also been reported by previous research [8, 52, 53], underscores the importance of considering subtypes separately when examining their relationship with early weight status, as different risk pathways may be involved. Prior evidence has also shown that adolescents with overweight or obesity are more likely to experience body dissatisfaction and high weight concern, both of which are recognized risk factors for the development of disordered eating behaviors and ED [17, 54, 55]. Furthermore, interventions that are narrowly focused on weight loss or that promote restrictive dieting may inadvertently contribute to the emergence of unhealthy eating behaviors [51, 56]. Therefore, our findings align with the ongoing shift from a weight-centric approach toward obesity prevention strategies incorporating psychological and behavioral components to promote healthy eating habits and positive body image among children and adolescents.
Our study is among the first to consider SSBs and ASBs, weight status, and ED symptoms in combination in a large population-based cohort of children and adolescents aged 11 and 18. While our findings were statistically significant, they also suggest that the association between weight status and ED symptoms remained largely independent of SSBs and ASBs consumption. Nevertheless, the promotion of reduced consumption remains a relevant public health objective in the context of the prevention of overweight and obesity, as well as the improvement of overall dietary quality among youth.
This study has considerable strengths, such as the large study population, the prospective data collection including information from both children and parents, and the linkage to registry data. Questions on SSBs and ASBs consumption were carefully designed and deliberately included in the DNBC-11 in a specific questionnaire, enabling to uniquely assess their potential contribution on health, based on reported weekly intake in the most consumed formats. Incorporation of self-reported measures of ED symptoms in the DNBC may have facilitated a more comprehensive identification of underdiagnosed individuals still facing significant health challenges. Nevertheless, the present study is also subject to certain limitations. Self-reported data is typically susceptible to misreporting errors and inaccuracies. However, the data on SSBs and ASBs were reported directly by the children, in contrast to being reported by their parents, which could have mitigated some risks of underreporting bias. Moreover, although the dietary questions have not yet undergone formal validation in the present study, in the field of nutritional epidemiology, simple frequency-based dietary items are generally considered appropriate for the grouping of individuals, particularly in large cohorts [57]. Similarly, self- and parent-reported height and weight are usually accepted as suitable for BMI classification and have been previously utilised in DNBC publications to examine weight status categories [23]. Given that detailed data on food consumption and total energy and nutrient intake were not collected at age 11, the possibility of unmeasured dietary confounding still exists. In particular, the impossibility of adjusting for total energy intake may introduce some residual confounding, as SSBs consumption is commonly associated with higher overall caloric intake [58], and ASBs consumption may reflect other aspects of a children’s diet or lifestyle that affect body weight [59]. Nevertheless, this gap was addressed by adjusting for all the available intake data of specific food groups from the same questionnaire. Furthermore, repeated measures of SSBs and ASBs consumption at age 18 were not available, thus preventing the assessment of changes in beverage consumption over time or potential reciprocal relationships with subsequent weight outcome. Register-based diagnoses covered the period from ages 6 to 18, whereas self-reported ED reflected symptoms experienced in the past year at the DNBC-11 and DNBC-18, complicating direct comparisons. Nevertheless, both sources complemented each other: registries captured clinically treated cases, whereas self-reports reflected current symptoms and broader behavioral patterns. While the DNBC includes a substantial sample of Danish children and adolescents, the number of cases of overweight and obesity in this cohort was relatively low compared with WHO estimates [1], and the extent to which these findings can be generalized to other populations with different ages, sociodemographic, and cultural characteristics should be interpreted with caution. In fact, recent global estimates indicate that Denmark has among the lowest levels of SSB consumption in children and adolescents within the group of high-income countries which may further limit the generalizability of these findings to contexts with higher consumption patterns [60]. In addition, the low number of ED symptom cases may have limited statistical power to detect small effects or to conduct detailed subgroup analyses. Consumption patterns of SSBs and ASBs among children have likely shifted since the data were collected nearly 15 years ago [61], with changes in quantity and type potentially contributing to stronger associations with health outcomes. BMI was employed to determine weight status categories, but despite its widespread acceptance and practicality in large-scale studies, this method does not accurately capture differences in body composition or fat distribution [62]. Finally, as generally occurs in any observational study, causal inferences are certainly limited, and unmeasured confounders cannot be entirely disregarded.
Conclusion
In conclusion, consumption of SSBs and ASBs in childhood was associated with overweight and obesity in the short- and long-term in a large Danish population-based cohort. Childhood overweight or obesity was associated with higher odds of developing ED symptoms in early adulthood. Although the associations were modest, children with overweight or obesity who reported higher SSBs and ASBs consumption at age 11 had slightly higher odds of ED symptoms at age 18 than peers with similar weight status but lower consumption. These findings suggest that the interplay between beverage consumption and weight status deserves further investigation. Therefore, the observed associations in this study underscore the importance of early preventive strategies aimed at reducing the consumption of SSBs and ASBs and addressing overweight and obesity at a young age. These findings contribute to the existing literature and support the implementation of integrated public health strategies that consider both nutritional and psychological dimensions during this critical developmental period.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The Danish National Birth Cohort was established with a significant grant from the Danish National Research Foundation. Additional support was obtained from the Danish Regional Committees, the Pharmacy Foundation, the Egmont Foundation, the March of Dimes Birth Defects Foundation, the Health Foundation and other minor grants. The DNBC Biobank has been supported by the Novo Nordisk Foundation and the Lundbeck Foundation. Follow-up of mothers and children have been supported by the Danish Medical Research Council (SSVF 0646, 271-08-0839/06-066023, O602-01042B, 0602-02738B), the Lundbeck Foundation (195/04, R100-A9193), The Innovation Fund Denmark 0603-00294B (09-067124), the Nordea Foundation (02-2013-2014), Aarhus Ideas (AU R9-A959-13-S804), University of Copenhagen Strategic Grant (IFSV 2012), and the Danish Council for Independent Research (DFF-4183-00594 and DFF-4183-00152). The 18-year follow-up was funded by the Danish Council for Independent Research (DFF-4183-00594B; Close to Adult: 17-year follow-up of the Danish National Birth Cohort). The authors would like to thank all the mothers and children who participated in the Danish National Birth Cohort (DNBC), as well as the founding bodies.
Abbreviations
- ASBs
Artificially sweetened beverages
- BMI
Body mass index
- CI
Confidence interval
- DNBC
Danish National Birth Cohort
- ED
Eating disorders
- IOTF
International Obesity Task Force
- ORs
Odds ratios
- SD
Standard deviation
- SDQ
Strengths and difficulties questionnaire
- SSBs
Sugar-sweetened beverages
- WHO
World Health Organization
Authors’ contributions
CVH, MGF and KSL conceived and designed the study. CVH conducted the analyses, supervised by MGF and KSL. CVH prepared all tables and figures, supervised by MGF and KSL. CVH, AJ, MGF and KSL interpreted the results and wrote the first draft of the manuscript. All authors contributed to the analytical approach and interpretation of the data, revisions of the manuscript, and approved final version of manuscript before submission. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted. CVH, MGF and KSL are the guarantor of the manuscript.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. The analyses for this publication were initiated during CV-H’s research stay, which was supported by the predoctoral grant from the Generalitat de Catalunya (2022 FI_B100108) and the URV-AURORA program 2023–2024 mobility grant. M.G.-F. was supported by the Novo Nordisk Foundation grant NNF23SA0084103. A.J and K.S-L were salaried by The Independent Research Fund Denmark (DFF-8045-00047B). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Data availability
According to European law (General Data Protection Regulation), data containing potentially identifying or sensitive personal information are restricted. However, for academic research, data can be available subsequent to approval. Any request for DNBC data needs to follow the outlined procedures, see: https://www.dnbc.dk/access-to-dnbc-data. For access to the underlying person-level data in Danish registers, researchers need to apply to the Danish Health Data Authority and Statistics Denmark. The programming codes used on DNBC and register data in this study are available by contacting the corresponding author.
Declarations
Conflict of interest
All authors declare no conflict of interest.
Ethical approval
The DNBC cohort is approved by the Danish Data Protection Agency under the general approval granted to Statens Serum Institut (fællesfortegnelse), reference number 18/04608 and by the Committee on Health Research Ethics (case number (KF) 01-471/94). Participants in the DNBC were enrolled with informed consent and individuals born into the cohort were notified about their participation, rights and the option to opt out upon turning 18. Data approval for the analyses conducted in this study, which involved linking DNBC data with register data accessed on Statistics Denmark’s server, was obtained from the DNBC managerial team (2018-15), Statistics Denmark and registered under the Danish Data Protection Agency’s general approval for the Faculty of Health and Medical Sciences at the University of Copenhagen (514–0400/19-3000). This study was conducted with compliance with the Declaration of Helsinki.
Footnotes
Katrine Strandberg-Larsen and Marta Guasch-Ferré have contributed equally as co last-authors.
Contributor Information
Cristina Valle-Hita, Email: cristina.valle@sund.ku.dk.
Marta Guasch-Ferré, Email: marta.guasch@sund.ku.dk.
References
- 1.World Health Organization (WHO) (2025) Obesity and overweight. https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight. Accessed 3 Jun 2025
- 2.GBD 2021 Adolescent BMI Collaborators (2025) Global, regional, and national prevalence of child and adolescent overweight and obesity, 1990-2021, with forecasts to 2050: a forecasting study for the Global Burden of Disease Study 2021. Lancet 405:785–812. 10.1016/S0140-6736(25)00397-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lister NB, Baur LA, Felix JF et al (2023) Child and adolescent obesity. Nat Rev Dis Primers 9:24. 10.1038/s41572-023-00435-4 [DOI] [PubMed] [Google Scholar]
- 4.Chaves E, Jeffrey DT, Williams DR (2023) Disordered eating and eating disorders in pediatric obesity: assessment and next steps. Int J Environ Res Public Health. 10.3390/ijerph20176638 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wiklund CA, Kuja-Halkola R, Thornton LM et al (2018) Childhood body mass index and development of eating disorder traits across adolescence. Eur Eat Disord Rev 26:462–471. 10.1002/erv.2612 [DOI] [PubMed] [Google Scholar]
- 6.Reed ZE, Micali N, Bulik CM et al (2017) Assessing the causal role of adiposity on disordered eating in childhood, adolescence, and adulthood: a Mendelian randomization analysis. Am J Clin Nutr 106:764–772. 10.3945/ajcn.117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.López-Gil JF, García-Hermoso A, Smith L et al (2023) Global proportion of disordered eating in children and adolescents: a systematic review and meta-analysis. JAMA Pediatr 177:363–372. 10.1001/jamapediatrics.2022.5848 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Camacho-Barcia L, Giel KE, Jiménez-Murcia S et al (2024) Eating disorders and obesity: bridging clinical, neurobiological, and therapeutic perspectives. Trends Mol Med 30:361–379 [DOI] [PubMed] [Google Scholar]
- 9.Khalid R, Lister NB, Paxton SJ et al (2025) Potential pathways to the onset and development of eating disorders in people with overweight and obesity: a scoping review. Obes Rev. 10.1111/obr.13840 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Arafa A, Yasui Y, Kokubo Y et al (2024) Lifestyle Behaviors of childhood and adolescence: contributing factors, health consequences, and potential interventions. Am J Lifestyle Med 0(0):1-19 10.1177/15598276241245941 [DOI] [PMC free article] [PubMed]
- 11.Blundell JE, Baker JL, Boyland E et al (2017) Variations in the prevalence of obesity among European countries, and a consideration of possible causes. Obes Facts 10:25–37. 10.1159/000455952 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Diaz C, Rezende LFM, Sabag A et al (2023) Artificially sweetened beverages and health outcomes: an umbrella review. Adv Nutr 14:710–717 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Costa D, Warkentin S, Oliveira A (2022) Sugar-sweetened beverages, effects on appetite and public health strategies to reduce the consumption among children: a review. Porto Biomed J 7:e172. 10.1097/j.pbj.0000000000000172 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Dai J, Soto MJ, Dunn CG, Bleich SN (2021) Trends and patterns in sugar-sweetened beverage consumption among children and adults by race and/or ethnicity, 2003–2018. Public Health Nutr 24:2405–2410. 10.1017/S1368980021001580 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Nguyen M, Jarvis SE, Tinajero MG et al (2023) Sugar-sweetened beverage consumption and weight gain in children and adults: a systematic review and meta-analysis of prospective cohort studies and randomized controlled trials. Am J Clin Nutr 117:160–174. 10.1016/j.ajcnut.2022.11.008 [DOI] [PubMed] [Google Scholar]
- 16.Abbasalizad Farhangi M, Mohammadi Tofigh A, Jahangiri L et al (2022) Sugar-sweetened beverages intake and the risk of obesity in children: an updated systematic review and dose–response meta-analysis. Pediatr Obes. 10.1111/ijpo.12914 [DOI] [PubMed] [Google Scholar]
- 17.Voelker D, Reel J, Greenleaf C (2015) Weight status and body image perceptions in adolescents: current perspectives. Adolesc Health Med Ther 6:149. 10.2147/ahmt.s68344 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Strandberg-Larsen K, Cederkvist L, Aakjær A et al (2025) Cohort profile: the Danish National Birth Cohort from foetal life to young adulthood. Int J Epidemiol 54:dyaf083. 10.1093/ije/dyaf083 [DOI] [PubMed] [Google Scholar]
- 19.Erlangsen A, Fedyszyn I (2015) Danish nationwide registers for public health and health-related research. Scand J Public Health 43:333–339 [DOI] [PubMed] [Google Scholar]
- 20.Yurgelun-Todd D (2007) Emotional and cognitive changes during adolescence. Curr Opin Neurobiol 17:251–257 [DOI] [PubMed] [Google Scholar]
- 21.Henderson ZB, Fox JRE, Trayner P, Wittkowski A (2019) Emotional development in eating disorders: a qualitative metasynthesis. Clin Psychol Psychother 26:440–457 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Larsen PS, Strandberg-Larsen K, Olsen EM et al (2018) Parental characteristics in association with disordered eating in 11- to 12-year-olds: a study within the Danish National Birth Cohort. Eur Eat Disord Rev 26:315–328. 10.1002/erv.2599 [DOI] [PubMed] [Google Scholar]
- 23.Stemann Larsen P, Nybo Andersen AM, Olsen EM et al (2019) Weight trajectories and disordered eating behaviours in 11- to 12-year-olds: a longitudinal study within the Danish National Birth Cohort. Eur Eat Disord Rev 27:436–444. 10.1002/erv.2680 [DOI] [PubMed] [Google Scholar]
- 24.Ferruzzi MG, Tanprasertsuk J, Kris-Etherton P et al (2020) Perspective: the role of beverages as a source of nutrients and phytonutrients. Adv Nutr 11:507–523 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Cole TJ, Lobstein T (2012) Extended international (IOTF) body mass index cut-offs for thinness, overweight and obesity. Pediatr Obes 7:284–294. 10.1111/j.2047-6310.2012.00064.x [DOI] [PubMed] [Google Scholar]
- 26.Joensen A, Elsenburg LK, Olsen EM et al (2025) Trajectories of childhood adversity and eating disorders in adolescence. Int J Eat Disord. 10.1002/eat.24514 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lynge E, Sandegaard JL, Rebolj M (2011) The Danish national patient register. Scand J Public Health 39:30–33. 10.1177/1403494811401482 [DOI] [PubMed] [Google Scholar]
- 28.Mors O, Perto GP, Mortensen PB (2011) The Danish psychiatric central research register. Scand J Public Health 39:54–57. 10.1177/1403494810395825 [DOI] [PubMed] [Google Scholar]
- 29.Kann L, Warren CW, Harris WA et al (1996) Youth risk behavior surveillance- United States, 1995. J Sch Health 66:365–377. 10.1111/j.1746-1561.1996.tb03394.x [DOI] [PubMed] [Google Scholar]
- 30.Shisslak CM, Renger R, Sharpe T et al (1999) Development and evaluation of the McKnight risk factor survey for assessing potential risk and protective factors for disordered eating in preadolescent and adolescent girls. Int J Eat Disord 25:195–214. 10.1002/(SICI)1098-108X(199903)25:2%3c195::AID-EAT9%3e3.0.CO;2-B [DOI] [PubMed] [Google Scholar]
- 31.Micali N, De Stavola B, Ploubidis G et al (2015) Adolescent eating disorder behaviours and cognitions: gender-specific effects of child, maternal and family risk factors. Br J Psychiatry 207:320–327. 10.1192/bjp.bp.114.152371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.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. J Abnorm Psychol 126:38–51. 10.1037/abn0000219 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Micali N, Solmi F, Horton NJ et al (2015) Adolescent eating disorders predict psychiatric, high-risk behaviors and weight outcomes in young adulthood. J Am Acad Child Adolesc Psychiatry 54:652–659 10.1016/j.jaac.2015.05.009 [DOI] [PMC free article] [PubMed]
- 34.Danielsen S, Joensen A, Andersen PK et al (2023) Self-injury, suicidality and eating disorder symptoms in young adults following COVID-19 lockdowns in Denmark. Nat Hum Behav 7:411–419. 10.1038/s41562-022-01511-7 [DOI] [PubMed] [Google Scholar]
- 35.Joergensen AC, Strandberg-Larsen K, Andersen PK et al (2021) Spinal pain in pre-adolescence and the relation with screen time and physical activity behavior. BMC Musculoskelet Disord. 10.1186/s12891-021-04263-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Goodman R (1997) The strengths and difficulties questionnaire: a research note. J Child Psychol Psychiatry 38:581–586. 10.1111/j.1469-7610.1997.tb01545.x [DOI] [PubMed] [Google Scholar]
- 37.Keel PK (2019) Purging disorder: recent advances and future challenges. Curr Opin Psychiatry 32:518–524 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Jakobsen DD, Brader L, Bruun JM (2023) Association between food, beverages and overweight/obesity in children and adolescents—a systematic review and meta-analysis of observational studies. Nutrients. 10.3390/nu15030764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Malik VS, Pan A, Willett WC, Hu FB (2013) Sugar-sweetened beverages and weight gain in children and adults: a systematic review and meta-analysis. Am J Clin Nutr 98:1084–1102. 10.3945/ajcn.113.058362 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Zheng M, Rangan A, Olsen NJ et al (2014) Sugar-sweetened beverages consumption in relation to changes in body fatness over 6 and 12 years among 9-year-old children: the European youth heart study. Eur J Clin Nutr 68:77–83. 10.1038/ejcn.2013.243 [DOI] [PubMed] [Google Scholar]
- 41.Nissinen K, Mikkilä V, Männistö S et al (2009) Sweets and sugar-sweetened soft drink intake in childhood in relation to adult BMI and overweight. the cardiovascular risk in young finns study. Public Health Nutr 12:2018–2026. 10.1017/S1368980009005849 [DOI] [PubMed] [Google Scholar]
- 42.Gračner T, Marquez-Padilla F, Hernandez-Cortes D (2022) Changes in weight-related outcomes among adolescents following consumer price increases of taxed sugar-sweetened beverages. JAMA Pediatr 176:150–158. 10.1001/jamapediatrics.2021.5044 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Chiang WL, Azlan A, Mohd Yusof BN (2022) Effectiveness of education intervention to reduce sugar-sweetened beverages and 100% fruit juice in children and adolescents: a scoping review. Expert Rev Endocrinol Metab 17:179–200 [DOI] [PubMed] [Google Scholar]
- 44.Montgomery M, Johnson P, Ewell P (2022) A comparative analysis of rural versus urban preschool children’s Sugar-sweetened beverage consumption, body mass index and parent’s weight status. SAGE Open Nurs 8:23779608221082960. 10.1177/23779608221082962 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Nielsen J, Hulman A, Narayan KMV, Cunningham SA (2022) Body mass index trajectories from childhood to adulthood and age at onset of overweight and obesity: the influence of parents’ weight status. Am J Epidemiol 191:1877–1885. 10.1093/aje/kwac124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Bolt-Evensen K, Vik FN, Stea TH et al (2018) Consumption of sugar-sweetened beverages and artificially sweetened beverages from childhood to adulthood in relation to socioeconomic status–15 years follow-up in Norway. Int J Behav Nutr Phys Act. 10.1186/s12966-018-0646-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Lane MM, Travica N, Gamage E et al (2025) Sugar-sweetened beverages and adverse human health outcomes: an umbrella review of meta-analyses of observational studies. Annu Rev Nutr 45 44(1):383-404.10.1146/annurev-nutr-062322-020650 [DOI] [PubMed]
- 48.Chakravartti SP, Jann K, Veit R et al (2025) Non-caloric sweetener effects on brain appetite regulation in individuals across varying body weights. Nat Metab 7:574–585. 10.1038/s42255-025-01227-8 [DOI] [PubMed] [Google Scholar]
- 49.Sun Y, Xu B (2025) A critical review on effects of artificial sweeteners on gut microbiota and gastrointestinal health. J Sci Food Agric. 10.1002/jsfa.14148 [DOI] [PubMed] [Google Scholar]
- 50.Jebeile H, Lister NB, Baur LA et al (2021) Eating disorder risk in adolescents with obesity. Obes Rev. 10.1111/obr.13173 [DOI] [PubMed] [Google Scholar]
- 51.Neumark-Sztainer D (2015) Higher weight status and restrictive eating disorders: an overlooked concern. J Adolesc Health 56:1–2 [DOI] [PubMed] [Google Scholar]
- 52.Treasure J, Antunes Duarte T, Schmidt U (2020)Eating disorders. Lancet 395(10227):899-911.10.1016/S0140-6736(20)30059-3 [DOI] [PubMed]
- 53.Leth-Møller KB, Hebebrand J, Strandberg-Larsen K et al (2023) Childhood body mass index and the subsequent risk of anorexia nervosa and bulimia nervosa among women: a large Danish population-based study. Int J Eat Disord 56:1614–1622. 10.1002/eat.23986 [DOI] [PubMed] [Google Scholar]
- 54.Hughes EK, Kerr JA, Patton GC et al (2019) Eating disorder symptoms across the weight spectrum in Australian adolescents. Int J Eat Disord 52:885–894. 10.1002/eat.23118 [DOI] [PubMed] [Google Scholar]
- 55.Neumark-Sztainer D, Story M, Hannan PJ et al (2002) Weight-related concerns and behaviors among overweight and nonoverweight adolescents implications for preventing weight-related disorders. Arch Pediatr Adolesc Med. 10.1001/archpedi.156.2.171 [DOI] [PubMed] [Google Scholar]
- 56.Fan J, Nagata JM, Cuccolo K, Ganson KT (2024) Associations between dieting practices and eating disorder attitudes and behaviors: results from the Canadian study of adolescent health behaviors. Eat Behav 54:101886. 10.1016/j.eatbeh.2024.101886 [DOI] [PubMed] [Google Scholar]
- 57.Willett W (2013) Nutritional epidemiology. Oxford University Press, Oxford [Google Scholar]
- 58.Malik VS, Hu FB (2022) The role of sugar-sweetened beverages in the global epidemics of obesity and chronic diseases. Nat Rev Endocrinol 18:205–218 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Brown RJ, De Banate MA, Rother KI (2010) Artificial sweeteners: a systematic review of metabolic effects in youth. Int J Pediatr Obes 5:305–312 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Lara-Castor L, Micha R, Cudhea F et al (2024) Intake of sugar sweetened beverages among children and adolescents in 185 countries between 1990 and 2018: population based study. BMJ. 10.1136/bmj-2024-079234 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Matthiessen J, Christensen T, Biltoft-Jensen AP (2024) Lightdrikke: Sammenhæng med kostkvalitet og drikkevaner i Danmark. DTU Fødevareinstituttet, Lyngby [Google Scholar]
- 62.Sweatt K, Garvey WT, Martins C (2024) Strengths and limitations of BMI in the diagnosis of obesity: What is the path forward? Curr Obes Rep 13:584–595 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
According to European law (General Data Protection Regulation), data containing potentially identifying or sensitive personal information are restricted. However, for academic research, data can be available subsequent to approval. Any request for DNBC data needs to follow the outlined procedures, see: https://www.dnbc.dk/access-to-dnbc-data. For access to the underlying person-level data in Danish registers, researchers need to apply to the Danish Health Data Authority and Statistics Denmark. The programming codes used on DNBC and register data in this study are available by contacting the corresponding author.