Title of project: Influence of Immigration on Child Feeding by R2k7n3e5


									Title of project: Influence of Immigration on Child Feeding Practices that Promote Obesity
Lucia Lynn Kaiser, Co-PI University of California at Davis, Department of Nutrition
Luz Elvia Vera Becerra, Co-PI University of Guanajuato, CA of Nutrition, Dept. Of Medicine and Nutrition
Martha Lopez, Emeritus University of California Cooperative Extension, Ventura County

Obesity increases the risk of chronic health problems, including heart disease, cardiovascular disease, and
type 2 diabetes. Even in children who are overweight, risk factors for chronic disease, including insulin
resistance and high blood pressure, is present 1. Moreover, childhood obesity tracks into adulthood.

Mexican immigrant children may be particularly vulnerable to obesigenic factors. Immigration and
acculturation to the US food system has been associated with many dietary and lifestyle changes,
including higher frequency of consuming restaurant food2, higher fat intake, more sweetened beverages
and less physical activity3. Mexican immigrant parents who experienced food deprivation in the past may
be more indulgent or permissive in their child feeding practices, compared to parents who did not
experience hunger 4-5.Iin other words household food security status may be related to parenting attitudes6
and the nutritional status of Mexican American children7.

Parents may not be aware that their children are becoming overweight or concerned about the potential
consequences. Research suggests that minority mothers’ perceptions of children’s body size may be
biased, resulting in a misperception of their child’s weight status-9.

Since obesity in adults is very difficult to treat, early prevention is critical. Effective prevention efforts
are based on timely data regarding child feeding practices, cultural attitudes about body weight and
health, and current weight status. Intervention researchers may want to focus on teaching parents how to
monitor and positively reinforce their children’s health behavior10. The purpose of this research was to
compare cultural attitudes and beliefs, child feeding practices, and overweight status of children (ages one
to six years) who are living in immigrant households in California with a similar cohort of children, living
in Mexican communities with high levels of out-migration to the US. This study addresses the next
research questions:
       How does prevalence of overweight differ among Mexican children living in California and
       Are immigrant mothers in California more likely to identify correctly their child’s weight status
        (as normal or overweight), compared to mothers in Mexico?
       How does use of permissive and controlling child feeding practices differ in the two populations?
       How are child feeding practices related to overweight in the two populations?


Design and Subjects
Overview of the Study and the Collaboration: The study involved a cross-sectional design, with
interviews conducted in Ventura County, California and the municipality of Cueramaro, Guanajuato
Mexico from April to December, 2006.

Human Subjects Protocols: The protocol was reviewed and approved by the Institutional Review Board at
the UC Davis and the Committee of Bioethics at the University of Guanajuato. All subjects provided
informed consent and signed approved consent forms.

Selection of Sites in Guanajuato: The Social Research Center of the University of Guanajuato provided
important information on the municipalities in the state of Guanajuato with high index of migration, as
well as indicators of socioeconomics status. With this information, the Co-PIs asked for support from the
Ministry of Health in Guanajuato, explaining the purpose of project and asking their opinion about which
municipalities would best meet the criteria for this project. Specifically, these criteria included
marginalized and with high-out migration, particularly to areas near Oxnard in Ventura County
California. The Co-PIs visited in the candidate municipality, where they talked with staff from the local
hospital and the health center. In a later meeting, the Mexican Co-PI obtained authorization for work the
fieldwork of the project. Finally, the village of San Gregorio and the nearby town of Cueramaro were
selected for the survey in Mexico.

Recruitment and Interviewing Procedures:
United States: We contacted community agencies to assist in reaching immigrant families living in
Ventura County. UC Cooperative Extension staff contacted more than 40 different agencies in Ventura
County to recruit subjects for the study. These contacts included schools, apartment complexes, public
health programs, clinics, community centers, street fairs, and immigrant services agencies. More than
2200 fliers were distributed through these contacts, and several presentations were made at parent
meetings. We were able to recruit some subjects from the Food Stamp Nutrition Education Program.
The final sample included 96 immigrant households from Ventura County.

Mexico The Mexican researchers presented the project to the local hospital and health center staff,
including those in charge of social work and community nursing so that they could inform and promote
the participation in the community. Strategies within the village and town varied somewhat due to the size
of the community.
San Gregorio: This community has 180 families and a health center. All members of the community
meeting the criteria for this study (have at least one child 1-6 years and a family member in the US) were
included (91 families). For recruitment the following strategies were used:
    1. The nurse in charge of the health, with 25 years of working and living in the community, helped
        identify many families;
    2. Project staff visited the kindergarten and the elementary school. With the teacher’s help, they
        were able to identify children of the proper age for the study and speak to their mothers to invite
        them to participate;
    3. Staff went house by house to invite people to participate; and
    4. Some came to the health center when they were referred by relatives or neighbors.

Cueramaro: Being the main town in this municipality, this community has approximately 1500 families.
The local hospital provided a room to conduct the interviews and record measurements. The final sample
included 109 families in this town. The recruitment strategies included the following:
    1. Placing flyers on the main places of the city;
    2. Visiting the four pre-elementary and three primary schools of community to talk with the teachers
        and the mothers of the children;
    3. Asking any mothers who attended the hospital if they had children within the age range and then
        referring them to the staff project;
    4. Reviewing hospital files to identify families with children within the age range and with fathers
        who are migrants to the US; and
    5. Reaching those who came to the hospital, based on referrals of relatives or neighbors.

Staff Training: The PIs participated in both staff training events to standardize procedures across the two
countries. They prepared a training manual, with detailed instructions on recruitment, interview
techniques, anthropometry, human subjects, and general procedures. In California one bilingual staff
member was hired and trained, along with her supervisor (ML) and collaborator on this project, in a two-
day training held in March 2006. LK made an additional trip one month later to observe and train during
additional household interviews. In Mexico four Mexican nutrition undergraduate students participated in
five day training event, held in May at the School of Medicine in the University of Guanajuato. The
training covered all procedures, including the recruitment of subjects, content of the interview, and the
anthropometric evaluation. We performed a training exercise to standardize the anthropometric measures
with children between the 2 to 6 years of age (WHO, 2006). The part related to interview techniques was
conducted by a psychologist in the School of Medicine.

Data Collection: Data were collected mostly at home in California or a Health Center in Mexico.

Interview instrument: The interview instrument had a mixed response format with some answers being
YES or NO, some of numerical rank (eg, 2 to 3 times per week, 1 time per month), and some open-ended
with very short answers (eg, how many hours did your child play outside yesterday?). The food security
questions were taken from instruments previously validated in California (Kuyper, 2006 and Kaiser et al,
2003). New questions related to child feeding practices and maternal attitudes or concerns regarding
weight were developed and reviewed by all members of the research team. LVB tested the wording of the
questions with ten Spanish-speaking immigrant families living in Sacramento. The nutrition students in
Mexico made further suggestions on revision of the wording after testing the instrument in the village. All
interviews in both countries were done in Spanish. The interview includes different sections briefly
described next.
1) Selection questions, six items that explored who were eligible to be a study participant, in California as
much Guanajuato, questions adjusted according to the case for both cities, inclusion criteria in both sides
of the border includes families with at least one children 1 to 6 years, where father or mother were born in
Guanajuato or Jalisco, besides in Mexico those families where father or other member of the family in he
household has been working in United States, exclusion criteria include children whit some current
disease that could influence in children’s weight or food behavior.
2) Demographics, four items requested information of the interviewed, the family and the people who live
in their home;
3) Dietetic evaluation, four items related to the current children feeding behavior and 24 hour dietary
recall hours were asked twice, at the first encounter and during the fallow-up one week or lately;
4) Infant feeding practices, eight items that explored child’s birth weight and maternal perception of their
child’s birth weight. Items explored also past of infant feeding practices as breast-feeding history and
introduction of solid foods;
5)Child feeding practices, nine items that explored parent feeding style and general child feeding
practices, were including in this section maternal perception of their current child’s weight and also
requested medical advice about child’s weight.;
6) Toddlers feeding practices, six items exclusive for small children behaviors (children 12 to 24 months
at the survey date), for this group were included details concerning weaning;
7) Young children feeding practices, ten items exclusive for young children food behaviors (children
older than 24 months), in this section questions about physical activity were included;
8) Food frequency questionnaire, thirty four items requested how often some Mexican and United States
food items were consumed. In this section were included some familiar food behaviors;
9) United Stated Department of Agriculture (USDA).Household food security was measured using the
household food security instrument of the USDA, Spanish Version (18 items) and
10) Lastly more information about household were assessed, were included maternal educational status,
monthly income, remittances and social aid.

Anthropometry: The research staff weighed and measured all children, ages one to six years, living in the
households at the time of the study. Portable Seca scales and stadiometers were used in the study
(Perspective Enterprises, Kalamazoo MI). In the training (described below), staff standardized their
measurements, following recommended procedures by the World Health Organization (WHO). For
children younger than 60 months old the new WHO growth standards software
(http://www.who.int/childgrowth/en/) was used to calculate the Body Mass Index-for-age (BMI) z-scores
and the Height-for-age (HA) z-scores, which are indicators of body fatness and linear growth,
respectively. Length, rather than height, was measured for all children less than 2 years. For children
older than 60 months Centers for Disease Control (CDC) charts were used to evaluate child growth.
Height and weight of the mothers (or caregiver) were measured and used to calculate the maternal BMI.

Data Handling and Analysis: Data analysis, using SAS software (version 8, SAS Institute, Cary, NC)
was done. Initial analyses, using descriptive statistics including t-tests, chi-square, and Pearson’s
correlations , have been done to explore general infant and child feeding practices. We have used
analysis of covariance to compare BMI-for age and Height-for-age z-scores between the two countries
and among the three sites, adjusting for differences in age structure of the samples. We will use
multivariate logistic regression to determine possible risk factors for been classified as overweight in
either sides of the border.
Preliminary results:

Characteristics of the sample: Table 1 shows the characteristics of the families in Ventura County and
Guanajuato, Mexico.

Prevalence of overweight and stunting :( Figures 1-2) Using the new World Health Organization growth
standards, we examined the weight status of children, ages 1-2 years (n=47). Based on the body mass
index for age, Figure 1 shows the mean zscores by country. Overweight is defined as BMI zscore between
1.036 - 1.640; obese is BMI zscore > 1.645). There are no significant differences between the toddler-age
children from Ventura and Guanajuato based on height-for-age Z-score (HAZ), weight-for-height Z-
score (WHZ) or BMI Z-score (BMIZ). Although the prevalence of children with low HAZ is not
significantly different between the children from Guanajuato and Ventura, the mean height-for-age tends
to be lower in Mexico.
The prevalence of children 2 to 6 years who are classified as overweight (BMI between 1.036 - 1.640), is
significantly higher among the Ventura children, compared to the Mexican children, but there are no
significant differences in the HAZ.

Maternal perception of weight status: As shown in Table 2, there were no significant differences between
Mexican and Californian mothers in their ability to correctly identify their overweight child as being
overweight. Moreover, there were no differences in the percentage of mothers reporting ever having been
told by a health professional that their child is overweight. However, Mexican mothers are more worried
about their child becoming either underweight or overweight, compared to the California mothers.

Infant and child feeding practices: Infant and child feeding practices are summarized in Tables 3.
Breastfeeding rates (any and exclusive breastfeeding) are similar in the two populations but the California
women stopped breastfeeding earlier than the Mexican mothers. The practice of giving sweetened
beverages or teas to infants is significantly more common in Mexico than in California.

In the future, we will analyze the relationships between feeding practices and overweight. An initial
observation is that exclusive breastfeeding in the first month of life is marginally related to lower
prevalence of weight >97th percentile at 12-60 months of age (p=0.10).
Table 1: Characteristics of the sample (n=296)
                                          CALIFORNIA             MEXICO
                                              (n = 94)           (n = 200)
Mean age of study child (mo)               50.8 (±19.59)         45.5 (±20)      (NS)

Gender of target child (% male)              51.06 %              45.05 %        (NS)

Household size (no.)                        4.6 ( + 1.1)         6.0 (+ 2.4)     P=0.0001

Educational level of mother (yrs)           8.4 (± 2.8)          5.9 (± 3.0)     P=0.0001

Food assistance
WIC (%)                                      57.9 (55)              NA
Food stamps (%)                              21.0 (20)              NA
School meals (%)                             56.8 (54)              NA

Table 2. Maternal perceptions and concern related to child’s growth and size for children
                                                CALIFORNIA             MEXICO
                                                     % (n)               % (n)
Worried about child becoming underweight           58.6 (34)           94.4(134)             p=.0001
Worried about child becoming overweight            68.0 (40)           88.0 (125)             P=.001
Among mothers with overweight children (as         28.89(13)           26.67(12)               (NS)
measured in study), % of mothers recognizing
child as being overweight (n=90)
Among mothers with overweight children (as         17.78 (8)            17.78(8)               (NS)
measured in study), % of mothers ever told by
MD that child is overweight (n=90)

Table 3. Infant feeding practices
                                                    CALIFORNIA           MEXICO
                                                         n =95            n = 201
Any breastfeeding % (n)                             82.1 (78)        87.0(174)              (NS)
Exclusive breastfeeding (EBF) for 1 month % (n)     77.9 (74)        82.6 (161)             (NS)
Any formula in first 12 months % (n)                71.6 (68)        58.5(116)              P=0.12
Any juice in first 12 months % (n)                  76.8 (73)        82.0(164)              (NS)
Cow’s milk before 12 mo of age % (n)                15.8 (15)        34.3(68)               p=.01
Teas/sweetened beverages (including soda) % (n)     30.5(29)         76.0(152)              P=.0001
Duration of breastfeeding (mo) Mean + SD            8.3(+7.4)        11.0( +7.7)            P=.04
Figure 1. Growth status of children 1 to 2 years (n=47)

                    1                            NS            NS
  Z-score                                                                            CA
                  -0.6           NS
                  -0.8           S
                            HAZ              WHZ                BMIZ

                                                                          •   2005 WHO References

Figure 2. Growth status of children 2 to 6 years (n=352)

                  1.6                                          P < .001
                  0.6                                                                    MX
Z-score           0.4                NS                                                  CA
                                 HAZ                       BMIZ
                                                           •    WHO References for children 24 to 60 mo
                                                               •  CDC References for children > 60 mo

Although the prevalence of overweight appears greater among young Mexican immigrant children in
California than their counterparts in Mexico, US health professionals also need to consider the emerging
problem of childhood obesity in marginalized Mexican communities with high levels of out-migration. In
poor communities of Mexico, high rates of both overweight and stunting are occurring, and these trends
have implications for US programs serving newly arriving families from Mexico.

Among immigrant women in California and their counterparts in Mexico, early infant feeding practices
deviate substantially from expert recommendations, as is exclusive breast feeding during the first six
months. Between 30-76% of mothers give sweetened beverages to their infants. Between 15-34% give
cow’s milk before 12 months of age. Exclusive breastfeeding is recommended for the first 6 months of
life, but even in the month, 78-83% of mothers are already providing other fluids to their babies.

Concern about children becoming overweight appears to exceed concerns about underweight. Despite our
expectations, California mothers are not more likely to be worried about overweight or able to correctly
identify their child as overweight, compared to Mexican mothers. Most mothers of overweight children,
in either country, had not been informed that their child is overweight. Further analysis will determine
how practices, concerns, and awareness may differ among the Mexican and Californian samples. We also
will explore in the US sample the effects of participation in the WIC programs on these variables.

In the future, we will analyze the relationships between feeding practices and household food security
status, and overweight.

Stunting prevalence is not significant different among both sides of the border in this sample, while the
prevalence of overweight in children 2 to 6 years is significant greater in Mexican immigrant children in
California than their counterparts in Mexico. There are no differences among mothers in either country
in being able to correctly identify their children as overweight, but Mexican mothers in California are less
likely to be worried about their children becoming overweight. Most mothers of overweight children, in
either country, had not been informed that their child is overweight.

   1. Perichart-Perera O, Balas-Nakash M, Schiffman-Selechnik E, Barbaro-Dosal A, Vadillo-Ortega
      F. Obesity increases metabolic syndrome risk factors in school aged children from urban school
      in Mexico City. J Am Diet Assoc 2007; 107 (1): 81-91.

   2. Durkensen, S.C., Elder J.P., Arredondo, E.M., Ayala, G.X., Slymen, D.J., Campbell, N.R., et al.
      (2007). Family Restaurant choices are associated with child and adult overweight status in
      Mexican American families. Journal of the American Dietetic Association, 107(5), 849-853.

   3. Ariza, A.J., Chen, E.H., Binns, H.J., & Christoffel, K.K. (2004). Risk factors for overweight in
      five-to-six year-old Hispanic American children: A pilot study. Journal of Human Health, 81
      (11), 150-161.

   4. Kuyper EM, Espinosa-Hall G, Lamp CL, Martin AC, Metz DL, Smith D, Townsend MS, Kaiser
      LL. Development of a tool to assess past food insecurity of immigrant Latino mothers. JNEB
      2006; 38: 378-382

   5. Kaiser LL, Melgar-Quiñonez HR, Townsend MS, Nicholson Y, Fujii ML, Martin AC, Lamp
      CL.2003. Food insecurity and food supplies in Latino households with young children. Journal of
      Nutrition Education and Behavior 35: 148-153.

   6. Mathensen, D.M., Robinson, T.N., Varady, A., & Killen, J.D. (2006). Do Mexican American
      mother’s food-related parenting practices influence their children’s weight and dietary intake?
      Journal of the American Dietetic Association, 106 (11), 1861-1865.

   7. Mathensen, D.M., Varady, J.,Varady,A., & Killen, J.D.,(2002) Household food security and
      nutritional status of Hispanic children in the fifth grade, American Journal of Clinical Nutrition,
      76(1), 210-217.

   8. Baughcum, A.E., Chamberlin, L.A., Deeks, C.M., Powers, S.W., & Whitaker, R.C. (2000).
      Maternal perceptions of overweight preschool children. Pediatrics, 106(6), 1380-1387.

   9. Killion, L., Hughe, S.O., Wendt, J.C., Pease, D., & Nicklas, T.A. (2006). Minority mothers’
      perceptions of children’s body size. International Journal of Pediatric Obesity, 1(2), 96-102.

   10. Arredondo EM, Elder JP, Ayala Guadalupe X, Campbell N, Baquero B and Duerksen S. Is
       parenting style related to children’s healthy eating and physical activity in Latino families?
       Health education research 2006; 21(6), 862-871.

To top