摘要：Objectives. We examined the impact of a maternal–child home visitation program on birth spacing for first-time Latina mothers, focusing on adolescents and women who identified as Mexican or Puerto Rican. Methods. This was a retrospective cohort study. One thousand Latina women enrolled in the Pennsylvania Nurse–Family Partnership between January 1, 2003, and December 31, 2007, were matched to nonenrolled Latina women using propensity scores. The primary outcome was the time to second pregnancy that resulted in a live birth (interpregnancy interval). Proportional hazards models and bootstrap methods compared the time to event. Results. Home visitation was associated with a small decrease in the risk of a short interpregnancy interval (≤ 18 months) among Latina women (hazards ratio [HR] = 0.86; 95% confidence interval [CI] = 0.75, 0.99). This effect was driven by outcomes among younger adolescent women (HR = 0.80; 95% CI = 0.65, 0.96). There was also a trend toward significance for women of Mexican heritage (HR = 0.74; 95% CI = 0.49, 1.07), although this effect might be attributed to individual agency performance. Conclusions. Home visitation using the Nurse–Family Partnership model had measurable effects on birth spacing in Latina women. Maternal and child home visitation, which uses trained professionals or paraprofessionals to provide in-home support to families from the prenatal period through the first several years of a child’s life, is a key component of public health services for at-risk pregnant women and families in the United States. 1,2 Maternal and child home visitation programs provide support for an estimated 400 000 children annually. 3 This number is expected to rise following the appropriation of US$ 1.5 billion over 5 years to improve and expand home visitation under the Affordable Care Act. 4 As maternal and child home visitation expands, programs must meet the needs of Latino families. Latino children constitute 25% of children younger than 5 years, are the fastest growing sector of the US child population, and make up the largest group of children living in poverty. 5,6 Vulnerable Latino families experience persistent maternal–child health inequalities, including disproportionately high rates of adolescent pregnancy (128.9/1000 pregnancies vs 44.0/1000 among non-Hispanic Whites) 7 and rapid subsequent pregnancy during adolescence. 8 Adverse outcomes associated with very short interpregnancy intervals (< 6 months) include premature birth, 9 low birth weight, 10 and maternal mortality. 11 In general, Latina women are less likely to have adverse birth outcomes relative to other women from historically marginalized ethnic/racial groups, a phenomenon known as the “epidemiologic paradox.” 12 However, they are nonetheless vulnerable to the negative effects of very short interpregnancy intervals. 13 Data from the Nurse–Family Partnership (NFP), a widely implemented maternal and child home visitation program, suggests that home visitation using the NFP model increases birth spacing among participating Latina mothers. Support for this hypothesis comes from a randomized controlled trial (RCT) conducted in Denver, Colorado, in 1994 and 1995 that included a high proportion of Latina women (45%) who were predominantly of Mexican heritage. Investigators found that randomization to NFP reduced the overall number of second pregnancies during the 24-month follow-up (29% vs 41%) and lengthened the interpregnancy interval for all racial/ethnic groups, including Latina participants. 14 Other data on the effectiveness of maternal and child home visitation in preventing rapid repeat pregnancies among Latina mothers are limited. The Latino community is diverse, and most studies have included participants from a single ethnic/national group (e.g., Mexican or Mexican American women) or have not disaggregated data for participants from different Latino ethnic/national groups. Additionally, many studies have been pilot initiatives with small sample sizes. Most importantly, published studies have rarely included program evaluation following wide-scale implementation. 15–18 At this time, program effectiveness is likely to differ from the efficacy observed under experimental conditions. 19–21 Data on real-world program effectiveness are critical for program administrators, public health officials, and policymakers who seek to improve maternal–child health outcomes. To document the effectiveness of maternal child home visitation for Latina mothers, we conducted a focused analysis of birth spacing for Latina women enrolled in the Pennsylvania NFP program after state-wide implementation, with attention to adolescent clients and women who identified their ethnicity as Mexican or Puerto Rican. NFP uses trained, registered nurses to deliver a series of structured home visits intended to promote healthy prenatal behaviors, improve parents’ care of their children, and enhance parents’ life course development. This includes planning future pregnancies to optimize both maternal and infant health (e.g., promoting a safe interpregnancy interval). NFP nurses visit low-income, first-time mothers beginning no later than the end of the 28th week of pregnancy and continuing for up to 2 years postpartum. Women are typically referred to NFP from other services (including other health and public health providers), participate voluntarily, and ideally have one-to-one interactions with the same nurse visitor through program enrollment. NFP was widely implemented across Pennsylvania in 2003. Latina women constituted approximately one fifth of NFP’s clients during the study period. Latinos in Pennsylvania are young (median age = 24 years), native-born or naturalized citizens (83%), and predominantly of Caribbean (e.g., Puerto Rico, Cuba; 63%) or Mexican heritage (18%). Most speak Spanish at home (60%) but speak English well (73%). 22 The demographic characteristics of low-income, first-time Latina mothers who gave birth in Pennsylvania during the study period are described as data available as a supplement to the online version of this article at http://www.ajph.org . Among this group, 1 in 3 were adolescents (≤ 18 years), and half had not completed high school. The majority identified as Puerto Rican (59%), and nearly all lived in urban areas (98%).
