Attachment processes in Early Head Start families

Established in 1994, Early Head Start is the first and only federal program designed to support child and family development in low-income families with infants and toddlers. Put another way, Early Head Start is the first and only federal program designed to support vulnerable infants and toddlers during the time that crucial infant–caregiver attachments form and develop. The Secretary's Advisory Committee on Services for Families with Infants and Toddlers, convened to design Early Head Start, in fact emphasized the need for services to support child–caregiver relationships (US Department of Health and Human Services, 1994). Moreover, Early Head Start staff affirm that supporting early attachments is a key goal (US Department of Health and Human Services, 2002).

There are currently over 700 Early Head Start programs across the US, serving over 70,000 children (Administration for Children and Families, 2010). Early Head Start services can be home-based, center-based, mixed, or otherwise configured (e.g., including family child care), with most programs providing a home-based (40%) or center-based (50%) option. Regardless of how services are provided, all programs are governed by a set of performance standards, and all must address the four program “cornerstones” of child development, family development, community development, and staff development (Administration for Children and Families, 2010). Program services include parenting education, child care, medical screenings, and the active linking of families to a wide range of community-based health and social services. Parenting education services are provided to 65% of Early Head Start families (Hoffmann & Ewen, 2006).

The Congressionally-mandated Early Head Start National Research and Evaluation Project is an ongoing, longitudinal trial evaluating Early Head Start services in 17 of the first 68 programs (Administration for Children and Families, 2002; Love et al., 2005). This Project began in 1995 with random assignment of 3001 low-income families to receive either Early Head Start services or community services as usual. In-depth data on child and family development are collected from interviews, observations, and direct child assessments of all participants. Each of the 17 sites also collected site-specific data. Findings to date indicate numerous, consistent, modest positive program effects on observed parenting behavior and toddler cognition, language, and behavior problems at age three (Love et al., 2005). For example, compared to randomly assigned control participants, Early Head Start mothers were observed to be more “supportive” (sensitive, positive, and cognitively stimulating) towards their children, who, similarly, were observed to be more engaged with their mothers. Moreover, Early Head Start children received fewer spankings than control children, and were reported by their mothers to be less aggressive.

From an attachment perspective, both the emphasis on infant–caregiver relationships of the 1994 Advisory Committee, and the research evidence that Early Head Start can indeed improve important aspects of infant–mother interaction, are heartening. At the same time, despite Early Head Start's programmatic emphasis on early attachments, there is much about attachment processes in Early Head Start families that is not well understood. For example, relatively little is known about parents' attachment security at program intake, about children's attachment security at program completion, or about attachment-relevant mediators or moderators of program effects. (For some important inroads into these topics, see Hoffman, Marvin, Cooper, & Powell, 2006; Robinson & Emde, 2004; Spieker, Nelson, DeKlyen, & Staerkel, 2005.) This Special Issue of Attachment & Human Development was assembled to address such issues.

First, in addition to providing a rigorous evaluation of Early Head Start services, the National Research and Evaluation Project data set offers an unprecedented opportunity to study important developmental issues in a large, exclusively low-income sample of infants and toddlers. In the initial article in this Special Issue, Howard et al. (pp. 5–26) draw on these data (N = 2080) to examine the attachment-relevant yet under-studied question of the effects of mother–child separation between birth and age two (Howard, Martin, Berlin, & Brooks-Gunn, 2010). As Bowlby might have predicted, Howard et al. report negative effects of early separation up to three years later.

In the second article in this Special Issue, Green et al. (pp. 27–47) longitudinally investigated the associations between mothers' self-reported attachment style and social support in 181 African American participants from one of the 17 sites of the National Research and Evaluation Project (Green, Furrer, & McAllister, 2010). Interestingly, mothers' self-reported attachment avoidance was relatively stable, and related to decreasing perceptions of social support. Mothers' self-reported attachment anxiety appeared more malleable, however, at least under conditions of low stress, with increased social support predicting decreasing attachment anxiety. Green and her colleagues thoughtfully discuss the implications of their findings for Early Head Start and similar programs, including the importance of service providers considering mothers' attachment styles.

The importance of programs' consideration of mothers' attachment styles is also highlighted by the third article in this Special Issue, in which I and my colleagues (pp. 49–67) examine maternal depression and self-reported attachment avoidance and attachment anxiety as moderators of Early Head Start's effects on parenting (Berlin et al., 2010). Using a sample of 947 mothers drawn from six sites of the National Research and Evaluation Project, we present some evidence of more positive program effects for mothers with less self-reported attachment avoidance or attachment anxiety at program intake.

Last, it is notable that, to date, there have been no standard assessments of attachment collected for the full sample of the Early Head Start National Research and Evaluation Project. This lamentable omission is due mainly to the infeasibility of conducting relatively labor-intensive procedures (such as the laboratory-based Strange Situation) with 3001 low-income families in 17 locations across the country. It also reflects the need for more manageable, less labor-intensive assessments of attachment. 1 The fourth article in this Special Issue (pp. 69–90) responds directly to this need by testing a new, home-based observational measure of toddler–parent attachment, the TAS-45, in a sample of 59 Early Head Start families (Spieker, Nelson, & Condon, 2010). Providing some initial evidence of this measure's validity, Spieker and her colleagues discuss its potential utility for both researchers and practitioners.

This Special Issue concludes with commentaries by Thompson (pp. 91–98, 2010) and Boris and Zeanah (pp. 99–104, 2010). Taken as a whole, the issue provides new findings about attachment processes in Early Head Start families, with implications for both research and service provision, in not only Early Head Start programs but also other widely disseminated early intervention initiatives, such as the Nurse-Family Partnership and Healthy Families programs.

As this Special Issue goes to press, the Obama administration is expanding the number of Early Head Start programs, and states across the US are scrambling to respond to the recently passed health care reform legislation, which authorizes US$1.5 billion over five years for Maternal, Infant, and Early Childhood Home Visiting Programs. The latest attachment theory, research, and evidence-based attachment interventions can and should be increasingly leveraged to inform Early Head Start and home visiting services (Berlin, Zeanah, & Lieberman, 2008). For example, programs might conduct brief assessments of mothers' attachment styles at enrollment, and use these assessments to inform program approaches; and/or incorporate brief theory- and research-based intervention protocol to support infant–caregiver attachment security (e.g., Attachment and Biobehavioral Catch-Up [Dozier, Peloso, Lewis, Laurenceau, & Levine, 2008], the Circle of Security [Hoffman et al., 2006]). Such innovations, though potentially challenging to implement, are also potentially rich in rewards.

Notes

1. This was especially the case when the Early Head Start National Research and Evaluation Project was designed in 1994.