Cross-national comparison of prenatal methamphetamine exposure on infant and early child physical growth: a natural experiment

Abstract

The current study seeks to compare the effects of prenatal methamphetamine exposure (PME) on infant and child physical growth between the USA and New Zealand (NZ). This cross-national comparison provides a unique opportunity to examine the potential impact of services provided to drug using mothers on child health. The longitudinal Infant Development, Environment and Lifestyle study of PME from birth to 36 months was conducted in the USA and NZ. The US cohort included 204 children with PME and 212 non-PME matched comparisons (NPME); the NZ cohort included 108 children with PME and 115 NPME matched comparisons. Latent growth curve models were used to examine effects of PME, country of origin, and the country × PME interaction on growth in length/height and weight. In regard to length/height, PME and country of origin were associated with initial length and growth over time. There was also a significant interaction effect, such that children with PME in the USA were shorter at birth than children with PME in NZ after controlling for other prenatal exposures, infant set, socioeconomic status, and maternal height. In regard to weight, there was only an effect of country of origin. Effects of PME on infant and child growth were shown to differ across countries, with exposed children in NZ faring better than exposed children in the USA. Implications for prevention programs and public policy are discussed.

Figure 1. Unconditional latent growth curve of child length/height

Chronometric factor loadings from the intercept, linear slope, and quadratic trend were fixed. Correlations between factors are presented. Model fit the data well, χ² (1) = 4.75, p = 0.03; CFI = 0.99; RMSEA < 0.08. * p < 0.05, ** p < 0.01, ***p < 0.001

Figure 2. Unconditional latent growth curve of child weight

Chronometric factor loadings from the intercept, linear slope, and quadratic trend were fixed. The correlation between intercept and linear slope is presented. The quadratic trend was held constant to facilitate model convergence with a positive definite psi matrix. Model fit the data well, χ² (4) = 13.70, p < 0.01; CFI = 0.96; RMSEA = 0.06. * p < 0.05, ** p < 0.01, ***p < 0.001

Figure 3. Conditional latent growth curve of child length/height using PME, country of origin, PME × country interaction, and covariates

Chronometric factor loadings from the intercept, linear slope, and quadratic trend were fixed. Correlations between factors are presented. All associations between predictors/covariates and growth factors were modeled, but only statistically significant paths are presented in the interest of clarity and parsimony. Prenatal exposure to tobacco, alcohol, and marijuana, child sex, and SES were included as covariates of growth parameters, but none of these effects were statistically significant. As such, these variables were not included in the figure. Model fit the data well, χ² (11) = 20.52, p = 0.04; CFI = 0.98; RMSEA < 0.05. * p < 0.05, ** p < 0.01, ***p < 0.001

Figure 4. Child length/height by country and exposure status

Values presented in the figure represent the unadjusted mean standardized height/length values using World Health Organization standards for birth through 36 months.

Figure 5. Conditional latent growth curve of child weight using PME, country of origin, PME × country interaction, and covariates

Chronometric factor loadings from the intercept, linear slope, and quadratic trend were fixed. The correlation between intercept and linear slope is presented. The quadratic trend was held constant to facilitate model convergence with a positive definite psi matrix. All associations between predictors/covariates and growth factors were modeled, but only statistically significant paths are presented in the interest of clarity and parsimony. Prenatal exposure to alcohol and tobacco and SES were included as covariates of growth parameters, but none of these effects were statistically significant. As such, these variables were not included in the figure. Model fit the data well, χ² (22) = 58.34, p < 0.001; CFI = 0.94; RMSEA = 0.05. * p < 0.05, ** p < 0.01, ***p < 0.001

Figure 6. Child weight by country and exposure status

Values presented in the figure represent the unadjusted mean standardized weight values using World Health Organization standards for birth through 36 months.