Maternal Cultural Orientation and Child Growth in New Zealand Pacific Families

Abstract

Background:

In New Zealand (NZ), children of Pacific heritage are generally born heavier and gain weight more quickly than non-Pacific children. Immigrants' health is commonly expected to converge with the host population through acculturation. The aim of this analysis was to examine whether mothers' acculturation is associated with less-rapid weight gain in NZ Pacific children, and whether this differs by mothers' nativity.

Methods:

In a birth cohort of 1249 children followed 2000–2011, birth weight and weight and standing height, measured at years 2, 4, 6, 9, and 11, were quantified as sex- and age-specific weight (weight-for-age; WFA) and BMI z-scores. Maternal acculturation (range, 11–54) was assessed at baseline and years 4, 6, and 11.

Results:

In adjusted models using generalized estimating equations to account for repeated measures, maternal acculturation was not significantly associated with children's WFA or BMI z-scores overall. In stratified analyses, change in maternal acculturation score was inversely associated with WFA z-score change among children of NZ-born, but not immigrant, mothers (beta=−0.021; 95% confidence interval, −0.036 to −0.007; p=0.006; interaction, p=0.005).

Conclusions:

Our study provides the first evidence in a longitudinal sample that changes in maternal acculturation can influence children's growth, suggesting the importance of lifestyle or behavioral factors related to a mother's cultural orientation. Given the high risk of obesity and its related conditions in the NZ Pacific population, critical next steps are to identify mediating factors, as well as to understand the processes influencing growth among children of immigrant mothers.

Introduction

In New Zealand (NZ), children of Pacific heritage generally are born heavier and gain weight more quickly than non-Pacific children,¹ likely contributing to greater morbidity and substantial health disparities throughout life.²

The Pacific population, representing 7% of the total NZ population, includes a diversity of cultures and migration histories, with most originating from Samoa, Tonga, and Fiji, as well as the Cook Islands, Niue, and Tokelau—three nations whose residents also have held NZ citizenship and unrestricted settlement rights.³ Despite the diversity of origins, prevalence of adult overweight or obesity in these populations is uniformly high, with prevalence in Samoa, Tonga, and the Cooks Islands being well above 80%.⁴

Although a genetic basis is possible,⁵ the rapid increase in overweight and obesity prevalence among Pacific populations in the latter half of the 20th century⁶ suggests the importance of nongenetic factors. Among infants and young children, these include the child's diet and activity level, as well as characteristics and behaviors of the mother, including her adiposity level, smoking status, and breastfeeding practices.^7,8 Maternal acculturation may therefore influence a child's growth trajectory in immigrant populations.

Although the impact of acculturation on health outcomes among immigrants may depend on economic development of the country of origin as well as of the destination country,⁹ immigrants' health status is commonly expected to converge with the host population because of acculturation, in which immigrants adopt the host population's lifestyle, values, and behaviors. Numerous studies indicate that diet and overweight/obesity prevalence differ with acculturation, represented by nativity, length of residence, or scales considering language and other social aspects.^10–12 Further, studies among US immigrant women suggest that child feeding practices change after mothers' migration,¹³ and that mothers' acculturation is associated with feeding practices as well as children's dietary habits and intake,^14–17 although findings for children's risk of overweight are inconsistent.^14,18–26 All previous studies were cross-sectional, however, and could not examine whether maternal acculturation is associated with child's growth over time. An association of acculturation with overweight/obesity may also not be linear; studies among US immigrants, including children, have shown the highest risk of obesity in the second generation of immigrants (i.e., first US-born generation).^27–32

The aim of this analysis was to determine whether the mother's acculturation is related to child growth in a longitudinal sample of Pacific children in NZ.³³ Growth was quantified as weight-for-age (WFA) and BMI-for-age z-scores based on World Health Organization (WHO) growth reference curves.³⁴ According to conventional understanding of immigrant health (i.e., convergence with acculturation), mothers' child-rearing and feeding behaviors should change with acculturation to converge with the host population. Thus, in this sample of Pacific children, we hypothesized that mothers' greater NZ cultural orientation would be associated with children's less-rapid growth. Because adopting a new culture is commonly assumed to be at the expense of the traditional culture, we also examined the association of mothers' Pacific cultural orientation with children's growth. Finally, because of evidence of different patterns by generational status, we conducted additional analyses to examine NZ-born and immigrant mothers separately.

Methods

Sample

The Pacific Islands Families Study established a cohort of Pacific infants born at the Middlemore Hospital in South Auckland between March and December 2000.³³ An infant was eligible if at least one parent identified as being of Pacific ethnicity and was a permanent NZ resident. Notably, 66% of the NZ Pacific population resides in the Auckland region,³⁵ and South Auckland contains the highest density of Pacific residents in NZ.^3,36 Of 1477 eligible mothers, 1376 (93%) agreed to participate, for a total of 1398 infants. All data collection procedures were reviewed and approved by the National Ethics Advisory Committee, in accord with the Helsinki Declaration of 1975, as revised in 2008. Mothers gave their written informed consent for their own and for their children's participation.

One twin from each of 22 twin pairs was randomly selected for inclusion in these analyses. Analyses excluded infants born <37 weeks of gestation (n=107), missing birth weight data (n=19), or whose mother did not provide her age at baseline, an adjustment variable in regression models (n=1), leaving a sample of 1249.

Data Collection

At each of seven measurement waves (baseline [child's age ∼6 weeks] and 1, 2, 4, 6, 9, and 11 years), researchers conducted detailed, in-person interviews with mothers on sociodemographic, psychosocial, and health factors, including education, employment, ethnic and cultural identification, place of birth, length of NZ residence, smoking during pregnancy, and breastfeeding at 6 weeks. Maternal acculturation scores were obtained at baseline and years 4, 6, and 11, using an adaptation of the General Ethnicity Questionnaire,³⁷ modified to assess NZ or Pacific cultural orientation. The original 38-item scale was shortened to exclude items of limited relevance to NZ and reduce participant burden. Three items relating to social affiliation (contact with Pacific/non-Pacific family and relatives as well as attendance at Pacific/non-Pacific church) and activities (participation in Pacific/NZ sports and recreation) were added for cultural relevance. In all, 11 items were retained (Table 1), each scored from 1 to 5 based on the respondent's level of agreement with the statement, for a possible total scale range from 11 to 55. The internal consistency of both scales was acceptable (Cronbach's alpha=0.81 and 0.83 for NZ and Pacific cultural orientation scales, respectively).³⁸ Further, prestudy focus groups, input from the study's Pacific Advisory Board, and subsequent analyses using these variables^38–40 support the scales' face and construct validity.

Table 1.

Items Included in General Ethnicity Questionnaire, Used to Assess New Zealand and Pacific Cultural Orientation^a

1. I was brought up the New Zealand/Pasifika way.

2. I am familiar with New Zealand/Pasifika cultural practices and customs.

3. I can understand English/a Pasifika language well.

4. I have several non-Pasifika/Pasifika friends.

5. Most of my friends speak English/a Pasifika language.

6. I participate in NZ/Pasifika sports and recreation.

7. I speak English/a Pasifika language.

8. I have contact with non-Pasifika/Pasifika families and relatives.

9. I eat non-Pasifika/Pasifika food.

10. I visit Western-trained doctors/a traditional Pasifika healer when I have an illness.

11. I go to a church that is mostly attended by non-Pasifika/Pasifika people.

Response options for items 1–5 were 1 (strongly disagree), 2 (disagree), 3 (neither disagree nor agree), 4 (agree), or 5 (strongly agree). Response options for items 6–11 were 1 (not at all), 2 (a little), 3 (somewhat), 4 (quite a lot), or 5 (a lot). Responses were summed over all 11 items for a minimum score of 11 and a possible maximum score of 55.

NZ, New Zealand.

Child's birth weight was obtained from hospital records. Beginning in year 2, trained research staff recorded the child's weight in light clothing and standing height without shoes using standardized procedures⁴¹ (PIFS Child Measurement Manual available upon request). Scales (Wedderburn, Soehlne) and stadiometers (Mentone) were calibrated at the beginning of each measurement wave and checked routinely. Research staff also underwent periodic retraining during the data collection period, including standardization exercises beginning in year 4.

Statistical Analysis

Sex- and age-specific z-scores for weight and BMI were calculated based on the WHO Growth Standards⁴² for ages 0–5 years and the WHO Growth Reference⁴³ for ages 5+. Because they are a less-useful indicator in older children,⁴⁴ WFA z-scores were calculated only from birth to 9 years. BMI-for-age z-scores were calculated for ages 2–11 years because length at birth was not recorded for the study. Maternal acculturation at years 2 and 9 was imputed using the average of each mother's cultural orientation scores at birth and year 4 for year 2 and at years 6 and 11 for year 9. Sensitivity analyses excluding imputed values did not change the major findings.

In descriptive analyses, NZ-born and immigrant mothers were compared using Cochran-Mantel-Haenszel's test statistics for categorical variables and t-tests for continuous variables. Participants with data from at least one follow-up visit (n=1002) were also compared with those with no follow-up data (n=247) in a similar fashion.

We used generalized estimating equations (GEEs) with an exchangeable correlation matrix to model associations between mothers' cultural orientation (NZ and Pacific separately) and children's WFA and BMI z-scores, using 4968 observations from 1249 participants at baseline and over the 2-, 4-, 6-, 9-, and 11-year follow-ups. To assess change over time, mothers' cultural orientation scores were modeled in GEEs as a function of years in study, included as a time-dependent, continuous, linear predictor, controlling for mother's age at baseline. Children's WFA and BMI z-scores were modeled in GEEs as a function of the child's age in years.

Next, we modeled children's WFA z-scores (birth to year 9) as a function of mother's NZ cultural orientation scores and child's age (years) and sex, mother's age (years) and level of education (no formal qualifications, secondary school, or postsecondary) at baseline, and mother's nativity (NZ or elsewhere) and ethnicity (Samoan, Tongan, Cook Island or other Pacific ethnicity, or non-Pacific).

To distinguish between intraindividual changes in cultural orientation over time and existing, interindividual differences in cultural orientation among mothers, we also modeled, among participants with at least one follow-up, change in child's z-score between time points as a function of change in mother's cultural orientation score between time points. Change was calculated by subtracting the value of each earlier time point from the value at the subsequent time point.

Additional analyses considered infant's low birth weight (<2500 g), breast or formula feeding at 6 weeks, maternal smoking during pregnancy, mother's parity at baseline, and baseline household income as potential confounders; adjusted for NZ and Pacific cultural orientation scores; and excluded non-Pacific mothers or mothers with diagnosed gestational or existing diabetes. We used a similar strategy to model children's WFA z-scores as a function of mother's Pacific cultural orientation score and their BMI z-scores (ages 2–11 years) as a function of maternal NZ or Pacific cultural orientation score with the same covariates.

We explored differences by mother's nativity by including a mother's cultural orientation score (continuous)×place of birth (NZ or elsewhere) interaction term in models and by comparing effect estimates stratified on mother's place of birth. In additional analyses, we stratified further on immigrant mothers' age at migration to distinguish women who migrated earlier or later in life, using cutoffs ranging from 5 to 10 years of age. All statistical analyses were conducted using SAS software (version 9.2, 2008; SAS Institute Inc., Cary, NC).

Results

Sample Description

Most (67%) mothers were born outside of NZ (Table 2). Half were of Samoan ethnicity, 21% Tongan, with another 25% reporting other Pacific ethnicities, especially Cook Island. Of the 7% non-Pacific mothers, more than half (50 of 85; 59%) identified as Maori. NZ-born mothers were, on average, younger than immigrant mothers and more likely to be of non-Pacific ethnicity and to have achieved postsecondary educational qualifications (all p<0.0001). They also had higher mean NZ cultural orientation scores and lower mean Pacific cultural orientation scores throughout the study (all p<0.0001). Among the children, mean WFA and BMI z-scores were above 1 for the overall sample at every measurement wave. Children of NZ-born mothers were lighter at birth (p<0.0001), but similar to children of immigrant mothers with respect to WFA and BMI z-scores from age 2 years on.

Table 2.

Description of Mothers and Children in the Pacific Islands Families Cohort (n=1249)

	Overall sample	Immigrant mothers	New Zealand–born mothers
	(n=1249)	(n=834)	(n=415)	p value
Mothers
Mean (SD) age (years) at baseline	28.5 (6.1)	29.9 (6.0)	25.4 (5.3)	<0.0001
Born in New Zealand, % (n)	33 (415)	—	—
Ethnicity, % (n)
Samoan	47 (586)	51 (427)	39 (159)	<0.0001
Tongan	21 (258)	27 (228)	7 (30)
Cook Island or other Pacific ethnicity	25 (313)	21 (175)	33 (138)
Non-Pacific	7 (92)	0.5 (4)	21 (88)
Education level, % (n)				<0.0001
No formal qualifications	39 (483)	43 (357)	30 (126)
Secondary school qualification	34 (423)	34 (284)	33 (139)
Postsecondary qualification	28 (343)	23 (193)	36 (150)
Partnered or married, % (n)	80 (1007)	83 (693)	75 (314)	0.002
Mean (SD) cultural orientation score
NZ
Birth	(n=1237)35 (8)	(n=824)32 (7)	(n=413)41 (6)	<0.0001
Year 2	(n=797)35 (8)^a	(n=527)32 (7)^a	(n=270)41 (5)^a	<0.0001
Year 4	(n=774)35 (9)	(n=518)31 (9)	(n=256)41 (6)	<0.0001
Year 6	(n=763)36 (10)	(n=509)33 (9)	(n=254)43 (6)	<0.0001
Year 9	(n=585)37 (8)^a	(n=392)34 (8)^a	(n=193)44 (5)^a	<0.0001
Year 11	(n=793)38 (9)	(n=526)35 (9)	(n=267)44 (6)	<0.0001
Pacific
Birth	(n=1241)43 (8)	(n=829)46 (6)	(n=412)37 (9)	<0.0001
Year 2	(n=801)43 (8)^a	(n=533)46 (6)^a	(n=268)36 (7)^a	<0.0001
Year 4	(n=776)42 (9)	(n=521)46 (7)	(n=255)36 (8)	<0.0001
Year 6	(n=763)43 (9)	(n=509)45 (7)	(n=254)37 (9)	<0.0001
Year 9	(n=585)42 (8)^a	(n=392)45 (6)^a	(n=193)36 (8)^a	<0.0001
Year 11	(n=793)42 (8)	(n=526)45 (6)	(n=267)36 (8)	<0.0001
Children
Female, % (n)	48 (599)	49 (408)	46 (191)	0.33
Mean (SD) weight-for-age z-scores
Birth	(n=1244)0.70 (1.05)	(n=831)0.82 (1.02)	(n=413)0.46 (1.05)	<0.0001
Year 2	(n=804)1.34 (1.15)	(n=534)1.35 (1.09)	(n=270)1.32 (1.25)	0.79
Year 4	(n=777)1.53 (1.23)	(n=521)1.51 (1.18)	(n=256)1.57 (1.34)	0.55
Year 6	(n=763)1.82 (1.44)	(n=509)1.78 (1.36)	(n=254)1.90 (1.60)	0.32
Year 9	(n=537)2.12 (1.38)	(n=367)2.12 (1.36)	(n=170)2.13 (1.43)	0.98
Mean (SD) BMI-for-age z-scores
Year 2	(n=794)1.75 (1.41)	(n=529)1.76 (1.38)	(n=265)1.73 (1.48)	0.74
Year 4	(n=767)1.91 (1.40)	(n=515)1.88 (1.36)	(n=252)1.97 (1.49)	0.44
Year 6	(n=761)1.84 (1.49)	(n=508)1.79 (1.42)	(n=253)1.93 (1.64)	0.23
Year 9	(n=584)1.98 (1.31)	(n=391)1.97 (1.28)	(n=193)2.00 (1.36)	0.81
Year 11	(n=794)1.79 (1.26)	(n=527)1.80 (1.27)	(n=267)1.78 (1.25)	0.87

Values at year 2 were imputed as average of scores reported at birth and year 4. Values at year 9 were imputed as average of scores reported at years 6 and 11.

SD, standard deviation; NZ, New Zealand.

In comparisons of participants with data from at least one follow-up visit (n=1002) versus participants with baseline data only (n=247; results not shown), those with follow-up data were older at baseline (mean [SD], 28.7 [6.1] vs. 27.6 [6.1] years; p=0.02), more likely to have achieved at least secondary school qualification (63% vs. 55%; p=0.02), had higher NZ cultural orientation scores (mean [SD], 35 [8] vs. 33 [8]; p=0.0006), and had heavier children at birth (mean WFA z-score [SD], 0.73 [1.04] vs. 0.56 [1.07]; p=0.02), but nativity, ethnicity, and marital status as well as Pacific cultural orientation scores were not significantly different between the two groups.

Change in Mothers' Cultural Orientation and Children's z-Scores Over Time

In the overall sample, mean NZ cultural orientation score increased from 35 (range, 11–54; n=1237) at baseline to 38 (range, 13–55; n=793) at year 11, whereas mean Pacific cultural orientation scores remained relatively stable at 42–43 (range, 11–55). In linear regression models, NZ cultural orientation scores increased in the overall sample by 0.24 each year (p<0.0001); Pacific cultural orientation declined slightly (−0.06 per year; p=0.002). Change over time did not differ significantly between NZ-born and immigrant mothers.

Mean WFA z-score increased from 0.70 (range, −3.53 to 3.78) at birth to 2.12 (range, −2.05 to 7.35) at year 9 (Table 2). Mean BMI z-scores did not increase linearly between ages 2–11 years, but ranged from 1.75 to 1.98 over the course of the study. In linear regression modeling, WFA z-scores increased significantly as a function of child's age, with a steeper increase in boys (0.19 per year; p<0.0001) than girls (0.11 per year; p<0.0001; interaction, p<0.0001). BMI z-scores increased significantly as a function of time only in boys (0.02 per year; p=0.02), but not girls (−0.005; p=0.53; interaction, p=0.03).

Mothers' Cultural Orientation in Relation to Children's z-Scores

In the overall sample, neither NZ nor Pacific cultural orientation scores in mothers were significantly associated with children's WFA z-scores (Table 3). Stratified analyses suggested a difference depending on mothers' nativity. Among NZ-born mothers, each one-unit increment in mother's NZ cultural orientation score was associated with a lower child's WFA z-score of 0.014 (95% confidence interval [CI], −0.028 to −0.005; p=0.048). The association was more pronounced when change in WFA z-scores was modeled as a function of change in mothers' cultural orientation scores. Each one-unit change in mother's NZ cultural orientation score was associated with a 0.021 (95% CI, −0.036 to −0.007; p=0.006) smaller change in child's WFA z-score over time, but again, only in NZ-born mothers (interaction, p=0.005). In further stratified analyses among immigrant mothers, neither NZ nor Pacific cultural orientation score was associated with children's WFA z-scores, even if mothers had migrated at an early age, whether defined as <10, 7, or 5 years old.

Table 3.

Associations between Mothers' Cultural Orientation Scores and Children's WFA and BMI-for-Age z-Scores, and between Change in Mothers' Cultural Orientation Scores and Change in Children's WFA and BMI-for-Age z-Scores between Measurement Waves, Estimated Using Generalized Estimating Equations

	Associations between mother's cultural orientation scores and child growth measures^a			Associations between change in mother's cultural orientation scores and change in child growth between measurement waves^b
	All	Immigrant	NZ-born	All	Immigrant	NZ-born
WFA z-scores^c
N	1249	834	415	991	652	339
NZ cultural orientation	−0.0003 (0.003)0.94	0.003 (0.004)0.39	−0.014 (0.007) 0.05	−0.003 (0.003)0.43	0.003 (0.004)0.34	−0.021 (0.008) 0.006
Interaction p value		0.14			0.005
Pacific cultural orientation	0.003 (0.004)0.47	0.002 (0.005)0.62	0.003 (0.006)0.57	−0.001 (0.004)0.77	0.002 (0.005)0.69	−0.008 (0.007)0.26
Interaction p value		0.84			0.32
BMI z-scores^d
N	1055	692	363	928	611	317
NZ cultural orientation	−0.005 (0.003)0.16	−0.004 (0.004)0.37	−0.010 (0.006)0.11	<0.0001 (0.004)0.99	0.002 (0.004)0.58	−0.008 (0.007)0.25
Interaction p value		0.27			0.24
Pacific cultural orientation	0.005 (0.004)0.14	0.003 (0.005)0.53	0.011 (0.006)0.09	−0.003 (0.004)0.45	−0.003 (0.005)0.49	−0.002 (0.007)0.77
Interaction p value		0.31			0.91

Adjusted for child's age (years) and sex, mother's age (years) and level of education (no formal qualifications, secondary school, or postsecondary) at baseline, mother's ethnicity (Samoan, Tongan, Cook Island or other Pacific ethnicity, or non-Pacific), and (in unstratified analyses) mother's nativity (NZ or other).

Adjusted for change in child's age between measurement waves, child's sex, mother's age (years) and level of education (no formal qualifications, secondary school, or postsecondary) at baseline, mother's ethnicity (Samoan, Tongan, Cook Island or other Pacific ethnicity, or non-Pacific), and (in unstratified analyses) mother's nativity (NZ or other).

Calculated for birth to age 9 years.

Calculated for ages 2–11 years.

WFA, weight-for-age; NZ, New Zealand.

Bold items indicate statistically significant findings.

Additional adjustment for low birth weight, mother's smoking during pregnancy, mother's parity at baseline, breast or formula feeding at 6 weeks, or baseline household income did not meaningfully change these results, nor did estimates change among NZ-born mothers when both NZ and Pacific cultural orientation scores were included in the model. Results were also not different when we excluded imputed values or excluded 85 mothers not of Pacific ethnicity.

Mothers' NZ cultural orientation score was not associated with children's BMI z-score (Table 3). We limited analyses on WFA z-scores to years 2–9 to determine whether the lack of association for BMI might be because of changes that occurred before year 2. The inverse association between change in NZ-born mothers' NZ cultural orientation score and change in children's WFA z-scores was still apparent, but attenuated from −0.021 to −0.010 (95% CI, −0.025 to 0.004) and no longer statistically significant (p=0.15).

Discussion

In this longitudinal sample, greater NZ cultural orientation in mothers was associated with lower WFA z-scores in children, but only for mothers born in NZ. This appeared to be owing to change in cultural orientation over time rather than to existing differences among individual mothers. Though the findings provide some evidence for an acculturation effect, the lack of association among children of immigrants, despite changes in NZ cultural orientation comparable to that in NZ-born mothers, suggests more-complex processes.

Possibly, NZ cultural orientation scores in immigrant mothers were not high enough or did not change enough over the course of the study to have an effect. However, NZ cultural orientation score was not associated with WFA, even among mothers who had migrated as children. Further analyses (not shown) also indicated that NZ cultural orientation scores increased over time in immigrant mothers as well as in NZ-born mothers, and regardless of age at migration.

A second possible explanation is that NZ cultural orientation score is associated with a different constellation of values and behaviors in immigrant, as compared to NZ-born, mothers. Body-size perceptions vary by ethnicity,⁴⁵ for example, and may change with level of acculturation, a phenomenon that may manifest more clearly in subsequent generations.⁴⁶ Changes in feeding practices with acculturation^13–17 may also be more pronounced among NZ-born mothers, although previous studies have not examined this. Conversely, for mothers not born in NZ, the affiliation with “NZ culture” may be more related to the environment and society in which they live (i.e., socioeconomically deprived neighborhoods in South Auckland with poor access to healthy food outlets) than to the values and behaviors they adopt.

Studies of US immigrants suggest that the immigrant generation is particularly vulnerable to overweight and obesity,^28–32,47 providing another possible explanation for the lack of association among immigrant mothers in this study. Prenatal influences may increase immigrant mothers' risk of adiposity or metabolic disorders, for example, which may affect her own child's birth weight and growth trajectory.^7,8 We excluded mothers with diagnosed gestational or existing diabetes in additional analyses, but under-reporting is likely,^48,49 and data on mothers' anthropometry were not available to evaluate the effect of maternal obesity.

Finally, it is possible that, for immigrant mothers, other family members play a greater role in child care so that the mother's cultural orientation has a lesser effect on the child's growth. In qualitative studies among Latina mothers, Sussner and colleagues¹³ noted the strong influence of grandparents on cultural beliefs surrounding child feeding practices, as well as resistance from older relatives with a transition away from traditional beliefs. A greater understanding of processes affecting child growth among immigrant mothers of Pacific background, such as social and family structure, feeding practices, and attitudes toward and opportunities for physical activity, merits attention in future work.

In this analysis, cultural orientation scores were not associated with BMI z-scores. Notably, the length or height measurements needed to calculate BMI were available only from age 2 years on, and other studies have shown that differences in growth trajectories emerge before age 2 years.^50,51 When we limited the WFA analysis to ages 2–9 years, associations with mothers' cultural orientation were largely attenuated, suggesting that the period of growth soon after birth is critical. Another possibility is that associations might have been harder to detect for BMI if it had been measured with more error than was WFA, particularly given that it is dependent on accurate measurement of children's height as well as weight. Finally, the difference in findings between WFA and BMI may point to differences between the two indicators. Mothers' NZ cultural orientation likely influences height as well as weight, and previous analyses show greater height-for-age in this cohort relative to international standards.^1,52 Thus, rapid growth in both of these does not manifest as rapid change in BMI. Additional research is needed to clarify the implications of rapid change in WFA, as opposed to BMI, for subsequent health. Indeed, whereas both WFA and BMI are commonly used measures to monitor child growth, both are indicators only and not health outcomes in and of themselves. Further research on Pacific child health would benefit from the direct measurement of functional outcomes, such as measures of physical function and ability.

Further work should also explore the specific behaviors, resources, or environments associated with mothers' NZ cultural orientation that more directly influence child growth. Studies in Mexican Americans have found differences in child feeding strategies and dietary intake with US acculturation,^13,14,16,17 but uncertain effects on growth and adiposity.^14,18–26 Acculturative changes may also be accompanied by changes in social or environmental resources, or in perceptions of ideal weight and other social norms. Moreover, children of immigrant mothers were heavier and had the fastest growth trajectories, but their growth was not associated with mothers' cultural orientation. Among these children, factors that influence growth trajectories remain to be identified.

The findings have implications for expectations with respect to convergence in Pacific immigrants. Previously observed patterns in other immigrant groups suggest that Pacific populations should experience a decrease in overweight and obesity with migration to NZ. Our findings add insight by providing evidence of less-rapid growth in children with mothers' change toward a NZ cultural orientation. Whether risk of obesity and its related conditions will eventually converge completely, however, is unclear.²⁷

A limitation of our study is that the sample size did not allow for more-detailed examination of individual Pacific ethnicities and cultures despite differences in their migration histories.^53,54 Residual confounding is also possible. Although many women reported a mix of ethnicities, we controlled only for the one with which the mother identified most. Further, we controlled for baseline education and household income as indicators of social class or deprivation, but detailed follow-up measures were not available. Finally, information on mother's anthropometry and metabolic status would have permitted a more-detailed examination of maternal factors in relation to children's growth. Strengths of the study include its longitudinal design and detailed, repeated measures of maternal cultural orientation as well as children's anthropometry.

Conclusions

In summary, our study provides the first evidence in a longitudinal sample that changes in maternal cultural orientation, or acculturation, can influence growth trajectories in children. Thus, whereas intervening at a young age is a sound strategy, improving trajectories in children may call for intervening with the characteristics and circumstances of the mother. Lower WFA with greater NZ cultural orientation is consistent with a theory of convergence, so that we might expect less-rapid growth in subsequent generations, and with acculturation. However, to inform public health strategies, we need a better understanding of reasons for such a convergence, with the goal of identifying when and how to intervene to facilitate the decline in risk. Given the high risk of obesity and its related conditions in the NZ Pacific population, identifying potential mediating factors and clarifying the processes that influence child growth among children of immigrant mothers are critical next steps toward improving growth trajectories among NZ children of Pacific heritage.

Footnotes

Acknowledgments

The authors thank Dr. Debbie Ryan for valuable input from her Aniva Pacific Nurses Master Class on preliminary results. This work was supported by grants from the Foundation for Research Science and Technology, the Health Research Council, the Ministry of Business, Innovation and Employment, and Fulbright New Zealand.

Author Disclosure Statement

No competing financial interests exist.

References

Rush

, Gao

, Funaki-Tahifote

, et al. Birth weight and growth trajectory to six years in Pacific children. Int J Pediatr Obes, 2010; 5:192–199.

Ministry of Health. Tupu Ola Moui: Pacific Health Chart Book 2012. Ministry of Health: Wellington, New Zealand, 2012.

Statistics New Zealand, Ministry of Pacific Island Affairs. Demographics of New Zealand's Pacific population. Statistics New Zealand: Wellington, New Zealand: 2010.

World Health Organization. Overweight and obesity, 2008. World Health Organization: Geneva, Switzerland, 2008.

Stewart

, Reed

, Rush

. Children of the outer Cook Islands have lower BMI compared to their urban peers. N Z Med J, 2014; 127:96–98.

Hodge

, Dowse

, Zimmet

, et al. Prevalence and secular trends in obesity in Pacific and Indian Ocean island populations. Obes Res, 1995; 3(Suppl 2):77s–87s.

, Han

, Zhu

, et al. Pre-pregnancy body mass index in relation to infant birth weight and offspring overweight/obesity: A systematic review and meta-analysis. PLoS One, 2013; 8:e61627.

Weng

, Redsell

, Swift

, et al. Systematic review and meta-analyses of risk factors for childhood overweight identifiable during infancy. Arch Dis Child, 2012; 97:1019–1026.

Van Hook

, Balistreri

. Immigrant generation, socioeconomic status, and economic development of countries of origin: A longitudinal study of body mass index among children. Soc Sci Med, 2007; 65:976–989.

10.

Goel

, McCarthy

, Phillips

, et al. Obesity among US immigrant subgroups by duration of residence. JAMA, 2004; 292:2860–2867.

11.

Delavari

, Sønderlund

, Swinburn

, et al. Acculturation and obesity among migrant populations in high income countries—A systematic review. BMC Public Health, 2013; 13:458.

12.

Lauderdale

, Rathouz

. Body mass index in a US national sample of Asian Americans: Effects of nativity, years since immigration and socioeconomic status. Int J Obes Relat Metab Disord, 2000; 24:1188–1194.

13.

Sussner

, Lindsay

, Greaney

, et al. The influence of immigrant status and acculturation on the development of overweight in Latino families: A qualitative study. J Immigr Minor Health, 2008; 10:497–505.

14.

Kaiser

, Melgar-Quinonez

, Lamp

, et al. Acculturation of Mexican-American mothers influences child feeding strategies. J Am Diet Assoc, 2001; 101:542–547.

15.

Park

, Paik

, Skinner

, et al. Mothers' acculturation and eating behaviors of Korean American families in California. J Nutr Educ Behav, 2003; 35:142–147.

16.

Morello

, Madanat

, Crespo

, et al. Associations among parent acculturation, child BMI, and child fruit and vegetable consumption in a Hispanic sample. J Immigr Minor Health, 2012; 14:1023–1029.

17.

Dave

, Evans

, Saunders

, et al. Associations among food insecurity, acculturation, demographic factors, and fruit and vegetable intake at home in Hispanic children. J Am Diet Assoc, 2009; 109:697–701.

18.

Chen

, Weiss

, Heyman

, et al. Risk factors for obesity and high blood pressure in Chinese American children: Maternal acculturation and children's food choices. J Immigr Minor Health, 2011; 13:268–275.

19.

Elder

, Arredondo

, Campbell

, et al. Individual, family, and community environmental correlates of obesity in Latino elementary school children. J Sch Health, 2010; 80:20–30; quiz, 53–55.

20.

Sussner

, Lindsay

, Peterson

. The influence of maternal acculturation on child body mass index at age 24 months. J Am Diet Assoc, 2009; 109:218–225.

21.

Wojcicki

, Schwartz

, Jimenez-Cruz

, et al. Acculturation, dietary practices and risk for childhood obesity in an ethnically heterogeneous population of Latino school children in the San Francisco bay area. J Immigr Minor Health, 2012; 14:533–539.

22.

Fuentes-Afflick

, Hessol

. Overweight in young Latino children. Arch Med Res, 2008; 39:511–518.

23.

Hernández-Valero

, Wilkinson

, Forman

, et al. Maternal BMI and country of birth as indicators of childhood obesity in children of Mexican origin. Obesity (Silver Spring), 2007; 15:2512–2519.

24.

Wiley

, Cloutier

, Wakefield

, et al. Acculturation determines BMI percentile and noncore food intake in Hispanic children. J Nutr, 2014; 144:305–310.

25.

Ariza

, Chen

, Binns

, et al. Risk factors for overweight in five- to six-year-old Hispanic-American children: A pilot study. J Urban Health, 2004; 81:150–161.

26.

Barroso

, Roncancio

, Hinojosa

, et al. The association between early childhood overweight and maternal factors. Child Obes, 2012; 8:449–454.

27.

Creighton

, Goldman

, Pebley

, et al. Durational and generational differences in Mexican immigrant obesity: Is acculturation the explanation?. Soc Sci Med, 2012; 75:300–310.

28.

Balistreri

, Van Hook

. Trajectories of overweight among US school children: A focus on social and economic characteristics. Matern Child Health J, 2011; 15:610–619.

29.

Hamilton

, Teitler

, Reichman

. Mexican American birthweight and child overweight: Unraveling a possible early life course health transition. J Health Soc Behav, 2011; 52:333–348.

30.

Martinson

, McLanahan

, Brooks-Gunn

. Race/ethnic and nativity disparities in child overweight in the United States and England. Ann Am Acad Pol Soc Sci, 2012; 643:219–238.

31.

Van Hook

, Baker

. Big boys and little girls: Gender, acculturation, and weight among young children of immigrants. J Health Soc Behav, 2010; 51:200–214.

32.

Van Hook

, Baker

, Altman

, et al. Canaries in a coalmine: Immigration and overweight among Mexican-origin children in the US and Mexico. Soc Sci Med, 2012; 74:125–134.

33.

Paterson

, Percival

, Schluter

, et al. Cohort profile: The Pacific Islands Families (PIF) Study. Int J Epidemiol, 2008; 37:273–279.

34.

World Health Organization. WHO Child Growth Standards: length/height-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development. World Health Association: Geneva, Switzerland, 2006.

35.

Statistics New Zealand. 2013 Census QuickStats about culture and identity. Statistics New Zealand: Wellington, New Zealand, 2014.

36.

Auckland Regional Council. Immigration and ethnicity in the Auckland region. Auckland Regional Council: Auckland, New Zealand, 2007.

37.

Tsai

, Ying

, Lee

. The meaning of “being Chinese” and “being American”: Variation among Chinese American young adults. J Cross-Cult Psychol, 2000; 31:302–332.

38.

Borrows

, Williams

, Schluter

, et al. Pacific Islands Families Study: The association of infant health risk indicators and acculturation of Pacific Island mothers living in New Zealand. J Cross-Cult Psychol, 2011; 42:699–724.

39.

Schluter

, Tautolo el

, Paterson

. Acculturation of Pacific mothers in New Zealand over time: Findings from the Pacific Islands Families Study. BMC Public Health, 2011; 11:307.

40.

Tautolo

, Schluter

, Paterson

, et al. Acculturation status has a modest effect on smoking prevalence among a cohort of Pacific fathers in New Zealand. Aust N Z J Public Health, 2011; 35:509–516.

41.

World Health Organization. Training course on child growth assessment. World Health Organization: Geneva, Switzerland, 2008.

42.

World Health Organization. Child growth standards: Software. World Health Organization: Geneva, Switzerland, 2011.

43.

World Health Organization. Growth reference 5–19 years: Application tools. World Health Organization: Geneva, Switzerland, 2007.

44.

World Health Organization. WHO AnthroPlus for personal computers manual: Software for assessing growth of the world's children and adolescents. World Health Organization: Geneva, Switzerland, 2009.

45.

O'Dea

. Gender, ethnicity, culture and social class influences on childhood obesity among Australian schoolchildren: Implications for treatment, prevention and community education. Health Soc Care Community, 2008; 16:282–290.

46.

Olvera

, Suminski

, Power

. Intergenerational perceptions of body image in hispanics: role of BMI, gender, and acculturation. Obes Res, 2005; 13:1970–1979.

47.

Buttenheim

, Pebley

, Hsih

, et al. The shape of things to come? Obesity prevalence among foreign-born vs. US-born Mexican youth in California. Soc Sci Med, 2013; 78:1–8.

48.

Smith

, Papa

, Jackson

. Diabetes in CMDHB and northern region: Estimation using routinely collected data. Counties Manukau District Health Board: Manukau City, New Zealand, 2008.

49.

Coppell

, Mann

, Williams

, et al. Prevalence of diagnosed and undiagnosed diabetes and prediabetes in New Zealand: Findings from the 2008/09 Adult Nutrition Survey. N Z Med J, 2013; 126:23–42.

50.

Min

, Li

, et al. Impacts of infancy rapid weight gain on 5-year childhood overweight development vary by age and sex in China. Pediatr Obes, 2012; 7:365–373.

51.

Crume

, Ogden

, Mayer-Davis

, et al. The impact of neonatal breast-feeding on growth trajectories of youth exposed and unexposed to diabetes in utero: The EPOCH Study. Int J Obes (Lond), 2012; 36:529–534.

52.

Rush

, Obolonkin

, Savila

. Growth centiles of Pacific children living in Auckland, New Zealand. Ann Hum Biol, 2013; 40:406–412.

53.

Abel

, Park

, Tipene-Leach

, et al. Infant care practices in New Zealand: A cross-cultural qualitative study. Soc Sci Med, 2001; 53:1135–1148.

54.

Bedford

, Ho

, Lidgard

. International migration in New Zealand: Context, components and policy issues. In: Population Studies Centre Discussion Papers. The University of Waikato: Hamilton, New Zealand, 2000.