We examined the relationship between prenatal tobacco smoke exposure and infant neurobehavior. The relatively normative sample of infants was derived from women who smoked at levels reflective of nationally reported levels [3
]. We found significant differences in infant neurobehavior at age five weeks in relation to exposure to tobacco smoke during pregnancy; however, the relationships between exposure and neurobehavioral outcome differed dramatically by race.
Among White infants, as prenatal tobacco smoke exposure increased, infants were more disturbed and agitated during the assessment. In addition, they were less able to self-regulate and calm themselves, requiring more intervention from the examiner. These infants also showed a trend toward a reduced ability to focus on the examiner and to participate in portions of the neurobehavioral assessment that require active engagement and socialization that became significant when we accounted for SHS exposure concurrent with the 5-week assessment. These results are consistent with previous studies reporting increased excitability [34
], decreased attention [35
], and a need for additional examiner assistance [13
] during the NNNS in association with maternal smoking during pregnancy. They are also consistent with studies that have used other infant assessments reporting increased nervous system excitation [14
] and decreased alertness [15
] in relation to maternal smoking during pregnancy.
When controlling for reported postnatal SHS exposure, this behavioral profile of agitation and inattention was strengthened in relation to prenatal exposure suggesting that in general, prenatal exposure remained as influential to neurobehavior as postnatal SHS for as long as five weeks after birth in these White infants. When limiting our analyses to infants with low-level exposure during pregnancy, determined by serum cotinine levels ≤ 10 ng/mL, we found that infants still exhibited decreased attention and self-regulation and increased excitability, but the association with increased arousal was no longer statistically significant. This suggests that among White infants, higher levels of tobacco smoke exposure are necessary to impact infant arousal, but that even very low levels of prenatal exposure that may result from minimal maternal SHS exposure can negatively impact the infant’s ability to handle stimulation, self-regulate, and participate in focused interaction when stressed.
In contrast, among Black infants, as prenatal tobacco smoke exposure increased, infants were more relaxed and self-controlled during the assessment, requiring less assistance from the examiner to complete the exam. These results in aggregate could suggest that the neurobehavior among Black infants was less affected by exposure to tobacco during pregnancy and that they were much better equipped neurologically and behaviorally to participate in the assessment than White infants. However, there was no indication that they participated at an elevated level as would be revealed by higher scores in alertness. It appears that as exposure increased among Black infants, they actually became less responsive during the exam, were less able to interact socially, and were less affected of the manipulation of the examiner.
When controlling for SHS exposure at the time of the exam, these Black infants remained less aroused and better self-regulated, but all other relationships were attenuated. This suggests that among Black infants, prenatal exposure remains critical in influencing a low arousal level and better self-regulation, but that SHS exposure at the time of the assessment overrides other neurobehavioral effects of prenatal exposure. When limiting the analyses to infants with very low-level exposure during pregnancy, only the relationship with increased self-regulation remained. This suggests that self-regulation among Black infants continues to be affected by prenatal exposure, but that in other areas of neurobehavior, Black infants are more likely to be affected by higher levels of tobacco during pregnancy and may be less sensitive to lower levels of exposure, and that the impact of postnatal SHS exposure overpowers the effect of prenatal exposure in Blacks. However, we suggest that these findings should be interpreted with caution due to the small sample size and potentially reduced analytic power.
It is not surprising that we found racial differences in the effects of prenatal tobacco exposure on infant neurobehavior. There is substantial evidence from research on adults indicating that Blacks often have higher cotinine levels than whites and Hispanics despite similar smoking prevalence rates [46
] and fewer reported cigarettes per day [47
]. Similarly, racial differences in cotinine levels have been reported among adolescent smokers [48
], pregnant smokers [50
], and children exposed to SHS [51
]. Further study has found slower metabolism of nicotine among Blacks when compared with Whites [52
]. Given these reports, it is possible that racial differences in fetal and infant metabolism of nicotine also exist, and that subsequent neurobehavioral outcomes may also vary by race. Perera [54
] found higher umbilical cord serum cotinine in Black versus Hispanic newborns, but neurobehavioral outcomes are not reported. We know of no other studies of cotinine levels in early infancy. Previously published studies have included small samples and have included race solely as a covariate rather than a major contributor to outcome. An advantage of the present study is that we had a larger sample of both White and Black infants to explore potential racial differences. However, as with most subgroup analyses, the study was underpowered to fully examine differences in gradients of exposure. For example, among Black infants, the coefficient for self regulation (β=0.077, se=0.027, p=.006) falls only slightly when considering SHS exposure at 5 weeks (β=0.062, se=0.032, p=.077). Because of the change in sample size, we see an associated change in standard error and thus effect size and significance. For this reason, we report both betas and standard errors in our tables so that the relative effect sizes for our population may be useful for planning future studies. Despite these power limitations, our study provides convincing evidence that future studies of relationships between tobacco smoke exposure and infant neurobehavior should include sufficient samples of diverse racial backgrounds and exposure levels.
The study has several limitations. First, although we had serial measures of prenatal tobacco smoke exposure, we lacked a biomarker of SHS exposure concordant with the neurobehavioral assessment at five weeks of age. Second, although this is the first known study to examine racial differences in infant neurobehavior using serial cotinine levels, the stratified analysis of the data based on race resulted in smaller sample sizes, was associated with reduced effect sizes as discussed above, and diminished our ability to fully detect neurobehavioral differences. Third, the sample included very few women who actively smoked during pregnancy. Nevertheless, the exposure patterns resemble those in nationally reported female smoking rates [55
] and rates of smoking during pregnancy [3
]. Lastly, although there are clear racial differences in the metabolism of nicotine, it is important to acknowledge that race may be acting as a proxy for other sociodemographic differences that could impact infant neurobehavior. Factors such as maternal depression [56
] and socioeconomic strain [58
] have been linked with diminished attention, alertness, and interaction patterns in infants, as well as increased arousal and stress responses. It is imperative that future investigations examining the link between prenatal tobacco smoke and infant neurobehavior include samples that are more sociodemographically balanced to allow clarity in illuminating potential racial differences in outcome distinct from those that may be confounded by sociodemographic characteristics.
In a sample of infants whose exposure to tobacco smoke during pregnancy resembled nationally-reported rates, we found significant differences in neurobehavioral outcomes associated with higher tobacco smoke exposure. These outcomes varied by race. With higher levels of tobacco smoke exposure, White infants were more excitable, aroused, and less able to calm themselves, whereas Black infants were less responsive during the assessment. These racial differences in neurobehavior could be related to metabolic differences in nicotine metabolism as disparities in the consequences of nicotine exposure. The findings of this study suggest that even low levels of exposure to tobacco smoke may impact infant neurobehavior. Additional study is required to further investigate these relationships and differentiate the influences that prenatal tobacco smoke and postnatal SHS exposure may have on neurobehavioral profiles of infants of different races.