Prenatal Cocaine Exposure and Motor Performance at 4 Months

Conceptual Framework

Global measures of motor development, such as the Psychomotor Development Index of the Bayley Scales of Infant Development–II (Bayley, 1993) and the Peabody Developmental Motor Scales (Folio & Fewell, 2000), although commonly used in studies of PCE, are not sufficiently sensitive to detect subtle differences in motor development. These assessments were originally constructed using a hierarchical neuromaturation model, which assumes that the rate and sequence of normal motor development are fixed and directly tied to central nervous system (CNS) maturation. Studies of motor control have demonstrated that flexible organization is the critical aspect of motor performance that leads to proficiency (Lockman & Thelen, 1993). Current research indicates that basic units of motor control are organized as functional synergies or coordinated structures rather than as individual actions by specific muscles or muscle groups (Turvey & Fitzpatrick, 1993). Functional synergies cannot be attributed to particular nerve tracts. Instead, these coordinated motor structures are produced through activity in multiple regions of the brain in concert with certain unique characteristics of the individual, such as sensory processing ability, muscle tone, strength, body weight, cognition, motivation (Thelen & Spencer, 1998), and the affordances within the environment (Berger & Adolph, 2007; de Campos, Rocha, & Savelsbergh, 2009). As such, functional movement synergies are exquisitely complex and endlessly adaptable.

In addition, a basic premise of the current conceptual framework for motor development is the role of discontinuity in the developmental process. Acquisition of motor milestones does not follow a linear trajectory (Trevarthan, 1982). Instead, typical motor development shows periods of positive linear change interrupted by dips in performance. These periods of regression, referred to as phase shifts (Smith & Thelen, 2003), introduce an element of disequilibrium, expanding the options for movement and providing opportunities to practice movement structures that may be more flexible in their organization. This process, by virtue of its dynamic nature, fuels adaptation, which ultimately fuels motor performance (Hedberg, Carlberg, Forssberg, & Hadders-Algra, 2005; Meyer-Lindenberg, Ziemann, Hajak, Cohen, & Berman, 2002).

The influence of dynamical theory in motor development has been seen in test construction. The Bayley Scales of Infant and Toddler Development–III (Bayley, 2005) includes a motor scale that encompasses some qualitative aspects of movement. To capture the dynamic nature of the developing motor system, however, an assessment tool must allow infants the opportunity to elaborate on movements and to demonstrate emerging skills. When motor assessment captures emerging skills (marked by instability in performance) as well as established skills (stable performance), clinicians can analyze both factors that are constraining development and those that promote new skills. Because the purpose of assessment is often to determine the need for and direction of intervention, understanding the variables that constrain or facilitate new skills is a worthy goal (Case-Smith & Bigsby, 2000, p. 10). Standardized tests do not lend themselves to this aspect of motor assessment as well as criterion-referenced or play-based assessments, in which variability can become the focus of observation instead of an indicator of abnormality (Kelly-Vance, Needelman, Troia, & Ryalls, 1999).

Although infants with PCE have been described as having postural tone and movements that may appear qualitatively different from those of unexposed infants during the newborn period and early infancy (Chiriboga, Kuhn, & Wasserman, 2007), over time their scores on scales of motor development are reported to be largely within age expectations. Even when the extent of PCE is taken into consideration, no persistent effects of PCE appear in overall scores of motor development (Frank et al., 2002). In fact, the trajectory of motor development for infants with PCE has been shown to be one of improvement in motor development scores over the first 18 mo of life, even for infants with prenatal exposure to heavy cocaine use, and even when the potential effects of tobacco, low SES, and negative environmental factors are entered into the model (Miller-Loncar et al., 2005). Such findings could be interpreted as evidence for a model of recovery from the initial negative effects of PCE—even high levels of exposure—over time. However, the current expanded conceptual framework for motor development, which goes beyond acquisition of milestones and includes qualitative aspects of movement, has not yet been tested.

The purpose of this study was to examine potentially subtle differences in the quality of movement of term and preterm infants with PCE at 4 mo and 4-mo corrected age, respectively, and that of unexposed comparison groups, using the Posture and Fine Motor Assessment of Infants (PFMAI–I; Case-Smith & Bigsby, 2000) within the context of multiple potentially intervening factors, including gestational age at birth; maternal characteristics; SES; and level of exposure to cocaine, tobacco, alcohol, and marijuana. We hypothesized that qualitative evaluation of posture and fine motor performance can reveal the negative impact of PCE.

Description of the Posture and Fine Motor Assessment of Infants

The PFMAI–I was designed for infants ages 2–12 mo as a criterion-referenced tool. The goal of the PFMAI–I is to assess the acquisition not only of age-related motor milestones but also of motor skills characteristic of phase shifts—transitional periods when new skills emerge. Posture items include qualitative aspects of movement such as the infant’s weight-bearing position, postural alignment, and ability to adapt to challenges imposed by gravity. Fine motor items include components of reach, grasp, manipulation, and release using objects that offer differing affordances for action. For this study, a research version of the PFMAI–I including 45 items was administered to term infants at 4 mo and to preterm infants at 4-mo corrected age; each item was rated on a 5-point scale (0–4); higher scores indicated better quality of posture and fine motor performance.

Interobserver reliability for the PFMAI–I is .97 for the Posture scale and .99 for the Fine Motor scale (Case-Smith & Bigsby, 1997); test–retest reliability ranges from .49 for the Fine Motor scale to .90 for the Posture scale (Case-Smith, 1992); and Cronbach’s α, a measure of internal consistency, is high at .97 for the Posture scale and .99 for the Fine Motor scale (n = 59; Case-Smith & Bigsby, 1997). Strong correlations have been demonstrated between the PFMAI–I and the Psychomotor Scale of the Bayley Scales of Infant Development–II (n = 23; r = .75; Case-Smith, 1992) and infant age (n = 59; Posture r = .89; Fine Motor r = .92; Case-Smith & Bigsby, 1997). In addition, the performance results of 211 preterm infants and 162 term infants were combined (N = 373) and submitted to Rasch analysis to determine the spread of scores (item difficulty) within the context of ability level. The Rasch analysis resulted in elimination of three items from the Posture scale and three items from the Fine Motor scale to obtain a final “fair to good” spread of scores and restructuring to a 4-point scale (Case-Smith & Bigsby, 1997).

Method

This study was part of the Maternal Lifestyle Study (MLS), a longitudinal investigation of the effects of prenatal cocaine exposure on child outcome. The MLS was conducted under the auspices of the National Institute of Child Health and Human Development Neonatal Research Network at four sites (Brown University, Providence, RI; University of Miami, Miami, FL; University of Tennessee, Memphis, TN; and Wayne State University, Detroit, MI). MLS participants were recruited postpartum at four hospital sites from 1993 to 1995. The institutional review board at each site approved the study. A National Institute on Drug Abuse Certificate of Confidentiality was obtained to ensure confidentiality of information regarding participants’ drug use, and the mother provided written informed consent. The MLS was conducted in two phases: (1) an acute phase (Bauer et al., 2005), which extended through hospital discharge, and (2) the longitudinal follow-up of a subset of mothers and infants. The PFMAI–I was included in the battery of tests administered at 4 mo (age corrected for prematurity; Lester et al., 2002).

Results

Discussion

This study represents an advance in assessment of the potential effects of PCE on infant motor outcomes. It contrasts with previous studies in two ways. First, instead of focusing on acquisition of motor milestones, which had not been found to be significantly affected by PCE (Mayes, Cicchetti, Acharyya, & Zhang, 2003), we incorporated into our study model the current conceptual framework for motor development emphasizing qualitative aspects of motor performance. Second, embedded within this conceptual framework is the concept that characteristics of the environment, including the affordances of stimulus objects, may have a direct impact on the quality of movement that is observed. The PFMAI–I affords infants the opportunity to demonstrate their full repertoire of movements in the least restrictive fashion by offering stimulus objects for fine motor exploration with three distinctly different sets of characteristics. Thus, use of the PFMAI–I allows the full range of potential responses to be elicited and subtle differences to be detected.

Our initial hypothesis, that the PFMAI–I would detect subtle differences between infants with PCE and an unexposed comparison group, was partially supported. A statistically significant main effect of PCE on posture scores, but not on fine motor scores, was demonstrated. The PFMAI–I may not have been sufficiently sensitive to detect a fine motor difference. However, because the tool was sensitive enough to detect subtle differences in both posture and fine motor performance among infants born at <33 wk gestation, a more likely interpretation is that PCE has a greater impact on areas of the brain that organize posture and gross motor functioning than on the areas that organize fine motor activity. For decades, researchers have known that the neuronal tracts directly involved in posture and gross motor activity are distinct from those involved in fine motor activity (Hadders-Algra, Bogren, & Forssberg, 1998; Kuypers, 1981). Thus, it is important to assess gross motor and fine motor performance separately, particularly when attempting to discern subtle effects on CNS function.

When fine motor performance is specifically examined and the potential effects of coinciding substances such as tobacco, alcohol, and marijuana are included in the model, evidence suggests that factors outside the infant, such as maternal vocabulary and features of the home environment, may have a stronger effect on fine motor scores than PCE (Arendt et al., 2004; de Campos et al., 2009). Our finding of a direct effect of PCE on posture but not on fine motor scores concurs with those observations and stands in contrast to results obtained using a presumably less sensitive global measure such as the Motor Scale of the Bayley Scales of Infant Development–II (Mayes et al., 2003).

Additional strengths of this study are the multisite sampling methodology, the size of the sample, and the statistical power these features afford. The main effect of cocaine exposure on infant posture was obtained after adjusting for the possibility of site differences and maternal characteristics in the statistical model. This approach was necessary because of significant differences between the exposure groups with regard to age, education, marital status, and SES, all of which are factors with the potential to influence the affordances for growth and development that a mother is able to provide within the home environment. These characteristics of the sample impose certain limitations on the interpretation of our findings, however. Substance-abusing women are clearly a diverse group with varying degrees of risk and protection. Discretion is advised when generalizing the results of any study of PCE, including our own, to other groups of women and their infants.

Conclusion

In this study, we incorporated the context for motor development into the study model and used the PFMAI–I, a tool that focuses on specific aspects of motor performance, to detect subtle differences in posture development between infants prenatally exposed to cocaine and an unexposed comparison group. The PFMAI–I was sufficiently sensitive to reveal a main effect of prenatal cocaine exposure on postural performance and differences in posture and fine motor development among infants born at or before 33 weeks gestation, even after correcting for prematurity. Independent contributions of prenatal cocaine exposure and prematurity to infant risk status are important for clinicians to consider when conducting assessments and intervening with infants at risk. These findings further demonstrate the validity of the PFMAI–I as a measure of motor competence in early infancy.