Maternal Risk of Cardiovascular Disease Is Associated With Higher BMI Among Preschool-Aged Children: A Cross-Sectional Study

Abstract

Childhood obesity is an ongoing concern in the United States. Higher weight status in early childhood is associated with higher weight status at older ages. The Maternal Obesity Matters (MOMs) Study investigated associations between maternal risk of cardiovascular disease (CVD) and child BMI z-scores (BMIz) among preschool-aged children. This cross-sectional exploratory study included mothers and their 3- to 5-year-old children in Colorado, United States. Maternal nonfasting blood samples, blood pressure, and maternal and child anthropometrics were collected. Maternal CVD risk was assessed on a scale of 0–5 using five health measures. Multivariate regression tested associations between maternal CVD risk and child BMIz. Each 1-point increase in maternal CVD risk was associated with a 0.18 increase in child BMIz, controlling for maternal employment. Intervening upon maternal health may be an important strategy for addressing childhood obesity.

Introduction

Childhood obesity remains a major concern in the United States because of its continued high prevalence and associations with later health status.¹ Preschool-aged children living in rural areas may tend to have higher BMI z-scores (BMIz),² although there is relatively little information on children living in rural communities. Higher BMI at age 3 years is a strong predictor of BMI at age 14 years,³ and higher BMI in children has been associated with increased risk of hypertension.⁴

Child and parental health are also closely related. For example, among families with children 5–7 years old, child and parent weight status, physical activity levels, and fruit and vegetable intake were found to be similar.⁵ Because similar household factors and exposures may affect both parental and child weight status and health, this study investigated whether current maternal health—specifically risk of cardiovascular disease (CVD)—was associated with child weight status among preschool-aged children.

Materials and Methods

The Maternal Obesity Matters (MOMs) Study was a cross-sectional exploratory study. Mothers of children aged 3–5 years were recruited from 16 Colorado preschools and Head Start centers across two studies between 2015–2016 and 2019–2020. Inclusion criteria were (1) mothers <50 years old, (2) not having an illness or condition that limited eating or physical activity in the mother or child, and (3) attesting to being comfortable reading and completing forms in English. Data were collected in person at the child's preschool in a private room.

Study staff collected maternal blood samples, blood pressure, and maternal and child anthropometrics. This research project was approved by the Colorado State University Institutional Review Board. Mothers provided written consent, and children provided assent for collection of anthropometric measures.

Measures

Demographics

Mothers (N = 119) completed demographic surveys that included questions about maternal and child age, race, ethnicity, maternal education, employment, household income, number of children in the household, and child gender.

Biomarkers, blood pressure, and anthropometrics

A full description of methods and equipment is available in Supplementary Data S1. Nonfasting maternal fingerstick samples were collected and processed using point-of-care units to assess HbA_1c, total cholesterol, HDL-C, non-HDL-C, LDL-C, triglycerides, and glucose. Measures were excluded from analyses if total cholesterol <100 mg/dL and triglycerides <45 mg/dL.

Maternal blood pressure, height, weight, and waist circumference were collected by trained study staff using National Health and Nutrition Examination Survey techniques.^6,7 Maternal blood pressure was collected in triplicate. Maternal and child height and maternal waist circumference were collected in duplicate to the nearest 0.1 cm. A single weight measurement was collected to the nearest 0.1 pound and converted to kg to calculate BMI (kg/m²) and child BMI-for-age z-scores (BMIz; EpiInfo software, v.7 CDC, Atlanta, GA).⁸ Means for each measure were calculated and used in statistical analyses.

Maternal CVD risk

CVD risk was calculated as a sum score (0–5) based on the presence of up to five nonfasting CVD risk factors: HDL-C < 50 mg/dL, triglycerides ≥200 mg/dL, HbA_1c ≥ 5.6%, blood pressure ≥130/85 mmHg, and waist circumference ≥35 inches (88 cm).^9,10 Because nonfasting measures were collected, the cut point for triglycerides was revised from ≥150 to ≥200 mg/dL,¹¹ and fasting blood glucose ≥100 mg/dL was substituted with HbA_1c ≥ 5.6%, the screening cut point recommended by the American Diabetes Association for prediabetes.¹²

Data Management and Analysis

Study data were collected and managed through Research Electronic Data Capture, a Health Insurance Portability and Accountability Act of 1996 (HIPAA)-compliant web-based research database hosted by the Colorado Clinical & Translational Sciences Institute.^13,14 Continuous data were inspected for normal distribution using the Shapiro–Wilk test and histograms. Descriptive statistics were computed for participant characteristics, anthropometric measurements, and maternal CVD risk factors. t-Tests, one-way analysis of variance, and Pearson's correlations tested associations between child BMIz and maternal CVD risk, maternal and child characteristics (Supplementary Table S1).

Multivariate linear regression models tested for associations between maternal CVD risk scores and child BMIz. Covariate effects were tested using forward selection to identify the most parsimonious model. Covariates were added based on the strength of bivariate associations with child BMIz: maternal employment (employed vs. stayed at home), child ethnicity (Hispanic vs. non-Hispanic), maternal education (high school education or less, some college or associate's degree, 4-year degree or higher), household income status (low income or higher), and number of children in the household. Model fit statistics (F values, log likelihood) were compared to select the final model.

Analyses were performed using Stata (v 16; StataCorp, College Station, TX). Significance was set at p < 0.05. Mothers who had one or more missing or excluded risk factor values were excluded from final analyses (N = 6 with missing values; N = 13 excluded values; Supplementary Fig. S1).

Results

Of the 119 participants, 117 dyads included mothers who were <50 years old (Table 1). Mothers had a mean ± SD age of 32.4 ± 6.1 years, 34% had a high school education or less, and 50% identified as Hispanic. Children had a mean age of 4.5 ± 0.6 years, 51% were female, and 53% were identified as Hispanic. Most families had limited resources: 73% had an annual income <185% of the US Federal Poverty Line.¹⁵

Table 1.

Sociodemographic and Health Characteristics of Mothers and Caregivers in the Maternal Obesity Matters Study (N = 117)

	Mean ± SD or N (%)
Maternal characteristics
Age (years)	32.4 ± 6.1
Identified as Hispanic	59 (50)
Employed outside the home	72 (62)
Level of education
High school or less	39 (34)
Some college, associate's degree	48 (41)
4-Year degree or higher	29 (25)
Low-income household (≤185% FPL)	86 (73)
Very low-income household (≤100% FPL)	51 (44)
No. of children in household	2.5 ± 1.1
BMI	29.9 ± 6.8
HDL <50 mg/dL	57 (49)
Triglycerides >200 mg/dL	29 (25)
HbA_1c > 5.6%	22 (19)
Blood pressure ≥130/85 mmHg	7 (6)
Waist circumference ≥88 cm	88 (77)
CVD risk score	1.87 ± 1.2
Child characteristics
Age (years)	4.5 ± 0.6
Female	60 (51)
Hispanic	62 (53)
BMI z-score	0.5 ± 1.0
BMI 85th to <95th percentile	29 (25)
BMI ≥95th percentile	13 (11)

CVD, cardiovascular disease; FPL, federal poverty level¹⁵; HbA_1c, hemoglobin A_1C; HDL, high-density lipoprotein.

Few bivariate associations approached significance (Supplementary Table S1), yet there were substantial differences between absolute values: child BMIz differed by maternal employment (BMIz Mean ± SD, 0.37 ± 1.10 employed vs. 0.73 ± 0.93 not employed, p = 0.07), child ethnicity (0.66 ± 1.00 for Hispanic children vs. 0.34 ± 1.09 for non-Hispanic children, p = 0.10), and maternal ethnicity (0.65 ± 1.02 for children of Hispanic mothers vs. 0.36 ± 1.07 for children of non-Hispanic mothers, p = 0.13).

The final model (N = 98) adjusted for maternal employment (Table 2). Child ethnicity, maternal education, and household income status were tested because of their theoretical links to child BMIz based on previous research.^16,17 Inclusion of these variables did not improve model fit and were excluded from the final model. In the final model, higher maternal CVD risk score was associated with higher child BMIz (β = 0.18, SE = 0.09, p = 0.04, 95% CI 0.01–0.35), and maternal employment for wages was associated with lower child BMIz (β = −0.43, SE = 0.21, p = 0.047, 95% CI −0.85 to −0.01). The model accounted for 7% of the variance in child BMIz (F [3, 94] = 4.7, p = 0.01).

Table 2.

Model Selection Using Likelihood Ratio Tests

Outcome: child BMI z-score	Model 1	Model 2	Model 3a	Model 3b	Model 3c	Model 3d	Model 3e
Maternal CVD risk score	0.20 (0.09) ^a [0.02, 0.37]	0.18 (0.08) ^a [0.01, 0.35]	0.15 (0.09)[−0.02, 0.33]	0.17 (0.09)[0.004, 0.34]	0.18 (0.09)^a[0.01, 0.36]	0.19 (0.09)^a[0.02, 0.36]	0.17 (0.09)[−0.01, 0.35]
Mother employed		−0.43 (0.21) ^a [−0.85, −0.01]	−0.43 (0.21)^a[−0.85, −0.01]	−0.34 (0.25)[−0.78, 0.10]	−0.43 (0.21)^a[−0.86, −0.003]	−0.49 (0.22)^a[−0.92, −0.06]	−0.41 (0.23)[−0.86, 0.04]
Hispanic child			0.29 (0.21)[−0.13, 0.70]				0.16 (0.24)[−0.30, 0.63]
Maternal education
High school or less				(ref)			(ref)
Some college/associate's degree				−0.18 (0.25)[−0.67, 0.31]			−0.15 (0.26)[−0.66, 0.36]
4-Year degree or higher				−0.44 (0.28)[−0.99, 0.11]			−0.43 (0.32)[−1.07, 0.21]
Low-income household^b					−0.07 (0.23)[−0.52, 0.39]		−0.21 (0.25)[−0.71, 0.28]
No. of children in the household						−0.14 (0.10)[−0.33, 0.05]	−0.13 (0.10)[−0.32, 0.07]
F-statistic, p-value	5.17, 0.03	4.69, 0.01	3.78, 0.01	2.87, 0.03	3.12, 0.03	3.84, 0.01	2.09, 0.05
Adjusted R²	0.04	0.07	0.08	0.07	0.06	0.08	0.07
Log likelihood	−139.77	−137.73	−136.76	−135.33	−137.68	−136.68	−133.65
Degrees of freedom	2	3	4	5	4	4	8
−2L (compared with model n − 1)		4.08	1.94	4.80	0.10	2.10	8.16
Likelihood ratio, p		<0.05	>0.10	>0.05	>0.10	>0.10	>0.10

p < 0.05.

Low-income household defined as income <185% of the Federal Poverty Line for the year in which data were collected.²⁵

Discussion

This study found that current CVD risk in mothers was associated with higher BMIz in preschool-aged children in low-income rural communities in Colorado, United States. The observed effect was cumulative: each 1-point increase in CVD risk was associated with a 0.18 increase in BMIz. These findings are consistent with a 2012 study that observed higher rates of adult obesity and diabetes were associated with higher rates of childhood obesity among rural populations.¹⁸ Because the present data are cross-sectional, it is not clear what the implications may be for children's later metabolic risk.

However, higher BMI at age 3 years is a strong predictor of higher BMI in adolescence,³ and both higher parental BMI and child baseline BMI (at age 3–5 years) have been associated with crossing the 95th BMI percentile (a marker for obesity) over a 36-month period.¹⁹ Furthermore, child body size is associated with a variety of factors, including genetic, gestational, feeding, and environmental, and the small but significant observed variance suggests that understanding and intervening on factors that drive the health of parents living in rural communities may also be important for understanding and improving children's health outcomes.

Even when lifestyle interventions do not yield short-term changes in BMI, interventions often demonstrate significant changes in target behaviors, which may lead to longer term health gains.²⁰ By targeting the whole family in an intervention, it may be possible to improve the health of multiple family members.

Maternal employment was associated with lower child BMIz, which differs from other study findings.^21,22 It is possible that, compared with their peers, participating children whose mothers were employed spent more time in organized care settings and had access to different nutrition and physical activity opportunities, which may have had an impact on their BMIz. Surprisingly, lower household income was not associated with child BMIz. Other researchers in the United States have identified an association: BMI was less differentiated between children from low- and middle-income households than between children from low- and high-income households, suggesting a proportional effect of increased income.²³

It is possible that the present study failed to detect differences by household income because of the high proportion of low-income households and limited sample size. However, the high proportion of dyads from low-income households may also help explain why so many women exhibited multiple CVD risk factors (n = 59 or 60% of mothers exhibited ≥2 CVD risk factors), as lower household income has previously been associated with higher CVD risk.²⁴

Despite its contributions, this study has some limitations, including the cross-sectional design. Therefore, it is unclear how maternal CVD risk may be associated with child growth trajectories in this population. The small sample size limited our ability to test many covariates within a single model. Although there were no significant differences between maternal, household, and child characteristics, a greater proportion of participants in the final sample had high waist circumference compared with participants who were excluded (Supplementary Table S2), which could have biased CVD risk scores upward.

Medications taken were not collected for all participants and, therefore, could not be included in final analyses. In addition, because nonfasting blood samples were collected, alternative cut points for CVD risk factors had to be considered, although some argue that nonfasting samples may be important to examine.²⁵ Finally, participants were recruited from local preschools, and families who do not attend a preschool-based center may differ from the participating families. All participants included in the final sample were biologically related, so the study does not represent diversity in family structures.

This study does have its strengths. Multiple noninvasive biomarkers were collected in a rural community setting, and a composite measure assessed current maternal CVD risk. These maternal CVD risk scores were compared with concurrent child BMIz in a diverse group of mothers and preschool-aged children.

Conclusion

Higher maternal CVD risk was associated with higher preschooler BMI z-scores. This association accounted for significant, though limited, variance and is only one of many environmental and genetic risks associated with child obesity. The findings suggest that it may be important to study parental health and household- or community-level factors that may affect both parent and child health, which can in turn inform future childhood obesity interventions.

Impact Statement

Maternal and child health are closely linked during the preschool years: mothers who exhibited higher cardiovascular disease risk tended to have children with higher BMI z-scores. Interventions to improve maternal health may also help to improve the health of their children.

Footnotes

Acknowledgments

The authors thank the participants as well as Head Start and preschool staff at the 16 centers who worked with us on recruitment and provided on-site space and support to conduct this study.

Authors' Contributions

K.J.B. contributed methodology, formal analyses, writing—original draft. S.H.-S. was involved in conceptualization, methodology, investigation, and writing—review and editing. S.H. carried ou conceptualization, methodology, investigation, and writing—review and editing. R.E.B. was involved in methodology and writing—review and editing. T.L.N. was in charge of methodology and writing—review and editing. S.L.J. took care of conceptualization, funding acquisition, methodology, supervision, aand writing—review and editing. L.L.B. took care of conceptualization, funding acquisition, methodology, supervision, and writing—review and editing.

Funding Information

This protocol was jointly funded by the Colorado Agricultural Experiment Station (USDA Hatch funds) Grant Number COL00640, the Colorado State University School of Public Health Faculty Seed Grant, and the Agriculture and Food Research Initiative Grant number 2015-68001-23240 from the US Department of Agriculture, National Institute of Food and Agriculture. The data management portion of this project was supported by NIH/NCRR Colorado CTSI, Grant Number UL1 RR025780.

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

References

Ogden

, Fryar

, Martin

, et al. Trends in obesity prevalence by race and hispanic origin—1999–2000 to 2017–2018. JAMA, 2020; 324(12):1208–1210.

Contreras

, Martoccio

, Brophy-Herb

, et al. Rural-urban differences in body mass index and obesity-related behaviors among low-income preschoolers. J Public Health (Oxf), 2021; 43(4):e637–e644.

Freedman

, Goodwin-Davies

, Kompaniyets

, et al. Interrelationships among age at adiposity rebound, BMI during childhood, and BMI after age 14years in an electronic health record database. Obesity (Silver Spring, Md), 2022; 30(1):201–208.

Giussani

, Antolini

, Brambilla

, et al. Cardiovascular risk assessment in children: role of physical activity, family history and parental smoking on BMI and blood pressure. J Hypertens, 2013; 31(5):983–992.

Coto

, Pulgaron

, Graziano

, et al. Parents as role models: Associations between parent and young children's weight, dietary intake, and physical activity in a minority sample. Matern Child Health J, 2019; 23(7):943–950.

Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey (NHANES): Health Tech/Blood Pressure Procedures Manual. U.S. Department of Health and Human Services: Atlanta, GA, 2009.

Centers for Disease Control and Prevention. National Health and Examination Survey (NHANES): Anthropometry Procedures Manual. Centers for Disease Control and Prevention: Atlanta, GA, USA; 2017.

National Institutes of Health. Clinical guidelines for the identification, evaluation, and treatment of overweight and obesity in adults-the evidence report. Obes Res, 1998; 6(2):51S–209S.

Alberti

KGMM

, Eckel

, Grundy

, et al. Harmonizing the metabolic syndrome. Circulation, 2009; 120(16):1640–1645.

10.

Driver

, Martin

, Gluckman

, et al. Fasting or nonfasting lipid measurements. J Am Coll Cardiol, 2016; 67(10):1227–1234.

11.

Miller

, Stone

, Ballantyne

, et al. Triglycerides and Cardiovascular Disease. Circulation, 2011; 123(20):2292–2333.

12.

Higgins

. HbA1c for screening and diagnosis of diabetes mellitus. Endocrine, 2013; 43(2):266–273.

13.

Harris

, Taylor

, Thielke

, et al. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inf, 2009; 42(2):377–381.

14.

Harris

, Taylor

, Minor

, et al. The REDCap consortium: Building an international community of software platform partners. J Biomed Inf, 2019; 95:103208.

15.

U.S. Department of Health and Human Services Office of the Assistant Secretary for Planning and Evaluation. Prior HHS Poverty Guidelines and Federal Register References 2022. Available from: https://aspe.hhs.gov/topics/poverty-economic-mobility/poverty-guidelines/prior-hhs-poverty-guidelines-federal-register-references [Last accessed: January 2, 2023].

16.

Lê-Scherban

, Moore

, Headen

, et al. Are there birth cohort effects in disparities in child obesity by maternal education?. Int J Obes, 2021; 45(3):599–608.

17.

Ogden

, Carroll

, Fakhouri

, et al. Prevalence of obesity among youths by household income and education level of head of household—United States 2011–2014. Morb Mortal Wkly Rep, 2018; 67(6):186.

18.

Salois

. The built environment and obesity among low-income preschool children. Health Place, 2012; 18(3):520–527.

19.

Heerman

, Sommer

, Slaughter

, et al. Predicting early emergence of childhood obesity in underserved preschoolers. J Pediatr, 2019; 213:115–120.

20.

Kamath

, Vickers

, Ehrlich

, et al. Behavioral interventions to prevent childhood obesity: A systematic review and metaanalyses of randomized trials. J Clin Endocrinol Metab, 2008; 93(12):4606–4615.

21.

Speirs

, Liechty

, Wu

. Sleep, but not other daily routines, mediates the association between maternal employment and BMI for preschool children. Sleep Med, 2014; 15(12):1590–1593.

22.

Fitzsimons

, Pongiglione

. The impact of maternal employment on children's weight: Evidence from the UK. SSM Popul Health, 2019; 7:100333.

23.

Weaver

, Beets

, Brazendale

, et al. Disparities by household income and race/ethnicity: The utility of BMI for surveilling excess adiposity in children. Ethn Health, 2021; 26(8):1180–1195.

24.

Kibria

GMA

, Crispen

, Chowdhury

MAB

, et al. Disparities in absolute cardiovascular risk, metabolic syndrome, hypertension, and other risk factors by income within racial/ethnic groups among middle-aged and older US people. J Hum Hypertens, 2021. Available from: https://www-nature-com-s.web.bisu.edu.cn/articles/s41371-021-00513-8 [Last accessed: February 22, 2023].

25.

Langsted

, Nordestgaard

. Nonfasting versus fasting lipid profile for cardiovascular risk prediction. Pathology, 2019; 51(2):131–141.

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

0.02 MB