Home Visiting Effects on Breastfeeding and Bedsharing in a Low-Income Sample

Abstract

Background

Research suggests that home visiting interventions can promote breastfeeding initiation, though their effects on breastfeeding continuation are unclear. No known studies have assessed the impact of home visiting on bedsharing.

Aims

To test the effects of home visiting on breastfeeding and bedsharing in a low-income, urban sample in the United States.

Methods

During a field trial conducted in Milwaukee, Wisconsin, from April 2014 to March 2017, referrals to a public health department were randomized to a Healthy Families America (HFA) program or a prenatal care and coordination (PNCC) program. Of the 204 women who accepted services, 139 consented to the study and were allocated to the two treatment groups, which were compared with each other and a third quasi-experimental group of 100 women who did not accept services. Data were collected at four time points up to 12 months postpartum.

Results

Breastfeeding initiation was higher among 72 HFA participants (88.4%; odds ratio [OR] = 2.7) and 67 PNCC participants (88.5%; OR = 2.2) than 100 comparison participants (76.5%). Similar results emerged for breastfeeding duration, though group differences were not statistically significant. Unexpectedly, bedsharing prevalence was higher among HFA participants (56.5%) than PNCC participants (31.1%; OR = 2.9) and comparison group participants (38.8%; OR = 2.0).

Discussion

Home visiting was linked to increased breastfeeding, while effects on bedsharing varied by program. Progress toward precision home visiting will be advanced by identifying program components that promote breastfeeding and safe sleep.

Conclusion

Further research is needed to examine whether home visiting reduces disparities in breastfeeding and safe sleep practices.

Keywords

bedsharing breastfeeding disparities home visiting low income

Breastfeeding is associated with many positive outcomes, including reduced child morbidity and mortality (Victora et al., 2016). The American Academy of Pediatrics (AAP) recommends exclusive breastfeeding for the first 6 months because benefits accrue over this period (Eidelman, 2012), though breastfeeding for a briefer duration may still confer protective effects (Ip et al., 2009; Victora et al., 2016). Despite these recommendations, it is estimated that only one quarter (25.4%) of infants are exclusively breastfed for this long (Centers for Disease Control and Prevention, 2018). There is significant variation in breastfeeding rates across racial/ethnic and socioeconomic strata; non-Hispanic Black, less educated, and low-income mothers tend to initiate and continue breastfeeding at lower rates than the general population (Hardison-Moody et al., 2018; Li et al., 2005; Merewood et al., 2019; Pugh et al., 2010; Thomson et al., 2016).

The AAP also advises that infants should sleep on a separate surface in the same room as their caregivers for the first 6 months (Moon & Task Force on Sudden Infant Death Syndrome, 2016). While some studies indicate that bedsharing facilitates breastfeeding (McKenna & Gettler, 2016; Mileva-Seitz et al., 2017), parents are discouraged from sleeping with their infants due to the increased risk of sleep-related death (Hussain et al., 2018; Moon & Task Force on Sudden Infant Death Syndrome, 2016). Nevertheless, infant bedsharing remains a common practice in the United States. For example, an analysis of data from the National Infant Sleep Position Study showed that nearly 45% of parents bedshared with their infants (Colson et al., 2013; Willinger et al., 2003). There are also disparities in bedsharing, with higher rates observed among racial/ethnic minorities and low-income households (Colson et al., 2013; Lahr et al., 2007; Salm Ward & Doering, 2014).

Population health strategies are needed to reduce disparities in breastfeeding and safe sleep practices. Home visiting programs are well positioned to achieve this goal as they typically provide prenatal and postpartum support services to socioeconomically disadvantaged households (McGinnis et al., 2018). Home visiting interventions for pregnant women and new mothers vary in their timing, duration, and curricular focus, though they typically share common practice elements such as screening, assessment, and parent education, as well as common outcome goals such as improved maternal and child health outcomes and enhanced parenting attitudes and practices. Research on breastfeeding outcomes of women who receive home visiting is underdeveloped, but the available evidence suggests that these programs can increase the likelihood that a new mother will initiate breastfeeding (Edwards et al., 2013; Kronborg et al., 2012; LeCroy & Lopez, 2018; Shah & Austin, 2014). Few studies have reported the effects of home visiting on breastfeeding duration (Edwards et al., 2013; McGinnis et al., 2018), and no known studies have assessed home visiting effects on bedsharing. This study aims to address these knowledge gaps.

Our specific aims were to test the effects of home visiting on breastfeeding initiation and duration and infant bedsharing. Data were derived from a field trial of two home visiting programs at an urban health department that serve predominantly low-income, racial/ethnic minority families. We hypothesized that participants who were randomized to an evidence-based, long-term home visiting program would breastfeed more and bedshare less than would participants randomized to a locally developed program that offers briefer, less intensive services. We also hypothesized that, when compared with a third group of mothers who did not accept the services, both treatment groups would exhibit higher rates of breastfeeding and lower rates of bedsharing.

Methods

Sample and Design

A field trial of two existing home visiting programs was conducted at a large, urban health department in Milwaukee, Wisconsin. Study protocols were developed collaboratively between program staff and university researchers to maximize rigor while also minimizing program burden and disruption. The protocols were approved by a university institutional review board prior to engaging human participants. All participants completed a written informed consent procedure with a research team member before participating in the study. No families were denied home visiting services, and participation in the study was voluntary.

The sample consisted of low-income, pregnant women who were referred to the health department’s centralized intake unit for home visiting services. Most referrals originate from a network of prenatal health care providers that serve low-income families in the city. The department’s home visiting programs serve predominantly poor families (98.5% Medicaid eligible) who reside in zip code areas of the city with the highest rates of premature birth and infant mortality. Referrals were eligible for the study unless (a) they did not speak English or Spanish, (b) they previously received home visiting services from the health department, or (c) they were referred to the health department’s Nurse Family Partnership home visiting program. Nurse Family Partnership was not included in the trial because the program serves only first-time mothers (i.e., primiparas). In the current study sample, 25.7% of women were primiparas.

Randomization

From April 2014 to March 2017, eligible women were randomly assigned to one of two home visiting programs (Figure 1). One program was accredited by Healthy Families America (HFA), a widely disseminated evidence-based model, and the other program was developed by the health department. Referrals were randomized in blocks of 6, with equal numbers allocated to the two treatment groups within each block. Randomization occurred after a service referral was received and before program recruitment was initiated. Referral information was relayed using a secure listserv from centralized intake staff to the research team, who returned assignment allocation information within 48 hours.

Figure 1.

Participant flowchart.

Quasi-Experimental Comparison Group

The study includes a third quasi-experimental group of women who did not enroll in home visiting services at the health department after referral. From this pool, a convenience sample was recruited that matched the demographic composition of the experimental groups.

Intervention Procedures

The HFA program was part of a statewide network of agencies that are subsidized by the federal Maternal Infant and Early Childhood Home Visiting (MIECHV) program. Consistent with MIECHV and HFA program standards (2017), the program aims to support low-income families beginning prenatally and up to a child’s third birthday with a service array that includes developmental and functional assessments of children and caregivers, parenting guidance, and referrals to community services. Target outcomes of the program include improved maternal and child health, parenting practices, and child development and school readiness. HFA guidelines stipulate that families should be offered at least one visit per week until infants are 6 months of age. Home visits average about 1 hour in duration. Home visitors were human service professionals who received support from public health nurses. On average, mothers enrolled during the 24th week of pregnancy, and their mean length of service was 26.8 weeks.

The second home visiting program, which was developed by the health department, aims to enroll women during pregnancy and serve them up to 60 days postpartum. The program is largely funded by a state-administered Prenatal Care Coordination (PNCC) program that reimburses for medical, educational, and social services provided to Medicaid-eligible women at risk of adverse pregnancy outcomes. Reimbursable services include screening and assessment, health and nutrition education, and care coordination. In addition to breastfeeding and safe infant sleep practices, target outcomes include reductions in tobacco and substance use and improvements in family planning. PNCC services are provided by public health nurses. Visits lasting approximately 1 hour are typically scheduled to occur once every 2 weeks, although the nurses have discretion to vary the frequency and duration of appointments according to family needs. The average PNCC enrollment in this study occurred during the 26th week of pregnancy, and their mean length of service was 21.8 weeks.

Data Collection

This investigation uses data collected at four time points, including a prenatal baseline assessment of participant demographics and functioning. Outcomes were measured during a second assessment 14 to 60 days postpartum (Time 2), a third assessment approximately 6 months postpartum (Time 3), and a fourth assessment approximately 12 months postpartum (Time 4). Data were collected from program participants by HFA providers at all time points and by PNCC providers at the first two time points. Research staff collected data from PNCC participants at Times 3 and 4 and from HFA and PNCC participants who dropped out of services. The research team collected all data from the nontreatment comparison group. Participants were not compensated for assessments completed by their home visitors; all other participants received a $25 gift card for each assessment completed.

Outcomes

Mothers responded to questions about breastfeeding at assessment Time Points 2, 3, and 4. The data were used to create two outcomes, one of which is a dichotomous measure of breastfeeding initiation. Participating mothers were asked if the baby was ever breastfed, and those who responded yes were recorded as having initiated breastfeeding with the index child. Mothers who indicated that they had initiated breastfeeding were also asked how long they had breastfed. These data were used to create a second outcome, breastfeeding duration, which is a continuous measure ranging from 1 day to 12 months postpartum.

Infant sleep conditions were assessed at Time Points 2 and 3, including information about bedsharing. Participants reported the frequency of infant bedsharing in response to the following question: How often does your new baby sleep in the same bed with you or anyone else? Response options ranged on a 5-point Likert-type scale from never to always. A dichotomous outcome was created indicating any bedsharing by 6 months postpartum, with participants who responded never at both time points coded 0 and participants who provided any affirmative response coded 1.

Statistical Analysis

We analyzed sample characteristics at baseline using means and standard deviations for continuous data and proportions for categorical data (Table 1). Omnibus and pairwise tests revealed that there were no statistical differences between the study groups in race/ethnicity, age, or postsecondary education.

Table 1.

Characteristics and Outcomes in a Sample of Low-Income Women in Milwaukee, Wisconsin, 2014–2017.

Characteristic	Total sample (N = 237)	HFA (n = 72)	PNCC (n = 65)	Comparison (n = 100)
Maternal race/ethnicity, n (%)
Hispanic	96 (40.5)	27 (37.5)	29 (44.6)	40 (40.0)
African American/Black	112 (47.3)	37 (51.4)	26 (40.0)	49 (49.0)
Other	29 (12.2)	8 (11.1)	10 (15.4)	11 (11.0)
Maternal age, M (SD)	26.8 (6.1)	26.4 (6.1)	27.2 (5.8)	26.8 (6.4)
Maternal postsecondary education, n (%)	55 (23.2)	16 (22.2)	15 (23.1)	24 (24.0)
Any breastfeeding, n (%)	180 (83.7)	61 (88.4)	54 (88.5)	65 (76.5)
Any bedsharing, n (%)	91 (42.3)	39 (56.5)	19 (31.1)	33 (38.8)
Duration of breastfeeding, M (SD)	11.0 (13.3)	13.4 (16.9)	10.8 (11.6)	9.1 (10.7)

Note. HFA = Healthy Families America; PNCC = Prenatal Care and Coordination program.

We conducted logistic regressions to test for group differences in the likelihood of breastfeeding and bedsharing; odds ratios and 95% confidence intervals are reported as estimates of association between predictors and study outcomes. We tested group differences in breastfeeding duration, which is a time-to-event outcome, via Cox regression, with results reported as hazard ratios and 95% confidence intervals. Although results suggested that there was acceptable balance between study groups at baseline, unobserved differences are possible due to the modest sample size and use of a quasi-experimental comparison group. To guard against chance imbalances between study groups (Kahan et al., 2014), we performed regression analyses controlling for race/ethnicity, age, and education—three maternal characteristics that have been linked to differences in breastfeeding and bedsharing (e.g., Colson et al., 2013; Lahr et al., 2007; Li et al., 2005; Thomson et al., 2016). All statistical analyses were performed in SPSS, version 25 (IBM Corporation, 2017).

Analyses followed intention-to-treat parameters. Of the 139 treatment group participants that enrolled in the study, 130 (93.5%) provided valid outcome data; 85% of the 100 comparison group participants provided outcome data. Reasons for participant loss included inability to locate participants and study discontinuation (Figure 1).

Power estimations were performed a priori in G*Power 3.1.92 assuming a two-tailed test (α = .05) with 80% power. We derived anticipated effect sizes (Cohen’s d = 0.21 to 0.37; odds ratio [OR] = 1.46-1.96) from meta-analyses (Nievar et al., 2010; Sweet & Applebaum, 2004), which indicate that home visiting programs tend to achieve small average effects in most outcome domains. Sample size targets (n = 110 per experimental group) were not achieved, however, due to lower program enrollments than expected. As a result, the study is not powered to detect effect sizes at the lower end of the projected range.

Results

As shown in Figure 1, 910 referrals were randomized to the HFA program (n = 456) or PNCC program (n = 454). Services were not accepted by 254 (55.7%) participants assigned to HFA and 280 (61.7%) participants assigned to PNCC. Seventy-five (16.4%) participants assigned to HFA and 78 (17.2%) participants assigned to PNCC were ineligible for the study. There were 12 (2.6%) HFA assignments and 6 (1.3%) PNCC assignments in program outreach at the end of study recruitment. Among study-eligible participants, 115 accepted HFA services and 89 accepted PNCC services; 72 out of 115 (62.6%) HFA participants and 67 out of 89 PNCC participants (75.3%) consented to the study. Out of 534 women who did not accept services, 311 were recruited, and of these, 100 (32.2%) consented to the study.

Table 1 describes the sample. The mean age of participants was 26.8 years (standard deviation = 6.1). The sample was 47.3% non-Hispanic Black, 40.5% Hispanic, and 12.2% other race/ethnicity (6.8% non-Hispanic White, 1.3% non-Hispanic American Indian, and 4.2% multiracial or unknown). Less than one fourth (23.2%) had any postsecondary education. The prevalence of breastfeeding initiation was 88.4% for HFA participants, 88.5% for PNCC participants, and 76.5% for comparison group participants.

Logistic regression results shown in Table 2 indicate that HFA mothers were more likely than comparison group mothers to have breastfed (OR = 2.7, p = .04). The odds of breastfeeding were 2.2 times greater for PNCC mothers than for comparison group mothers (p = .11). The odds of breastfeeding did not differ significantly between the HFA and the PNCC treatment groups (OR = 1.2, p = .73).

Table 2.

Treatment Effects on Breastfeeding and Bedsharing in Milwaukee, Wisconsin, 2014-2017.

Model 1: treatment versus treatment	Any breastfeeding, OR [95% CI]	N = 130	p	Any bedsharing, OR [95% CI]	N = 130	p
Maternal race/ethnicity^a
Hispanic	15.1 [1.8, 129.2]	n = 55	.01	0.9 [0.4, 2.0]	n = 55	.80
Other	1.4 [0.3, 6.0]	n = 18	.66	0.6 [0.2, 1.8]	n = 18	.32
Maternal age	1.0 [0.9, 1.1]	n = 130	.62	1.0 [<1.0, 1.1]	n = 130	.49
Maternal education	1.5 [0.4, 5.6]	n = 30	.58	0.6 [0.2, 1.5]	n = 30	.29
HFA program^b	1.2 [0.4, 3.9]	n = 69	.73	2.9 [1.4, 5.9]	n = 69	<.01
Model 2: treatments versus comparison	Any breastfeeding, OR [95% CI]	N = 215	p	Any bedsharing, OR [95% CI]	N = 215	p
Maternal race/ethnicity^a
Hispanic	7.6 [2.6, 22.4]	n = 90	.00	0.9 [0.5, 1.7]	n = 90	.81
Other	2.5 [0.8, 8.1]	n = 29	.13	0.7 [0.3, 1.8]	n = 29	.46
Maternal age	1.0 [0.9, 1.0]	n = 215	.41	1.0 [<1.0, 1.1]	n = 215	.78
Maternal education	1.4 [0.5, 3.5]	n = 51	.51	0.5 [0.3, 1.1]	n = 51	.08
HFA program^c	2.7 [1.1, 7.0]	n = 69	.04	2.0 [1.1, 3.9]	n = 69	.03
PNCC program^c	2.2 [0.8, 5.9]	n = 61	.11	0.7 [0.4, 1.5]	n = 61	.36

Note. HFA = Healthy Families America; PNCC = Prenatal Care and Coordination program.

Reference group is African American/Black. ^bReference group is PNCC program. ^cReference group is quasi-experimental comparison sample.

The observed mean duration of breastfeeding was 13.4 weeks for HFA participants, 10.8 weeks for PNCC participants, and 9.1 weeks for the comparison group. Cox regression results presented in Figure 2 show that HFA and PNCC participants had similar cumulative probabilities of breastfeeding during the study period (hazard ratio [HR]_{PNCC vs. HFA} = 1.02, p = .91). Pairwise contrasts indicated that, relative to the comparison group, the HFA and PNCC groups had higher, albeit nonsignificant, cumulative probabilities of breastfeeding duration (HR_{comparison vs. HFA} = 1.30, p = .11; HR_{comparison vs. PNCC} = 1.23, p = .20).

Figure 2.

Survival rate of breastfeeding duration in weeks stratified by study groups.

Results (not shown) indicated that the point-biserial correlation between breastfeeding duration and bedsharing was nonsignificant (r = .06, p =.40). Nonetheless, planned sensitivity analyses of program effects on breastfeeding duration were performed with bedsharing modeled as a covariate. Results were largely unchanged from primary analyses reported in the preceding paragraph. Cumulative probabilities of breastfeeding did not differ between treatment groups (HR_{PNCC vs. HFA} = 1.0, p = .97). HFA and PNCC mothers had higher cumulative probabilities of breastfeeding relative to comparison mothers, though the differences were not statistically significant (HR_{comparison vs. HFA} = 1.30, p = .12; HR_{comparison vs. PNCC} = 1.23, p = .20).

Results showed that infant bedsharing was reported by 42.3% of study participants. Bedsharing was reported by 56.5% of HFA mothers, 31.1% of PNCC mothers, and 38.8% of nontreatment mothers. The HFA group was more likely to report bedsharing than the PNCC group (OR = 2.9; p < .01) and the nontreatment group (OR = 2.0, p = .03). The odds of bedsharing did not differ between the nontreatment group and the PNCC group (OR = 0.7, p = .36) or the combined group of HFA and PNCC participants (OR = 1.3, p = .40). Sensitivity analyses of program effects on bedsharing controlling for breastfeeding duration yielded comparable results (not shown). The HFA group was more likely to bedshare than the PNCC group (OR = 2.9, p < .01) and the nonservice comparison group (OR = 2.0, p < .05). Bedsharing odds did not differ between the PNCC and the nontreatment groups (OR = 0.7, p = .34).

Discussion

Breastfeeding and bedsharing outcomes are unequally distributed along socioeconomic and racial/ethnic lines in the United Sates. Home visiting programs may help address these disparities by providing targeted psychoeducation, guidance, and support to pregnant women and new mothers. Research indicates that this broad class of interventions is effective in promoting breastfeeding initiation (Kronborg et al., 2012; LeCroy & Lopez, 2018; Shah & Austin, 2014), and at least two studies have shown that they can promote breastfeeding continuation as well (Edwards et al., 2013; McGinnis et al., 2018). Little is known about the impact of home visiting on bedsharing, and it is uncertain what home visiting approaches are most effective at promoting breastfeeding and preventing bedsharing.

Our study adds to the literature by examining breastfeeding initiation and duration as well as infant bedsharing among low-income and predominantly racial/ethnic minority mothers who were randomized to one of two home visiting programs at an urban health department. One program was accredited by HFA—a widely disseminated, long-term home visiting model—and the other was a less intensive nurse visiting model developed by the health department. The two treatment conditions were compared with each other and to a third group of mothers who were referred for services but did not enroll.

Results showed breastfeeding initiation exceeded 88% in both home visiting programs, above the national average of 81% (Thomson et al., 2016). There were no differences in breastfeeding initiation or duration between program groups. Compared with mothers who did not enroll in either home visiting program, the odds of breastfeeding were 2.7 times higher in the HFA group and 2.2 times higher in the PNCC group. The cumulative probability of breastfeeding continuation during the study period was 30% higher among HFA mothers and 23% higher among PNCC mothers than among comparison mothers, though the differences were not statistically significant.

Approximately 42% of sample mothers reported infant bedsharing by 6 months postpartum, supporting studies that have shown that bedsharing is a common practice in the United States (Colson et al., 2013). Contrary to our hypothesis, the prevalence of bedsharing was significantly higher in households that received HFA services (56.5%) than those that received PNCC services (31.1%) or that did not enroll in services (38.8%). Secondary analyses showed that controlling for breastfeeding did not alter the estimated program effects on bedsharing. In fact, breastfeeding was not correlated with bedsharing in this sample. These results are consistent with another study of breastfeeding and bedsharing among women receiving home visiting services (Hussain et al., 2018). Prior studies in the United Sates that link bedsharing to increased breastfeeding have primarily studied the general population or socioeconomically advantaged samples (Hauck et al., 2008; Huang et al., 2013); our study points to the need for further research with low-income and racially/ethnically diverse groups.

Taken together, the findings suggest that home visiting programs can increase breastfeeding initiation among predominantly low-income, racial/ethnic minority women in an urban setting. The evidence of comparable benefits between two different programs can be attributed to many factors, including that both programs serve women prenatally. Similar results emerged for breastfeeding duration, though program effects were not statistically significant, which may be partly due to the study’s limited sample size and power. Alternatively, because the breastfeeding continuation analysis includes only women who breastfed, program effects on initiation could have suppressed program effects on continuation. That is, higher rates of breastfeeding initiation in the home visiting groups may have emerged because the programs modified the behaviors of some women who were hesitant to breastfeed. Compared with women who were inclined to breastfeed, women who breastfed despite their reluctance may have been more likely to discontinue. Supporting this assertion, research has shown that prenatal intentions to breastfeed predict the likelihood of continuation (DiGirolamo et al., 2005). Thus, future research may generate more precise estimates of program effects by accounting for preprogram differences in breastfeeding motivations.

The bedsharing findings are open to multiple interpretations, one of which pertains to program differences. HFA is a comprehensive service model that targets many child, caregiver, and family outcomes, whereas the PNCC program focuses more narrowly on maternal and child health outcomes. In addition, PNCC services were delivered by public health nurses, whereas HFA services were delivered by other human service providers with support from nurses. Nurses may be more likely than generalist practitioners to communicate bedsharing messages consistent with AAP recommendations. Or, clients may have been more willing to disclose bedsharing practices to HFA providers than to PNCC nurses. Future studies should examine if perceptions of bedsharing vary among home visitors from different professional backgrounds and how home visitors communicate safe sleep messages to clients who bedshare.

Strengths of this field trial include a design that permitted intention-to-treat analyses of home visiting effects under usual conditions with a sample at risk of poor breastfeeding and bedsharing outcomes. The study also has multiple limitations, including low statistical power. Despite the 3-year study recruitment window, the sample size was restricted because program enrollment numbers were lower than projected. We randomized participants prior to program enrollment, which may have resulted in an unobserved imbalance between treatments due to differences in program recruitment strategies and enrollment rates. Quasi-experimental comparisons between the treatment groups and the nontreatment group should also be interpreted with caution due to potential differences between conditions. For instance, selection effects may have resulted in unobserved variation between the treatment groups and the comparison group. In addition, assessors varied across study conditions and were not blinded to participant assignments, which may have introduced bias. Last, generalizability is limited because the study was conducted at an urban health department that serves mainly low-income, racial/ethnic minority clients.

Despite these limitations, the study results support research indicating home visiting programs have the potential to promote breastfeeding (e.g., Edwards et al., 2013; LeCroy & Lopez, 2018; McGinnis et al., 2018). Although breastfeeding tends to be less prevalent in poorer and less educated populations (Hardison-Moody et al, 2018; Merewood et al., 2019), more than 88% of economically disadvantaged women in two different home visiting programs initiated breastfeeding, above the estimated average of 81% in the U.S. population (Thomson et al., 2016). Yet we also found that infant bedsharing was reported by well over half the HFA program participants, significantly exceeding the rates reported by participants in a briefer nurse home visiting program and by women who did not receive home visiting services. Results suggested that the group differences in bedsharing were not due to differences in breastfeeding. This is the first known study to examine the effects of home visiting on bedsharing, and further research along these lines is warranted.

Home visiting is a promising strategy to address socioeconomic and racial/ethnic disparities in breastfeeding initiation, though further research is needed to determine whether these interventions promote breastfeeding continuation. Little is known about the effects of home visiting on bedsharing, and results from this study suggest that program effects may be variable. In keeping with rising interest in precision home visiting (Home Visiting Applied Research Collaborative, 2018), future studies should examine whether the impacts of home visiting on breastfeeding and bedsharing differ across population subgroups and community contexts. In addition, research could be advanced by examining potential moderators of program effects, including client variables such as service satisfaction and engagement as well as program features such as dosage, curricular focus, model fidelity, and provider type.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Joshua P. Mersky

References

Centers for Disease Control and Prevention. (2018). Breastfeeding among U.S. children born 2009-2015: CDC National Immunization Survey. https://www.cdc.gov/breastfeeding/data/nis_data/results.html

Colson

E. R.

Willinger

Rybin

Heeren

Smith

L. A.

Lister

Corwin

M. J.

(2013). Trends and factors associated with infant bed sharing, 1993-2010: The National Infant Sleep Position Study. JAMA Pediatrics, 167(11), 1032–1037. https://doi.org/10.1001/jamapediatrics.2013.2560

DiGirolamo

Thompson

Martorell

Fein

Grummer-Strawn

(2005). Intention or experience? Predictors of continued breastfeeding. Health Education & Behavior, 32(2), 208–226. https://doi.org/10.1177/1090198104271971

Edwards

R. C.

Thullen

M. J.

Korfmacher

Lantos

J. D.

Henson

L. G.

Hans

S. L.

(2013). Breastfeeding and complementary food: Randomized trial of community doula home visiting. Pediatrics, 132(Suppl. 2), S160–S166. https://doi.org/10.1542/peds.2013-1021P

Eidelman

A. I.

(2012). Breastfeeding and the use of human milk: An analysis of the American Academy of Pediatrics 2012 Breastfeeding policy statement. Breastfeeding Medicine, 7(5), 323–324. https://doi.org/10.1089/bfm.2012.0067

Hardison-Moody

MacNell

Elliott

Bowen

(2018). How social, cultural, and economic environments shape infant feeding for low-income women: A qualitative study in North Carolina. Journal of the Academy of Nutrition and Dietetics, 118(10), 1886–1894. https://doi.org/10.1016/j.jand.2018.01.008

Hauck

F. R.

Signore

Fein

S. B.

Raju

T. N.

(2008). Infant sleeping arrangements and practices during the first year of life. Pediatrics, 122(Suppl. 2), S113–S120. https://doi.org/10.1542/peds.2008-1315o

Healthy Families America. (2017). HFA best practice standards. Prevent Child Abuse America.

Home Visiting Applied Research Collaborative. (2018, June). Introduction to precision home visiting (Brief). https://www.hvresearch.org/wp-content/uploads/2018/06/PHV-Summary-Brief_Final.pdf

10.

Huang

Hauck

F. R.

Signore

Raju

T. N.

Huang

T. T. K.

Fein

S. B.

(2013). Influence of bedsharing activity on breastfeeding duration among US mothers. JAMA Pediatrics, 167(11), 1038–1044. https://doi.org/10.1001/jamapediatrics.2013.2632

11.

Hussain

Lowell

G. S.

Roehler

D. R.

Quinlan

K. P.

Tandon

S. D.

Schwartz

(2018). You can have your breastmilk and safe sleep too: A preliminary analysis of infant safe sleep data in a Midwestern home visiting program. Injury Epidemiology, 5(Suppl. 1), 14. https://doi.org/10.1186/s40621-018-0138-y

12.

IBM Corporation. (2017). IBM SPSS Statistics for Windows (Version 25.0) [Computer software].

13.

Chung

Raman

Trikalinos

T. A.

Lau

(2009). A summary of the Agency for Healthcare Research and Quality’s evidence report on breastfeeding in developed countries. Breastfeeding Medicine, 4(S1), S17-S30. https://doi.org/10.1089/bfm.2009.0050

14.

Kahan

B. C.

Jairath

Doré

C. J.

Morris

T. P.

(2014). The risks and rewards of covariate adjustment in randomized trials: An assessment of 12 outcomes from 8 studies. Trials, 15(1), 139. https://doi.org/10.1186/1745-6215-15-139

15.

Kronborg

Væth

Kristensen

(2012). The effect of early postpartum home visits by health visitors: A natural experiment. Public Health Nursing, 29(4), 289–301. https://doi.org/10.1111/j.1525-1446.2012.01019.x

16.

Lahr

M. B.

Rosenberg

K. D.

Lapidus

J. A.

(2007). Maternal-infant bedsharing: Risk factors for bedsharing in a population-based survey of new mothers and implications for SIDS risk reduction. Maternal and Child Health Journal, 11(3), 277–286. https://doi.org/10.1007/s10995-006-0166-z

17.

LeCroy

C. W.

Lopez

(2018). A randomized controlled trial of Healthy Families: 6-month and 1-year follow-up. Prevention Science, 33, 1761–1766. https://doi.org/10.1007/s11121-018-0931-4

18.

Darling

Maurice

Barker

Grummer-Strawn

L. M.

(2005). Breastfeeding rates in the United States by characteristics of the child, mother, or family: The 2002 National Immunization Survey. Pediatrics, 115(1), e31–e37. https://doi.org/10.1542/peds.2004-0481

19.

McGinnis

Lee

Kirkland

Miranda-Julian

Greene

(2018). Let’s talk about breastfeeding: The importance of delivering a message in a home visiting program. American Journal of Health Promotion, 32(4), 989–996. https://doi.org/10.1177/0890117117723802

20.

McKenna

J. J.

Gettler

L. T.

(2016). There is no such thing as infant sleep, there is no such thing as breastfeeding, there is only breastsleeping. Acta Paediatrica, 105(1), 17–21. https://doi.org/10.1111/apa.13161

21.

Merewood

Bugg

Burnham

Krane

Nickel

Broom

Edwards

Feldman-Winter

(2019). Addressing racial inequities in breastfeeding in the southern United States. Pediatrics, 143(2), e20181897. https://doi.org/10.1542/peds.2018-1897

22.

Mileva-Seitz

V. R.

Bakermans-Kranenburg

M. J.

Battaini

Luijk

M. P.

(2017). Parent-child bed-sharing: The good, the bad, and the burden of evidence. Sleep Medicine Reviews, 32, 4–27. https://doi.org/10.1016/j.smrv.2016.03.003

23.

Moon

R. Y

., & Task Force on Sudden Infant Death Syndrome. (2016). SIDS and other sleep-related infant deaths: Evidence base for 2016 updated recommendations for a safe infant sleeping environment. Pediatrics, 138(5), e20162940. https://doi.org/10.1542/peds.2016-2940

24.

Nievar

M. A.

Van Egeren

L. A.

Pollard

(2010). A meta-analysis of home visiting programs: Moderators of improvements in maternal behavior. Infant Mental Health Journal, 31(5), 499–520. https://doi.org/10.1002/imhj.20269

25.

Pugh

L. C.

Serwint

J. R.

Frick

K. D.

Nanda

J. P.

Sharps

P. W.

Spatz

D. L.

Milligan

R. A.

(2010). A randomized controlled community-based trial to improve breastfeeding rates among urban low-income mothers. Academic Pediatrics, 10(1), 14–20. https://doi.org/10.1016/j.acap.2009.07.005

26.

Salm Ward

T. C.

Doering

J. J

. (2014). Application of a socio-ecological model to mother-infant bed-sharing. Health Education & Behavior, 41(6), 577–589. https://doi.org/10.1177/1090198114543010

27.

Shah

M. K.

Austin

K. R.

(2014). Do home visiting services received during pregnancy improve birth outcomes? Findings from Virginia PRAMS 2007–2008. Public Health Nursing, 31(5), 405–413. https://doi.org/10.1111/phn.12103

28.

Sweet

M. A.

Appelbaum

M. I.

(2004). Is home visiting an effective strategy? A meta-analytic review of home visiting programs for families with young children. Child Development, 75(5), 1435–1456. https://doi.org/10.1111/j.1467-8624.2004.00750.x

29.

Thomson

J. L.

Tussing-Humphreys

L. M.

Goodman

M. H.

Landry

A. S.

Olender

S. E.

(2016). Low rate of initiation and short duration of breastfeeding in a maternal and infant home visiting project targeting rural, Southern, African American women. International Breastfeeding Journal, 12(1), 15. https://doi.org/10.1186/s13006-017-0108-y

30.

Victora

C. G.

Bahl

Barros

A. J.

França

G. V.

Horton

Krasevec

Murch

Sankar

M. J.

Walker

Rollins

N. C.

, & Lancet Breastfeeding Series Group. (2016). Breastfeeding in the 21st century: Epidemiology, mechanisms, and lifelong effect. Lancet, 387(10017), 475–490. https://doi.org/10.1016/S0140-6736(15)01024-7

31.

Willinger

C. W.

Hoffman

H. J.

Kessler

R. C.

Corwin

M. J.

(2003). Trends in infant bed sharing in the United States, 1993-2000: The National Infant Sleep Position Study. Archives of Pediatrics & Adolescent Medicine, 157(1), 43–49. https://doi.org/10.1001/archpedi.157.1.43