- Email: [email protected]
Sensitive Timing for Risk of Overweight Among Infants of Low-Income Hispanic Immigrants
ARTICLE
Sensitive Timing for Risk of Overweight Among Infants of Low-Income Hispanic Immigrants Kathleen F. Gaffney, PhD, RN, F/PNP-BC, Albert V. Brito, MD, Deborah A. Kermer, MA, & Panagiota Kitsantas, PhD
ABSTRACT Introduction: This study examined the association between early weight gain (0–6 months) and risk for overweight, as defined by the Institute of Medicine, at 1 year among infants of low-income Hispanic immigrant mothers. Method: Weight-for-age data were extracted from electronic medical records of 335 infants with gestations of 37 weeks or longer and birthweights appropriate for gestational age and without medical problems likely to interfere with growth or feeding. Logistic regression models were constructed to examine the impact of early weight changes on weight status at 1 year.
Kathleen F. Gaffney, Professor, School of Nursing, George Mason University, Fairfax, VA. Albert V. Brito, Medical Director, Inova Cares Clinic for Children, Falls Church, VA, and Assistant Professor, School of Medicine, Virginia Commonwealth University, Richmond, VA. Deborah A. Kermer, Data Services Research Consultant, University Libraries Data Services, George Mason University, Fairfax, VA. Panagiota Kitsantas, Professor, Health Administration and Policy Department, George Mason University, Fairfax, VA. Conflicts of interest: None to report. Correspondence: Kathleen F. Gaffney, PhD, RN, F/PNP-BC, School of Nursing, 3C4, George Mason University, Fairfax, VA 22030; e-mail: [email protected]. 0891-5245/$36.00 Copyright Q 2017 by the National Association of Pediatric Nurse Practitioners. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pedhc.2017.08.001
www.jpedhc.org
Results: By 12 months, 36.7% of infants had crossed weightfor-age of 84.1% or greater on World Health Organization growth charts. In adjusted models, infants had 20.8 (95% confidence interval = [19.8, 44.0]) times the odds of reaching this benchmark at 1 year for each z score increase at 0 to 6 months. Discussion: The study highlights a time-sensitive opportunity for interventions to reduce risk for overweight for this vulnerable population. J Pediatr Health Care. (2017) -, ---.
KEY WORDS Infant weight, Hispanic, immigrant
A recent population-based, national U.S. survey of childhood obesity found that among 2- to 5-year-olds, Hispanic children have higher rates of obesity and extreme obesity than any other racial/ethnic group (Ogden et al., 2016). This finding has important health policy implications, because current estimates indicate that 1 in 4 of our nation’s young children is Hispanic, with more than half of these children having at least one parent who is an immigrant to the United States (Murphey, Guzman, & Torres, 2014). The survey findings also have substantial clinical implications based on the well-established evidence that childhood obesity leads to elevated blood pressure, diabetes, musculoskeletal disorders, and other cumulative effects that may extend from childhood into adolescence and adulthood. A report of the World Health Organization (2017) pointed out that the risk for these morbidities depends partly on the age of onset and the duration of the overweight or obesity. The national survey findings, however, did not address weight status differences for children younger than 2 years of age. -/- 2017
1
This is an important omission, because recent evidence suggests that the underpinnings of childhood obesity may begin early in the life cycle. One retrospective study (Harrington et al., 2010) of overweight and obese older children found that half had been overweight by 2 years of age and 90% had become overweight by age 5 years. Further examination of the children’s weight trajectories led these researchers to conclude that the critical period for preventing childhood obesity may have been during the first 2 years of life and that for many the tipping point may have occurred between birth and 3 months of age. This latter finding is consistent with the research by Taveras et al. (2011), who found that obesity prevalence at ages 5 and 10 years of age was highest among children who crossed two or more major growth percentiles in the first 6 months of life. To date, studies of the relationship between weight gain in early infancy and later obesity have been limited by an underrepresentation of Hispanic children (Dennison, Edmunds, Stratton, & Pruzek, 2006; Goodell, Wakefield, & Ferris, 2009; Taveras et al., recent evidence 2009). Therefore, the purpose of this study suggests that the was to examine the underpinnings of association between childhood obesity early weight gain (0–6 months) and risk may begin early in for overweight, as the life cycle. defined by the Institute of Medicine (IOM; 2011), at 1 year of age among infants of low-income Hispanic immigrant mothers.
METHODS Sample and Setting Data were extracted from the medical records of 335 infants at a large pediatric, primary care setting in Virginia that serves low-income families. After approval of health care site and university institutional review boards, demographic and growth data were extracted for maternal–infant dyads who met study inclusion criteria. To address the study purpose, the sample was restricted to infants of mothers who were Hispanic immigrants to the United States and who had birthweights within 1 standard deviations of the mean, based on World Health Organization (WHO) growth standards (Grummer-Strawn, Reinold, Krebs, & Centers for Disease Control and Prevention, 2010). Thus, their infants weighed between 2,800 and 3,850 g at birth (2,800–3,695 g for females; 2,900– 3,850 g for males), 37 weeks’ gestation or longer, without medical problems that would interfere with feeding or growth and for whom weight-for-age (WFA) data were available for birth, 6 months, and 12 months of age. 2
Volume - Number -
Measures Infant WFA data from measurements taken at birth and at 2, 4, 6, and 12 months of age during regularly scheduled well visits were extracted from electronic medical records. Because preliminary analyses indicated that hospital-based measurements of birth length were not consistent with later measurements at first well child visits at the study site, we used WHO sex- and agespecific WFA z scores and percentiles to describe growth changes during these timeframes, as has been done in prior research (Li, Fein, & Grummer-Strawn, 2008; Stettler, Kumanyika, Katz, Zamel, & Stallings, 2003). We excluded z scores greater or less than 5 as biologically implausible, as also has been the procedure in prior research (Li et al., 2008). z scores correspond to major percentile scores on the WHO growth charts for infants 0 to 2 years of age. The z score is the standard deviation above or below the mean, such that a z score of 0 is the same as the 50th percentile, z + 1 is a percentile of 84.1, and z + 2 a percentile of 97.7 on the WHO WFA growth charts for children 0 to 2 years of age. Main Outcome Based on IOM (2011) designations of weight status categories, 12-month-old infants with a WFA percentile of 84.1 or greater were identified as being at risk for overweight. In addition to the description of z score differences, growth percentiles were used to describe findings because this approach mirrors clinical practice, where growth percentiles are used at regularly scheduled well-child visits. Statistical Analysis Descriptive statistics were used to assess the distribution of maternal–infant characteristics and WFA changes in early infancy (0–6 months). Logistic regression models were constructed to examine the impact of weight change in early infancy on weight status at 12 months of age (WFA percentile < vs. $ 84.1 at 12 months), adjusting for factors that have been found to be significant in prior child obesity research. These factors included maternal age, education, prepregnancy body mass index (weight/height squared, kg/m2), parity, and infant sex and birth weight (Gaffney, Kitsantas, & Cheema, 2012; Taveras, Gillman, Kleinman, Rich-Edwards, & Rifas-Shiman, 2010). For our sample, we controlled for mother’s length of time in the United States as a marker for acculturation. Further, we converted the odds ratios obtained from the logistic regression models into probabilities using Stata’s margins and marginsplot command (StataCorp, 2015) to gain a better understanding of the results. The probability estimates were constructed for the sample median birth weight of 3,340 g 450 g (2,890 g; 3,340 g; and 3,790 g) to illustrate its statistically significant effect on crossing the benchmark for risk for Journal of Pediatric Health Care
overweight (WFA percentile $ 84.1) at 12 months of infant age. Values for other variables were assumed to match the sample characteristics (the Strata default). RESULTS Table 1 displays maternal–infant characteristics. In this sample, 48% of mothers were between the ages of 26 and 33 years old (mean = 28.4, standard deviation = 5.5). Ninety percent of them had a high school education or less, with one third having none to 6th grade education level. One half of mothers had pre-pregnancy body mass indexes in the overweight/obese range. For approximately one third of the sample, the infant who was the focus of this study was their first child. Most
TABLE 1. Maternal–infant characteristics (N = 335) Characteristics Maternal characteristics Age in years #25 26–33 $34 Education in years 0–6 7–11 12 $13 Pre-pregnancy body mass index Normal (18 to <25) Overweight (25 to <30) Obese (30) Number of living children 1 2 $3 Country of origin El Salvador Honduras Guatemala Mexico Other Years in the United States <5 5 to <10 $10 Infant characteristics Sex Male Female WFA percentile at birth 15 to <50 50 to <84.1 WFA percentile at 12 months 0 to <15 >15 to <50 50 to <84.1 $84.1
n
%
110 162 63
33 48 19
112 90 98 30
33 27 29 9
131 100 67
39 30 20
114 114 106
34 34 32
151 40 52 37 55
45 12 16 11 16
126 144 60
38 43 18
169 166
50 50
142 193
42 58
6 55 151 123
2 16 45 36
Note. Some cells are less than 100% because of missing values or rounding. WFA, weight-for-age.
www.jpedhc.org
of the mothers (73%) were from Central America, and most (81%) had lived in the United States for less than 10 years. By 12 months of age, 36.7% of the infants had crossed the IOM (2011) risk for overweight designation. For the present study, we categorized all infants who crossed the threshold for risk for overweight and compared them with those who did not make this crossing (WHO WFA percentile < vs. $ 84.1) by 12 months of age (n = 123, 36.7% vs. n = 212, 63.3%). As illustrated in Figure 1, infants who crossed the 84.1 percentile growth line at 12 months on average were found to have increased their average WFA percentile at each 2-month period in early infancy, with the most substantial increase occurring between birth and 2 months of age. Conversely, those who did not reach this benchmark risk for overweight at 12 months showed a more stable weight gain trajectory on average throughout early infancy (Table 2). Based on our finding of a potential tipping point for weight gain at 0 to 2 months among infants who were in the WFA percentile of 84.1 or greater at 12 months, logistic regression models were built for weight change at both 0 to 2 months and 0 to 6 months. In the adjusted logistic regression models, increases in WFA z scores in early infancy were statistically significant predictors of WFA percentile greater than or equal to 84.1 at 12 months. Specifically, for each z score increase between birth and 2 months, infants were found to have 4.98 times (95% confidence interval = [2.99, 8.29]) the odds of reaching this threshold at 12 months. Furthermore, for each z score increase between birth and 6 months, infants were found to have 20.8 times (95% confidence interval = [9.88, 44.0]) the odds of being at risk for overweight at 1 year of age. Birth weight and number of living children were also found to be statistically significant predictors (Table 3). To better understand the effects of WFA change in the first 6 months on crossing the WFA percentile of 84.1 or greater at 1 year of age for our sample, we then converted the odds into expected probabilities. These estimates are stratified by the sample median birthweight of 3,340 450 g to account for its effect on WFA percentile at 12 months of infant age. Figure 2 thus displays the estimated likelihood of an infant crossing the threshold WFA percentile of 84.1 or greater at 12 months based on WFA z score differences between birth and 6 months at three different birth weight designations (2,890 g; 3,340 g; and 3,790 g). These findings show that an infant born at the median birth weight for our sample (3,340 g; Figure 2, solid middle line) and who had no z score change in early infancy (0–6 months) had an 8% probability of exceeding the benchmark risk for overweight at 12 months. If that infant matching our sample characteristics had a WFA z score increase of 1 (the equivalent of crossing one major growth percentile on the WHO WFA growth chart) during the same period, the -/- 2017
3
FIGURE 1. Early infancy growth trajectories by weight-for-age percentile less than versus greater than or equal to 84.1 at 12 months. mo, months.
probability increased to 60%. However, in the case of an infant born at 3,790 g (top line in Figure 2), no z score change during early infancy (0–6 months) resulted in a 51% probability of reaching that benchmark, whereas a +1 z score change was associated with a 95% probability of crossing the threshold for risk for overweight at 12 months of age. DISCUSSION In this study of weight gain among the infants of low income Hispanic immigrant mothers, more than one third of the sample crossed the WFA percentile of 84.1 percentile at 12 months of age and, consequently, was categorized as being at risk for overweight. Furthermore, infants who did and did not cross this threshold differed significantly in average WFA percentile increases between each of the timeframes of regularly scheduled 0- to 6-month well-child checks, with the most substantial difference occurring between birth and 2 months of age. The Life Course Health Development (LCHD) model offers insight into understanding these findings. Specifically, the conceptual model posits that an individual’s long-term health trajectory is shaped by the cumulative effect of factors programmed into the human system during sensitive times in development. According to the LCHD, early infancy (0–6 months) is one of these sensitive times. Exposures such as rapid weight gain during this time have a heightened potential for both near-term and lifelong impact (Devereaux, 2011; Halfon & Hochstein,
4
Volume - Number -
2002; Halfon, Russ, & the conceptual Regalado, 2005). Furmodel posits that thermore, the present an individual’s study findings suggest that the cumulative long-term health impact described by trajectory is shaped the LCHD model beby the cumulative comes evident as early as the infant’s first effect of factors birthday. programmed into The LCHD model the human system also specifies that multiple factors occurring during sensitive simultaneously during times in these sensitive times development. have a nested effect on an individual’s health trajectory. These factors include behavioral, social, cultural, and economic experiences. All mothers in our study were of low income, and most came to the United States from low-resource countries in Central America. These countries have a widespread problem with infant malnutrition (CDC, 2017). The mothers’ own life experiences in observing the impact of malnutrition on family and community may have prompted a conviction that ‘‘bigger is better’’ in terms of infant growth patterns. The current study adds to the accumulating body of evidence that early infant weight gain is associated with later childhood obesity. Polk, Thornton, Caufield, and Mu~ noz (2015) found that accelerated
Journal of Pediatric Health Care
TABLE 2. Weight-for-age percentiles in early infancy, by risk for overweight percentile (< vs. $ 84.1) status at 12 months of age Percentile $ 84.1 at 12 months
Percentile < 84.1 at 12 months Age
x
SD
n
x
SD
n
p
Birth 2 months 4 months 6 months
50.9 51.7 54.6 57.6
18.8 20.9 22.0 21.2
212 205 200 212
55.1 72.0 80.7 86.2
18.8 20.9 22.0 21.2
123 119 116 123
.045 <.001 <.001 <.001
Note. SD, Standard deviation.
growth during the first 2 years of life increased the risk of childhood obesity at 3 years of age. Their study also found that Hispanic children had a higher incidence of early childhood obesity than non-Hispanic children in their low-income sample. A systematic review by Woo Baidal et al. (2016) found 45 studies that reported an association between higher early childhood weight or weight gain and later obesity. Most studies in the review
examined weight status over a broad span of time, for example, over the first 2 years of life, and reported prediction of obesity among older children. The current study contributes to this body of research by using a narrower time frame for prediction, namely, the prediction of risk at 1 year of age based on weight changes between 0 and 6 months of age. The rationale for this more refined scope was to help identify a tipping point
TABLE 3. Effects of weight-for-age change in early infancy on weight status at 12 months of age Change in WFA z score between birth and 2/6 months
0–2 months Percentile $ 84.1 at 12 months OR 4.98*
Maternal characteristics Age in years #25 26–33 $34 Education in years 0–6 7–11 12 $13 Pre-pregnancy body mass index in kg/m2 Normal (18 to <25) Overweight (25 to <30) Obese ($30) Number of living children 1 2 $3 Years in the United States <5 5 to <10 $10 Infant characteristics Sex Male Female Birth weight in grams Observations
95% CI [2.99, 8.29]
1.13 Reference 0.83
[0.57, 2.24]
1.00 1.71 Reference 1.33
[0.47, 2.13] [0.82, 3.58]
Reference 1.83 2.02
OR 20.8*
95% CI [9.88, 44.0]
0.84 Reference 1.01
[0.36, 1.93]
[0.38, 2.25] [0.92, 5.17]
[0.47, 3.75]
0.92 2.18 Reference 1.18
[0.93, 3.63] [0.91, 4.46]
Reference 1.49 1.76
[0.35, 1.97]
2.11** Reference 0.72
[1.01, 4.42]
0.63 Reference 0.90
[0.33, 1.22]
Reference 1.12 1.00* 284
0-6 months Percentile $ 84.1 at 12 months
[0.35, 1.49]
[0.39, 2.06]
[0.60, 2.08] [1.00, 1.00]
[0.36, 2.83]
[0.31, 4.59]
[0.67, 3.29] [0.68, 4.59]
1.32 Reference 0.54
[0.55, 3.14]
0.91 Reference 0.79
[0.41, 2.00]
Reference 1.38 1.01* 293
[0.22, 1.33]
[0.30, 2.09]
[0.66, 2.88] [1.00, 1.01]
Note. CI, Confidence interval; OR, odds ratio. *p < .001. **p < .05.
www.jpedhc.org
-/- 2017
5
FIGURE 2. Estimates for probability of WFA percentile $ 84.1 at 1 year, based on birthweight and WFA change at 0 to 6 months. mo, months; WFA, weight-for-age.
in early weight gain that may lead to implementation of earlier, evidence-based interventions to prevent childhood obesity. Also, Woo Baidal et al. (2016) found in their review that despite substantial disparities in childhood obesity, most of the studies they extracted from the literature took place in high-income economies with predominantly non-Hispanic White participants. In the United States, few of the studies included children from lowincome families of minority ethnicities. This observation is consistent with a report by the National Institutes of Diabetes and Digestive and Kidney Diseases that overall most studies of childhood obesity have focused on school-age children, but insufficient emphasis has been placed on the earlier stages of the life cycle. The report called for studies that reflect the population of U.S. infants characterized by a high prevalence of poverty and non-White or Hispanic race/ethnicity (Lumeng, Taveras, Birch, & Yanovski, 2015). Hispanic children are predicted to make up 35% of U.S. children by the year 2050 (Passel & Cohn, 2008). Health disparities that affect them need to be addressed now with preventive interventions implemented at sensitive times in the life cycle for greatest impact. Future Research Primary care settings that provide a medical home for low-income Hispanic immigrant families, including Federally Qualified Health Centers, are ideally suited for intervention studies that address excessive early infant weight gain because of their multidisciplinary
6
Volume - Number -
approach to caring for children who are disproportionately affected by obesity and its lifelong health consequences. Machuca et al. (2016) conducted research in such a setting to examine the preliminary efficacy of Well Baby Group Care as an alternative to traditional one-on-one well-child care for infants during the first 18 months of life. Content for group sessions included responsive feeding practices. The children in the intervention group were found to be significantly less likely to be overweight/obese at 2 years than those in the control group. Their study sample was small (n = 187) but included 64% who were Hispanic. Future replication and extension studies are needed to inform the direction of evidence-based practice for this growing area of concern in primary pediatric health care. Implications for Practice As pediatric health care providers, we monitor growth during well-baby visits and routinely point out WHO growth curve changes to parents of infants in our care. The conceptual underpinnings of the LCHD model and the present study findings add support to this well-established clinical practice (Hagan, Shaw, & Duncan, 2017). The clinical observation of early and rapid weight gain points to a modifiable risk factor occurring at a sensitive time in the infant’s life cycle, with the potential for cumulative risk on long-term health and development. Concomitantly, the shared observation of growth curve changes by provider and parent also may be occurring at a sensitive time for anticipatory guidance that bigger is not always
Journal of Pediatric Health Care
better in terms of infant size. The visual presentation of rapid weight The visual gain, accompanied presentation of by information about risk for future metarapid weight gain, bolic and cardiovascuaccompanied by lar consequences, may information about enhance receptivity to age-appropriate oberisk for future sity prevention meametabolic and sures, including the cardiovascular promotion of breastfeeding and feeding consequences, styles that are sensitive may enhance to infant cues of hunreceptivity to ageger and satiety. Involving parents in appropriate obesity early one-to-one or prevention group sessions about measures. how best to implement these recommendations within the context of their current family life circumstances may lead to culturally appropriate and effective methods that reduce the incidence rapid weight gain that leads to long-term health disparities among vulnerable populations, including the infants of low-income Hispanic immigrants.
REFERENCES Centers for Disease Control and Prevention. (2017). Central American refugee health profile. Atlanta, GA: Author. Retrieved from https://www.cdc.gov/immigrantrefugeehealth/profiles/ central-american/health-information/nutrition/index.html Dennison, B. A., Edmunds, L. S., Stratton, H. H., & Pruzek, R. M. (2006). Rapid infant weight gain predicts childhood overweight. Obesity, 14, 491-499. Devereaux, J. (2011). The life course approach to obesity: A focus on Latino youth-An interview with Elsie M. Taveras, MD, MPH. Childhood Obesity, 7, 8-10. Gaffney, K. F., Kitsantas, P., & Cheema, J. (2012). Clinical practice guidelines for feeding behaviors and weight-for-age at 12 months: A secondary analysis of Infant Feeding Practices Study II. Worldviews on Evidence-Based Nursing, 9, 234-242. Goodell, L. S., Wakefield, D. B., & Ferris, A. M. (2009). Rapid weight gain during the first year of life predicts obesity in 2-3 year olds from a low-income, minority population. Journal of Community Health, 34, 370-375. Grummer-Strawn, L. M., Reinold, C., Krebs, N. F., & Centers for Disease Control and Prevention. (2010). Use of World Health Organization and CDC growth charts for children aged 0–59 months in the United States. MMWR Recommendations and Reports, 59(RR09), 1-15. Hagan, J. F., Shaw, J. S. & Duncan, P. M. (Eds.). (2017). Bright futures: Guidelines for health supervision of infants, children, and adolescents (4th ed.). Elk Grove Village, IL: American Academy of Pediatrics. Halfon, N., & Hochstein, M. (2002). Life Course Health Development: An integrated framework for developing health, policy, and research. The Milbank Quarterly, 80, 433-479.
www.jpedhc.org
Halfon, N., Russ, S., & Regalado, M. (2005). The Life Course Health Development model: A guide to children’s health care policy and practice. Zero to Three 4-12. Harrington, J. W., Nguyen, V. Q., Paulson, J. F., Garland, R., Pasquinelli, L., & Lewis, D. (2010). Identifying the ‘‘Tipping Point’’ age for overweight pediatric patients. Clinical Pediatrics, 49, 638643. Institute of Medicine. (2011). Early childhood obesity prevention policies. Washington, DC: National Academies Press. Lumeng, L. C., Taveras, E. M., Birch, L. L., & Yanovski, S. Z. (2015). Prevention of obesity in infancy and early childhood: A National Institutes of Health workshop. JAMA Pediatrics, 169, 484-490. Li, R., Fein, S. B., & Grummer-Strawn, L. M. (2008). Association of breastfeeding and bottle-emptying behaviors at early infancy with infants’ risk for excess weight at late infancy. Pediatrics, 122(Suppl. 2), S77-S84. Machuca, H., Arevalo, S., Hackley, B., Applebaum, J., Miskin, A., Heo, M., & Shapiro, A. (2016). Well baby group care: Evaluation of a promising intervention for primary obesity prevention in toddlers. Childhood Obesity, 12(3), 171-178. Murphey, D., Guzman, L., & Torres, A. (2014). America’s Hispanic children: Gaining ground, looking forward. Bethesda, MD: Child Trends Hispanic Institute. Retrieved from https://www. childtrends.org/wp-content/uploads/2014/09/2014-38America HispanicChildren.pdf Ogden, C. L., Caroll, M. D., Lawman, H. G., Fryar, C. D., KruszonMoran, D., Kit, B. K., & Flegal, K. M. (2016). Trends in obesity prevalence among children and adolescents in the United States, 1988-1994 through 2013-2014. JAMA, 315, 2292-2299. Passel, J. S., & Cohn, D. (2008). US Population Projections: 20052050. Washington, DC: Pew Research Center. ~oz, A. (2015). Rapid infant Polk, S., Thornton, R. J., Caulfield, L., & Mun weight gain and early childhood obesity in low-income Latinos and non-Latinos. Public Health Nutrition, 19, 1777-1784. StataCorp. (2015). Stata Statistical Software: Release 14. College Station, TX: StataCorp LP. Stettler, N., Kumanyika, S. K., Katz, S. H., Zemel, B. S., & Stallings, V. A. (2003). Rapid weight gain during infancy and obesity in young adulthood in a cohort of African Americans. American Journal of Clinical Nutrition, 77, 1374-1378. Taveras, E. M., Gilman, M. W., Kleinman, K., Rich-Edwards, J. W., & Rifas-Shiman, S. L. (2010). Racial/ethnic differences in early life-risk factors for childhood obesity. Pediatrics, 125, 686-695. Taveras, E. M., Rifas-Shiman, S. L., Belfort, M. B., Kleinman, K. P., Oken, E., & Gillman, M. W. (2009). Weight status in the first 6 months of life and obesity at 3 years of age. Pediatrics, 123, 1177-1183. Taveras, E. M., Rifas-Shiman, S. L., Sherry, B., Oken, E., Haines, J., Kleinman, K., Rich-Edwards, J. W., & Gillman, M. W. (2011). Crossing growth percentiles in infancy and risk of obesity in childhood. Archives of Pediatrics and Adolescent Medicine, 165(11), 993-998. Woo Baidal, J. A., Locks, L. M., Cheng, E. R., Blake-Lamb, T. L., Perkins, M. E., & Taveras, E. M. (2016). Risk factors for childhood obesity in the first 1,000 days: A systematic review. American Journal of Preventive Medicine, 50, 761779. World Health Organization. (2017). Global strategy on diet, physical activity and health. Why does childhood overweight and obesity matter? Geneva, Switzerland: Author. Retrieved from http:// www.who.int/dietphysicalactivity/childhood_consequences/ en/
-/- 2017
7
Sensitive Timing for Risk of Overweight Among Infants of Low-Income Hispanic Immigrants Kathleen F. Gaffney, PhD, RN, F/PNP-BC, Albert V. Brito, MD, Deborah A. Kermer, MA, & Panagiota Kitsantas, PhD
ABSTRACT Introduction: This study examined the association between early weight gain (0–6 months) and risk for overweight, as defined by the Institute of Medicine, at 1 year among infants of low-income Hispanic immigrant mothers. Method: Weight-for-age data were extracted from electronic medical records of 335 infants with gestations of 37 weeks or longer and birthweights appropriate for gestational age and without medical problems likely to interfere with growth or feeding. Logistic regression models were constructed to examine the impact of early weight changes on weight status at 1 year.
Kathleen F. Gaffney, Professor, School of Nursing, George Mason University, Fairfax, VA. Albert V. Brito, Medical Director, Inova Cares Clinic for Children, Falls Church, VA, and Assistant Professor, School of Medicine, Virginia Commonwealth University, Richmond, VA. Deborah A. Kermer, Data Services Research Consultant, University Libraries Data Services, George Mason University, Fairfax, VA. Panagiota Kitsantas, Professor, Health Administration and Policy Department, George Mason University, Fairfax, VA. Conflicts of interest: None to report. Correspondence: Kathleen F. Gaffney, PhD, RN, F/PNP-BC, School of Nursing, 3C4, George Mason University, Fairfax, VA 22030; e-mail: [email protected]. 0891-5245/$36.00 Copyright Q 2017 by the National Association of Pediatric Nurse Practitioners. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pedhc.2017.08.001
www.jpedhc.org
Results: By 12 months, 36.7% of infants had crossed weightfor-age of 84.1% or greater on World Health Organization growth charts. In adjusted models, infants had 20.8 (95% confidence interval = [19.8, 44.0]) times the odds of reaching this benchmark at 1 year for each z score increase at 0 to 6 months. Discussion: The study highlights a time-sensitive opportunity for interventions to reduce risk for overweight for this vulnerable population. J Pediatr Health Care. (2017) -, ---.
KEY WORDS Infant weight, Hispanic, immigrant
A recent population-based, national U.S. survey of childhood obesity found that among 2- to 5-year-olds, Hispanic children have higher rates of obesity and extreme obesity than any other racial/ethnic group (Ogden et al., 2016). This finding has important health policy implications, because current estimates indicate that 1 in 4 of our nation’s young children is Hispanic, with more than half of these children having at least one parent who is an immigrant to the United States (Murphey, Guzman, & Torres, 2014). The survey findings also have substantial clinical implications based on the well-established evidence that childhood obesity leads to elevated blood pressure, diabetes, musculoskeletal disorders, and other cumulative effects that may extend from childhood into adolescence and adulthood. A report of the World Health Organization (2017) pointed out that the risk for these morbidities depends partly on the age of onset and the duration of the overweight or obesity. The national survey findings, however, did not address weight status differences for children younger than 2 years of age. -/- 2017
1
This is an important omission, because recent evidence suggests that the underpinnings of childhood obesity may begin early in the life cycle. One retrospective study (Harrington et al., 2010) of overweight and obese older children found that half had been overweight by 2 years of age and 90% had become overweight by age 5 years. Further examination of the children’s weight trajectories led these researchers to conclude that the critical period for preventing childhood obesity may have been during the first 2 years of life and that for many the tipping point may have occurred between birth and 3 months of age. This latter finding is consistent with the research by Taveras et al. (2011), who found that obesity prevalence at ages 5 and 10 years of age was highest among children who crossed two or more major growth percentiles in the first 6 months of life. To date, studies of the relationship between weight gain in early infancy and later obesity have been limited by an underrepresentation of Hispanic children (Dennison, Edmunds, Stratton, & Pruzek, 2006; Goodell, Wakefield, & Ferris, 2009; Taveras et al., recent evidence 2009). Therefore, the purpose of this study suggests that the was to examine the underpinnings of association between childhood obesity early weight gain (0–6 months) and risk may begin early in for overweight, as the life cycle. defined by the Institute of Medicine (IOM; 2011), at 1 year of age among infants of low-income Hispanic immigrant mothers.
METHODS Sample and Setting Data were extracted from the medical records of 335 infants at a large pediatric, primary care setting in Virginia that serves low-income families. After approval of health care site and university institutional review boards, demographic and growth data were extracted for maternal–infant dyads who met study inclusion criteria. To address the study purpose, the sample was restricted to infants of mothers who were Hispanic immigrants to the United States and who had birthweights within 1 standard deviations of the mean, based on World Health Organization (WHO) growth standards (Grummer-Strawn, Reinold, Krebs, & Centers for Disease Control and Prevention, 2010). Thus, their infants weighed between 2,800 and 3,850 g at birth (2,800–3,695 g for females; 2,900– 3,850 g for males), 37 weeks’ gestation or longer, without medical problems that would interfere with feeding or growth and for whom weight-for-age (WFA) data were available for birth, 6 months, and 12 months of age. 2
Volume - Number -
Measures Infant WFA data from measurements taken at birth and at 2, 4, 6, and 12 months of age during regularly scheduled well visits were extracted from electronic medical records. Because preliminary analyses indicated that hospital-based measurements of birth length were not consistent with later measurements at first well child visits at the study site, we used WHO sex- and agespecific WFA z scores and percentiles to describe growth changes during these timeframes, as has been done in prior research (Li, Fein, & Grummer-Strawn, 2008; Stettler, Kumanyika, Katz, Zamel, & Stallings, 2003). We excluded z scores greater or less than 5 as biologically implausible, as also has been the procedure in prior research (Li et al., 2008). z scores correspond to major percentile scores on the WHO growth charts for infants 0 to 2 years of age. The z score is the standard deviation above or below the mean, such that a z score of 0 is the same as the 50th percentile, z + 1 is a percentile of 84.1, and z + 2 a percentile of 97.7 on the WHO WFA growth charts for children 0 to 2 years of age. Main Outcome Based on IOM (2011) designations of weight status categories, 12-month-old infants with a WFA percentile of 84.1 or greater were identified as being at risk for overweight. In addition to the description of z score differences, growth percentiles were used to describe findings because this approach mirrors clinical practice, where growth percentiles are used at regularly scheduled well-child visits. Statistical Analysis Descriptive statistics were used to assess the distribution of maternal–infant characteristics and WFA changes in early infancy (0–6 months). Logistic regression models were constructed to examine the impact of weight change in early infancy on weight status at 12 months of age (WFA percentile < vs. $ 84.1 at 12 months), adjusting for factors that have been found to be significant in prior child obesity research. These factors included maternal age, education, prepregnancy body mass index (weight/height squared, kg/m2), parity, and infant sex and birth weight (Gaffney, Kitsantas, & Cheema, 2012; Taveras, Gillman, Kleinman, Rich-Edwards, & Rifas-Shiman, 2010). For our sample, we controlled for mother’s length of time in the United States as a marker for acculturation. Further, we converted the odds ratios obtained from the logistic regression models into probabilities using Stata’s margins and marginsplot command (StataCorp, 2015) to gain a better understanding of the results. The probability estimates were constructed for the sample median birth weight of 3,340 g 450 g (2,890 g; 3,340 g; and 3,790 g) to illustrate its statistically significant effect on crossing the benchmark for risk for Journal of Pediatric Health Care
overweight (WFA percentile $ 84.1) at 12 months of infant age. Values for other variables were assumed to match the sample characteristics (the Strata default). RESULTS Table 1 displays maternal–infant characteristics. In this sample, 48% of mothers were between the ages of 26 and 33 years old (mean = 28.4, standard deviation = 5.5). Ninety percent of them had a high school education or less, with one third having none to 6th grade education level. One half of mothers had pre-pregnancy body mass indexes in the overweight/obese range. For approximately one third of the sample, the infant who was the focus of this study was their first child. Most
TABLE 1. Maternal–infant characteristics (N = 335) Characteristics Maternal characteristics Age in years #25 26–33 $34 Education in years 0–6 7–11 12 $13 Pre-pregnancy body mass index Normal (18 to <25) Overweight (25 to <30) Obese (30) Number of living children 1 2 $3 Country of origin El Salvador Honduras Guatemala Mexico Other Years in the United States <5 5 to <10 $10 Infant characteristics Sex Male Female WFA percentile at birth 15 to <50 50 to <84.1 WFA percentile at 12 months 0 to <15 >15 to <50 50 to <84.1 $84.1
n
%
110 162 63
33 48 19
112 90 98 30
33 27 29 9
131 100 67
39 30 20
114 114 106
34 34 32
151 40 52 37 55
45 12 16 11 16
126 144 60
38 43 18
169 166
50 50
142 193
42 58
6 55 151 123
2 16 45 36
Note. Some cells are less than 100% because of missing values or rounding. WFA, weight-for-age.
www.jpedhc.org
of the mothers (73%) were from Central America, and most (81%) had lived in the United States for less than 10 years. By 12 months of age, 36.7% of the infants had crossed the IOM (2011) risk for overweight designation. For the present study, we categorized all infants who crossed the threshold for risk for overweight and compared them with those who did not make this crossing (WHO WFA percentile < vs. $ 84.1) by 12 months of age (n = 123, 36.7% vs. n = 212, 63.3%). As illustrated in Figure 1, infants who crossed the 84.1 percentile growth line at 12 months on average were found to have increased their average WFA percentile at each 2-month period in early infancy, with the most substantial increase occurring between birth and 2 months of age. Conversely, those who did not reach this benchmark risk for overweight at 12 months showed a more stable weight gain trajectory on average throughout early infancy (Table 2). Based on our finding of a potential tipping point for weight gain at 0 to 2 months among infants who were in the WFA percentile of 84.1 or greater at 12 months, logistic regression models were built for weight change at both 0 to 2 months and 0 to 6 months. In the adjusted logistic regression models, increases in WFA z scores in early infancy were statistically significant predictors of WFA percentile greater than or equal to 84.1 at 12 months. Specifically, for each z score increase between birth and 2 months, infants were found to have 4.98 times (95% confidence interval = [2.99, 8.29]) the odds of reaching this threshold at 12 months. Furthermore, for each z score increase between birth and 6 months, infants were found to have 20.8 times (95% confidence interval = [9.88, 44.0]) the odds of being at risk for overweight at 1 year of age. Birth weight and number of living children were also found to be statistically significant predictors (Table 3). To better understand the effects of WFA change in the first 6 months on crossing the WFA percentile of 84.1 or greater at 1 year of age for our sample, we then converted the odds into expected probabilities. These estimates are stratified by the sample median birthweight of 3,340 450 g to account for its effect on WFA percentile at 12 months of infant age. Figure 2 thus displays the estimated likelihood of an infant crossing the threshold WFA percentile of 84.1 or greater at 12 months based on WFA z score differences between birth and 6 months at three different birth weight designations (2,890 g; 3,340 g; and 3,790 g). These findings show that an infant born at the median birth weight for our sample (3,340 g; Figure 2, solid middle line) and who had no z score change in early infancy (0–6 months) had an 8% probability of exceeding the benchmark risk for overweight at 12 months. If that infant matching our sample characteristics had a WFA z score increase of 1 (the equivalent of crossing one major growth percentile on the WHO WFA growth chart) during the same period, the -/- 2017
3
FIGURE 1. Early infancy growth trajectories by weight-for-age percentile less than versus greater than or equal to 84.1 at 12 months. mo, months.
probability increased to 60%. However, in the case of an infant born at 3,790 g (top line in Figure 2), no z score change during early infancy (0–6 months) resulted in a 51% probability of reaching that benchmark, whereas a +1 z score change was associated with a 95% probability of crossing the threshold for risk for overweight at 12 months of age. DISCUSSION In this study of weight gain among the infants of low income Hispanic immigrant mothers, more than one third of the sample crossed the WFA percentile of 84.1 percentile at 12 months of age and, consequently, was categorized as being at risk for overweight. Furthermore, infants who did and did not cross this threshold differed significantly in average WFA percentile increases between each of the timeframes of regularly scheduled 0- to 6-month well-child checks, with the most substantial difference occurring between birth and 2 months of age. The Life Course Health Development (LCHD) model offers insight into understanding these findings. Specifically, the conceptual model posits that an individual’s long-term health trajectory is shaped by the cumulative effect of factors programmed into the human system during sensitive times in development. According to the LCHD, early infancy (0–6 months) is one of these sensitive times. Exposures such as rapid weight gain during this time have a heightened potential for both near-term and lifelong impact (Devereaux, 2011; Halfon & Hochstein,
4
Volume - Number -
2002; Halfon, Russ, & the conceptual Regalado, 2005). Furmodel posits that thermore, the present an individual’s study findings suggest that the cumulative long-term health impact described by trajectory is shaped the LCHD model beby the cumulative comes evident as early as the infant’s first effect of factors birthday. programmed into The LCHD model the human system also specifies that multiple factors occurring during sensitive simultaneously during times in these sensitive times development. have a nested effect on an individual’s health trajectory. These factors include behavioral, social, cultural, and economic experiences. All mothers in our study were of low income, and most came to the United States from low-resource countries in Central America. These countries have a widespread problem with infant malnutrition (CDC, 2017). The mothers’ own life experiences in observing the impact of malnutrition on family and community may have prompted a conviction that ‘‘bigger is better’’ in terms of infant growth patterns. The current study adds to the accumulating body of evidence that early infant weight gain is associated with later childhood obesity. Polk, Thornton, Caufield, and Mu~ noz (2015) found that accelerated
Journal of Pediatric Health Care
TABLE 2. Weight-for-age percentiles in early infancy, by risk for overweight percentile (< vs. $ 84.1) status at 12 months of age Percentile $ 84.1 at 12 months
Percentile < 84.1 at 12 months Age
x
SD
n
x
SD
n
p
Birth 2 months 4 months 6 months
50.9 51.7 54.6 57.6
18.8 20.9 22.0 21.2
212 205 200 212
55.1 72.0 80.7 86.2
18.8 20.9 22.0 21.2
123 119 116 123
.045 <.001 <.001 <.001
Note. SD, Standard deviation.
growth during the first 2 years of life increased the risk of childhood obesity at 3 years of age. Their study also found that Hispanic children had a higher incidence of early childhood obesity than non-Hispanic children in their low-income sample. A systematic review by Woo Baidal et al. (2016) found 45 studies that reported an association between higher early childhood weight or weight gain and later obesity. Most studies in the review
examined weight status over a broad span of time, for example, over the first 2 years of life, and reported prediction of obesity among older children. The current study contributes to this body of research by using a narrower time frame for prediction, namely, the prediction of risk at 1 year of age based on weight changes between 0 and 6 months of age. The rationale for this more refined scope was to help identify a tipping point
TABLE 3. Effects of weight-for-age change in early infancy on weight status at 12 months of age Change in WFA z score between birth and 2/6 months
0–2 months Percentile $ 84.1 at 12 months OR 4.98*
Maternal characteristics Age in years #25 26–33 $34 Education in years 0–6 7–11 12 $13 Pre-pregnancy body mass index in kg/m2 Normal (18 to <25) Overweight (25 to <30) Obese ($30) Number of living children 1 2 $3 Years in the United States <5 5 to <10 $10 Infant characteristics Sex Male Female Birth weight in grams Observations
95% CI [2.99, 8.29]
1.13 Reference 0.83
[0.57, 2.24]
1.00 1.71 Reference 1.33
[0.47, 2.13] [0.82, 3.58]
Reference 1.83 2.02
OR 20.8*
95% CI [9.88, 44.0]
0.84 Reference 1.01
[0.36, 1.93]
[0.38, 2.25] [0.92, 5.17]
[0.47, 3.75]
0.92 2.18 Reference 1.18
[0.93, 3.63] [0.91, 4.46]
Reference 1.49 1.76
[0.35, 1.97]
2.11** Reference 0.72
[1.01, 4.42]
0.63 Reference 0.90
[0.33, 1.22]
Reference 1.12 1.00* 284
0-6 months Percentile $ 84.1 at 12 months
[0.35, 1.49]
[0.39, 2.06]
[0.60, 2.08] [1.00, 1.00]
[0.36, 2.83]
[0.31, 4.59]
[0.67, 3.29] [0.68, 4.59]
1.32 Reference 0.54
[0.55, 3.14]
0.91 Reference 0.79
[0.41, 2.00]
Reference 1.38 1.01* 293
[0.22, 1.33]
[0.30, 2.09]
[0.66, 2.88] [1.00, 1.01]
Note. CI, Confidence interval; OR, odds ratio. *p < .001. **p < .05.
www.jpedhc.org
-/- 2017
5
FIGURE 2. Estimates for probability of WFA percentile $ 84.1 at 1 year, based on birthweight and WFA change at 0 to 6 months. mo, months; WFA, weight-for-age.
in early weight gain that may lead to implementation of earlier, evidence-based interventions to prevent childhood obesity. Also, Woo Baidal et al. (2016) found in their review that despite substantial disparities in childhood obesity, most of the studies they extracted from the literature took place in high-income economies with predominantly non-Hispanic White participants. In the United States, few of the studies included children from lowincome families of minority ethnicities. This observation is consistent with a report by the National Institutes of Diabetes and Digestive and Kidney Diseases that overall most studies of childhood obesity have focused on school-age children, but insufficient emphasis has been placed on the earlier stages of the life cycle. The report called for studies that reflect the population of U.S. infants characterized by a high prevalence of poverty and non-White or Hispanic race/ethnicity (Lumeng, Taveras, Birch, & Yanovski, 2015). Hispanic children are predicted to make up 35% of U.S. children by the year 2050 (Passel & Cohn, 2008). Health disparities that affect them need to be addressed now with preventive interventions implemented at sensitive times in the life cycle for greatest impact. Future Research Primary care settings that provide a medical home for low-income Hispanic immigrant families, including Federally Qualified Health Centers, are ideally suited for intervention studies that address excessive early infant weight gain because of their multidisciplinary
6
Volume - Number -
approach to caring for children who are disproportionately affected by obesity and its lifelong health consequences. Machuca et al. (2016) conducted research in such a setting to examine the preliminary efficacy of Well Baby Group Care as an alternative to traditional one-on-one well-child care for infants during the first 18 months of life. Content for group sessions included responsive feeding practices. The children in the intervention group were found to be significantly less likely to be overweight/obese at 2 years than those in the control group. Their study sample was small (n = 187) but included 64% who were Hispanic. Future replication and extension studies are needed to inform the direction of evidence-based practice for this growing area of concern in primary pediatric health care. Implications for Practice As pediatric health care providers, we monitor growth during well-baby visits and routinely point out WHO growth curve changes to parents of infants in our care. The conceptual underpinnings of the LCHD model and the present study findings add support to this well-established clinical practice (Hagan, Shaw, & Duncan, 2017). The clinical observation of early and rapid weight gain points to a modifiable risk factor occurring at a sensitive time in the infant’s life cycle, with the potential for cumulative risk on long-term health and development. Concomitantly, the shared observation of growth curve changes by provider and parent also may be occurring at a sensitive time for anticipatory guidance that bigger is not always
Journal of Pediatric Health Care
better in terms of infant size. The visual presentation of rapid weight The visual gain, accompanied presentation of by information about risk for future metarapid weight gain, bolic and cardiovascuaccompanied by lar consequences, may information about enhance receptivity to age-appropriate oberisk for future sity prevention meametabolic and sures, including the cardiovascular promotion of breastfeeding and feeding consequences, styles that are sensitive may enhance to infant cues of hunreceptivity to ageger and satiety. Involving parents in appropriate obesity early one-to-one or prevention group sessions about measures. how best to implement these recommendations within the context of their current family life circumstances may lead to culturally appropriate and effective methods that reduce the incidence rapid weight gain that leads to long-term health disparities among vulnerable populations, including the infants of low-income Hispanic immigrants.
REFERENCES Centers for Disease Control and Prevention. (2017). Central American refugee health profile. Atlanta, GA: Author. Retrieved from https://www.cdc.gov/immigrantrefugeehealth/profiles/ central-american/health-information/nutrition/index.html Dennison, B. A., Edmunds, L. S., Stratton, H. H., & Pruzek, R. M. (2006). Rapid infant weight gain predicts childhood overweight. Obesity, 14, 491-499. Devereaux, J. (2011). The life course approach to obesity: A focus on Latino youth-An interview with Elsie M. Taveras, MD, MPH. Childhood Obesity, 7, 8-10. Gaffney, K. F., Kitsantas, P., & Cheema, J. (2012). Clinical practice guidelines for feeding behaviors and weight-for-age at 12 months: A secondary analysis of Infant Feeding Practices Study II. Worldviews on Evidence-Based Nursing, 9, 234-242. Goodell, L. S., Wakefield, D. B., & Ferris, A. M. (2009). Rapid weight gain during the first year of life predicts obesity in 2-3 year olds from a low-income, minority population. Journal of Community Health, 34, 370-375. Grummer-Strawn, L. M., Reinold, C., Krebs, N. F., & Centers for Disease Control and Prevention. (2010). Use of World Health Organization and CDC growth charts for children aged 0–59 months in the United States. MMWR Recommendations and Reports, 59(RR09), 1-15. Hagan, J. F., Shaw, J. S. & Duncan, P. M. (Eds.). (2017). Bright futures: Guidelines for health supervision of infants, children, and adolescents (4th ed.). Elk Grove Village, IL: American Academy of Pediatrics. Halfon, N., & Hochstein, M. (2002). Life Course Health Development: An integrated framework for developing health, policy, and research. The Milbank Quarterly, 80, 433-479.
www.jpedhc.org
Halfon, N., Russ, S., & Regalado, M. (2005). The Life Course Health Development model: A guide to children’s health care policy and practice. Zero to Three 4-12. Harrington, J. W., Nguyen, V. Q., Paulson, J. F., Garland, R., Pasquinelli, L., & Lewis, D. (2010). Identifying the ‘‘Tipping Point’’ age for overweight pediatric patients. Clinical Pediatrics, 49, 638643. Institute of Medicine. (2011). Early childhood obesity prevention policies. Washington, DC: National Academies Press. Lumeng, L. C., Taveras, E. M., Birch, L. L., & Yanovski, S. Z. (2015). Prevention of obesity in infancy and early childhood: A National Institutes of Health workshop. JAMA Pediatrics, 169, 484-490. Li, R., Fein, S. B., & Grummer-Strawn, L. M. (2008). Association of breastfeeding and bottle-emptying behaviors at early infancy with infants’ risk for excess weight at late infancy. Pediatrics, 122(Suppl. 2), S77-S84. Machuca, H., Arevalo, S., Hackley, B., Applebaum, J., Miskin, A., Heo, M., & Shapiro, A. (2016). Well baby group care: Evaluation of a promising intervention for primary obesity prevention in toddlers. Childhood Obesity, 12(3), 171-178. Murphey, D., Guzman, L., & Torres, A. (2014). America’s Hispanic children: Gaining ground, looking forward. Bethesda, MD: Child Trends Hispanic Institute. Retrieved from https://www. childtrends.org/wp-content/uploads/2014/09/2014-38America HispanicChildren.pdf Ogden, C. L., Caroll, M. D., Lawman, H. G., Fryar, C. D., KruszonMoran, D., Kit, B. K., & Flegal, K. M. (2016). Trends in obesity prevalence among children and adolescents in the United States, 1988-1994 through 2013-2014. JAMA, 315, 2292-2299. Passel, J. S., & Cohn, D. (2008). US Population Projections: 20052050. Washington, DC: Pew Research Center. ~oz, A. (2015). Rapid infant Polk, S., Thornton, R. J., Caulfield, L., & Mun weight gain and early childhood obesity in low-income Latinos and non-Latinos. Public Health Nutrition, 19, 1777-1784. StataCorp. (2015). Stata Statistical Software: Release 14. College Station, TX: StataCorp LP. Stettler, N., Kumanyika, S. K., Katz, S. H., Zemel, B. S., & Stallings, V. A. (2003). Rapid weight gain during infancy and obesity in young adulthood in a cohort of African Americans. American Journal of Clinical Nutrition, 77, 1374-1378. Taveras, E. M., Gilman, M. W., Kleinman, K., Rich-Edwards, J. W., & Rifas-Shiman, S. L. (2010). Racial/ethnic differences in early life-risk factors for childhood obesity. Pediatrics, 125, 686-695. Taveras, E. M., Rifas-Shiman, S. L., Belfort, M. B., Kleinman, K. P., Oken, E., & Gillman, M. W. (2009). Weight status in the first 6 months of life and obesity at 3 years of age. Pediatrics, 123, 1177-1183. Taveras, E. M., Rifas-Shiman, S. L., Sherry, B., Oken, E., Haines, J., Kleinman, K., Rich-Edwards, J. W., & Gillman, M. W. (2011). Crossing growth percentiles in infancy and risk of obesity in childhood. Archives of Pediatrics and Adolescent Medicine, 165(11), 993-998. Woo Baidal, J. A., Locks, L. M., Cheng, E. R., Blake-Lamb, T. L., Perkins, M. E., & Taveras, E. M. (2016). Risk factors for childhood obesity in the first 1,000 days: A systematic review. American Journal of Preventive Medicine, 50, 761779. World Health Organization. (2017). Global strategy on diet, physical activity and health. Why does childhood overweight and obesity matter? Geneva, Switzerland: Author. Retrieved from http:// www.who.int/dietphysicalactivity/childhood_consequences/ en/
-/- 2017
7