Abstract
Background
Despite associations of dietary added sugar with excess weight gain and chronic disease risk, intake among most Americans exceeds the recommended limits (<10% total energy). Maternal diet plays an important role in pregnancy-related outcomes, but little is known about the extent of added sugar intake in pregnancy.
Objective
To assess added sugar intake among pregnant and non-pregnant women in the United States.
Design
Cross-sectional analysis of data from the National Health and Nutrition Examination Survey (NHANES), 2003–2004 to 2011–2012.
Participants
4,179 pregnant and non-pregnant women (20–39 years) who completed a dietary recall.
Statistical Analysis
Survey-weighted analyses were used to estimate means (95% confidence intervals) in total grams and as % total energy for added sugar intake by pregnancy status and by demographic subgroup, and to identify leading sources of added sugar.
Results
Added sugar intake trended towards being higher in pregnant compared to non-pregnant women in absolute grams, 85.1 g (95% CI: 77.4, 92.7) vs. 76.7 g (95% CI: 73.6, 79.9), respectively (p=0.06); but was lower among pregnant women when total energy intake was accounted for, 14.8% (95% CI: 13.8%, 15.7%) vs. 15.9% (95% CI: 15.2%, 16.6%) of total energy, respectively (p=0.03). Among pregnant women, added sugar intake was similar among demographic subgroups. However, in multivariable regression pregnancy status significantly modified the associations of education and income with added sugar intake, whereby less educated and lower income females had lower added sugar intakes if pregnant compared to if non-pregnant, but higher educated or higher income females did not exhibit this pattern. The top five sources of added sugar for all women were sugar-sweetened beverages; cakes, cookies, and pastries; sugars and sweets; juice drinks and smoothies; and milk desserts.
Conclusions
Although pregnant women had higher energy intakes, this was not attributed to higher intakes of added sugar. While education and income impacted consumption during pregnancy, intake of added sugar among all women, regardless of pregnancy status, exceeded recommendations.
Keywords: Pregnancy, Dietary Intakes, Added Sugar, Free Sugar, Carbohydrates, Gestational Weight Gain
Introduction/Background
Maternal over-nutrition is an emerging public health issue worldwide. Women who are obese prior to pregnancy or who gain gestational weight above the Institute of Medicine’s recommendations are at increased risk for adverse health outcomes, especially gestational diabetes (GDM) and delivery complications.1–6 Additionally, offspring of these mothers are at higher risk of several birth outcomes, including macrosomia, stillbirth, congenital defects, and shoulder dystocia,7–9 and may be predisposed to obesity and cardiometabolic diseases later in life.10–12 Together, these risks make it vital that health professionals are equipped with interventions to promote healthy weight control among women of reproductive age both before and during pregnancy.
In the general population, high consumption of added sugar has been associated with increased risk of weight gain and obesity.13–15 The 2015–2020 Dietary Guidelines for Americans recommended that all Americans limit their intakes of added sugar to less than 10% of total calories.16 No specific limits have been set for pregnant women, but a position paper from the Academy of Nutrition and Dietetics advised that pregnant women reduce intake of high-added sugar foods as a strategy to avoid excess calories without compromising nutrient adequacy.17 The potential benefits of following this guideline have been supported by several studies of pregnant women that found associations of high dietary intakes of added sugar with increased offspring fat mass and maternal gestational weight gain (GWG), and increased odds of GDM.18–21 However, few studies, if any, have investigated how added sugar intakes among pregnant women compare to intakes among non-pregnant women, as well as to recommendations for the general population.
Using surveillance data from the National Health and Nutrition Examination Survey (NHANES), the objective of this study was to describe patterns of added sugar intake among pregnant women in the United States overall and by demographic subgroup, and to determine the major food or beverage sources contributing to added sugar intake in pregnancy. Based on qualitative studies of pregnant women that have identified several barriers to achieving appropriate GWG,22,23 in particular limited knowledge of what a “healthy” diet is, the misunderstanding that pregnant mothers are “eating for two”, and the belief that overeating is necessary to meet the fetus’ nutritional needs, we hypothesized that pregnant women on average consume a higher percentage of calories from added sugar compared to non-pregnant as a result of these misconceptions. The insights gained from this analysis will inform efforts to identify possible risk factors in pregnancy that can be modified by nutritional interventions to improve health outcomes among women of reproductive age and their offspring.
Materials and Methods
Study Population
The National Health and Nutrition Examination Survey (NHANES) is conducted cross-sectionally in 2-year cycles by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC) to assess the health status of non-institutionalized U.S. residents. It utilizes a complex, multistage, probability sampling design to obtain a nationally-representative sample, the details of which are described elsewhere.24 For this analysis, we merged five survey cycles (2003–2004, 2005–2006, 2007–2008, 2009–2010, and 2011–2012) to achieve greater sample size within subgroups and to ensure statistically reliable estimates. The NCHS Ethics Review Board approved all study protocols for each NHANES cycle included in this report.
The sample for this analysis was limited to females aged 20–39 years old with valid data for the urine pregnancy test at the mobile examination center (MEC) exam. Females 19 years or younger were excluded for continuity between cycles because starting in 2007–2008, pregnancy exam results were only released for females age 20–44 years. Females 40 years or older were also excluded due to the small number of pregnant women in this age range (n=11). Subjects were also excluded if they had incomplete or unreliable day 1 dietary recall data, or if they were missing values for family poverty-income ratio (PIR), education level, or marital status. (Figure 1) This resulted in a final sample of 4,179 pregnant (n=650) and non-pregnant women (n=3529).
Figure 1.
Flow chart of inclusion and exclusion criteria for the final analytical sample of pregnant and non-pregnant women in the United States (ages 20–39 years), NHANES 2003–2012
Dietary Intakes
What We Eat in America (WWEIA) is the dietary assessment component of NHANES and includes two 24-hr recalls performed by trained interviewers using the Automated Multiple-Pass Method. For this analysis, we used the day 1 dietary recall because a single recall for a large population is adequate to yield unbiased estimates for mean dietary intakes.25 For each subject, total energy intake [kilocalories (kcal)] per day was calculated using the USDA’s Food and Nutrient Database for Dietary Studies (FNDDS). Teaspoon equivalents of added sugars consumed per subject per day and per food/beverage item were calculated using the USDA’s Food Patterns Equivalence Database (FPED) corresponding to each cycle. In this database, “added sugar” was defined as all sugars added to foods during preparation, processing, or at the table, but not naturally occurring sugars present in milk and fruit. Teaspoon equivalents was converted to grams of sugar using the conversion factor of 4.2 grams of sugar per teaspoon, as suggested by the NCHS. Added sugar intakes were also expressed as a percentage of calories consumed per day by converting grams of sugar to kcals using the conversion factor 4 kcal/gram of sugar, and then dividing by total energy intake to yield a percentage.
Assessment of Other Covariates
Sociodemographic information, including age, race-ethnicity, education, and family poverty income ratio (PIR), was collected during in-person interviews. Body measurements, including weight (kg) and height (cm), were measured by trained NHANES staff during the MEC exam, and used to calculate current body mass index (BMI). For pregnant women, pre-pregnancy BMI was estimated using self-reported weight one year prior to the NHANES examinations, which was collected in the weight history questionnaire, and the same height measured at the MEC exam. To estimate the trimester of pregnancy, month of pregnancy was collected from the reproductive health questionnaire administered at the MEC exam.
Statistical Methods
All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). NHANES sampling weights and complex survey procedures were used to obtain nationally-representative estimates and to account for different sampling probabilities, non-response and non-coverage. Descriptive statistics of the sample were reported as counts and weighted frequencies by pregnancy status in Table 1. Least squares means (LS-means) and 95% confidence intervals were calculated for added sugar in grams and as a percentage of total energy intake per day overall for pregnant versus non-pregnant women, and compared using pairwise t-tests. The analysis was conducted separately on both crude and adjusted LS-mean intakes controlling for potential confounders including age, race/ethnicity, family PIR, education, and survey year.
Table 1.
Sociodemographic and health characteristics of the sample of pregnant and non-pregnant women in the United States (ages 20–39 years), NHANES 2003–2012
Category | Pregnant (n=650) | Non-pregnant (n=3529) | p-value | ||
---|---|---|---|---|---|
n | Weighted % | n | Weighted % | ||
Age at Screening | |||||
20–24 | 194 | 28.1% | 912 | 26.1% | 0.001 |
25–29 | 225 | 33.1% | 821 | 23.6% | |
30–34 | 161 | 23.9% | 905 | 24.4% | |
35–39 | 70 | 15.0% | 891 | 25.9% | |
Race/Ethnicity | |||||
Non-Hispanic White | 285 | 53.8% | 1519 | 63.9% | 0.002 |
Mexican American | 176 | 17.9% | 661 | 10.5% | |
Non-Hispanic Black | 113 | 17.1% | 769 | 13.5% | |
Other | 76 | 11.1% | 580 | 12.1% | |
Poverty Level | |||||
PIR < 100% | 171 | 20.4% | 1007 | 22.6% | 0.64 |
PIR 101–185% | 144 | 20.6% | 817 | 19.7% | |
PIR 186–350% | 135 | 21.6% | 812 | 24.2% | |
PIR > 350% | 200 | 37.3% | 893 | 33.5% | |
Education Level | |||||
Less than High School | 171 | 20.1% | 699 | 15.0% | 0.16 |
HS Diploma or GED | 130 | 18.7% | 719 | 20.1% | |
Some College | 192 | 33.0% | 1266 | 36.9% | |
College Degree or Above | 157 | 28.2% | 845 | 28.1% | |
Marital Status | |||||
Married | 428 | 66.3% | 1481 | 44.0% | <0.001 |
Not married | 222 | 33.7% | 2048 | 56.0% | |
Trimester | |||||
Missing | 51 | 10.8% | – | – | n/a |
1st (1–3 Months) | 149 | 28.4% | – | – | |
2nd (4–6 Months) | 226 | 28.3% | – | – | |
3rd (7–9 Months) | 224 | 32.5% | – | – | |
Pre-pregnancy BMI2 | |||||
Missing | 14 | 2.5% | – | – | n/a |
Underweight | 35 | 5.2% | – | – | |
Normal weight | 324 | 49.7% | – | – | |
Overweight | 154 | 21.7% | – | – | |
Obese | 123 | 20.9% | – | – | |
Survey Year3 | |||||
2003–2004 | 202 | 24.5% | 572 | 19.8% | <0.001 |
2005–2006 | 299 | 30.7% | 618 | 18.8% | |
2007–2008 | 49 | 13.1% | 752 | 21.0% | |
2009–2010 | 54 | 17.1% | 866 | 20.0% | |
2011–2012 | 46 | 14.6% | 721 | 20.3% |
P-values calculated using Rao-Scott modified chi-square test for each category.
Underweight (<18.5kg/m2), normal (18.5–24.9 kg/m2), overweight (25–29.9 kg/m2), obese (≥ 30 kg/m2)
Survey cycles 2003–2004 and 2005–2006 oversampled pregnant women, thus resulting in a larger sample size of pregnant women compared to the subsequent survey cycles.
Subgroup analyses focused on estimating added sugar intake as a proportion of total energy intake. To evaluate differences among pregnant women by subgroup, added sugar intakes were stratified by several covariates including: age (20–24, 25–29, 30–34, or 35–40 years), race/ethnicity (non-Hispanic White, non-Hispanic Black, Mexican American, or other/mixed race), education level (less than high school, high school diploma or GED, some college, or college degree and above), income level (family PIR ≤100%, 100 to 185%, 185 to 350%, or ≥ 350%), marital status (married or non-married), pre-pregnancy BMI [underweight (< 18.5 kg/m2), normal weight (18.5–24.9 kg/m2), overweight (25–29.9 kg/m2), or obese (> 30 kg/m2)] and trimester. Pairwise t-tests assessed differences in LS-mean intakes between each stratified level and a reference level among pregnant women.Tukey-Kramer adjustments were applied to account for multiple pairwise comparisons.
To assess heterogeneity in the association of each subgroup with added sugar intake by pregnancy status, an interaction term of pregnancy status (yes or no) × each covariate applicable for both pregnant and non-pregnant women (i.e., age, race/ethnicity, education, income, and marital status) was added to separate survey-weighted regression models with added sugar intake as the dependent variable. Models were adjusted for all other covariates to differentiate the independent association of each variable. Significance of interaction terms was tested by the survey-weighted Wald test.
Finally, the relative contribution of different sources of added sugar in the diet were assessed by first creating mutually-exclusive food groupings adapted from the USDA’s Food Research Survey Group (FSRG)-defined food groups. To create these food groups, we assigned all food/beverage items that contained added sugar and that were consumed in any of the combined cycles to one food grouping (Table S1; online supplementary material). For each subject, the total grams of added sugar consumed from each food grouping was calculated and divided by their total added sugar consumed from all groupings. This yielded several ratios representing the percent contribution of each food grouping to each subject’s daily added sugar intake. A mean for each ratio was calculated for pregnant versus non-pregnant women and compared using pairwise t-tests. Only the top ten food groupings that contributed to added sugar intake were reported. For all statistical tests, significance was set at alpha=0.05.
Results
Sample Characteristics
The sociodemographic and health traits of the sample of pregnant and non-pregnant women are summarized in Table 1. The proportion of pregnant vs. non-pregnant women in each sub-group level was balanced, except for age-group and race/ethnicity. Specifically, there was a higher percentage of Mexican American women and lower percentage of non-Hispanic white women and older women, ages 35–39 years old, among pregnant versus non-pregnant women, based on Rao-Scott Modified Chi-square. It should also be noted that there was a higher unweighted frequency of pregnant women in 2003–2004 and 2005–2006, compared to later cycles, because NHANES over-sampled pregnant women during these two cycles.
Added Sugar Intake by Pregnancy Status
In multivariable analysis estimating mean daily intake of added sugar, pregnant women consumed 85.1 g (340 kcal) and non-pregnant women 76.7 g (307 kcal; p=0.06). When the higher total energy intake in pregnancy was accounted for (2220 kcal/day for pregnant women vs. 1923 kcal/day for non-pregnant women, p<0.001), we found intake as a percent of total energy to be lower among pregnant women, 14.8% vs. 15.9% of total energy (p=0.03). (Table 2)
Table 2.
Mean consumption of sugar and total energy per day among pregnant and non-pregnant women in the United States (ages 20–39 years), NHANES 2003–2012
Pregnant (n=650) | Non-pregnant (n=3529) | P-value | P-value | |||
---|---|---|---|---|---|---|
Mean | 95% CI | Mean | 95% CI | (crude)1 | (adjusted)2 | |
Added Sugar | ||||||
Grams/day | 85.1 | (77.4, 92.7) | 76.7 | (73.6, 79.9) | 0.035 | 0.055 |
Kcal/day | 340.2 | (309.7, 370.7) | 306.9 | (294.2, 319.5) | 0.035 | 0.055 |
% kcal/day | 14.8% | (13.8%, 15.7%) | 15.9% | (15.2%, 16.6%) | 0.027 | 0.034 |
Total Energy | ||||||
Kcal/day | 2220 | (2126, 2314) | 1923 | (1895, 1952) | <0.001 | <0.001 |
SSB Sugar | ||||||
Grams/day | 38.6 | (32.8, 44.4) | 39.6 | (36.7, 42.5) | 0.737 | 0.366 |
Kcal/day | 158.3 | (146.8, 169.8) | 154.4 | (131.7, 370.7) | 0.737 | 0.366 |
% kcal/day | 6.7% | (5.7%, 7.6%) | 8.4% | (7.8%, 9.1%) | <0.001 | <0.001 |
Unadjusted mean intake among the eligible sample by pregnancy status.
P-value for pairwise difference for unadjusted mean intakes by pregnant vs. non-pregnant.
P-value for pairwise difference for mean intakes by pregnant vs. non-pregnant adjusted for age, race/ethnicity, family PIR, education, marital status, and survey year.
Intakes by Stratified Subgroups
Among pregnant women, there were no significant differences in adjusted mean intakes of added sugar as a percent of calories by any of the stratified subgroup levels compared to the reference categories. (Table 3) In multivariable analysis, there was significant effect modification of education and income level on the association of pregnancy status with added sugar intake (p=0.028 for pregnancy × income interaction term; and p<0.001 for pregnancy × education interaction term). (Figure 2) Specifically, among women with an educational level of a high school diploma or less, those who were pregnant consumed a lower percent of calories from added sugar compared to those who were non-pregnant (12.9% vs. 17.3%, p<0.05 for less than high school; 13.5% vs. 18.6%, p<0.001 for high school diploma or GED), but there was no difference in consumption among higher educated women who were pregnant versus non-pregnant. Similarly, among women in the lowest poverty bracket in this sample, those who were pregnant consumed a lower percent of calories from added sugar compared to those who were non-pregnant (13.9% vs. 17.3%, p<0.05). The interaction of pregnancy and age, race/ethnicity, or marital status were not significant (not shown).
Table 3.
Mean consumption of added sugar by sub-group among pregnant women in the United States (ages 20–39 years), NHANES 2003–2012.
Category | n | Mean Intake (% kcal) |
95% Confidence Interval |
P-value1 (Adjusted) |
---|---|---|---|---|
Age at Screening | ||||
20–24 | 194 | 15.4% | (13.6%, 17.2%) | ref |
25–29 | 225 | 14.3% | (12.2%, 16.3%) | 0.801 |
30–34 | 161 | 13.8% | (11.5%, 16.1%) | 0.702 |
35–39 | 70 | 12.9% | (10.4%, 15.5%) | 0.433 |
Race/Ethnicity | ||||
Non-Hispanic White | 285 | 15.8% | (13.8%, 17.8%) | ref |
Mexican American | 176 | 13.8% | (11.4%, 16.2%) | 0.460 |
Non-Hispanic Black | 113 | 14.2% | (11.8%, 16.6%) | 0.571 |
Other | 76 | 12.5% | (9.9%, 15.2%) | 0.207 |
Poverty Level | ||||
PIR < 100% | 171 | 14.8% | (12.5%, 17.0%) | 0.919 |
PIR 101–185% | 144 | 15.8% | (13.5%, 18.1%) | 0.444 |
PIR 186–350% | 135 | 12.1% | (9.9%, 14.3%) | 0.676 |
PIR > 350% | 200 | 13.7% | (11.4%, 15.9%) | ref |
Education Level | ||||
Less than High School | 171 | 12.6% | (9.9%, 15.3%) | 0.730 |
HS Diploma or GED | 130 | 13.6% | (11.6%, 15.7%) | 0.979 |
Some College | 192 | 15.9% | (14.0%, 17.8%) | 0.443 |
College Degree or Above | 157 | 14.2% | (12.4%, 16.0%) | ref |
Marital Status | ||||
Married | 428 | 13.2% | (11.4%, 15.0%) | ref |
Not married | 222 | 15.0% | (13.0%, 17.0%) | 0.195 |
Trimester | ||||
Missing | 51 | 13.6% | (11.7%, 15.6%) | 1.00 |
1st (1–3 Months) | 149 | 14. 5% | (12.4%, 16.6%) | 0.934 |
2nd (4–6 Months) | 226 | 14.8% | (13.7%, 16.9%) | 0.662 |
3rd (7–9 Months) | 224 | 13.5% | (11.4%, 15.6%) | ref |
Pre-pregnancy BMI2 | ||||
Missing | 14 | 14.1% | (10.1%, 18.1%) | 0.999 |
Underweight | 35 | 15.1% | (11.0%, 19.1%) | 0.939 |
Normal Weight | 324 | 13.5% | (12.5%, 14.5%) | ref |
Overweight | 154 | 14.3% | (12.2%, 16.4%) | 0.944 |
Obese | 123 | 13.7% | (11.9%, 15.4%) | 0.999 |
Survey Year2 | ||||
2003–2004 | 202 | 15.8% | (13.8%, 18.0%) | ref |
2005–2006 | 299 | 14.1% | (12.0%, 16.2%) | 0.58 |
2007–2008 | 49 | 13.5% | (11.4%, 15.7%) | 0.41 |
2009–2010 | 54 | 13.1% | (10.6%, 15.5%) | 0.31 |
2011–2012 | 46 | 13.9% | (11.7%, 16.1%) | 0.66 |
P-values for pairwise differences comparing the stratified subgroup level to the reference level in multivariable regression adjusting for other covariates (age, race/ethnicity, PIR, education, marital status, pre-pregnancy BMI, trimester, and survey year). Tukey-Kramer adjustments were applied to account for multiple comparisons
Figure 2.
Percent added sugar intake as a percentage of total energy intake (kilocalories) among pregnant and non-pregnant women in the United States (ages 20–39 years), NHANES 2003–2012, by educational level (A) and income level (B). Percent estimates are LS-mean intakes from multivariable models adjusted for race/ethnicity, age, marital status, and survey year. (*) indicates statistically significant difference at p<0.05 between pregnant and non-pregnant women per category.
Sources of Added Sugar Intake
Figure 3 summarizes the LS-means and 95% confidence intervals for the percentage of added sugar from the predefined food groupings relative to total added sugar intake for pregnant versus non-pregnant women. In pairwise t-tests comparing the percent contribution of each food grouping between pregnant and non-pregnant women, there were no statistically significant differences, but there were borderline trends suggesting that pregnant women consumed on average a higher percentage of added sugar from the juice drinks and smoothies (P=0.07), cereals and pasta (P=0.05), and milk drinks (P=0.05) food groupings compared to non-pregnant women. We also calculated per person added sugar intake from all sugar-sweetened beverages (SSBs) besides milk-based drinks (i.e., sodas, coffee and tea, flavored waters, and juice drinks and smoothies). In multivariable analysis adjusted for age, race/ethnicity, family PIR, education, marital status, and survey year, we again found that pregnant women consumed a similar percentage of added sugar from SSBs compared to non-pregnant women (39.2% vs. 41.4%of added sugar intake, respectively, p=0.33).
Figure 3.
Mean percent of added sugar intake from pre-defined food groups among pregnant and non-pregnant women in the United States (ages 20–39 years), NHANES 2003–2012. Definitions of the pre-defined food groups based on their food codes can be found in Table S1. Error bars represent 95% confidence intervals. Only the top ten contributors of added sugar intake among pregnant or non-pregnant group are shown above. Sodas and other sugar sweetened beverages (SSBs) were defined as all beverages containing added sugar except milk-based drinks.
Discussion
This study was the first to describe dietary intake patterns for added sugar specific to pregnant women in the United States. Importantly, the findings showed that both pregnant and non-pregnant women consume on average a percentage of calories from added sugar approximately 50% higher than the limit recommended in the 2015–2020 Dietary Guidelines for Americans (< 10% of calories per day) and more than three times the American Heart Association limit of < 100 calories from added sugar per day for women.16,26 However, contrary to our hypothesis and despite a greater but non-significant absolute intake in added sugars, we found that pregnant women consumed less added sugar as a percentage of total calories per day compared to non-pregnant women. This can be explained by the finding that pregnant women consume on average 300 kilocalories per day more than non-pregnant women, and suggests that these additional calories are coming from sources other than high-sugar foods.
While previous research by Ervin & Ogden demonstrated clear differences in added sugar intake patterns by demographic factors among non-pregnant women, specifically, higher intake among African American and low-income females27 we did not find these differences by subgroup among pregnant women. This is similar to the finding from an NHANES analysis by Gamba et al. that found that diet quality in pregnancy is not associated with household food security.28 Surprisingly, we also found no differences in added sugar intake by pre-pregnancy BMI. This conflicts with the results of a study by Shin et al., which that found that women with obese pre-pregnancy BMI had significantly lower Healthy Eating Index scores compared to those with normal pre-pregnancy BMI;29 however, while added sugar wasn’t examined specifically in this study, when investigated further it appears that the difference in diet quality was explained by other unhealthy dietary behaviors, such as the lower scores for fruit or sodium intake in females who had an obese pre-pregnancy BMI.29 In the present study, we also found no significant differences in added sugar intake by trimester of pregnancy, suggesting that – despite increased energy needs in the third trimester of pregnancy – pregnant women are not meeting these increased needs through foods that are high in added sugar.
Our study did find a significant interaction between pregnancy status (pregnant versus non-pregnant) and both educational level and income level, defined by family PIR. Women with a high-school education or less and women in the lowest income category (< 100% PIR) reportedly consumed a lower percentage of calories from added sugar if pregnant, compared to if non-pregnant women, but there was no difference in added sugar percentage between pregnant and non-pregnant females in higher educated and higher income categories. Although the reason for these differences remain unclear, it is an overall positive finding indicating that women in low-resource settings may be exhibiting more healthful dietary choices during pregnancy than their non-pregnant counterparts. While further research is needed to understand these differences, they could be the result of low income women’s access to nutrition counseling and support services through federal social programs such as the Supplemental Program for Women, Infant and Children (WIC).
Finally, examining the relative contribution of different food and beverage sources of added sugar for pregnant versus non-pregnant women, we found that the largest percentage of added sugar intake came from sugar-sweetened beverages for both pregnant and non-pregnant women, though this percentage was slightly lower among pregnant women. We also found borderline significant trends suggesting that pregnant women consumed a higher proportion of added sugar from fruit-based drinks/smoothies and milk drinks, behaviors which may reflect acknowledgement of the need for adequate intake of key nutrients, including calcium, during pregnancy, though this cannot be confirmed from the data available.
There are several limitations of this study that should be recognized. Because NHANES is a cross-sectional survey, we cannot assess temporality or causality in any of the associations found. Females younger than 20 or older than 40 years were not included due to limited access to the data or limited sample size, respectively, therefore the results cannot be generalized to women outside the United States or this age range. Our analysis used only one 24-hour recall, which may not accurately reflect the usual intake for each individual but provides a valid means for estimating intake of a group. Our finding also relied on self-reported dietary data, which may be subject to recall bias or systematic underreporting, particularly by those who are overweight or obese.30 A related weakness was that we were unable to account for pregnancy-related nausea and vomiting that typically occurs in the first trimester; thus, for the this subgroup it is possible that self-reported dietary intake was an overestimate of actual, digested intake. Additionally, although the 2003–2004 and 2005–2006 cycles oversampled for pregnant women, the subsequent cycles did not, resulting in a small number of pregnant women (< 60) per cycle, and only 650 overall, which was small relative to non-pregnant women. This may have limited the precision and/or reliability of our estimates, as well as our ability to detect significant differences in stratified analyses among pregnant women.
Despite these limitations, the study has several strengths. It is the first known study to describe added sugar intake among pregnant women in the United States. By using the continuous NHANES dataset, we were able to estimate added sugar consumption in a nationally representative sample of pregnant and non-pregnant women of reproductive age in the United States. Although 24-hour recalls have inherent biases, as described above, the dietary recalls in NHANES were performed by trained interviewers utilizing the USDA’s automated multiple pass method, which has been validated to ensure better accuracy of dietary assessment. Finally, by examining dietary intake stratified by important maternal sociodemographic characteristics, as well as by food and beverage groupings, our findings provide actionable insights into added sugar trends among women of reproductive age.
Conclusions
From this cross-sectional analysis we found that pregnant women, similar to non-pregnant women of the same age, consume added sugar far above recommended limits. This has important implications for public health efforts focused on helping women achieve a healthy diet during pregnancy. Our findings support the need for additional efforts toward educating women of reproductive age, both prior to and during pregnancy, to reduce foods and beverages high in added sugar, while promoting more nutrient-dense options. Further research is needed to fully understand the impact of a diet high in added sugar on pregnancy and birth outcomes.
Supplementary Material
Research Snapshot.
Research Question
How much added sugar do pregnant women in the United States consume and how does their consumption compare to non-pregnant women?
Key findings
Using cross-sectional data from 4,179 women participants in the National Health and Nutrition Examination Survey (NHANES), we found a trend toward pregnant women consuming more added sugar in grams per day compared to non-pregnant women. However, this equated to a lower percentage of total calories when offset by the higher total energy intakes among pregnant women (P<0.05). By subgroup, the contribution of added sugars to total energy intake was less among less educated and lower income pregnant women compared to their non-pregnant counterparts. Nonetheless, mean intakes of added sugar for both pregnant and non-pregnant women exceeded recommended levels of less than 10% of total energy intake.
Acknowledgments
Funding/Financial Reports: CEC was funded by an institutional T32 training grant (NIH-DK007734) when the research was conducted.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Author Contributions: JAW contributed the research idea. CEC combined and analyzed the data. JF reviewed the data analysis. CEC wrote the first draft of the manuscript. JAW and JF reviewed and provided revisions for the manuscript. All authors read and commented on subsequent drafts of the manuscript.
Conflict of Interest Disclosure: no conflicts of interest
Contributor Information
Catherine E. Cioffi, Laney Graduate School, Emory University, Atlanta, GA, 30322.
Janet Figueroa, Children’s Healthcare of Atlanta, Atlanta, GA, 30322; Emory University School of Medicine, Atlanta, GA, 30322.
Jean A. Welsh, Laney Graduate School, Emory University, Atlanta, GA, 30322; Children’s Healthcare of Atlanta, Atlanta, GA, 30322; Department of Pediatrics, Emory University School of Medicine, Atlanta, GA, 30322.
References
- 1.Black RE, Victora CG, Walker SP, et al. Maternal and child undernutrition and overweight in low-income and middle-income countries. The Lancet. 2013;382(9890):427–451. doi: 10.1016/S0140-6736(13)60937-X. [DOI] [PubMed] [Google Scholar]
- 2.Mamun AA, Mannan M, Doi SA. Gestational weight gain in relation to offspring obesity over the life course: a systematic review and bias-adjusted meta-analysis. Obes Rev. 2014;15(4):338–347. doi: 10.1111/obr.12132. [DOI] [PubMed] [Google Scholar]
- 3.Hedderson MM, Gunderson EP, Ferrara A. Gestational weight gain and risk of gestational diabetes mellitus. Obstet Gynecol. 2010;115(3):597–604. doi: 10.1097/AOG.0b013e3181cfce4f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Catalano PM. Management of obesity in pregnancy. Obstet Gynecol. 2007;109(2 Pt 1):419–433. doi: 10.1097/01.AOG.0000253311.44696.85. [DOI] [PubMed] [Google Scholar]
- 5.Herring SJ, Oken E, Rifas-Shiman SL, et al. Weight gain in pregnancy and risk of maternal hyperglycemia. Am J Obstet Gynecol. 2009;201(1):61.e61–61.e67. doi: 10.1016/j.ajog.2009.01.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Denison FC, Norwood P, Bhattacharya S, et al. Association between maternal body mass index during pregnancy, short-term morbidity, and increased health service costs: a population-based study. BJOG. 2014;121(1):72–81. doi: 10.1111/1471-0528.12443. Commentary by JP Lavin Jr 82. [DOI] [PubMed] [Google Scholar]
- 7.Block SR, Watkins SM, Salemi JL, et al. Maternal pre-pregnancy body mass index and risk of selected birth defects: evidence of a dose-response relationship. Paediatr Perinat Epidemiol. 2013;27(6):521–531. doi: 10.1111/ppe.12084. [DOI] [PubMed] [Google Scholar]
- 8.Tyrrell J, Richmond RC, Palmer TM, et al. Genetic evidence for causal relationships between maternal obesity-related traits and birth weight. JAMA. 2016;315(11):1129–1140. doi: 10.1001/jama.2016.1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Aviram A, Hod M, Yogev Y. Maternal obesity: Implications for pregnancy outcome and long-term risks-a link to maternal nutrition. Int J Gynaecol Obstet. 2011;115(Suppl 1):S6–10. doi: 10.1016/S0020-7292(11)60004-0. [DOI] [PubMed] [Google Scholar]
- 10.O’Reilly JR, Reynolds RM. The risk of maternal obesity to the long-term health of the offspring. Clin Endocrinol (Oxf) 2013;78(1):9–16. doi: 10.1111/cen.12055. [DOI] [PubMed] [Google Scholar]
- 11.Barbour LA. Changing perspectives in pre-existing diabetes and obesity in pregnancy: Maternal and infant short- and long-term outcomes. Curr Opin Endocrinol Diabetes Obes. 2014;21(4):257–263. doi: 10.1097/MED.0000000000000079. [DOI] [PubMed] [Google Scholar]
- 12.Smith CJ, Ryckman KK. Epigenetic and developmental influences on the risk of obesity, diabetes, and metabolic syndrome. Diabetes Metab Syndr Obes. 2015;8:295–302. doi: 10.2147/DMSO.S61296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Vartanian LR, Schwartz MB, Brownell KD. Effects of soft drink consumption on nutrition and health: a systematic review and meta-analysis. Am J Public Health. 2007;97(4):667–675. doi: 10.2105/AJPH.2005.083782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Te Morenga L, Mallard S, Mann J. Dietary sugars and body weight: systematic review and meta-analyses of randomised controlled trials and cohort studies. BMJ. 2012;346:e7492. doi: 10.1136/bmj.e7492. [DOI] [PubMed] [Google Scholar]
- 15.Popkin BM, Hawkes C. Sweetening of the global diet, particularly beverages: patterns, trends, and policy responses. Lancet Diabetes Endocrinol. 2016;4(2):174–186. doi: 10.1016/S2213-8587(15)00419-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.U.S. Department of Health and Human Services and U.S. Department of Agriculture. Dietary Guidelines for Americans 2015-2020. (8th) 2015 Dec; Available at: https://health.gov/dietaryguidelines/2015/. Accessed October 18, 2017.
- 17.Procter SB, Campbell CG. Position of the Academy of Nutrition and Dietetics: Nutrition and lifestyle for a healthy pregnancy outcome. J Acad Nutr Diet. 2014;114(7):1099–1103. doi: 10.1016/j.jand.2014.05.005. [DOI] [PubMed] [Google Scholar]
- 18.Renault KM, Carlsen EM, Norgaard K, et al. Intake of carbohydrates during pregnancy in obese women is associated with fat mass in the newborn offspring. Am J Clin Nutr. 2015;102(6):1475–1481. doi: 10.3945/ajcn.115.110551. [DOI] [PubMed] [Google Scholar]
- 19.Renault KM, Carlsen EM, Nørgaard K, et al. Intake of sweets, snacks and soft drinks predicts weight gain in obese pregnant women: Detailed analysis of the results of a randomised controlled trial. PLoS ONE. 2015;10(7) doi: 10.1371/journal.pone.0133041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Maslova E, Halldorsson TI, Astrup A, Olsen SF. Dietary protein-to-carbohydrate ratio and added sugar as determinants of excessive gestational weight gain: a prospective cohort study. BMJ. 2015;5(2) doi: 10.1136/bmjopen-2014-005839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shin D, Lee KW, Song WO. Dietary patterns during pregnancy are associated with gestational weight gain. Matern Child Health J. 2016;20(12):2527–2538. doi: 10.1007/s10995-016-2078-x. [DOI] [PubMed] [Google Scholar]
- 22.Herring SJ, Henry TQ, Klotz AA, Foster GD, Whitaker RC. Perceptions of low-income African-American mothers about excessive gestational weight gain. Matern Child Health J. 2012;16(9):1837–1843. doi: 10.1007/s10995-011-0930-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Reyes NR, Klotz AA, Herring SJ. A qualitative study of motivators and barriers to healthy eating in pregnancy for low-income, overweight, African-American mothers. J Acad Nutr Diet. 2013;113(9):1175–1181. doi: 10.1016/j.jand.2013.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Centers for Disease Control and Prevention (CDC) National Health and Nutrition Examination Survey. Available at: http://www.cdc.gov/nchs/nhanes/about_nhanes.htm. Last updated September 15, 2017. Accessed October 18, 2017.
- 25.Dodd KW, Guenther PM, Freedman LS, et al. Statistical methods for estimating usual intake of nutrients and foods: a review of the theory. J Am Diet Assoc. 2006;106(10):1640–1650. doi: 10.1016/j.jada.2006.07.011. [DOI] [PubMed] [Google Scholar]
- 26.Johnson RK, Appel LJ, Brands M, et al. Dietary sugars intake and cardiovascular health a scientific statement from the American Heart Association. Circulation. 2009;120(11):1011–1020. doi: 10.1161/CIRCULATIONAHA.109.192627. [DOI] [PubMed] [Google Scholar]
- 27.Ervin RB, Ogden CL. Consumption of added sugars among U.S. adults, 2005-2010. NCHS Data Brief. 2013;(122) [PubMed] [Google Scholar]
- 28.Gamba R, Leung CW, Guendelman S, Lahiff M, Laraia BA. Household food insecurity is not associated with overall diet quality among pregnant women in NHANES 1999–2008. Matern Child Health J. 2016;20(11):2348–2356. doi: 10.1007/s10995-016-2058-1. [DOI] [PubMed] [Google Scholar]
- 29.Shin D, Lee KW, Song WO. Pre-Pregnancy weight status is associated with diet quality and nutritional biomarkers during pregnancy. Nutrients. 2016;8(3) doi: 10.3390/nu8030162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Heitmann BL, Lissner L. Dietary underreporting by obese individuals—is it specific or non-specific? BMJ. 1995;311(7011):986–989. doi: 10.1136/bmj.311.7011.986. [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.