Effect of Adolescent Body Mass Index Trajectories on Working Memory: A Prospective Birth Cohort in Brazil

Abstract

Most research on the association between high body mass index (BMI) and working memory (WM) has been cross-sectional in design, limiting conclusions about long-term effects of overweight and obesity on WM. The aim of this study was to examine the association of BMI trajectory from 11 to 22 years with WM at 22 years of age. Data from the 1993 Pelotas (Brazil) Birth Cohort Study were analyzed (N = 3,010). Information on BMI was collected at ages 11, 15, 18, and 22 years. Group-based trajectory modeling was used separately for each sex to identify BMI trajectories. Working Memory performance (Digit Span backward score) was examined at age 22. Multiple linear regression was used to assess the association between BMI trajectory from 11 to 22 years and WM at age 22. In both sexes, the trajectory groups were: stable normal weight, stable overweight, and stable obesity. In the adjusted analyses, women in the stable-obesity group had lower WM at 22 years (β = −.49; 95% CI: −0.75, −0.23; p < .001) than those in the stable-normal weight group. No associations were found between BMI trajectories and WM in men. This study may contribute to future investigations of possible explanatory avenues for the association between high BMI and WM.

Keywords

body mass index working memory longitudinal cognition adolescence

Introduction

Obesity is defined through several factors including body mass index (BMI) and excess peripheral fat accumulation, which is a risk factor for hypertension, inflammation, dyslipidemia, and insulin resistance (Bastard et al., 2006). The prevalence of obesity in childhood and adolescents is problematic and rising in both developed and developing nations (Widhalm, 2018). Worldwide, more than 340 million children and adolescents aged 5 to 19 are overweight or obese (World Health Organization [WHO], 2021). In Brazil, the prevalence of overweight or obese in this age group increased by 60% between 1990 and 2018 (Simões et al., 2018). Additionally, the 2021 Food and Nutrition Surveillance System has shown that 19.8% of Brazilian adolescents in the age group 10 to 20 years were overweight and 13.0% were obese (IBGE, 2022).

Recently, considerable attention has been paid to the relationship between BMI and executive functioning, including working memory (Mamrot & Hanć, 2019; Yang et al., 2018). Working memory (WM) is the ability to temporarily store and retain information, while a particular task is being performed, which can be accessed, manipulated, and reorganized (Baddeley & Hitch, 1994). Individual differences in WM capacity have been associated with a variety of cognitive and social outcomes including academic performance (Dumontheil & Klingberg, 2012; Gathercole et al., 2004).

A common finding in obesity research is that elevated BMI in childhood and adolescence is associated with lower WM performance (Mamrot & Hanć, 2019; Yang et al., 2018). However, although obese children and adolescents are around five times more likely to be obese in adulthood than those who were not obese (Simmonds et al., 2016), most research on the association between high BMI and WM has been cross-sectional in design, limiting conclusions about long-term effects of overweight and obesity on working memory (Mamrot & Hanć, 2019; Yang et al., 2018). In addition, the capacity of BMI to predict adiposity may vary across populations (Deurenberg et al., 1998). Thus, studies that use other measures of body composition, such as body fat percentage, are needed to confirm and extend these findings.

Understanding the effects of the BMI trajectory in adolescence on WM has important implications for clinical practice as well as for public health. Adolescence is a critical period in development for establishing healthy behaviors and implementing interventions, given the tremendous physiological and social changes occurring during this time that make the brain particularly receptive to adaptation (Lee et al., 2014). Thus, the aim of this study was to examine the association of BMI trajectory in adolescence with WM capacity at 22 years of age. Our main hypothesis was that poorer WM performance would be observed in the group of individuals who were consistently overweight or obese across adolescence than those who were always of normal weight.

Materials and Methods

Study Data and Cohort Formation

In 1993, all hospitals in the city of Pelotas (Brazil) were monitored daily, and mothers of newborns were invited to participate in a prospective study. From the 5,265 live births in the city, 5,249 were enrolled in the birth cohort study. Mothers were interviewed shortly after delivery on demographic, socioeconomic, and health-related variables and newborns were weighted and measured by the study team (Victora et al., 2006). All cohort members were followed when they reached the mean age of 11, 15, 18, and 22 years (Gonçalves et al., 2018). In this study, we included all the participants who had data available on body composition measures and WM.

Measurements

Working Memory

At the 22-year follow-up, we assessed WM using the Digit Span Backward subtest from the WAIS-III (Wechsler, 1997). This subtest requires the participant to repeat the numbers in the reverse order of that presented by the examiner. In contrast to digits forwards (repetition of digits in the same order presented), which involves only the temporary storage and maintenance of information in mind, digits backward require storage, maintenance and manipulation of information and thus qualifies as a WM task (Diamond, 2013).

In the present study, a trained psychologist recited a set of digits (at the rate of one digit per second) which the participant repeated in reverse order. The first set of digits consisted of two digits. The set size increased by one digit every two trials. The test stopped when the subject made two consecutive errors at any given set size. The total score was the sum of the item scores; the maximum backwards digit span score was 14 points. The test was administered individually using a standardized procedure in a private and quiet room. Both the participants and the psychologists were blinded to the purpose of the study.

The WAIS-III has been adapted and standardized for the Brazilian population (Nascimento, 2004). In the 20- to 29-year-old participants in the Brazilian standardization sample, the median was 5 points in the digits backwards (de Figueiredo & do Nascimento, 2007).

Body Mass Index

Information about BMI was collected when the cohort participants were 11, 15, 18, and 22 years of age. The BMI was calculated by dividing weight (in kilograms) by height (in meters squared), and was assessed continuously. We measured weight using a digital scale with precision of 100 g at ages 11 and 15 years (SECA; Birmingham, UK). At the 18- and 22-year follow-up visits, weight was measured with a scale attached to a BOD POD plethysmograph (COSMED; Albano Laziale, Italy), with 10 g precision. In all follow-up visits, height was measured with a free-standing stadiometer (aluminum and wood), with 0.1 cm precision. Interviewers underwent standardization testing (Lohman et al., 1988) before beginning fieldwork and every 2 months afterward to determine repeatability and validity of measurements.

Body Fat Percentage

At the 18- and 22-year follow-up visits, body fat percentage (BF%) was evaluated by air-displacement plethysmography using BOD POD (COSMED; Albano Laziale, Italy) handled by specifically trained technicians. Plethysmography is a safe, fast, and noninvasive method used in various population groups (obese individuals, children, adults, and older adults; Dempster & Aitkens, 1995). For this measure, participants remain inside the device, a closed chamber, for a few seconds without moving. It is necessary to eliminate the effect of the volume of clothing, hair, body surface, and lungs to measure with adequate accuracy and minimize disparities in body volume measurement, so all participants were provided with appropriate clothing. Sets of a rubber (swimming) cap and clothes specially made (shorts and elastane tank top) were provided to the cohorts’ participants. A predictor of the thoracic gas volume was used based on the participant’s age, sex, and height (Lohman et al., 2005). Standard equations were used to define BF% at ages 18 and 22 years (Siri, 1993).

Covariates

We selected covariates according to previous literature on BMI and WM (Mamrot & Hanć, 2019; Yang et al., 2018). Perinatal covariates included were: sex (female and male), skin color (white, black, brown, and others), household income (expressed in Brazilian minimum wages), birth weight (<2,500, 2,500–2,999, 3,000–3,499, and ≥3,500 g), maternal education (0–4, 5–8, 9–11, and ≥12 years) and smoking during pregnancy (no/yes). Birth weight was measured by trained interviewers using pediatric scales with a precision of 10 g, and the other information was self-reported by the mothers. Skin color was measured using the procedure adopted by the Brazilian Institute for Geography and Statistics.

From the 18-year follow-up, the following covariates were included: intelligence quotient (IQ), major depressive disorder (MDD; no/yes), and generalized anxiety disorder (GAD; no/yes). The estimated IQ was assessed using the Wechsler Adult Intelligence Scale, third version (WAIS-III) with the arithmetic, digit symbol, similarities, and picture completion subtests. The MDD and GAD were assessed using the Brazilian version of Mini International Neuropsychiatric Interview (MINI), version 5.0. Mental disorders definitions followed the DSM-5 criteria, and individuals were diagnosed with current MDD and GAD if they had presented symptoms in the 15 days and 6 months prior to the interview, respectively. Full details of the IQ and mental disorders measurements have been published previously (Gonçalves et al., 2014; Soares et al., 2022).

Other covariates included attention-deficit/hyperactivity disorder (ADHD; no/yes) and sleep duration. The presence of ADHD was assessed at the 11-year follow-up using the Brazilian version of the Strengths and Difficulties Questionnaire (SDQ, parent-reported version) applied to the parents of the participants in the form of an interview conducted by trained personnel. The cutoff point of 8 or more points on the SDQ hyperactivity scale was adopted (85.7% sensitivity and 67.4% specificity for the ADHD diagnosis)(Anselmi et al., 2010). Sleep duration (minutes/day) was calculated using the mean difference between the usual time to wake up and sleep, except on Saturday and Sunday, at the 11- and 22-year follow-ups.

Ethics

The study was approved by the Ethics Committee of the Faculty of Medicine of the Federal University of Pelotas. Before participating in the study, the parental consent of the participants was obtained. More details of the methods have been reported previously (Gonçalves et al., 2018).

Statistical Analysis

Analyses were performed using Stata, version 15.0. (Stata Corp., College Station, USA) and statistical significance was set at 5% (in interaction analyses 10%). For descriptive analysis, we used ANOVA to compare means between the groups and Pearson correlation to compare two quantitative variables. Chi-square tests were performed comparing the distribution of the variables sex, family income, birth weight, and maternal education of the perinatal period with those of the 22-year-old follow-up to determine whether losses from follow-up could affect the sample.

To estimate the BMI trajectories in adolescence, BMI data from each of the four waves (at ages 11, 15, 18, and 22 years) were used. Group-based trajectory analysis (GBTM) (D. S. Nagin, 2009; D. Nagin & Odgers, 2010) was the statistical technique adopted to construct the BMI trajectory groups. This is a specialized form of finite mixture modeling using a polynomial function to model the relationship between an attribute, such as BMI, and age or time (D. S. Nagin, 2009; D. Nagin & Odgers, 2010). The models were estimated separately for each sex with the Stata procedure “traj” (Jones & Nagin, 2013). The number and shape of trajectories were based both on the best fit of the model (lower Bayesian information criteria—BIC) and on the interpretability of the obtained trajectories. The selection of these trajectories was confirmed using the posterior probability score, which assesses the subject’s probability of belonging to each trajectory group. This probability should be higher than 0.70 for all groups (D. S. Nagin, 2009).

A plot showing the average trajectories superimposed on the World Health Organization (WHO) BMI classification for underweight, normal weight, overweight, and obesity ranges was produced (de Onis et al., 2007; WHO, 1995). At ages 11, 15, and 18 years, age-, and sex-specific cut points were used to define BMI categories (de Onis et al., 2007). At 22 years, BMI was classified as underweight (BMI < 18.5) normal weight (18.5 kg/m² ≤ BMI < 24 kg/m²), overweight (24 kg/m² ≤ BMI < 28 kg/m²), and obesity (BMI ≥ 28 kg/m²) (WHO, 1995).

Furthermore, unadjusted and adjusted linear regressions were used to examine the relationship between BMI trajectories and WM at 22 years. In the linear regression models, we assessed multicollinearity between the explanatory variables using the variance inflation factor. We based statistical comparisons between categories on tests of heterogeneity and linear trend, and we present the one with the lower p value.

In order to investigate whether BMI was a good indicator of body fat in our sample, additional sensitivity analyses were performed, by comparing participants’ BMI at ages 18 and 22 years with their measured BF% at the same ages. In addition, we examined the distribution of BF% at ages 18 and 22 years by BMI trajectories.

Results

General Characteristics of the Study Population

From the 5,249 original cohort members, 3,810 were followed up to the age of 22 years (representing a retention rate of 76.3%). The 3,010 who presented information about the body mass index measures and the WM were included in this study. Sociodemographic characteristics of the participants whose data were included in the analyses were similar to those in the original cohort (Supplemental Table 1).

The characteristics of the sample studied are shown in Table 1. Most of the participants were female (51.7%), white (61.7%), and had a perinatal household income up to three minimum wages (59.6%). About 10% had low birth weight (<2,500 g). Regarding the characteristics of the mothers, 47.4% had between five and eight successful complete years of schooling and 33.7% smoked during pregnancy. At 11 years of age, 19.0% of the cohort members had ADHD. The prevalence of MDD and GAD at 18 years was 4.5% and 9.8%, respectively. The average sleep duration between ages 11 and 22 was 530.1 (SD: ±71.5) minutes per day. The IQ average at 18 was 96.7 (SD: ±12.7) points. The average was 4.9 (SD: ±1.9) points in the digits backward at 22 years of age.

Table 1.

Sample Characteristics.

	Total	Men (N = 1,450)	Women (N = 1,560)	p-Value
	N (%)	N (%)	N (%)	p-Value
Parental data
Household income (in minimum wage)				.991^a
≤1	521 (17.3)	254 (17.5)	267 (17.1)
1.1–3	1,273 (42.3)	616 (42.5)	657 (42.1)
3.1–6	759 (25.2)	363 (25.1)	395 (25.3)
6.1–10	250 (8.3)	117 (8.1)	133 (8.5)
>10	208 (6.9)	99 (6.8)	109 (7.0)
Maternal education (years)				.623^a
0–4	783 (26.0)	364 (25.1)	419 (26.8)
5–8	1,448 (48.1)	708 (48.8)	740 (47.4)
9–11	551 (18.3)	262 (18.1)	289 (18.5)
≥12	229 (7.6)	116 (8.0)	113 (7.2)
Smoking during pregnancy (yes)	957 (31.8)	452 (31.2)	505 (32.4)	.570^a
Adolescent data
Skin color				.393^a
White	1,896 (63.0)	919 (63.4)	977 (62.6)
Black	473 (15.7)	228 (15.7)	245 (15.7)
Brown	527 (17.5)	241 (16.6)	286 (18.3)
Others	114 (3.8)	62 (4.3)	52 (3.4)
Birth weight				<.001^a
<2,500	277 (9.2)	106 (7.3)	171 (1.9)
2,500–2,999	716 (23.8)	310 (21.4)	406 (26.1)
3,000–3,499	1,189 (39.5)	572 (39.4)	617 (39.5)
≥3,500	828 (27.5)	462 (31.9)	366 (23.4)
ADHD at 11 years (yes)	578 (19.2)	333 (23.0)	244 (15.7)	<.001^a
MDD at 18 years (yes)	135 (4.5)	35 (2.4)	100 (6.4)	<.001^a
GAD at 18 years (yes)	295 (9.8)	76 (5.2)	219 (14.0)	<.001^a
Sleep duration (minutes/day)*, mean (SD)	53.1 (71.5)	517.9 (67.6)	542.1 (72.9)	<.001^b
BMI at 11 years, mean (SD)	18.8 (3.5)	18.8 (3.4)	18.8 (3.6)	<.001^b
BMI at 15 years, mean (SD)	21.4 (3.9)	21.3 (3.8)	21.5 (3.9)	.657^b
BMI at 18 years, mean (SD)	23.4 (4.5)	23.3 (4.1)	23.4 (4.8)	.166^b
BMI at 22 years, mean (SD)	25.4 (5.4)	25.2 (4.9)	25.5 (5.8)	.499^b
IQ at 18 years, mean (SD)	97.3 (12.5)	96.9 (13.3)	97.5 (11.7)	.155^b

Note. 1993 Pelotas Birth Cohort (N = 3,010). ADHD = attention-deficit/hyperactivity disorder; MDD = major depressive disorder; GAD = generalized anxiety disorder; BMI = Body mass index; IQ = intelligence quotient; SD = standard deviation.

Chi-square test.

ANOVA test.

Mean sleep duration between ages 11 and 22.

Association Between Adolescent BMI Trajectories and Subsequent WM

The GBTM identified three distinct BMI trajectories in both sexes. Due to the low change in BMI classification within each trajectory group (Supplemental Table 2), we classified them as follows: Stable normal weight, Stable overweight, and Stable obesity (Figure 1). All trajectories were best represented by a cubic term. The average posterior probability for each trajectory was 0.96, 0.93, and 0.98 in men and 0.95, 0.92, and 0.95 in women, indicating a good discrimination of our trajectory assignment. The distribution of BMI measures in each follow-up according to BMI trajectories is shown in Table 2.

Figure 1.

Body mass index (BMI) trajectories from 11 to 22 years of age superimposed on underweight, normal weight, overweight, and obesity ranges. 1993 Pelotas Birth Cohort Study, Brazil (N = 3,010): (a) men (N = 1,450) and (b) women (N = 1,560).

Table 2.

The Distribution of Body Mass Index (BMI) Measures in Each Follow-up According to Body Mass Index (BMI) Trajectories From 11 to 22 Years.

	BMI trajectory
	Men (N = 1,450)			Women (N = 1,560)
	Stable normal weight	Stable overweight	Stable obesity	Stable normal weight	Stable overweight	Stable obesity
BMI, mean (SD)
At 11 years	16.90 (1.66)	21.09 (2.66)	25.84 (3.19)	16.65 (1.80)	20.37 (2.54)	24.85 (3.89)
At 15 years	19.11 (1.74)	23.71 (2.58)	30.42 (3.46)	19.10 (1.80)	23.04 (2.25)	29.34 (3.87)
At 18 years	20.93 (1.84)	25.93 (2.42)	33.52 (3.91)	20.39 (1.98)	25.39 (2.39)	33.48 (4.12)
At 22 years	22.35 (2.48)	28.39 (2.63)	36.82 (4.59)	21.77 (2.47)	28.04 (2.95)	37.40 (4.98)

Note. 1993 Pelotas Birth Cohort (N = 3,010).

Table 3 shows the crude and adjusted analyses of the associations between BMI trajectories from 11 to 22 years and WM at 22 years of age. Adjusted estimates showed that women in the Stable obesity group had, on average, a WM score at 22 years 0.48 points lower than those in the Stable normal weight group. However, no associations were found between BMI trajectories and WM in men. In the adjusted models, IQ was positively associated with WM at 22 years in both sexes. In women, WM increased with maternal education while lower WM scores were observed in participants with GAD at 18 years. The average variance inflation factors (VIF) showed no indication of multicollinearity (VIF ranged from 1.04 to 1.51).

Table 3.

Association Between Body Mass Index (BMI) Trajectories From 11 to 22 Years and Working Memory at 22 Years.

	Working memory (digit span backward score)
	Men (N = 1,450)		Women (N = 1,560)
	β [95% CI]	p-Value	β [95% CI]	p-Value
Crude analysis
BMI trajectory (ref.: stable normal weight)		.237^a		<.001^a
Stable overweight	.18 [−0.04, 0.40]		−.18 [−0.37, 0.14]
Stable obesity	.15 [−0.27, 0.59]		−.71 [−1.01, −0.42]
Adjusted analysis*
BMI trajectory (ref.: stable normal weight)		.380^a		.002^a
Stable overweight	−.13 [−0.32, 0.07]		−.14 [−0.30, 0.03]
Stable obesity	−.15 [−0.52, 0.21]		−.48 [−0.74, −0.21]
Skin color (ref.: White)		.060^a		.465^a
Black	−.27 [−0.52; −0.02]		−.14 [−0.36; 0.09]
Brown	−.19 [−0.43, 0.05]		−.14 [−0.35; 0.07]
Others	.21 [−0.23, 0.64]		−.11 [−0.54, 0.32]
Household income (in minimum wage; ref.: ≤1)		.301^b		.383^b
1.1–3	.10 [−0.15, 0.35]		−.03 [−0.25, 0.20]
3.1–6	.07 [−0.21, 0.36]		−.00 [−0.26; 0.25]
6.1–10	.22 [−0.18, 0.62]		−.05 [−0.40; 0.30]
>10	.25 [−0.20, 0.71]		.32 [−0.08, 0.71]
Maternal education (years; ref.: 0–4)		.133^b		.049^b
5–8	.08 [−0.15, 0.30]		.23 [0.03, 0.42]
9–11	.25 [−0.04, 0.55]		.23 [−0.03; 0.49]
≥12	.19 [−0.24; 0.62]		.27 [−0.11; 0.66]
Birth weight (ref.: <2,500)		.159^b		.064^b
2,500–2,999	.22 [−0.15, 0.58]		.18 [−0.09, 0.46]
3,000–3,499	.23 [−0.11, 0.58]		.21 [−0.05, 0.48]
≥3,500	.30 [−0.05, 0.66]		.30 [0.01; 0.58]
Smoking during pregnancy (yes)	.07 [−0.12, 0.27]	.449^a	.02 [−0.16, 0.19]	.895^a
ADHD at 11 years (yes)	−.01 [−0.04, 0.02]	.527^a	−.02 [−0.04, 0.01]	.283^a
MDD at 18 years (yes)	−0.08 [−0.65, 0.50]	.838^a	.26 [−0.07; 0.60]	.115^a
GAD at 18 years (yes)	0.08 [−0.32, 0.47]	.677^a	−.28 [−0.52, −0.04]	.024^a
Sleep duration (minutes/day)**, mean (SD)	−0.08 [−0.15, 0.00]	.063^a	−.03 [−0.10, 0.03]	.329^a
IQ at 18 years	0.08 [0.07, 0.08]	<.001^b	.07 [0.06, 0.07]	<.001^b

Note. 1993 Pelotas Birth Cohort (N = 3,010). ADHD = attention-deficit/hyperactivity disorder; MDD = major depressive disorder; GAD = generalized anxiety disorder; IQ = intelligence quotient; CI = confidence intervals.

Test for heterogeneity.

Test for linear trend.

Multiple linear regression models including all covariates as potential predictors.

Mean sleep duration between ages 11 and 22.

Sensitivity Analyses

The sensitivity analysis showed a strong and positive correlation between BMI- BF% in both sexes. The Pearson correlation coefficients for age 18 and 22 were 0.84 and 0.83 in women and 0.83 and 0.84 in men (Supplemental Figure 1). As shown in Supplemental Table 3, the mean BF% of the females was higher than the males in all BMI trajectories and the participants in the Stable obesity group had higher mean BF% compared to those in the Stable normal weight group.

Discussion

To our knowledge, this is the first prospective study to investigate the association between BMI trajectories in adolescence and WM in early adulthood. We found that being obese during adolescence predicts lower WM score at 22 years in females, not males, even after adjusting for IQ at 18 years. However, in our analyses, being overweight was not associated with differences in working memory in either sex.

Our findings corroborate with previous cross-sectional studies, which noted that adolescents with obesity performed worse than normal-weight adolescents on WM tasks (Alarcón et al., 2016; Bauer & Manning, 2016; Maayan et al., 2011; Schwartz et al., 2013; Verdejo-García et al., 2010). However, there is no clear evidence in the literature of sex differences in the relation of obesity with WM. Some authors have reported that body fat (e.g., visceral fat) and high BMI are related to reduced WM performance in adolescent and young adult females, not males (Schwartz et al., 2013; Yang et al., 2019), while others have reported no differences between sexes (Alarcón et al., 2016).

In our study, adolescents of both sexes who were overweight did not differ from normal-weight adolescents in their performance on WM. The only study to our knowledge that investigated the condition of being overweight compared to normal-weight and obesity in adolescence suggested that overweight and obesity predicts lower WM performance, with a clear gradient in WM performance by BMI categories (Alarcón et al., 2016). More studies are required to clarify how WM is expressed at the different stages of overweight and obesity and to understand if weight loss could improve WM performance in obese adolescents.

Our study has several strengths. It was carried out in a large population-based sample with high rates of retention and follow-up, minimizing the likelihood of selection bias. Regarding the complexity of the phenomenon studied, our analyses were controlled for a wide variety of sociodemographic factors and psychosocial factors, such as maternal schooling and ADHD. In addition, the BMI trajectory was evaluated through group-based trajectory modeling, a robust method used to identify groups of individuals with similar developmental trajectories (D. S. Nagin, 2009; D. Nagin & Odgers, 2010).

It is also important to consider that the data collected imposes some limitations on the analysis. The most important limitation is the non-availability of data on WM capacity in childhood and early adolescence. Some studies have suggested that WM performance could be a predictor of food behaviors and, consequently, of body weight changes (Allom et al., 2018; Dassen et al., 2018; Raman et al., 2018; Verbeken et al., 2013). To minimize reverse causality bias, we adjusted the analysis for IQ; in cohort members, the correlation between IQ and WM was 0.46. However, we cannot rule out the possibility that our findings were biased by reverse causation. Another limitation of the current investigation is that we used BMI as a proxy for adiposity, with the advantage of being easy to measure, and disadvantage of inability to distinguish between fatness and above average muscular development. However, BMI at ages 18 and 22 years showed a strong correlation with BF% in our sample. Finally, there is a lack of data on reproductive events in the female population that influence changes in body fat composition and may be related to changes in cognition, such as the occurrence and number of pregnancies and the use of hormonal contraception.

Conclusion

In this large prospective birth cohort, obesity across adolescence predicted lower WM performance only in women, suggesting that the effects of high BMI are not the same between sexes. Our findings may contribute to future investigations of possible explanatory avenues for the association between high BMI and WM performance.

Supplemental Material

sj-docx-1-jad-10.1177_10870547231153973 – Supplemental material for Effect of Adolescent Body Mass Index Trajectories on Working Memory: A Prospective Birth Cohort in Brazil

Supplemental material, sj-docx-1-jad-10.1177_10870547231153973 for Effect of Adolescent Body Mass Index Trajectories on Working Memory: A Prospective Birth Cohort in Brazil by Pedro San Martin Soares, Otávio Amaral de Andrade Leão, Mariane da Silva Dias, Fernando César Wehrmeister, Ana Maria Baptista Menezes and Helen Gonçalves in Journal of Attention Disorders

Footnotes

Author Statement

Pedro San Martin Soares, Otávio Amaral de Andrade Leão, Mariane da Silva Dias, Fernando César Wehrmeister, Ana Maria Baptista Menezes, and Helen Gonçalves—Contributed in the conception and design of the study, and in drafting and revising the paper. Pedro San Martin Soares, Otávio Amaral de Andrade Leão, and Mariane da Silva Dias—Contributed in data analysis and interpretation. All authors approved the final version.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Pedro San Martin Soares

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Pedro San Martin Soares is a postdoctoral researcher in the Epidemiology Postgraduate Program at Universidade Federal de Pelotas (UFPel), and an epidemiologist for the Health Secretary of Rio Grande do Sul, Brazil. He received a PhD in Epidemiology from the Universidade Federal de Pelotas (UFPel). His research is currently focused on the effects that screen time have on cognitive skills.

Otávio Amaral de Andrade is a postdoctoral researcher in the Epidemiology Postgraduate Program at Universidade Federal de Pelotas (UFPel). He received a PhD in Epidemiology from the Universidade Federal de Pelotas (UFPel). His research focuses on the epidemiology of the Life Cycle.

Mariane da Silva Dias is a postdoctoral researcher in the Epidemiology Postgraduate Program at Universidade Federal de Pelotas (UFPel). She received a PhD in Epidemiology from the Universidade Federal de Pelotas (UFPel). Her research focuses on the epidemiology of the Life Cycle.

Fernando César Wehrmeister is an associate professor at the Postgraduate Program in Epidemiology at Universidade Federal de Pelotas (UFPel). He received a PhD in Epidemiology from the Universidade Federal de Pelotas (UFPel). Dr. Wehrmeister’s research focuses on the epidemiology of the Life Cycle.

Ana Maria Baptista Menezes is a permanent professor at the Postgraduate Program in Epidemiology at Universidade Federal de Pelotas (UFPel). She received a PhD in Medicine (Pulmonology) from the Universidade Federal do Rio Grande do Sul (UFRGS). Her research focuses on the epidemiology of the Life Cycle.

Helen Gonçalves is a permanent professor at the Postgraduate Program in Epidemiology at Universidade Federal de Pelotas (UFPel). She received a PhD in Social Anthropology from the Universidade Federal do Rio Grande do Sul (UFRGS). Her research focuses on the premature determinants of health and nutrition of adolescents and adults and Mental Health.

References

Alarcón

Ray

Nagel

B. J.

(2016). Lower working memory performance in overweight and obese adolescents is mediated by white matter microstructure. Journal of the International Neuropsychological Society, 22(3), 281–292.

Allom

Mullan

Smith

Hay

Raman

(2018). Breaking bad habits by improving executive function in individuals with obesity. BMC Public Health, 18(1), 505.

Anselmi

Fleitlich-Bilyk

Menezes

A. M. B.

Araújo

C. L.

Rohde

L. A.

(2010). Prevalence of psychiatric disorders in a Brazilian birth cohort of 11-year-olds. Social Psychiatry and Psychiatric Epidemiology, 45(1), 135–142.

Baddeley

A. D.

Hitch

G. J.

(1994). Developments in the concept of working memory. Neuropsychology, 8(4), 485–493. https://doi.org/10.1037/0894-4105.8.4.485

Bastard

J.-P.

Maachi

Lagathu

Kim

M. J.

Caron

Vidal

Capeau

Feve

(2006). Recent advances in the relationship between obesity, inflammation, and insulin resistance. European Cytokine Network, 17(1), 4–12.

Bauer

L. O.

Manning

K. J.

(2016). Challenges in the detection of working memory and attention decrements among overweight adolescent girls. Neuropsychobiology, 73(1), 43–51.

Dassen

F. C. M.

Houben

Van Breukelen

G. J. P.

Jansen

(2018). Gamified working memory training in overweight individuals reduces food intake but not body weight. Appetite, 124, 89–98.

de Figueiredo

V. L. M.

do Nascimento

(2007). Desempenhos nas duas tarefas do subteste dígitos do WISC-III e do WAIS-III. Psicologia Teoria E Pesquisa, 23(3), 313–318.

Dempster

Aitkens

(1995). A new air displacement method for the determination of human body composition. Medicine and Science in Sports and Exercise, 27(12), 1692–1697.

10.

de Onis

Onyango

A. W.

Borghi

Siyam

Nishida

Siekmann

(2007). Development of a WHO growth reference for school-aged children and adolescents. Bulletin of the World Health Organization, 85(9), 660–667. https://doi.org/10.2471/blt.07.043497

11.

Deurenberg

Yap

van Staveren

W. A.

(1998). Body mass index and percent body fat: A meta analysis among different ethnic groups. International Journal of Obesity and Related Metabolic Disorders: Journal of the International Association for the Study of Obesity, 22(12), 1164–1171.

12.

Diamond

(2013). Executive functions. Annual Review of Psychology, 64, 135–168.

13.

Dumontheil

Klingberg

(2012). Brain activity during a visuospatial working memory task predicts arithmetical performance 2 years later. Cerebral Cortex, 22(5), 1078–1085.

14.

Gathercole

S. E.

Pickering

S. J.

Ambridge

Wearing

(2004). The structure of working memory from 4 to 15 years of age. Developmental Psychology, 40(2), 177–190.

15.

Gonçalves

Assunção

M. C.

Wehrmeister

F. C.

Oliveira

I. O.

Barros

F. C.

Victora

C. G.

Hallal

P. C.

Menezes

A. M.

(2014). Cohort profile update: The 1993 Pelotas (Brazil) birth cohort follow-up visits in adolescence. International Journal of Epidemiology, 43(4), 1082–1088.

16.

Gonçalves

Wehrmeister

F. C.

Assunção

M. C. F.

Tovo-Rodrigues

Oliveira

I. O.

de Murray

Anselmi

Barros

F. C.

Victora

C. G.

Menezes

A. M. B.

(2018). Cohort profile update: The 1993 Pelotas (Brazil) Birth Cohort follow-up at 22 years. International Journal of Epidemiology, 47(5), 1389–1390e.

17.

IBGE. (2022). Sistema de Vigilância Alimentar e Nutricional. Relatórios Públicos - SISVAN. https://sisaps.saude.gov.br/sisvan/relatoriopublico/estadonutricional

18.

Jones

B. L.

Nagin

D. S.

(2013). A note on a stata plugin for estimating group-based trajectory models. Sociological Methods & Research, 42(4), 608–613.

19.

Lee

F. S.

Heimer

Giedd

J. N.

Lein

E. S.

Šestan

Weinberger

D. R.

Casey

B. J.

(2014). Mental health. Adolescent mental health–opportunity and obligation. Science, 346(6209), 547–549.

20.

Lohman

T. G.

Roche

A. F.

Martorell

(1988). Anthropometric standardization reference manual. Human Kinetics Books.

21.

Lohman

T. G

Wang

Going

S. B.

(2005). Human body composition. Human Kinetics.

22.

Maayan

Hoogendoorn

Sweat

Convit

(2011). Disinhibited eating in obese adolescents is associated with orbitofrontal volume reductions and executive dysfunction. Obesity, 19(7), 1382–1387.

23.

Mamrot

Hanć

(2019). The association of the executive functions with overweight and obesity indicators in children and adolescents: A literature review. Neuroscience and Biobehavioral Reviews, 107, 59–68.

24.

Nagin

Odgers

C. L.

(2010). Group-based trajectory modeling in clinical research. SSRN.

25.

Nagin

D. S.

(2009). Group-based modeling: An overview. In Krohn

Lizotte

Hall

(Eds.), Handbooks of sociology and social research (pp. 59–73). Springer. https://doi.org/10.1007/978-1-4419-0245-0_4

26.

Nascimento

(2004). Adaptação, validação e normatização do WAIS-III para uma amostra brasileira. WAIS-III: Manual Para Administração E Avaliação, 1, 161–192.

27.

Raman

Hay

Tchanturia

Smith

(2018). A randomised controlled trial of manualized cognitive remediation therapy in adult obesity. Appetite, 123, 269–279.

28.

Schwartz

D. H.

Leonard

Perron

Richer

Syme

Veillette

Pausova

Paus

(2013). Visceral fat is associated with lower executive functioning in adolescents. International Journal of Obesity, 37(10), 1336–1343.

29.

Simmonds

Llewellyn

Owen

C. G.

Woolacott

(2016). Predicting adult obesity from childhood obesity: A systematic review and meta-analysis. Obesity Reviews, 17(2), 95–107.

30.

Simões

C. F.

Lopes

W. A.

Remor

J. M.

Locateli

J. C.

Lima

F. B.

Santos

T. L. C. dos

Nardo Junior

(2018). Prevalence of weight excess in Brazilian children and adolescents: A systematic review. Revista Brasileira de Cineantropometria & Desempenho Humano, 20(4), 517–531.

31.

Siri

W. E.

(1993). Body composition from fluid spaces and density: Analysis of methods. 1961. Nutrition, 9(5), 480–491; discussion 480, 492.

32.

Soares

P. S. M.

de Oliveira

P. D.

Wehrmeister

F. C.

Menezes

A. M. B.

Rohde

L. A.

Gonçalves

(2022). Does IQ influence association between working memory and ADHD symptoms in young adults? Journal of Attention Disorders, 26(8), 1097–1105.

33.

Verbeken

Braet

Goossens

van der Oord

(2013). Executive function training with game elements for obese children: A novel treatment to enhance self-regulatory abilities for weight-control. Behaviour Research and Therapy, 51(6), 290–299.

34.

Verdejo-García

Pérez-Expósito

Schmidt-Río-Valle

Fernández-Serrano

M. J.

Cruz

Pérez-García

López-Belmonte

Martín-Matillas

Martín-Lagos

J. A.

Marcos

Campoy

(2010). Selective alterations within executive functions in adolescents with excess weight. Obesity, 18(8), 1572–1578.

35.

Victora

C. G.

Araújo

C. L. P.

Menezes

A. M. B.

Hallal

P. C.

de Vieira

M. F.

Neutzling

M. B.

Gonçalves

Valle

N. C.

Lima

R. C.

Anselmi

Behague

Gigante

D. P.

Barros

F. C.

(2006). Methodological aspects of the 1993 Pelotas (Brazil) Birth Cohort Study. Revista de Saude Publica, 40(1), 39–46.

36.

Wechsler

(1997). WMS-III: Wechsler Memory Scale administration and scoring manual. Psychological Corporation.

37.

Widhalm

(2018). Prevention of obesity in childhood and adolescence. Obesity Facts, 11(3), 232–233.

38.

World Health Organization. (1995). Physical status: The use and interpretation of anthropometry (World Health Organization Technical Report Series, 854, 1–452). Report of a WHO Expert Committee.

39.

World Health Organization. (2021). Obesity and overweight. https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight

40.

Yang

Shields

G. S.

Guo

Liu

(2018). Executive function performance in obesity and overweight individuals: A meta-analysis and review. Neuroscience and Biobehavioral Reviews, 84, 225–244.

41.

Yang

Shields

G. S.

Liu

Guo

(2019). Obesity is associated with poor working memory in women, not men: Findings from a nationally representative dataset of U.S. adults. Eating Behaviors, 35, 101338.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.50 MB