about child development, social support to parents, encouragement of positive parent–child interactions, and social and health services. Some also provide case management services and health and developmental screening for children (Sweet and Appelbaum, 2004).

Sweet and Appelbaum (2004) conducted a meta-analysis of experimental and quasi-experimental evaluations of 60 home visiting programs. Only a fourth of these programs included home visiting during pregnancy. The authors conclude that on average, families receiving home visiting did better than those in control conditions. Mothers were more likely to pursue education but did not differ in their employment, self-sufficiency, or welfare dependence. The programs produced better outcomes in three of five areas of children’s cognitive and social-emotional functioning. However, the authors also note that the significant variability across programs makes it difficult to evaluate them as a group. Aos, Lieb, and colleagues (2004) found that average benefits of the 25 programs reviewed exceeded costs.

The home visiting program with the best experimental evaluations and strongest results to date is the Nurse-Family Partnership (NFP), which has been evaluated in three randomized controlled trials. NFP is unique in targeting only first-time mothers. The theory of change is that women may be more open to support and guidance during their initial pregnancies (Olds, Hill, et al., 2003), which may contribute to the strength of the program’s outcomes. This theory is supported by a randomized controlled trial of another home visiting program, which had a significant impact on first-time mothers’ positive caregiving but not on that of women who were already mothers (Stolk, Mesman, et al., 2007). In the first two trials (in New York and Tennessee), the program improved pregnancy outcomes, maternal caregiving, and the maternal life course and prevented the development of antisocial behavior. The third trial (in Colorado) showed benefits as well.

NFP has other distinguishing features that may contribute to the strength of its outcomes. First, the program providers are nurses with both substantial training and credibility regarding pregnancy and infants. The Colorado trial experimentally evaluated the impact of nurses versus paraprofessionals and found that nurse visitation produced more benefits compared with the control condition (Olds, Robinson, et al., 2002, 2004). None of the other home visitation interventions reviewed by Gomby (1999) employed nurses as providers. Second, NFP uses well-established techniques to guide changes in specific behaviors, such as smoking, seeking an education, and getting social support. The focus on smoking in the New York study, in which more than 50 percent of mothers smoked, is especially noteworthy given the well-established relationship between smoking during pregnancy and children’s subsequent antisocial behavior and substance use (see Brennan, Grekin, et al., 2002; Wakschlag, Lahey, et al., 1997; Weissman, Warner, et al., 1999).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement