The Efficacy of Using Peer Mentors to Improve Maternal and Infant Health Outcomes in Hispanic Families: Findings from a Randomized Clinical Trial

The Efficacy of Using Peer Mentors to Improve Maternal and Infant Health Outcomes in Hispanic... Introduction The Maternal Infant Health Outreach Worker (MIHOW) program is a home visiting program, utilizing peer mentors to improve maternal/child health outcomes in underserved communities. Findings are presented from a randomized clinical trial (RCT) testing the efficacy of the MIHOW model in a sample of Hispanic women in Tennessee. We hypoth- esized maternal and infant outcomes would be better in women assigned to MIHOW than women assigned to the minimal education intervention (MEI) group (receipt of educational materials). Methods Women entered the study during pregnancy (< 26 weeks gestation) and were followed through 6 months postpartum. A total of 188 women were enrolled and randomly assigned (MEI = 94; MIHOW = 94), with 178 women completing the study (MEI = 87; MIHOW = 91). Results Positive and statistically significant (p < 0.01) effects of MIHOW were observed on breastfeeding self-efficacy and exclusivity, levels of depressive symptoms and parenting stress, safe sleep practices, and infant stimulation in the home. No statistically significant differences were noted in number of prenatal visits. Discussion Results expand limited empiric evidence and provide strong support of the effectiveness of MIHOW on improving health outcomes in this sample of Hispanic mothers and their infants. MIHOW is a viable option for providing culturally sensitive services to immigrant and underserved families. Keywords Hispanic · Home visit · Prenatal · Depressive symptoms · Peer mentors · Safe sleep Significance Peer mentors and home visiting may be an effective strategy for immigrant women and their infants but little rigorous evidence exists and findings are inconclusive. Findings from this RCT suggest that a series of home visits by peer mentors * Melanie Lutenbacher beginning during pregnancy until 6 months postpartum is an melanie.lutenbacher@vanderbilt.edu effective intervention in reducing depressive symptoms and Tonya Elkins parenting stress, and improving social and emotional support tonya.elkins@vanderbilt.edu in Hispanic women. Women who received MIHOW also Mary S. Dietrich exclusively breastfed their infants longer, had higher rates mary.dietrich@vanderbilt.edu of exclusive breastfeeding, placed their babies on their backs Anais Riggs more often and co-slept with their infants less frequently ariggs@cctenn.org than women in a minimal education intervention group. Schools of Nursing and Medicine (Pediatrics), Vanderbilt University, Nashville, TN, USA Introduction Maternal Infant Health Outreach Worker Program, Vanderbilt University, Nashville, TN, USA The Maternal Infant Health Outreach Worker (MIHOW) Schools of Medicine (Biostatistics, Psychiatry, VICC) and Nursing, Vanderbilt University, Nashville, TN, USA program began in 1982 as a program to address the lack of healthcare in low income, isolated communities in Catholic Charities of Tennessee, Inc., Nashville, TN, USA Vol:.(1234567890) 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S93 Appalachia. The program goals were to improve mater- Hispanic families, the largest ethnic minority in the US nal health and child development, combat isolation and (U.S. Census Bureau & Population Division 2013), do not increase access to health care. The key component of this have the same access to health care as non-Hispanic whites model is the outreach worker or peer mentor. These are (Kirby and Kaneda 2013). Lack of access can be influenced women recruited from the target community of the same by immigration status, socioeconomic status, low education, race, culture and language, who have strong problem- limited English proficiency, or other social determinants of solving and communication skills and familiarity with health (Velasco-Mondragon et al. 2016). resources (Elkins et al. 2013). The peer mentors receive The majority (68.3%) of Hispanic pregnant women in the intensive training to provide health education, social sup- US start prenatal care in the first trimester (U.S. Department port and linkage to community resources. Since its begin- of Health and Human Services 2013). Early prenatal care ning over 30 years ago, MIHOW has served an estimated can identify and rectify maternal and infant risks. Studies 15,000 families in the Southeastern United States (US) of prenatal care usage among Hispanic women in Califor- (Elkins et al. 2013). nia, New York, and Florida suggest that they are less likely Since the 1950s, Hispanic promotoras, similar to to adequately use prenatal services than US born citizens MIHOW peer mentors, have worked to reduce and elimi- (Fuentes-Afflick et al. 2006) but because the majority initi- nate health disparities (Andrews et al. 2004; Koniak-Grif- ate prenatal care early, pregnancy may be an excellent time fin et  al. 2015; O’Brien et  al. 2010; Tran et  al. 2014). to engage Hispanic families in a structured health promot- Promotor as have targeted a variety of health problems, ing program such as MIHOW. However, few studies have providing health education, helping families navigate sys- examined the effectiveness of home visitation programs in tems, making referrals, and sometimes even directly deliv- Hispanic families (O’Brien et al. 2010). ering medical services, but rigorous evaluation of efforts Hispanic women experience health disparities related to is limited (O’Brien et al. 2010). depression identification and treatment (Baker-Ericzen et al. Currently, MIHOW is considered a promising approach 2012). Rates of depressive symptoms have been reported as it meets the criteria defined by the federal Social Secu - between 20 and 38% in samples of Hispanic/Latino women rity Act, Title V, § 511 [42 U.S.C. § 711] (c). Specifically, in Florida, California, and Massachusetts (Gress-Smith MIHOW uses a research-based curriculum, was developed et al. 2012; Wassertheil-Smoller et al. 2014). Left untreated, by an institution of higher education, and the approach depression can have signic fi ant consequences for the woman works to achieve the benchmark areas and outcomes speci- and impact her ability to effectively parent her children. fied in the act. Further, the MIHOW program has demon- Empiric evaluation of the impact of a home visitation pro- strated some effectiveness based upon program evalua- gram on the mental health of mothers is lacking and is rare tion (Clinton 1992; Elkins and Clinton 2009). However, in Hispanic populations (Gomby 2005). empiric studies are needed to determine the program’s Breastfeeding benefits are widely known (Flores et al. impact and to be qualified as an evidence-based program 2016; Vaughn et al. 2010; Victora et al. 2016). While His- (HRSA 2016). For more information on MIHOW, visit panic mothers in the US tend to initiate breastfeeding at http://www.mihow .org. high rates, they also supplement with formula at high rates This report presents findings from a randomized clini- despite recommendations to exclusively breastfeed (Jones cal trial (RCT) that tested the home visiting model by peer et  al. 2015). Women who receive prenatal education and mentors in improving selected maternal and infant health home based postpartum support are more likely to initiate outcomes in a sample of pregnant Hispanic women liv- breastfeeding and continue to breastfeed for 6 months (Gill ing in a large city in Tennessee. Our global hypothesis et al. 2007). In a study of predominantly low-income Domin- was that women (and their infants) randomly assigned to ican women, higher breastfeeding self-efficacy scores were receive the MIHOW Program would have better health associated with more breastfeeding and exclusive breastfeed- outcomes than those women who were assigned to a mini- ing (Glassman et al. 2014). Strategies, such as MIHOW, that mal education intervention group (MEI—received printed encourage Hispanic mothers’ choice to breastfeed exclu- educational materials only). Specifically, the team hypoth- sively and through 6 months need to be evaluated. esized that the mothers receiving the MIHOW intervention While more than 90% of Hispanic children in the US are would be more likely than the comparison group to: (1) US citizens, Hispanic children disproportionately live in breastfeed longer; (2) delay feeding their infants solids; (3) poverty, suffer from health problems such as overweight/ put babies to sleep on their backs; (4) attend more prenatal obesity, and enter school inadequately prepared (Murphey care visits; (5) report lower levels of parental stress; (6) et al. 2014). Hispanic infants are less likely than White non- report fewer maternal depressive symptoms; (7) receive Hispanic children to be read to by their parents on a daily more referrals; (8) report higher levels of parental support; basis (Federal interagency forum on child and family sta- and, (9) read to their babies more often. tistics 2017). Despite many barriers to healthy outcomes, 1 3 S94 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 most Hispanic mothers report compliance with American study statistician via a computer-generated, permuted block Academy of Pediatrics (AAP) recommended infant safe program. Participants received their group assignment after sleep practices supporting back sleeping and discouraging the enrollment interview was completed. All data were col- co-sleeping practices (Provini et al. 2017). lected by trained study staff who were women hired specifi- cally for this project from the local communities who were both linguistically (i.e., native Spanish speakers and fluent Methods in English) and culturally competent. Data collectors were ‘blind’ to group assignment. All study staff completed exten- This single site, randomized clinical trial was approved by sive training related to the conduct of a randomized clinical the Institutional Review Board (IRB) of Vanderbilt Univer- trial and to study protocols. Data collectors used an inter- sity Medical Center in the Southeastern US. view guide at 5 points: enrollment (≤ 26 weeks pregnant), approximately 35 weeks pregnant, and 2 weeks, 2 months Participants and Sample Size Justification and 6 months postpartum. Interview guides were available in Spanish and English. To be enrolled in the study, women had to: be eligible to Each data collection interview took approximately 1 h. receive MIHOW services; self-identify as Hispanic; provide With the exception of the HOME measure, data collectors written confirmation of pregnancy ≤ 26 weeks gestation; read all questions and items aloud to women and used a reside within 30 miles of the study offices; and be willing to paper and pencil format to complete the interview guides. be randomized into one of two study groups. Women were The HOME was completed via observation of the data col- excluded from the study if they had previously received lector at each of the postpartum data collection points. All MIHOW services; had a severe mental or physical disability; participants received a $25 merchandise card at the end of or were under 18 years of age. each interview. Staff entered data into a REDCap database Study participants were 188 pregnant Hispanic women located on a secure password protected server at the associ- living in a large metropolitan area in (blinded) Tennessee. ated university. Monthly data fidelity checks of a random The original sample size for this study was justified by both selection of data forms were conducted by the project coor- the prior experience of the research team and the MIHOW dinator. Data were collected between July 2014 and Sep- program and a conservative estimate of the number of par- tember 2016. ticipants that could complete the protocol within the study period (based on typical MIHOW participation numbers). Comparison Condition That original number was 150 (75 per study group). A sta- tistical powering analysis revealed that groups of that size The comparison condition (minimal education intervention: were sufficient to detect common Cohen’s d effect size of MEI) consisted of distribution of printed educational materi- 0.46 (80% power, 2-sided alpha = 0.05). All effects would als about maternal and infant health and development at the be translated into this common index. A Cohen’s d of that end of each data collection interview to all study partici- magnitude was deemed clinically meaningful and therefore pants (i.e., women assigned to both study groups) in order to further justified our sample size goal. Recruitment and par - maintain the blind status of data collectors. Materials were ticipant interest as the study progressed was such that a deci- available in Spanish or English. sion was made by the research team to continue enrollment as long as the protocol could allow. Ultimately, a total of 188 Intervention Condition women were enrolled. Of those women, 178 completed the study. For study consort flow diagram, see Fig.  1. The intervention condition included the core elements of the MIHOW model. No adaptations were made to MIHOW Procedure content, level of intensity, or home visitor training require- ments. MIHOW interventionists were recruited from the Upon receiving IRB approval for the study protocol, recruit- local Hispanic community who completed 40 h of training ment began with a variety of recruitment strategies includ- to the MIHOW curriculum. The MIHOW model stresses ing distributing flyers at locations with a high volume of recognizing family strengths and utilizing those to address Hispanic customers (e.g., clinics, markets, apartment com- their own family needs (Elkins et al. 2013); relationships plexes, churches) and by word of mouth. All women inter- begin in pregnancy and consist of monthly home visits and ested in participating in the study were screened by trained periodic group gatherings. MIHOW protocols include lis- study staff to determine eligibility. All eligible women tening to maternal concerns, educating about objectives interested in participating in the study completed a written relevant to the woman’s stage of pregnancy or the age of informed consent. Group assignments were generated by the the child, such as healthy eating, developmental milestones, 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S95 Fig. 1 CONSORT flow diagram of the progress through the phases of the randomized trial attachment, and breastfeeding, and helping provide links to Measures needed medical and social services. Home visits typically last approximately 1 h. Due to the study’s limited funding, Primary outcomes were assessed with standardized meas- the duration of MIHOW services was limited to pregnancy ures and established questions from national sources (e.g., through 6 months of age rather than the typical duration to 2011/12 National Survey of Children’s Health (NSCH), the child’s third birthday. http://c hild healt hdat a .or g/lear n /NSCH/t opic s_q ues t ions; 1 3 S96 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Pregnancy Risk Assessment Monitoring System (PRAMS) Statistical Analyses Phase 6, https ://www .cdc.go v/prams /q ues t ionna ir e.htm). See Table 1 for study variables, measures and the time points for SPSS software was used to summarize study data and test data collection. Standardized measures are briefly described hypotheses. All analyses were done using intention-to-treat below. principles. Descriptive statistics and plots were used to sum- marize and initially inspect the distributions of demographic Breastfeeding Self‑Efficacy Scale‑Short Form (BSES‑SF) and study measures at each time of assessment. For those (Dennis 2003) measures assessed more than once over the course of the study, values of change in those measures were also gener- The 14-item measure assesses a mother’s confidence in her ated and summarized. Frequency distributions summarized ability to breastfeed her new infant and has been evaluated nominal and ordinal distributions; means and standard devi- among women from diverse cultures with adequate reliabil- ations summarized normal continuous distributions, median ity and validity (Dennis 2003; Vaughn et al. 2010). Cron- and inter-quartile range (IQR) skewed distributions. Infor- bach’s alpha of the scores in the current study ranged from mation from those descriptive and graphical evaluations was 0.93 to 0.95. used to determine the most appropriate test distribution to specify within the mixed-effects generalized linear models Parenting Stress Index‑Short Form (PSI) (Abidin 2012) used for testing study hypotheses (e.g., normal, log with a Tweedie, etc.). Within these models, the interaction effect of PSI was used to assess the level of stress in the parent–child study group and time of assessment (controlling for baseline system. This 36-item scale includes three domains: paren- values) provided the critical test of differences between the tal distress (PD), parent–child dysfunctional interaction study groups in the amount of change in an outcome meas- (P–CDI), and difficult child (DC) that when combined form ures. Effect sizes were generated for all of the comparisons a total stress scale. Available in Spanish, previous reports using Cohen’s d statistic. Statistical significance tests main- indicate strong reliability coefficients (Barroso et al. 2016). tained maximum type I error rates (alpha values) of < 0.05. Reliability coefficients of the scores in the current study ranged from 0.88 to 0.89. Results Edinburgh Postnatal Depression Scale (EPDS) (Cox et al. 1987) Sample Characteristics This widely used 10-item scale utilizes a 4-point response The average maternal age at enrollment was 29.6  years set to measure level of depressive symptoms with a pos- (SD = 6.5). Most women reported a Mexican heritage sible range of scores from 0 to 30. Most research on the (66.9%), less than a high school education (80.6%), never EPDS indicates a cut-off of 13 to indicate high depressive marrying (56.7%), and annual incomes less than $15,000 symptoms (Cox et al. 1996, 1987). Cronbach’s alphas of the (96.6%). Both study groups had similar sociodemographic scores in the current study ranged from 0.87 to 0.88. and scores for standardized measures used to assess out- comes. See Table 2 for detailed sample characteristics. Home Observation for Measurement of the Environment– Infant‑Toddler (HOME‑IT) (Caldwell, 1984) Primary Outcomes The 45-item observational HOME-IT Inventory assesses the Outcomes in the Child Health Domain quality and quantity of stimulation and support available to a child (birth to age three) in the home environment. The Summaries of infant feeding practices outcomes at 6 measure contains 6 subscales that assess specific parent- months postpartum are shown in Table 3. The strongest ing behaviors that support child learning and development effects of the MIHOW program were observed on the including: responsivity; acceptance; organization; learning BSES-SF scores and on the rates and duration of breast- materials; involvement; and variety in environment (Cald- feeding exclusivity (Cohen’s d = 0.38–0.76). Approxi- well and Bradley 1984). Adaptations to some items on the mately 80% (n = 68 of 86, 79.1%) of the women in the scale were made for infants < 6 months. MEI group reported never breastfeeding exclusively. That respective percentage was considerably lower in the MIHOW group (n = 50 of 90, 55.6%, d = 0.38, p = 0.011). The difference between the groups in duration was a median 1.4 weeks, with 25% of the MIHOW group 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S97 Table 1 Study variables, measures, and data collection time points by domain Variable Measure/question(s) Child health domain  Breastfeeding initiation • Did you ever breastfeed or pump breast milk to feed your new baby after delivery, even for a short period of time? (Source: PRAMS) Time point: ~2 weeks pp  Breastfeeding duration and exclusivity • Are you currently breastfeeding or feeding pumped milk to your new baby? (Source: PRAMS) • How old was your baby when she/he completely stopped breastfeeding or being fed breast milk? (Source: NSCH) • How old was your baby when she/he was first fed formula? • Over the last 24 h, how many times did you breastfeed your baby? • Over the last 24 h, how many times did the baby receive formula? (Source: NSCH) Time point: ~2 months pp, ~ 6 months pp  Breastfeeding self-efficacy • Breastfeeding self-efficacy scale (BSES-SF) (Dennis 2003) Time point: enrollment, ~ 2 weeks pp, ~ 6 months pp  Introduction of solid foods • How old was your baby when she/he was first fed anything other than breast milk of formula? (Source: NSCH) • How old was your new baby the first time he or she ate food, such as baby cereal, baby food, or any other food? (Source: PRAMS) • How often have you added cereal to your baby’s bottle in the past 2 weeks? (Source: IFPS II) Time point: ~2 weeks pp, ~ 2 months pp, ~ 6 months pp  Infant safe sleep • How do you most often lay your baby down to sleep now? • How often does your new baby sleep in the same bed with you or anyone else? (Source: PRAMS) Time point: ~2 weeks pp, ~ 2 months pp, ~ 6 months pp  Prenatal care visits • How many weeks or months pregnant were you when you had your first visit for prenatal care? (Source: PRAMS) • How many prenatal visits did you have during the entire pregnancy? (ask at ~ 2 weeks only) Time point: enrollment, ~ 35 weeks, ~ 2 weeks pp Maternal health domain  Parenting stress and support • Parenting stress index 4—short form (Abidin 2012) How often do you get the social and emotional support you need? Time point: enrollment, ~ 2 weeks pp, ~ 2 months pp, ~ 6 months pp  Maternal depression • Edinburgh postpartum depression scale (Cox et al. 1987) Time point: enrollment, ~ 35 weeks, ~ 2 weeks pp, ~ 2 months pp, ~ 6 months pp Linkages and referrals domain  Follow through with referrals • Since you started the study, has anyone talked to you about services or resources in your commu- nity you may qualify for (such as WIC, a food bank, legal or immigration services, or a children’s group)? • If yes, which services/resources did they talk to you about? • Have you called or visited any of the places they told you about? • Have you received any new services as a result of the referral? • If yes, list the services Time point: ~35 weeks, ~ 6 months pp Positive parenting domain  Parenting practices • HOME inventory (Caldwell and Bradley 1984) Time Point: ~2 weeks pp, ~ 2 months pp, ~ 6 months pp  Reading and singing • During the past week, how many days did you or other family members tell stories or sing songs to your baby? • During the past week, how many days did you or other family members read to your baby? (Source: NSCH) Time point: ~2 months pp, ~ 6 months pp 1 3 S98 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Table 2 Demographic characteristics at enrollment (N = 178) Overall MEI MIHOW p value (N = 178) (N = 87) (N = 91) Mean (SD) Age (years) 29.6 (6.5) 28.7 (6.3) 30.4 (6.6) 0.093 Nation of origin N (%) 0.281  Costa Rica 1 (0.6) 0 (0.0) 1 (1.1)  El Salvador 17 (9.6) 8 (9.2) 9 (9.9)  Guatemala 12 (6.7) 3 (3.4) 9 (9.9)  Honduras 28 (15.7) 17 (19.5) 11 (12.1)  Mexico 119 (66.9) 59 (67.8) 60 (65.9)  Peru 1 (0.6) 0 (0.0) 1 (1.1) Employment status* < 0.001 a b  Full-time 17 (9.6) 16 (18.4) 1 (1.1)  Part-time 28 (15.7) 16 (18.4) 12 (13.2)  Unemployed/looking 2 (1.1) 1 (1.1) 1 (1.1) a b  Unemployed/not looking 131 (73.6) 54 (62.1) 77 (84.6) Marital status 0.066  Married 70 (39.3) 40 (46.0) 30 (33.0)  Separated, divorced, widowed 7 (3.9) 0 (0.0) 7 (5.7)  Never married 101 (56.7) 47 (54.0) 54 (59.3) Highest grade completed N = 176 N = 89 0.334  8th grade or less 71 (40.3) 36 (41.4) 35 (39.3)  9th–12th grade, no diploma 71 (40.3) 31 (35.6) 40 (44.9)  High school diploma/GED 34 (19.3) 20 (23.0) 14 (15.7) Family income 0.599  < $10,000 122 (68.5) 57 (65.5) 65 (71.4)  $10,001–$15,000 50 (28.1) 27 (31.0) 23 (25.3)  $15,001–$40,000 6 (3.4) 3 (3.4) 3 (3.3) Median [IQR] (min, max) Months in U.S. 108.0 [36–156] (1, 408) 108.0 [48–144] (2, 318) 120.0 [36–156] (1, 408) 0.504 Months in (blinded) 89.5 [24–132] (1, 288) 84.0 [24–124] (1, 264) 96.0 [24–132] (1, 288) 0.526 Number of children in home 2.0 [1–3] (0, 5) 2.0 [1–3] (0, 5) 2.0 [1–3] (0, 5) 0.812 Number of adults and children in home 4.0 [3–6] (1, 9) 4.0 [3–6] (2, 9) 4.0 [4–5] (1, 9) 0.751 No respondents received unemployment or worker’s compensation *Superscripts indicate statistically significant post-hoc pairwise comparisons, Bonferroni-corrected, p < 0.05 exclusively breastfeeding for at least 6 weeks (d = 0.42, postpartum; that respective interval was 18–22 weeks for p = 0.005). The effects of the MIHOW program on breast- the women in the MIHOW group (see Table 3). feeding self-efficacy occurred between baseline and the Safe sleeping practices reported by the women in each initial postpartum assessment at 2-weeks with the MIHOW of the study groups are summarized in Table 4. The women group scores being higher at that assessment and main- in the MIHOW group were much more likely to report taining that difference throughout the 6 month postpartum positioning the infant on the back than did the women period (see Table 3). A related but secondary outcome was in the MEI group (~ 98 vs. 66–75%, d = 0.63, p < 0.001). noted to be significant. While the median time to initiation A related but secondary outcome was significant. The of other liquids was the same for both groups (20 weeks), MIHOW group of women reported ‘Never’ practicing a higher proportion of the MIHOW group delayed the ini- co-sleeping with the infant more than did the MEI group tiation of other liquids (d = 0.59, p < 0.001). 50% of the (81–86% vs. 28–33%, d = 1.17, p < 0.001, see Table 4). MEI women initiated other liquids between weeks 16–20 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S99 Table 3 Summaries of infant feeding practices by study group (N = 178) Overall MEI MIHOW p value N = 178 N = 87 N = 91 N (%) N = 177 N = 86 N = 91 Ever breastfed 149 (84.2) 71 (82.6) 78 (85.7) 0.565 (d = 0.13) Breastfeeding status N = 175 N = 85 N = 90 0.762 (d = 0.01)  Breastfeeding—6 months PP 87 (49.7) 42 (49.4) 45 (50.0)  Never breastfed 28 (16.0) 15 (17.6) 13 (14.4)  Stopped by 2 weeks PP 4 (2.3) 2 (2.4) 2 (2.2)  Stopped by 2 months PP 22 (12.6) 8 (9.4) 14 (15.6)  Stopped by 6 months PP 34 (19.4) 18 (21.2) 16 (17.8) Median [IQR] (Min, Max) N = 146 N = 70 N = 76 Breastfeeding duration in weeks 28.0 [12–28] 28.0 [12–28] 28.0 [12–28] 0.754 (1, 28) (1, 28) (2, 28) (d = 0.05) N (%) Breastfeeding exclusivity N = 176 N = 86 N = 90 0.011 a b (d = 0.38)  Never 118 (67.0) 68 (79.1) 50 (55.6) a b  Stopped by 2 weeks PP 31 (17.6) 10 (11.6) 21 (23.3) a b  Stopped by 2 months PP 24 (13.6) 7 (8.1) 17 (18.9)  Still exclusive 6 months PP 3 (1.7) 1 (1.2) 2 (2.2) Exclusive breastfeeding duration in weeks 0.4 [0–4] 0.3 [0–2] 1.4 [0–6] 0.005 (0, 28) 175 (0, 28) 86 (0, 28) 89 (d = 0.42) Time to first other liquid in weeks 20.0 [16–22] 20.0 [16–20] 20.0 [18–22] < 0.001 (2, 24) 161 (2, 24) 79 (4, 24) 82 (d = 0.59) Time to first food in weeks 20.0 [20–22] 20.0 [20–22] 21.0 [20–22] 0.201 (1, 24) 163 (1, 24) 80 (12, 24) 83 (d = 0.19) Median [IQR] N BSES scores < 0.001 (d = 0.76)  Baseline 54.0 [50–60] 127 53.0 [50–60] 66 54.0 [50–61] 61  2 weeks PP 56.0 [51–65] 145 52.0 [48–56] 69 61.0 [56–66] 76  2 months PP 56.0 [51–63] 121 51.0 [46–56] 60 62.0 [56–65] 61  6 months PP 57.5 [52–64] 88 53.0 [48–55] 42 64.0 [58–66] 46 Interaction effect of study group and time of assessment, no statistically significant difference at baseline, MIHOW > MEI 2 weeks, 2 and 6 months throughout the postpartum period (d = 0.57, p < 0.001). Par- Outcomes in the Maternal Health Domain enting stress and support were only assessed postpartum. Those scores are also summarized in Table 5. As shown, rel- Ninety-nine percent received prenatal care beginning at approximately 13 weeks gestation with about a total number ative to the women in the MEI group, women in the MIHOW group reported lower levels of parenting stress and higher of nine prenatal visits. No statistically significant differences between the groups (d = 0.04–0.12) were found. Median levels of available social and emotional help (d = 0.43 and 0.39 respectively, p < 0.001). maternal depressive symptom scores were 7.0 (of possible 30) at baseline with essentially equivalent group variability Outcomes in the Linkages and Referrals Domain in those scores (see Table 4). Compared to the MEI group, women in the MIHOW group demonstrated a statistically As shown in Table  6, there were statistically signifi- significant greater decrease in scores between the baseline and prenatal assessments with the values remaining lower cant differences between the groups in the receipt of and 1 3 S100 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Table 4 Summaries of sleeping practices by study group (N = 178) points, statistically significant differences between groups for the HOME-IT Inventory scores emerged. Compared Overall MEI MIHOW p value to observations of home environments of infants in the N = 178 N = 87 N = 91 MEI group, infants in the MIHOW group experienced a N (%) statistically significantly higher level of quality and quan- Sleep position < 0.001 tity of stimulation and support at each time of assessment (d = 0.63)  2 weeks PP N = 177 N = 86 N = 91 (d = 1.99–2.32, p < 0.001). Furthermore, women in the   On back 147 (83.1) 57 (66.3) 90 (98.9) MIHOW group reported greater frequency of singing songs   On side 30 (16.9) 29 (33.7) 1 (1.1) or telling stories to their child (d = 0.86, p < 0.001). Women   On stomach 0 0 0 in the MIHOW group reported greater frequency of read-  2 months PP N = 176 N = 86 N = 90 ing to their child (d = 1.53, p < 0.001). While more than   On back 145 (82.4) 57 (66.3) 88 (97.8) 90% of the women in the MIHOW group reported read-   On side 30 (17.0) 28 (32.6) 2 (2.2) ing three or more times per week to their child at both 2   On stomach 1 (0.6) 1 (1.2) 0 (0.0) and 6 months postpartum, fewer than 40% of the mothers  6 months PP in the MEI group reported doing so at either time of assess-   On back 154 (86.5) 65 (74.7) 89 (97.8) ment (p < 0.001). Rates decreased from 37 to 26% between   On side 23 (12.9) 21 (24.1) 2 (2.2) 2 and 6 months postpartum in the MEI group while the rate   On stomach 1 (0.6) 1 (1.1) 0 (0.0) increased from 92 to 97% during that period in the MIHOW Co-sleeping < 0.001 group (p < 0.039). (d = 1.17)  2 weeks PP N = 177 N = 86 N = 91   Always 21 (11.9) 21 (24.4) 0 (0.0)   Often 16 (9.0) 13 (15.1) 3 (3.3) Conclusions for Practice   Sometimes 17 (9.6) 13 (15.1) 4 (4.4)   Rarely 21 (11.9) 11 (12.8) 10 (11.0) Using an intent to treat approach, the majority of our hypoth-   Never 102 (57.6) 28 (32.6) 74 (81.3) eses were supported and provide strong evidence of the  2 months PP N = 176 N = 86 N = 90 effectiveness of MIHOW on improving health outcomes in   Always 18 (10.2) 17 (19.8) 1 (1.1) this sample of Hispanic mothers and their infants. Overall,   Often 12 (6.8) 12 (14.0) 0 (0.0) women assigned to the MIHOW group had fewer depres-   Sometimes 20 (11.4) 18 (20.9) 2 (2.2) sive symptoms and less parenting stress and more social   Rarely 24 (13.6) 15 (17.4) 9 (10.0) and emotional help, and better infant feeding and safe sleep   Never 102 (58.0) 24 (27.9) 78 (86.7) practices.  6 Months PP Similar to national statistics (U.S. Department of Health   Always 17 (9.6) 16 (18.4) 1 (1.1) and Human Services 2013), the majority of women in our   Often 15 (8.4) 13 (14.9) 2 (2.2) study entered prenatal care early, underscoring the oppor-   Sometimes 18 (10.1) 15 (17.2) 3 (3.3) tunity to engage with Hispanic families during this time   Rarely 22 (12.4) 15 (17.2) 7 (7.7) period. As noted in our findings, the women assigned to   Never 106 (59.6) 28 (32.2) 78 (85.7) MIHOW had many better outcomes than women who only received the MEI. Main effect of study group (on back: MIHOW > MEI, p < 0.001) Coupling a home visitation program by trained peer Main effect of study group (never: MIHOW > MEI, p < 0.001) mentors such as MIHOW with standard prenatal care has potential to improve maternal and child health outcomes. follow through with referrals. Women in the MIHOW group At all postpartum time points, women in the MIHOW group received more referrals for additional services from com- reported fewer depressive symptoms, less parenting stress munity providers than did those in the MEI group (80–100 and more social and emotional help than women in the MEI vs. 22–28%, d = 1.77, p < 0.001), connected with those group. These findings depict a mother who may be more resources (65–81 vs. 56–54%, d = 0.31, p = 0.028), and able to effectively engage with her infant (Gress-Smith et al. received more new services (64–80 vs. 56–54%, d = 0.30, 2012; Nelson et al. 2016). At all postpartum time points, p = 0.035) than women in the MEI group. data collectors (blinded to group assignment and using a standardized tool) observed a higher level of quality and Outcomes in the Parenting Practices Domain quantity of stimulation and support available to the child in the home environment in mothers assigned to MIHOW Measures and indicators of several types of parenting prac- than in women assigned to the MEI group. Women in the tices are summarized in Table  7. At all postpartum time MIHOW group also reported a greater frequency of singing 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S101 Table 5 Summaries of prenatal care, maternal depressive symptoms, stress, and support by study group (N = 178) Overall MEI MIHOW p value N = 178 N = 87 N = 91 Prenatal care N (%) N = 172 N = 83 N = 89 Receiving prenatal care 0.960 (d = 0.04)  No 2 (1.2) 1 (1.2) 1 (1.1)  Yes 170 (98.8) 82 (98.8) 88 (98.9) Mean (SD) (Min, Max) N Time to first prenatal visit (weeks) 13.0 (5.0) 12.7 (5.0) 13.3 (4.9) 0.397 (3, 26) 170 (3, 26) 82 (3, 25) 88 (d = 0.12) Median [IQR] N EPDS < 0.001 (d = 0.57)  Baseline 7.0 [2–10] 178 7.0 [3–9] 87 7.0 [2–10] 91  Prenatal 1.0 [0–6] 172 4.0 [0–7] 83 0.0 [0–2] 89  2 weeks PP 2.0 [0–5] 177 5.0 [2–8] 86 0.0 [0–1] 91  2 months PP 0.0 [0–3] 176 3.0 [0–6] 86 0.0 [0–0] 90  6 months PP 0.0 [0–1] 178 0.0 [0–4] 87 0.0 [0–0] 91 PSI total stress < 0.001 (d = 0.43)  2 weeks PP 75.0 [72–80] 177 76.0 [74–81] 86 74.0 [66–79] 91  2 months PP 75.0 [72–80] 176 76.0 [74–81] 86 74.0 [68–79] 90  6 months PP 75.0 [73–81] 178 77.0 [75–82] 87 74.0 [70–79] 91 N (%) Social and emotional help < 0.001 (d = 0.39)  2 weeks PP N = 177 N = 86 N = 91   Always 146 (82.5) 58 (67.4) 88 (96.7)   Usually 30 (16.9) 27 (31.4) 3 (3.3)   Sometimes 1 (0.6) 1 (1.2) 0 (0.0)   Never 0 (0.0) 0 (0.0) 0 (0.0)  2 months PP N = 176 N = 86 N = 90   Always 138 (78.4) 58 (67.4) 80 (88.9)   Usually 33 (18.8) 27 (24.9) 9 (10.0)   Sometimes 4 (2.3) 4 (4.7) 0 (0.0)   Never 1 (0.6) 0 (0.0) 1 (1.1)  6 months PP N = 178 N = 87 N = 91   Always 124 (69.7) 37 (42.5) 87 (95.6)   Usually 52 (29.2) 49 (56.3) 3 (3.3)   Sometimes 2 (1.1) 1 (1.1) 1 (1.1)   Never 0 (0.0) 0 (0.0) 0 (0.0) Interaction effect of study group and time of assessment, no statistically significant difference at baseline, MIHOW < MEI 2 weeks, 2- and 6-months: p < 0.001 Main effect of study group (MIHOW < MEI) Main effect of study group (MIHOW > MEI) songs, telling stories and reading books to their child than of Health and Human Services 2013), most women in the mothers in the comparison group. All of these activities study initiated breastfeeding. We did not find any differences are precursors to appropriate child development and school between groups for breastfeeding rates at 6 months postpar- readiness (Nelson et al. 2016). tum or duration of breastfeeding. Our findings of approxi- In regards to breastfeeding practices, as in other samples mately 84% of all study participants providing some breast- of Hispanic mothers (Flores et al. 2016; U.S. Department feeding is consistent with national data (U.S. Department 1 3 S102 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Table 6 Summaries of linkages and referrals by study group attaining the HP2020 objective related to exclusive breast- (N = 178) feeding at 6 months (i.e., 25.5%), women in the MIHOW group did report more breastfeeding exclusivity at 6 months Overall MEI MIHOW p-value postpartum and longer duration of exclusive breastfeeding. N, n (%) Breastfeeding self-efficacy also was higher in the interven- Referred to resources < 0.001 tion group at all postpartum time points. Longer duration (d = 1.77)  Prenatal 172, 89 83, 18 (21.7) 89, 71 (79.8) of breastfeeding exclusivity has many potential benefits for (51.7) both the mother and the infant (Victora et al. 2016).  6 months 178, 115 87, 24 (27.6) 91, 91 Infant mortality rates are generally low among Hispanic PP (64.6) (100.0) families (U.S. Department of Health and Human Services Of referrals, made appointments/visits 0.028 (d = 0.31) 2013) Safe sleep practices are critical to reducing infant  Prenatal 89, 56 (62.9) 18, 10 (55.6) 70, 46 (64.8) mortality, particularly sudden infant death syndrome (SIDS)  6 months 114, 86 24, 13 (54.2) 90, 73 (81.1) PP (75.4) (American Academy of Pediatrics 2000). While both groups Of referrals, received new services 0.035 of mothers met or exceeded a reported national rate (i.e., (d = 0.30)  Prenatal 88, 55 (62.5) 18, 10 (55.6) 70, 45 (64.3) 65%) for placing their infants on their back to sleep, almost  6 months 115, 86 24, 13 (54.2) 91, 73 (80.2) 100% of MIHOW mothers reported at all time points placing PP (74.8) their infants on their backs. MIHOW families reported much a less co-sleeping than families in the MEI group. None of the reported resources or service types included mental Identifying maternal and family needs and providing health services appropriate referrals are important aspects of maternal child health care. Follow through on referrals is often low in all of Health and Human Services 2013) and exceeds the racial and ethnic groups (Anisfeld et al. 2004). In this study, related Healthy People 2020 (Office of Disease Prevention both receipt of and follow through with referrals was greater and Health Promotion 2017) objective (Office of Disease in the MIHOW group than in the MEI group. Shared lan- Prevention and Health Promotion 2017). Although not yet guage and cultural background between study participants Table 7 Summaries of parenting practices by study group (N = 178) Overall MEI MIHOW p-value N = 178 N = 87 N = 91 Median [IQR] (min, max) N HOME score  2 weeks PP (max = 26) 19.0 [15–22] 15.0 [13–18] 21.0 [19–23] < 0.001 (6,25) 177 (6, 23) 86 (15, 25) 91  2 months PP (max = 32) 24.0 [19–27] 19.0 [17–23] 27.0 [24–28] < 0.001 (7, 30) 175 (7, 27) 85 (15, 30) 90  6 months PP (max = 45) 37.0 [33–40] 33.0 [30–36] 40.0 [38–42] < 0.001 (23, 44) 178 (23, 41) 87 (33, 44) 91 2 months PP assessment  Songs and stories 5.0 [5–6] 5.0 [4–5] 6.0 [5–7] < 0.001 (# Days past week) (2, 7) 176 (2, 7) 86 (3, 7) 90  Read 4.0 [0–5] 0.0 [0–4] 5.0 [4–6] < 0.001 (# Days past week) (0, 7) 176 (0, 6) 86 (0, 7) 90 N (%) Overall MEI only MIHOW N = 176 N = 86 N = 90  Reads stories ≥ 3 times per week 0.039   2 months PP 115 (65.3) 32 (37.2) 83 (92.2)   6 months PP 109 (61.9) 22 (25.6) 87 (96.7) Interaction effect of study group and time of assessment. MEI decreased at 6-months compared to rate at 2-months; MIHOW group remained at similar level; overall main effect of study group: p < 0.001 The Home is essentially a different measure at each time of assessment therefore only group differences at each time of assessment were con- ducted 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S103 Acknowledgements Research reported in this publication was sup- and peer mentors may have enhanced participant motivation ported by the Affordable Care Act Maternal, Infant and Early Child- and ability to follow through with referrals to new services hood Home Visiting Program under Award Number D89MC23542 and (Andrews et al. 2004). by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR000445. The content is solely the responsibility of the authors and does not neces- sarily represent the official views of the National Institutes of Health. Study Limitations The authors would like to thank Chrystal Fizer and Jodie Upchurch from Vanderbilt University and Deborah Narrigan for their help in Our study has both strengths and limitations. The use of a preparing this manuscript. randomized controlled design minimized potential for bias. Open Access This article is distributed under the terms of the Crea- This study had a very high retention rate of participants as tive Commons Attribution 4.0 International License (http://creat iveco compared to other home visitation studies with Hispanic mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- participants (Nguyen et  al. 2003). MIHOW extensively tion, and reproduction in any medium, provided you give appropriate trains women from the community being served to conduct credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. the home visits. We also used data collectors who spoke Spanish and were from the same community. We believe the match of peer mentors and data collectors with the study participants added to the successful retention of participants. References One notable limitation of the study design was that because Abidin, R. R. (2012). Parenting stress index. Odessa, FL: Psychological all participants received a standard packet of printed edu- Assessment Resources. cational materials, we did not have a true control group. American Academy of Pediatrics. (2000). Changing concepts of sud- While this may have made finding differences between den infant death syndrome: Implications for infant sleeping envi- groups more difficult, most differences between our groups ronment and sleep position. Pediatrics, 105(3), 650–656. Andrews, J. O., Felton, G., Wewers, M. E., & Heath, J. (2004). Use yielded large effect sizes. The length of the study is another of community health workers in research with ethnic minority limitation. While the MIHOW intervention is designed to women. Journal of Nursing Scholarship, 36(4), 358–365. continue until children reach 3 years of age, the duration of Anisfeld, E., Sandy, J., & Guterman, N. B. (2004). Best beginnings: A the study’s funding only allowed outcomes to be measured randomized controlled trial of a paraprofessional home visiting program. Final Report Columbia. Project Report. until 6 months postpartum. Baker-Ericzen, M. J., Connelly, C. D., Hazen, A. L., Duenas, C., Landsverk, J. A., & Horwitz, S. M. (2012). A collaborative care telemedicine intervention to overcome treatment barriers for Latina women with depression during the perinatal period. Fami- lies, Systems and Health, 30(3), 224–240. Conclusion Barroso, N. E., Hungerford, G. M., Garcia, D., Graziano, P. A., & Bag- ner, D. M. (2016). Psychometric properties of the Parenting Stress Findings from this study expand the limited empiric evi- Index-Short Form (PSI-SF) in a high-risk sample of mothers and dence related to home visitation services. It demonstrates their infants. Psychological Assessment, 28(10), 1331–1335. Caldwell, B., & Bradley, R. (1984). Home observation for measure- beneficial effects of a well-trained and supervised peer-to- ment of the environment (HOME)—Revised Edition. Little Rock: peer model for a sample of Hispanic mothers and infants. University of Arkansas. Understanding the nuances of providing services to both Census Bureau, U. S. & Population Division. (2016). Annual estimates majority and minority families in the context of a chang- of the resident population by sex, age, race, and Hispanic origin for the United States and states: April 1, 2010 to July 1, 2015 from ing medical care landscape is necessary to providing qual- https ://fact finder .censu s.gov/faces /table servi ces/jsf/pages /produ ity care and developing appropriate policy. Results provide ctvie w.xhtml ?src=bkmk. strong support of the efficacy of the MIHOW program and Clinton, B. (1992). The maternal infant health outreach worker project: the potentially high retention rates for participants. This Appalachian communities help their own English Dance and Song (pp. 23–45) M. Larner, R. Halpern, O. Harvaky. program should be considered when planning home visita- Cox, J. L., Chapman, G., Murray, D., & Jones, P. (1996). Validation of tion services for childbearing immigrant and underserved the Edinburgh postnatal depression scale (EPDS) in non-postnatal families. MIHOW has established standards of practice, a women. Journal of Affective Disorders, 39(3), 185–189. research-based curriculum, and an accreditation component, Cox, J. L., Holden, J. M., & Sagovsky, R. (1987). Detection of postna- tal depression. Development of the 10-item Edinburgh postnatal along with decades of experience in several states. It is also depression scale. British Journal of Psychiatry, 150, 782–786. cost effective. Even with the extensive training, intensive Dennis, C. L. (2003). The breastfeeding self-ec ffi acy scale: Psychomet - support, and supervision needed for peer mentors, program ric assessment of the short form. Journal of Obstetric, Gyneco- costs are usually less, overall, than models requiring early logic, and Neonatal Nursing, 32(6), 734–744. Elkins, T., & Clinton, B. (2009). Vanderbilt’s Maternal Infant Health childhood or medical professionals to conduct home visits. Outreach Worker (MIHOW) program: A collaborative program Additional longitudinal studies are needed to further under- stand the sustained impact of MIHOW on health outcomes. 1 3 S104 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 to improve perinatal health and child development in low-income Nelson, B. B., Dudovitz, R. N., Coker, T. R., Barnert, E. S., Biely, families. Nashville: Vanderbilt University. C., Li, N., et al. (2016). Predictors of poor school readiness in Elkins, T., del Pilar Aguinaga, M., Clinton-Selin, C., Clinton, B., & children without developmental delay at age 2. Pediatrics and Gotterer, G. (2013). The maternal infant health outreach worker Neonatology, e20154477. program in low-income families. Journal of Health Care for the Nguyen, J. D., Carson, M. L., Parris, K. M., & Place, P. (2003). A com- Poor and Underserved, 24(3), 995–1001. parison pilot study of public health field nursing home visitation Federal Interagency Forum on Child and Family Statistics. (2017). program interventions for pregnant Hispanic adolescents. Public America’s children: Key national indicators of well-being. from Health Nursing, 20(5), 412–418. https ://www.child stats .gov/pdf/ac201 7/ac_17.pdf. O’Brien, M. J., Halbert, C. H., Bixby, R., Pimentel, S., & Shea, J. A. Flores, A., Anchondo, I., Huang, C., Villanos, M., & Finch, C. (2016). (2010). Community health worker intervention to decrease cer- “Las Dos Cosas”, or Why Mexican American Mothers Breast- vical cancer disparities in Hispanic women. Journal of General Feed, But Not for Long. Southern Medical Journal, 109(1), Internal Medicine, 25(11), 1186–1192. 42–50. Office of Disease Prevention and Health Promotion. (2017). Maternal Fuentes-Afflick, E., Hessol, N. A., Bauer, T., O’Sullivan, M. J., Gomez- infant and child health. Healthy People 2020. from https ://www. Lobo, V., Holman, S., et al. (2006). Use of prenatal care by His-healt hypeo ple.gov/2020/topic s-objec tives /topic /mater nal-infan panic women after welfare reform. Obstetrics and Gynecology, t-and-child -healt h/objec tives . 107(1), 151–160. Provini, L. E., Corwin, M. J., Geller, N. L., Heeren, T. C., Moon, R. Gill, S. L., Reifsnider, E., & Lucke, J. F. (2007). Effects of support on Y., Rybin, D. V., et al. (2017). Differences in infant care practices the initiation and duration of breastfeeding. Western Journal of and smoking among Hispanic mothers living in the United States. Nursing Research, 29(6), 708–723. Journal of Pediatrics, 182, 321–326. Glassman, M. E., McKearney, K., Saslaw, M., & Sirota, D. R. (2014). Tran, A. N., Ornelas, I. J., Kim, M., Perez, G., Green, M., Lyn, M. J., Impact of breastfeeding self-efficacy and sociocultural factors on et al. (2014). Results from a pilot promotora program to reduce early breastfeeding in an urban, predominantly Dominican com- depression and stress among immigrant Latinas. Health Promo- munity. Breastfeeding Medicine, 9(6), 301–307. tion Practice, 15(3), 365–372. Gomby, D. S. (2005). Invest in kids working paper No. 7 Home visita- U.S. Department of Health and Human Services. (2013). Child Health tion in 2005: Outcomes for children and parents (Vol. 7). USA 2013. Rockville, Maryland. Gress-Smith, J. L., Luecken, L. J., Lemery-Chalfant, K., & Howe, R. Vaughn, L. M., Ireton, C., Geraghty, S. R., Diers, T., Nino, V., Fal- (2012). Postpartum depression prevalence and impact on infant ciglia, G. A., et al. (2010). Sociocultural influences on the deter - health, weight, and sleep in low-income and ethnic minority minants of breast-feeding by Latina mothers in the Cincinnati women and infants. Maternal and Child Health Journal, 16(4), area. Family and Community Health, 33(4), 318–328. 887–893. Velasco-Mondragon, E., Jimenez, A., Palladino-Davis, A. G., Davis, HRSA. (2016). Demonstrating improvement in the maternal, infant, D., & Escamilla-Cejudo, J. A. (2016). Hispanic health in the USA: and early childhood home visiting program: A report to congress. A scoping review of the literature. Public Health Reviews, 37(1), Jones, K. M., Power, M. L., Queenan, J. T., & Schulkin, J. (2015). 31. Racial and ethnic disparities in breastfeeding. Breastfeeding Medi- Victora, C. G., Bahl, R., Barros, A. J., Franca, G. V., Horton, S., Kra- cine, 10(4), 186–196. sevec, J., et al. (2016). Breastfeeding in the 21st century: Epide- Kirby, J. B., & Kaneda, T. (2013). ‘Double jeopardy’ measure suggests miology, mechanisms, and lifelong effect. Lancet, 387(10017), blacks and hispanics face more severe disparities than previously 475–490. indicated. Health Affairs, 32(10), 1766–1772. Wassertheil-Smoller, S., Arredondo, E. M., Cai, J., Castaneda, S. F., Koniak-Griffin, D., Brecht, M. L., Takayanagi, S., Villegas, J., Melen- Choca, J. P., Gallo, L. C., et al. (2014). Depression, anxiety, anti- drez, M., & Balcazar, H. (2015). A community health worker-led depressant use, and cardiovascular disease among Hispanic men lifestyle behavior intervention for Latina (Hispanic) women: Fea- and women of different national backgrounds: Results from the sibility and outcomes of a randomized controlled trial. Interna- Hispanic Community Health Study/Study of Latinos. Annals of tional Journal of Nursing Studies, 52(1), 75–87. Epidemiology, 24(11), 822–830. Murphey, D., Guzman, L., & Torres, A. (2014). America’s Hispanic Children: Gaining ground, looking forward. 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Maternal and Child Health Journal Springer Journals

The Efficacy of Using Peer Mentors to Improve Maternal and Infant Health Outcomes in Hispanic Families: Findings from a Randomized Clinical Trial

Free
13 pages

Loading next page...
 
/lp/springer_journal/the-efficacy-of-using-peer-mentors-to-improve-maternal-and-infant-VRYF96jDYc
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Medicine & Public Health; Public Health; Sociology, general; Population Economics; Pediatrics; Gynecology; Maternal and Child Health
ISSN
1092-7875
eISSN
1573-6628
D.O.I.
10.1007/s10995-018-2532-z
Publisher site
See Article on Publisher Site

Abstract

Introduction The Maternal Infant Health Outreach Worker (MIHOW) program is a home visiting program, utilizing peer mentors to improve maternal/child health outcomes in underserved communities. Findings are presented from a randomized clinical trial (RCT) testing the efficacy of the MIHOW model in a sample of Hispanic women in Tennessee. We hypoth- esized maternal and infant outcomes would be better in women assigned to MIHOW than women assigned to the minimal education intervention (MEI) group (receipt of educational materials). Methods Women entered the study during pregnancy (< 26 weeks gestation) and were followed through 6 months postpartum. A total of 188 women were enrolled and randomly assigned (MEI = 94; MIHOW = 94), with 178 women completing the study (MEI = 87; MIHOW = 91). Results Positive and statistically significant (p < 0.01) effects of MIHOW were observed on breastfeeding self-efficacy and exclusivity, levels of depressive symptoms and parenting stress, safe sleep practices, and infant stimulation in the home. No statistically significant differences were noted in number of prenatal visits. Discussion Results expand limited empiric evidence and provide strong support of the effectiveness of MIHOW on improving health outcomes in this sample of Hispanic mothers and their infants. MIHOW is a viable option for providing culturally sensitive services to immigrant and underserved families. Keywords Hispanic · Home visit · Prenatal · Depressive symptoms · Peer mentors · Safe sleep Significance Peer mentors and home visiting may be an effective strategy for immigrant women and their infants but little rigorous evidence exists and findings are inconclusive. Findings from this RCT suggest that a series of home visits by peer mentors * Melanie Lutenbacher beginning during pregnancy until 6 months postpartum is an melanie.lutenbacher@vanderbilt.edu effective intervention in reducing depressive symptoms and Tonya Elkins parenting stress, and improving social and emotional support tonya.elkins@vanderbilt.edu in Hispanic women. Women who received MIHOW also Mary S. Dietrich exclusively breastfed their infants longer, had higher rates mary.dietrich@vanderbilt.edu of exclusive breastfeeding, placed their babies on their backs Anais Riggs more often and co-slept with their infants less frequently ariggs@cctenn.org than women in a minimal education intervention group. Schools of Nursing and Medicine (Pediatrics), Vanderbilt University, Nashville, TN, USA Introduction Maternal Infant Health Outreach Worker Program, Vanderbilt University, Nashville, TN, USA The Maternal Infant Health Outreach Worker (MIHOW) Schools of Medicine (Biostatistics, Psychiatry, VICC) and Nursing, Vanderbilt University, Nashville, TN, USA program began in 1982 as a program to address the lack of healthcare in low income, isolated communities in Catholic Charities of Tennessee, Inc., Nashville, TN, USA Vol:.(1234567890) 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S93 Appalachia. The program goals were to improve mater- Hispanic families, the largest ethnic minority in the US nal health and child development, combat isolation and (U.S. Census Bureau & Population Division 2013), do not increase access to health care. The key component of this have the same access to health care as non-Hispanic whites model is the outreach worker or peer mentor. These are (Kirby and Kaneda 2013). Lack of access can be influenced women recruited from the target community of the same by immigration status, socioeconomic status, low education, race, culture and language, who have strong problem- limited English proficiency, or other social determinants of solving and communication skills and familiarity with health (Velasco-Mondragon et al. 2016). resources (Elkins et al. 2013). The peer mentors receive The majority (68.3%) of Hispanic pregnant women in the intensive training to provide health education, social sup- US start prenatal care in the first trimester (U.S. Department port and linkage to community resources. Since its begin- of Health and Human Services 2013). Early prenatal care ning over 30 years ago, MIHOW has served an estimated can identify and rectify maternal and infant risks. Studies 15,000 families in the Southeastern United States (US) of prenatal care usage among Hispanic women in Califor- (Elkins et al. 2013). nia, New York, and Florida suggest that they are less likely Since the 1950s, Hispanic promotoras, similar to to adequately use prenatal services than US born citizens MIHOW peer mentors, have worked to reduce and elimi- (Fuentes-Afflick et al. 2006) but because the majority initi- nate health disparities (Andrews et al. 2004; Koniak-Grif- ate prenatal care early, pregnancy may be an excellent time fin et  al. 2015; O’Brien et  al. 2010; Tran et  al. 2014). to engage Hispanic families in a structured health promot- Promotor as have targeted a variety of health problems, ing program such as MIHOW. However, few studies have providing health education, helping families navigate sys- examined the effectiveness of home visitation programs in tems, making referrals, and sometimes even directly deliv- Hispanic families (O’Brien et al. 2010). ering medical services, but rigorous evaluation of efforts Hispanic women experience health disparities related to is limited (O’Brien et al. 2010). depression identification and treatment (Baker-Ericzen et al. Currently, MIHOW is considered a promising approach 2012). Rates of depressive symptoms have been reported as it meets the criteria defined by the federal Social Secu - between 20 and 38% in samples of Hispanic/Latino women rity Act, Title V, § 511 [42 U.S.C. § 711] (c). Specifically, in Florida, California, and Massachusetts (Gress-Smith MIHOW uses a research-based curriculum, was developed et al. 2012; Wassertheil-Smoller et al. 2014). Left untreated, by an institution of higher education, and the approach depression can have signic fi ant consequences for the woman works to achieve the benchmark areas and outcomes speci- and impact her ability to effectively parent her children. fied in the act. Further, the MIHOW program has demon- Empiric evaluation of the impact of a home visitation pro- strated some effectiveness based upon program evalua- gram on the mental health of mothers is lacking and is rare tion (Clinton 1992; Elkins and Clinton 2009). However, in Hispanic populations (Gomby 2005). empiric studies are needed to determine the program’s Breastfeeding benefits are widely known (Flores et al. impact and to be qualified as an evidence-based program 2016; Vaughn et al. 2010; Victora et al. 2016). While His- (HRSA 2016). For more information on MIHOW, visit panic mothers in the US tend to initiate breastfeeding at http://www.mihow .org. high rates, they also supplement with formula at high rates This report presents findings from a randomized clini- despite recommendations to exclusively breastfeed (Jones cal trial (RCT) that tested the home visiting model by peer et  al. 2015). Women who receive prenatal education and mentors in improving selected maternal and infant health home based postpartum support are more likely to initiate outcomes in a sample of pregnant Hispanic women liv- breastfeeding and continue to breastfeed for 6 months (Gill ing in a large city in Tennessee. Our global hypothesis et al. 2007). In a study of predominantly low-income Domin- was that women (and their infants) randomly assigned to ican women, higher breastfeeding self-efficacy scores were receive the MIHOW Program would have better health associated with more breastfeeding and exclusive breastfeed- outcomes than those women who were assigned to a mini- ing (Glassman et al. 2014). Strategies, such as MIHOW, that mal education intervention group (MEI—received printed encourage Hispanic mothers’ choice to breastfeed exclu- educational materials only). Specifically, the team hypoth- sively and through 6 months need to be evaluated. esized that the mothers receiving the MIHOW intervention While more than 90% of Hispanic children in the US are would be more likely than the comparison group to: (1) US citizens, Hispanic children disproportionately live in breastfeed longer; (2) delay feeding their infants solids; (3) poverty, suffer from health problems such as overweight/ put babies to sleep on their backs; (4) attend more prenatal obesity, and enter school inadequately prepared (Murphey care visits; (5) report lower levels of parental stress; (6) et al. 2014). Hispanic infants are less likely than White non- report fewer maternal depressive symptoms; (7) receive Hispanic children to be read to by their parents on a daily more referrals; (8) report higher levels of parental support; basis (Federal interagency forum on child and family sta- and, (9) read to their babies more often. tistics 2017). Despite many barriers to healthy outcomes, 1 3 S94 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 most Hispanic mothers report compliance with American study statistician via a computer-generated, permuted block Academy of Pediatrics (AAP) recommended infant safe program. Participants received their group assignment after sleep practices supporting back sleeping and discouraging the enrollment interview was completed. All data were col- co-sleeping practices (Provini et al. 2017). lected by trained study staff who were women hired specifi- cally for this project from the local communities who were both linguistically (i.e., native Spanish speakers and fluent Methods in English) and culturally competent. Data collectors were ‘blind’ to group assignment. All study staff completed exten- This single site, randomized clinical trial was approved by sive training related to the conduct of a randomized clinical the Institutional Review Board (IRB) of Vanderbilt Univer- trial and to study protocols. Data collectors used an inter- sity Medical Center in the Southeastern US. view guide at 5 points: enrollment (≤ 26 weeks pregnant), approximately 35 weeks pregnant, and 2 weeks, 2 months Participants and Sample Size Justification and 6 months postpartum. Interview guides were available in Spanish and English. To be enrolled in the study, women had to: be eligible to Each data collection interview took approximately 1 h. receive MIHOW services; self-identify as Hispanic; provide With the exception of the HOME measure, data collectors written confirmation of pregnancy ≤ 26 weeks gestation; read all questions and items aloud to women and used a reside within 30 miles of the study offices; and be willing to paper and pencil format to complete the interview guides. be randomized into one of two study groups. Women were The HOME was completed via observation of the data col- excluded from the study if they had previously received lector at each of the postpartum data collection points. All MIHOW services; had a severe mental or physical disability; participants received a $25 merchandise card at the end of or were under 18 years of age. each interview. Staff entered data into a REDCap database Study participants were 188 pregnant Hispanic women located on a secure password protected server at the associ- living in a large metropolitan area in (blinded) Tennessee. ated university. Monthly data fidelity checks of a random The original sample size for this study was justified by both selection of data forms were conducted by the project coor- the prior experience of the research team and the MIHOW dinator. Data were collected between July 2014 and Sep- program and a conservative estimate of the number of par- tember 2016. ticipants that could complete the protocol within the study period (based on typical MIHOW participation numbers). Comparison Condition That original number was 150 (75 per study group). A sta- tistical powering analysis revealed that groups of that size The comparison condition (minimal education intervention: were sufficient to detect common Cohen’s d effect size of MEI) consisted of distribution of printed educational materi- 0.46 (80% power, 2-sided alpha = 0.05). All effects would als about maternal and infant health and development at the be translated into this common index. A Cohen’s d of that end of each data collection interview to all study partici- magnitude was deemed clinically meaningful and therefore pants (i.e., women assigned to both study groups) in order to further justified our sample size goal. Recruitment and par - maintain the blind status of data collectors. Materials were ticipant interest as the study progressed was such that a deci- available in Spanish or English. sion was made by the research team to continue enrollment as long as the protocol could allow. Ultimately, a total of 188 Intervention Condition women were enrolled. Of those women, 178 completed the study. For study consort flow diagram, see Fig.  1. The intervention condition included the core elements of the MIHOW model. No adaptations were made to MIHOW Procedure content, level of intensity, or home visitor training require- ments. MIHOW interventionists were recruited from the Upon receiving IRB approval for the study protocol, recruit- local Hispanic community who completed 40 h of training ment began with a variety of recruitment strategies includ- to the MIHOW curriculum. The MIHOW model stresses ing distributing flyers at locations with a high volume of recognizing family strengths and utilizing those to address Hispanic customers (e.g., clinics, markets, apartment com- their own family needs (Elkins et al. 2013); relationships plexes, churches) and by word of mouth. All women inter- begin in pregnancy and consist of monthly home visits and ested in participating in the study were screened by trained periodic group gatherings. MIHOW protocols include lis- study staff to determine eligibility. All eligible women tening to maternal concerns, educating about objectives interested in participating in the study completed a written relevant to the woman’s stage of pregnancy or the age of informed consent. Group assignments were generated by the the child, such as healthy eating, developmental milestones, 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S95 Fig. 1 CONSORT flow diagram of the progress through the phases of the randomized trial attachment, and breastfeeding, and helping provide links to Measures needed medical and social services. Home visits typically last approximately 1 h. Due to the study’s limited funding, Primary outcomes were assessed with standardized meas- the duration of MIHOW services was limited to pregnancy ures and established questions from national sources (e.g., through 6 months of age rather than the typical duration to 2011/12 National Survey of Children’s Health (NSCH), the child’s third birthday. http://c hild healt hdat a .or g/lear n /NSCH/t opic s_q ues t ions; 1 3 S96 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Pregnancy Risk Assessment Monitoring System (PRAMS) Statistical Analyses Phase 6, https ://www .cdc.go v/prams /q ues t ionna ir e.htm). See Table 1 for study variables, measures and the time points for SPSS software was used to summarize study data and test data collection. Standardized measures are briefly described hypotheses. All analyses were done using intention-to-treat below. principles. Descriptive statistics and plots were used to sum- marize and initially inspect the distributions of demographic Breastfeeding Self‑Efficacy Scale‑Short Form (BSES‑SF) and study measures at each time of assessment. For those (Dennis 2003) measures assessed more than once over the course of the study, values of change in those measures were also gener- The 14-item measure assesses a mother’s confidence in her ated and summarized. Frequency distributions summarized ability to breastfeed her new infant and has been evaluated nominal and ordinal distributions; means and standard devi- among women from diverse cultures with adequate reliabil- ations summarized normal continuous distributions, median ity and validity (Dennis 2003; Vaughn et al. 2010). Cron- and inter-quartile range (IQR) skewed distributions. Infor- bach’s alpha of the scores in the current study ranged from mation from those descriptive and graphical evaluations was 0.93 to 0.95. used to determine the most appropriate test distribution to specify within the mixed-effects generalized linear models Parenting Stress Index‑Short Form (PSI) (Abidin 2012) used for testing study hypotheses (e.g., normal, log with a Tweedie, etc.). Within these models, the interaction effect of PSI was used to assess the level of stress in the parent–child study group and time of assessment (controlling for baseline system. This 36-item scale includes three domains: paren- values) provided the critical test of differences between the tal distress (PD), parent–child dysfunctional interaction study groups in the amount of change in an outcome meas- (P–CDI), and difficult child (DC) that when combined form ures. Effect sizes were generated for all of the comparisons a total stress scale. Available in Spanish, previous reports using Cohen’s d statistic. Statistical significance tests main- indicate strong reliability coefficients (Barroso et al. 2016). tained maximum type I error rates (alpha values) of < 0.05. Reliability coefficients of the scores in the current study ranged from 0.88 to 0.89. Results Edinburgh Postnatal Depression Scale (EPDS) (Cox et al. 1987) Sample Characteristics This widely used 10-item scale utilizes a 4-point response The average maternal age at enrollment was 29.6  years set to measure level of depressive symptoms with a pos- (SD = 6.5). Most women reported a Mexican heritage sible range of scores from 0 to 30. Most research on the (66.9%), less than a high school education (80.6%), never EPDS indicates a cut-off of 13 to indicate high depressive marrying (56.7%), and annual incomes less than $15,000 symptoms (Cox et al. 1996, 1987). Cronbach’s alphas of the (96.6%). Both study groups had similar sociodemographic scores in the current study ranged from 0.87 to 0.88. and scores for standardized measures used to assess out- comes. See Table 2 for detailed sample characteristics. Home Observation for Measurement of the Environment– Infant‑Toddler (HOME‑IT) (Caldwell, 1984) Primary Outcomes The 45-item observational HOME-IT Inventory assesses the Outcomes in the Child Health Domain quality and quantity of stimulation and support available to a child (birth to age three) in the home environment. The Summaries of infant feeding practices outcomes at 6 measure contains 6 subscales that assess specific parent- months postpartum are shown in Table 3. The strongest ing behaviors that support child learning and development effects of the MIHOW program were observed on the including: responsivity; acceptance; organization; learning BSES-SF scores and on the rates and duration of breast- materials; involvement; and variety in environment (Cald- feeding exclusivity (Cohen’s d = 0.38–0.76). Approxi- well and Bradley 1984). Adaptations to some items on the mately 80% (n = 68 of 86, 79.1%) of the women in the scale were made for infants < 6 months. MEI group reported never breastfeeding exclusively. That respective percentage was considerably lower in the MIHOW group (n = 50 of 90, 55.6%, d = 0.38, p = 0.011). The difference between the groups in duration was a median 1.4 weeks, with 25% of the MIHOW group 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S97 Table 1 Study variables, measures, and data collection time points by domain Variable Measure/question(s) Child health domain  Breastfeeding initiation • Did you ever breastfeed or pump breast milk to feed your new baby after delivery, even for a short period of time? (Source: PRAMS) Time point: ~2 weeks pp  Breastfeeding duration and exclusivity • Are you currently breastfeeding or feeding pumped milk to your new baby? (Source: PRAMS) • How old was your baby when she/he completely stopped breastfeeding or being fed breast milk? (Source: NSCH) • How old was your baby when she/he was first fed formula? • Over the last 24 h, how many times did you breastfeed your baby? • Over the last 24 h, how many times did the baby receive formula? (Source: NSCH) Time point: ~2 months pp, ~ 6 months pp  Breastfeeding self-efficacy • Breastfeeding self-efficacy scale (BSES-SF) (Dennis 2003) Time point: enrollment, ~ 2 weeks pp, ~ 6 months pp  Introduction of solid foods • How old was your baby when she/he was first fed anything other than breast milk of formula? (Source: NSCH) • How old was your new baby the first time he or she ate food, such as baby cereal, baby food, or any other food? (Source: PRAMS) • How often have you added cereal to your baby’s bottle in the past 2 weeks? (Source: IFPS II) Time point: ~2 weeks pp, ~ 2 months pp, ~ 6 months pp  Infant safe sleep • How do you most often lay your baby down to sleep now? • How often does your new baby sleep in the same bed with you or anyone else? (Source: PRAMS) Time point: ~2 weeks pp, ~ 2 months pp, ~ 6 months pp  Prenatal care visits • How many weeks or months pregnant were you when you had your first visit for prenatal care? (Source: PRAMS) • How many prenatal visits did you have during the entire pregnancy? (ask at ~ 2 weeks only) Time point: enrollment, ~ 35 weeks, ~ 2 weeks pp Maternal health domain  Parenting stress and support • Parenting stress index 4—short form (Abidin 2012) How often do you get the social and emotional support you need? Time point: enrollment, ~ 2 weeks pp, ~ 2 months pp, ~ 6 months pp  Maternal depression • Edinburgh postpartum depression scale (Cox et al. 1987) Time point: enrollment, ~ 35 weeks, ~ 2 weeks pp, ~ 2 months pp, ~ 6 months pp Linkages and referrals domain  Follow through with referrals • Since you started the study, has anyone talked to you about services or resources in your commu- nity you may qualify for (such as WIC, a food bank, legal or immigration services, or a children’s group)? • If yes, which services/resources did they talk to you about? • Have you called or visited any of the places they told you about? • Have you received any new services as a result of the referral? • If yes, list the services Time point: ~35 weeks, ~ 6 months pp Positive parenting domain  Parenting practices • HOME inventory (Caldwell and Bradley 1984) Time Point: ~2 weeks pp, ~ 2 months pp, ~ 6 months pp  Reading and singing • During the past week, how many days did you or other family members tell stories or sing songs to your baby? • During the past week, how many days did you or other family members read to your baby? (Source: NSCH) Time point: ~2 months pp, ~ 6 months pp 1 3 S98 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Table 2 Demographic characteristics at enrollment (N = 178) Overall MEI MIHOW p value (N = 178) (N = 87) (N = 91) Mean (SD) Age (years) 29.6 (6.5) 28.7 (6.3) 30.4 (6.6) 0.093 Nation of origin N (%) 0.281  Costa Rica 1 (0.6) 0 (0.0) 1 (1.1)  El Salvador 17 (9.6) 8 (9.2) 9 (9.9)  Guatemala 12 (6.7) 3 (3.4) 9 (9.9)  Honduras 28 (15.7) 17 (19.5) 11 (12.1)  Mexico 119 (66.9) 59 (67.8) 60 (65.9)  Peru 1 (0.6) 0 (0.0) 1 (1.1) Employment status* < 0.001 a b  Full-time 17 (9.6) 16 (18.4) 1 (1.1)  Part-time 28 (15.7) 16 (18.4) 12 (13.2)  Unemployed/looking 2 (1.1) 1 (1.1) 1 (1.1) a b  Unemployed/not looking 131 (73.6) 54 (62.1) 77 (84.6) Marital status 0.066  Married 70 (39.3) 40 (46.0) 30 (33.0)  Separated, divorced, widowed 7 (3.9) 0 (0.0) 7 (5.7)  Never married 101 (56.7) 47 (54.0) 54 (59.3) Highest grade completed N = 176 N = 89 0.334  8th grade or less 71 (40.3) 36 (41.4) 35 (39.3)  9th–12th grade, no diploma 71 (40.3) 31 (35.6) 40 (44.9)  High school diploma/GED 34 (19.3) 20 (23.0) 14 (15.7) Family income 0.599  < $10,000 122 (68.5) 57 (65.5) 65 (71.4)  $10,001–$15,000 50 (28.1) 27 (31.0) 23 (25.3)  $15,001–$40,000 6 (3.4) 3 (3.4) 3 (3.3) Median [IQR] (min, max) Months in U.S. 108.0 [36–156] (1, 408) 108.0 [48–144] (2, 318) 120.0 [36–156] (1, 408) 0.504 Months in (blinded) 89.5 [24–132] (1, 288) 84.0 [24–124] (1, 264) 96.0 [24–132] (1, 288) 0.526 Number of children in home 2.0 [1–3] (0, 5) 2.0 [1–3] (0, 5) 2.0 [1–3] (0, 5) 0.812 Number of adults and children in home 4.0 [3–6] (1, 9) 4.0 [3–6] (2, 9) 4.0 [4–5] (1, 9) 0.751 No respondents received unemployment or worker’s compensation *Superscripts indicate statistically significant post-hoc pairwise comparisons, Bonferroni-corrected, p < 0.05 exclusively breastfeeding for at least 6 weeks (d = 0.42, postpartum; that respective interval was 18–22 weeks for p = 0.005). The effects of the MIHOW program on breast- the women in the MIHOW group (see Table 3). feeding self-efficacy occurred between baseline and the Safe sleeping practices reported by the women in each initial postpartum assessment at 2-weeks with the MIHOW of the study groups are summarized in Table 4. The women group scores being higher at that assessment and main- in the MIHOW group were much more likely to report taining that difference throughout the 6 month postpartum positioning the infant on the back than did the women period (see Table 3). A related but secondary outcome was in the MEI group (~ 98 vs. 66–75%, d = 0.63, p < 0.001). noted to be significant. While the median time to initiation A related but secondary outcome was significant. The of other liquids was the same for both groups (20 weeks), MIHOW group of women reported ‘Never’ practicing a higher proportion of the MIHOW group delayed the ini- co-sleeping with the infant more than did the MEI group tiation of other liquids (d = 0.59, p < 0.001). 50% of the (81–86% vs. 28–33%, d = 1.17, p < 0.001, see Table 4). MEI women initiated other liquids between weeks 16–20 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S99 Table 3 Summaries of infant feeding practices by study group (N = 178) Overall MEI MIHOW p value N = 178 N = 87 N = 91 N (%) N = 177 N = 86 N = 91 Ever breastfed 149 (84.2) 71 (82.6) 78 (85.7) 0.565 (d = 0.13) Breastfeeding status N = 175 N = 85 N = 90 0.762 (d = 0.01)  Breastfeeding—6 months PP 87 (49.7) 42 (49.4) 45 (50.0)  Never breastfed 28 (16.0) 15 (17.6) 13 (14.4)  Stopped by 2 weeks PP 4 (2.3) 2 (2.4) 2 (2.2)  Stopped by 2 months PP 22 (12.6) 8 (9.4) 14 (15.6)  Stopped by 6 months PP 34 (19.4) 18 (21.2) 16 (17.8) Median [IQR] (Min, Max) N = 146 N = 70 N = 76 Breastfeeding duration in weeks 28.0 [12–28] 28.0 [12–28] 28.0 [12–28] 0.754 (1, 28) (1, 28) (2, 28) (d = 0.05) N (%) Breastfeeding exclusivity N = 176 N = 86 N = 90 0.011 a b (d = 0.38)  Never 118 (67.0) 68 (79.1) 50 (55.6) a b  Stopped by 2 weeks PP 31 (17.6) 10 (11.6) 21 (23.3) a b  Stopped by 2 months PP 24 (13.6) 7 (8.1) 17 (18.9)  Still exclusive 6 months PP 3 (1.7) 1 (1.2) 2 (2.2) Exclusive breastfeeding duration in weeks 0.4 [0–4] 0.3 [0–2] 1.4 [0–6] 0.005 (0, 28) 175 (0, 28) 86 (0, 28) 89 (d = 0.42) Time to first other liquid in weeks 20.0 [16–22] 20.0 [16–20] 20.0 [18–22] < 0.001 (2, 24) 161 (2, 24) 79 (4, 24) 82 (d = 0.59) Time to first food in weeks 20.0 [20–22] 20.0 [20–22] 21.0 [20–22] 0.201 (1, 24) 163 (1, 24) 80 (12, 24) 83 (d = 0.19) Median [IQR] N BSES scores < 0.001 (d = 0.76)  Baseline 54.0 [50–60] 127 53.0 [50–60] 66 54.0 [50–61] 61  2 weeks PP 56.0 [51–65] 145 52.0 [48–56] 69 61.0 [56–66] 76  2 months PP 56.0 [51–63] 121 51.0 [46–56] 60 62.0 [56–65] 61  6 months PP 57.5 [52–64] 88 53.0 [48–55] 42 64.0 [58–66] 46 Interaction effect of study group and time of assessment, no statistically significant difference at baseline, MIHOW > MEI 2 weeks, 2 and 6 months throughout the postpartum period (d = 0.57, p < 0.001). Par- Outcomes in the Maternal Health Domain enting stress and support were only assessed postpartum. Those scores are also summarized in Table 5. As shown, rel- Ninety-nine percent received prenatal care beginning at approximately 13 weeks gestation with about a total number ative to the women in the MEI group, women in the MIHOW group reported lower levels of parenting stress and higher of nine prenatal visits. No statistically significant differences between the groups (d = 0.04–0.12) were found. Median levels of available social and emotional help (d = 0.43 and 0.39 respectively, p < 0.001). maternal depressive symptom scores were 7.0 (of possible 30) at baseline with essentially equivalent group variability Outcomes in the Linkages and Referrals Domain in those scores (see Table 4). Compared to the MEI group, women in the MIHOW group demonstrated a statistically As shown in Table  6, there were statistically signifi- significant greater decrease in scores between the baseline and prenatal assessments with the values remaining lower cant differences between the groups in the receipt of and 1 3 S100 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Table 4 Summaries of sleeping practices by study group (N = 178) points, statistically significant differences between groups for the HOME-IT Inventory scores emerged. Compared Overall MEI MIHOW p value to observations of home environments of infants in the N = 178 N = 87 N = 91 MEI group, infants in the MIHOW group experienced a N (%) statistically significantly higher level of quality and quan- Sleep position < 0.001 tity of stimulation and support at each time of assessment (d = 0.63)  2 weeks PP N = 177 N = 86 N = 91 (d = 1.99–2.32, p < 0.001). Furthermore, women in the   On back 147 (83.1) 57 (66.3) 90 (98.9) MIHOW group reported greater frequency of singing songs   On side 30 (16.9) 29 (33.7) 1 (1.1) or telling stories to their child (d = 0.86, p < 0.001). Women   On stomach 0 0 0 in the MIHOW group reported greater frequency of read-  2 months PP N = 176 N = 86 N = 90 ing to their child (d = 1.53, p < 0.001). While more than   On back 145 (82.4) 57 (66.3) 88 (97.8) 90% of the women in the MIHOW group reported read-   On side 30 (17.0) 28 (32.6) 2 (2.2) ing three or more times per week to their child at both 2   On stomach 1 (0.6) 1 (1.2) 0 (0.0) and 6 months postpartum, fewer than 40% of the mothers  6 months PP in the MEI group reported doing so at either time of assess-   On back 154 (86.5) 65 (74.7) 89 (97.8) ment (p < 0.001). Rates decreased from 37 to 26% between   On side 23 (12.9) 21 (24.1) 2 (2.2) 2 and 6 months postpartum in the MEI group while the rate   On stomach 1 (0.6) 1 (1.1) 0 (0.0) increased from 92 to 97% during that period in the MIHOW Co-sleeping < 0.001 group (p < 0.039). (d = 1.17)  2 weeks PP N = 177 N = 86 N = 91   Always 21 (11.9) 21 (24.4) 0 (0.0)   Often 16 (9.0) 13 (15.1) 3 (3.3) Conclusions for Practice   Sometimes 17 (9.6) 13 (15.1) 4 (4.4)   Rarely 21 (11.9) 11 (12.8) 10 (11.0) Using an intent to treat approach, the majority of our hypoth-   Never 102 (57.6) 28 (32.6) 74 (81.3) eses were supported and provide strong evidence of the  2 months PP N = 176 N = 86 N = 90 effectiveness of MIHOW on improving health outcomes in   Always 18 (10.2) 17 (19.8) 1 (1.1) this sample of Hispanic mothers and their infants. Overall,   Often 12 (6.8) 12 (14.0) 0 (0.0) women assigned to the MIHOW group had fewer depres-   Sometimes 20 (11.4) 18 (20.9) 2 (2.2) sive symptoms and less parenting stress and more social   Rarely 24 (13.6) 15 (17.4) 9 (10.0) and emotional help, and better infant feeding and safe sleep   Never 102 (58.0) 24 (27.9) 78 (86.7) practices.  6 Months PP Similar to national statistics (U.S. Department of Health   Always 17 (9.6) 16 (18.4) 1 (1.1) and Human Services 2013), the majority of women in our   Often 15 (8.4) 13 (14.9) 2 (2.2) study entered prenatal care early, underscoring the oppor-   Sometimes 18 (10.1) 15 (17.2) 3 (3.3) tunity to engage with Hispanic families during this time   Rarely 22 (12.4) 15 (17.2) 7 (7.7) period. As noted in our findings, the women assigned to   Never 106 (59.6) 28 (32.2) 78 (85.7) MIHOW had many better outcomes than women who only received the MEI. Main effect of study group (on back: MIHOW > MEI, p < 0.001) Coupling a home visitation program by trained peer Main effect of study group (never: MIHOW > MEI, p < 0.001) mentors such as MIHOW with standard prenatal care has potential to improve maternal and child health outcomes. follow through with referrals. Women in the MIHOW group At all postpartum time points, women in the MIHOW group received more referrals for additional services from com- reported fewer depressive symptoms, less parenting stress munity providers than did those in the MEI group (80–100 and more social and emotional help than women in the MEI vs. 22–28%, d = 1.77, p < 0.001), connected with those group. These findings depict a mother who may be more resources (65–81 vs. 56–54%, d = 0.31, p = 0.028), and able to effectively engage with her infant (Gress-Smith et al. received more new services (64–80 vs. 56–54%, d = 0.30, 2012; Nelson et al. 2016). At all postpartum time points, p = 0.035) than women in the MEI group. data collectors (blinded to group assignment and using a standardized tool) observed a higher level of quality and Outcomes in the Parenting Practices Domain quantity of stimulation and support available to the child in the home environment in mothers assigned to MIHOW Measures and indicators of several types of parenting prac- than in women assigned to the MEI group. Women in the tices are summarized in Table  7. At all postpartum time MIHOW group also reported a greater frequency of singing 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S101 Table 5 Summaries of prenatal care, maternal depressive symptoms, stress, and support by study group (N = 178) Overall MEI MIHOW p value N = 178 N = 87 N = 91 Prenatal care N (%) N = 172 N = 83 N = 89 Receiving prenatal care 0.960 (d = 0.04)  No 2 (1.2) 1 (1.2) 1 (1.1)  Yes 170 (98.8) 82 (98.8) 88 (98.9) Mean (SD) (Min, Max) N Time to first prenatal visit (weeks) 13.0 (5.0) 12.7 (5.0) 13.3 (4.9) 0.397 (3, 26) 170 (3, 26) 82 (3, 25) 88 (d = 0.12) Median [IQR] N EPDS < 0.001 (d = 0.57)  Baseline 7.0 [2–10] 178 7.0 [3–9] 87 7.0 [2–10] 91  Prenatal 1.0 [0–6] 172 4.0 [0–7] 83 0.0 [0–2] 89  2 weeks PP 2.0 [0–5] 177 5.0 [2–8] 86 0.0 [0–1] 91  2 months PP 0.0 [0–3] 176 3.0 [0–6] 86 0.0 [0–0] 90  6 months PP 0.0 [0–1] 178 0.0 [0–4] 87 0.0 [0–0] 91 PSI total stress < 0.001 (d = 0.43)  2 weeks PP 75.0 [72–80] 177 76.0 [74–81] 86 74.0 [66–79] 91  2 months PP 75.0 [72–80] 176 76.0 [74–81] 86 74.0 [68–79] 90  6 months PP 75.0 [73–81] 178 77.0 [75–82] 87 74.0 [70–79] 91 N (%) Social and emotional help < 0.001 (d = 0.39)  2 weeks PP N = 177 N = 86 N = 91   Always 146 (82.5) 58 (67.4) 88 (96.7)   Usually 30 (16.9) 27 (31.4) 3 (3.3)   Sometimes 1 (0.6) 1 (1.2) 0 (0.0)   Never 0 (0.0) 0 (0.0) 0 (0.0)  2 months PP N = 176 N = 86 N = 90   Always 138 (78.4) 58 (67.4) 80 (88.9)   Usually 33 (18.8) 27 (24.9) 9 (10.0)   Sometimes 4 (2.3) 4 (4.7) 0 (0.0)   Never 1 (0.6) 0 (0.0) 1 (1.1)  6 months PP N = 178 N = 87 N = 91   Always 124 (69.7) 37 (42.5) 87 (95.6)   Usually 52 (29.2) 49 (56.3) 3 (3.3)   Sometimes 2 (1.1) 1 (1.1) 1 (1.1)   Never 0 (0.0) 0 (0.0) 0 (0.0) Interaction effect of study group and time of assessment, no statistically significant difference at baseline, MIHOW < MEI 2 weeks, 2- and 6-months: p < 0.001 Main effect of study group (MIHOW < MEI) Main effect of study group (MIHOW > MEI) songs, telling stories and reading books to their child than of Health and Human Services 2013), most women in the mothers in the comparison group. All of these activities study initiated breastfeeding. We did not find any differences are precursors to appropriate child development and school between groups for breastfeeding rates at 6 months postpar- readiness (Nelson et al. 2016). tum or duration of breastfeeding. Our findings of approxi- In regards to breastfeeding practices, as in other samples mately 84% of all study participants providing some breast- of Hispanic mothers (Flores et al. 2016; U.S. Department feeding is consistent with national data (U.S. Department 1 3 S102 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 Table 6 Summaries of linkages and referrals by study group attaining the HP2020 objective related to exclusive breast- (N = 178) feeding at 6 months (i.e., 25.5%), women in the MIHOW group did report more breastfeeding exclusivity at 6 months Overall MEI MIHOW p-value postpartum and longer duration of exclusive breastfeeding. N, n (%) Breastfeeding self-efficacy also was higher in the interven- Referred to resources < 0.001 tion group at all postpartum time points. Longer duration (d = 1.77)  Prenatal 172, 89 83, 18 (21.7) 89, 71 (79.8) of breastfeeding exclusivity has many potential benefits for (51.7) both the mother and the infant (Victora et al. 2016).  6 months 178, 115 87, 24 (27.6) 91, 91 Infant mortality rates are generally low among Hispanic PP (64.6) (100.0) families (U.S. Department of Health and Human Services Of referrals, made appointments/visits 0.028 (d = 0.31) 2013) Safe sleep practices are critical to reducing infant  Prenatal 89, 56 (62.9) 18, 10 (55.6) 70, 46 (64.8) mortality, particularly sudden infant death syndrome (SIDS)  6 months 114, 86 24, 13 (54.2) 90, 73 (81.1) PP (75.4) (American Academy of Pediatrics 2000). While both groups Of referrals, received new services 0.035 of mothers met or exceeded a reported national rate (i.e., (d = 0.30)  Prenatal 88, 55 (62.5) 18, 10 (55.6) 70, 45 (64.3) 65%) for placing their infants on their back to sleep, almost  6 months 115, 86 24, 13 (54.2) 91, 73 (80.2) 100% of MIHOW mothers reported at all time points placing PP (74.8) their infants on their backs. MIHOW families reported much a less co-sleeping than families in the MEI group. None of the reported resources or service types included mental Identifying maternal and family needs and providing health services appropriate referrals are important aspects of maternal child health care. Follow through on referrals is often low in all of Health and Human Services 2013) and exceeds the racial and ethnic groups (Anisfeld et al. 2004). In this study, related Healthy People 2020 (Office of Disease Prevention both receipt of and follow through with referrals was greater and Health Promotion 2017) objective (Office of Disease in the MIHOW group than in the MEI group. Shared lan- Prevention and Health Promotion 2017). Although not yet guage and cultural background between study participants Table 7 Summaries of parenting practices by study group (N = 178) Overall MEI MIHOW p-value N = 178 N = 87 N = 91 Median [IQR] (min, max) N HOME score  2 weeks PP (max = 26) 19.0 [15–22] 15.0 [13–18] 21.0 [19–23] < 0.001 (6,25) 177 (6, 23) 86 (15, 25) 91  2 months PP (max = 32) 24.0 [19–27] 19.0 [17–23] 27.0 [24–28] < 0.001 (7, 30) 175 (7, 27) 85 (15, 30) 90  6 months PP (max = 45) 37.0 [33–40] 33.0 [30–36] 40.0 [38–42] < 0.001 (23, 44) 178 (23, 41) 87 (33, 44) 91 2 months PP assessment  Songs and stories 5.0 [5–6] 5.0 [4–5] 6.0 [5–7] < 0.001 (# Days past week) (2, 7) 176 (2, 7) 86 (3, 7) 90  Read 4.0 [0–5] 0.0 [0–4] 5.0 [4–6] < 0.001 (# Days past week) (0, 7) 176 (0, 6) 86 (0, 7) 90 N (%) Overall MEI only MIHOW N = 176 N = 86 N = 90  Reads stories ≥ 3 times per week 0.039   2 months PP 115 (65.3) 32 (37.2) 83 (92.2)   6 months PP 109 (61.9) 22 (25.6) 87 (96.7) Interaction effect of study group and time of assessment. MEI decreased at 6-months compared to rate at 2-months; MIHOW group remained at similar level; overall main effect of study group: p < 0.001 The Home is essentially a different measure at each time of assessment therefore only group differences at each time of assessment were con- ducted 1 3 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 S103 Acknowledgements Research reported in this publication was sup- and peer mentors may have enhanced participant motivation ported by the Affordable Care Act Maternal, Infant and Early Child- and ability to follow through with referrals to new services hood Home Visiting Program under Award Number D89MC23542 and (Andrews et al. 2004). by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR000445. The content is solely the responsibility of the authors and does not neces- sarily represent the official views of the National Institutes of Health. Study Limitations The authors would like to thank Chrystal Fizer and Jodie Upchurch from Vanderbilt University and Deborah Narrigan for their help in Our study has both strengths and limitations. The use of a preparing this manuscript. randomized controlled design minimized potential for bias. Open Access This article is distributed under the terms of the Crea- This study had a very high retention rate of participants as tive Commons Attribution 4.0 International License (http://creat iveco compared to other home visitation studies with Hispanic mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- participants (Nguyen et  al. 2003). MIHOW extensively tion, and reproduction in any medium, provided you give appropriate trains women from the community being served to conduct credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. the home visits. We also used data collectors who spoke Spanish and were from the same community. We believe the match of peer mentors and data collectors with the study participants added to the successful retention of participants. References One notable limitation of the study design was that because Abidin, R. R. (2012). Parenting stress index. Odessa, FL: Psychological all participants received a standard packet of printed edu- Assessment Resources. cational materials, we did not have a true control group. American Academy of Pediatrics. (2000). Changing concepts of sud- While this may have made finding differences between den infant death syndrome: Implications for infant sleeping envi- groups more difficult, most differences between our groups ronment and sleep position. Pediatrics, 105(3), 650–656. Andrews, J. O., Felton, G., Wewers, M. E., & Heath, J. (2004). Use yielded large effect sizes. The length of the study is another of community health workers in research with ethnic minority limitation. While the MIHOW intervention is designed to women. Journal of Nursing Scholarship, 36(4), 358–365. continue until children reach 3 years of age, the duration of Anisfeld, E., Sandy, J., & Guterman, N. B. (2004). Best beginnings: A the study’s funding only allowed outcomes to be measured randomized controlled trial of a paraprofessional home visiting program. Final Report Columbia. Project Report. until 6 months postpartum. Baker-Ericzen, M. J., Connelly, C. D., Hazen, A. L., Duenas, C., Landsverk, J. A., & Horwitz, S. M. (2012). A collaborative care telemedicine intervention to overcome treatment barriers for Latina women with depression during the perinatal period. Fami- lies, Systems and Health, 30(3), 224–240. Conclusion Barroso, N. E., Hungerford, G. M., Garcia, D., Graziano, P. A., & Bag- ner, D. M. (2016). Psychometric properties of the Parenting Stress Findings from this study expand the limited empiric evi- Index-Short Form (PSI-SF) in a high-risk sample of mothers and dence related to home visitation services. It demonstrates their infants. Psychological Assessment, 28(10), 1331–1335. Caldwell, B., & Bradley, R. (1984). Home observation for measure- beneficial effects of a well-trained and supervised peer-to- ment of the environment (HOME)—Revised Edition. Little Rock: peer model for a sample of Hispanic mothers and infants. University of Arkansas. Understanding the nuances of providing services to both Census Bureau, U. S. & Population Division. (2016). Annual estimates majority and minority families in the context of a chang- of the resident population by sex, age, race, and Hispanic origin for the United States and states: April 1, 2010 to July 1, 2015 from ing medical care landscape is necessary to providing qual- https ://fact finder .censu s.gov/faces /table servi ces/jsf/pages /produ ity care and developing appropriate policy. Results provide ctvie w.xhtml ?src=bkmk. strong support of the efficacy of the MIHOW program and Clinton, B. (1992). The maternal infant health outreach worker project: the potentially high retention rates for participants. This Appalachian communities help their own English Dance and Song (pp. 23–45) M. Larner, R. Halpern, O. Harvaky. program should be considered when planning home visita- Cox, J. L., Chapman, G., Murray, D., & Jones, P. (1996). Validation of tion services for childbearing immigrant and underserved the Edinburgh postnatal depression scale (EPDS) in non-postnatal families. MIHOW has established standards of practice, a women. Journal of Affective Disorders, 39(3), 185–189. research-based curriculum, and an accreditation component, Cox, J. L., Holden, J. M., & Sagovsky, R. (1987). Detection of postna- tal depression. Development of the 10-item Edinburgh postnatal along with decades of experience in several states. It is also depression scale. British Journal of Psychiatry, 150, 782–786. cost effective. Even with the extensive training, intensive Dennis, C. L. (2003). The breastfeeding self-ec ffi acy scale: Psychomet - support, and supervision needed for peer mentors, program ric assessment of the short form. Journal of Obstetric, Gyneco- costs are usually less, overall, than models requiring early logic, and Neonatal Nursing, 32(6), 734–744. Elkins, T., & Clinton, B. (2009). Vanderbilt’s Maternal Infant Health childhood or medical professionals to conduct home visits. Outreach Worker (MIHOW) program: A collaborative program Additional longitudinal studies are needed to further under- stand the sustained impact of MIHOW on health outcomes. 1 3 S104 Maternal and Child Health Journal (2018) 22 (Suppl 1):S92–S104 to improve perinatal health and child development in low-income Nelson, B. B., Dudovitz, R. N., Coker, T. R., Barnert, E. S., Biely, families. Nashville: Vanderbilt University. C., Li, N., et al. (2016). Predictors of poor school readiness in Elkins, T., del Pilar Aguinaga, M., Clinton-Selin, C., Clinton, B., & children without developmental delay at age 2. Pediatrics and Gotterer, G. (2013). The maternal infant health outreach worker Neonatology, e20154477. program in low-income families. Journal of Health Care for the Nguyen, J. D., Carson, M. L., Parris, K. M., & Place, P. (2003). A com- Poor and Underserved, 24(3), 995–1001. parison pilot study of public health field nursing home visitation Federal Interagency Forum on Child and Family Statistics. (2017). program interventions for pregnant Hispanic adolescents. Public America’s children: Key national indicators of well-being. from Health Nursing, 20(5), 412–418. https ://www.child stats .gov/pdf/ac201 7/ac_17.pdf. O’Brien, M. J., Halbert, C. H., Bixby, R., Pimentel, S., & Shea, J. A. Flores, A., Anchondo, I., Huang, C., Villanos, M., & Finch, C. (2016). (2010). Community health worker intervention to decrease cer- “Las Dos Cosas”, or Why Mexican American Mothers Breast- vical cancer disparities in Hispanic women. Journal of General Feed, But Not for Long. Southern Medical Journal, 109(1), Internal Medicine, 25(11), 1186–1192. 42–50. Office of Disease Prevention and Health Promotion. (2017). Maternal Fuentes-Afflick, E., Hessol, N. A., Bauer, T., O’Sullivan, M. J., Gomez- infant and child health. Healthy People 2020. from https ://www. Lobo, V., Holman, S., et al. (2006). Use of prenatal care by His-healt hypeo ple.gov/2020/topic s-objec tives /topic /mater nal-infan panic women after welfare reform. Obstetrics and Gynecology, t-and-child -healt h/objec tives . 107(1), 151–160. Provini, L. E., Corwin, M. J., Geller, N. L., Heeren, T. C., Moon, R. Gill, S. L., Reifsnider, E., & Lucke, J. F. (2007). Effects of support on Y., Rybin, D. V., et al. (2017). Differences in infant care practices the initiation and duration of breastfeeding. Western Journal of and smoking among Hispanic mothers living in the United States. Nursing Research, 29(6), 708–723. Journal of Pediatrics, 182, 321–326. Glassman, M. E., McKearney, K., Saslaw, M., & Sirota, D. R. (2014). Tran, A. N., Ornelas, I. J., Kim, M., Perez, G., Green, M., Lyn, M. J., Impact of breastfeeding self-efficacy and sociocultural factors on et al. (2014). Results from a pilot promotora program to reduce early breastfeeding in an urban, predominantly Dominican com- depression and stress among immigrant Latinas. Health Promo- munity. Breastfeeding Medicine, 9(6), 301–307. tion Practice, 15(3), 365–372. Gomby, D. S. (2005). Invest in kids working paper No. 7 Home visita- U.S. Department of Health and Human Services. (2013). Child Health tion in 2005: Outcomes for children and parents (Vol. 7). USA 2013. Rockville, Maryland. Gress-Smith, J. L., Luecken, L. J., Lemery-Chalfant, K., & Howe, R. Vaughn, L. M., Ireton, C., Geraghty, S. R., Diers, T., Nino, V., Fal- (2012). Postpartum depression prevalence and impact on infant ciglia, G. A., et al. (2010). Sociocultural influences on the deter - health, weight, and sleep in low-income and ethnic minority minants of breast-feeding by Latina mothers in the Cincinnati women and infants. Maternal and Child Health Journal, 16(4), area. Family and Community Health, 33(4), 318–328. 887–893. Velasco-Mondragon, E., Jimenez, A., Palladino-Davis, A. G., Davis, HRSA. (2016). Demonstrating improvement in the maternal, infant, D., & Escamilla-Cejudo, J. A. (2016). Hispanic health in the USA: and early childhood home visiting program: A report to congress. A scoping review of the literature. Public Health Reviews, 37(1), Jones, K. M., Power, M. L., Queenan, J. T., & Schulkin, J. (2015). 31. Racial and ethnic disparities in breastfeeding. Breastfeeding Medi- Victora, C. G., Bahl, R., Barros, A. J., Franca, G. V., Horton, S., Kra- cine, 10(4), 186–196. sevec, J., et al. (2016). Breastfeeding in the 21st century: Epide- Kirby, J. B., & Kaneda, T. (2013). ‘Double jeopardy’ measure suggests miology, mechanisms, and lifelong effect. Lancet, 387(10017), blacks and hispanics face more severe disparities than previously 475–490. indicated. Health Affairs, 32(10), 1766–1772. Wassertheil-Smoller, S., Arredondo, E. M., Cai, J., Castaneda, S. F., Koniak-Griffin, D., Brecht, M. L., Takayanagi, S., Villegas, J., Melen- Choca, J. P., Gallo, L. C., et al. (2014). Depression, anxiety, anti- drez, M., & Balcazar, H. (2015). A community health worker-led depressant use, and cardiovascular disease among Hispanic men lifestyle behavior intervention for Latina (Hispanic) women: Fea- and women of different national backgrounds: Results from the sibility and outcomes of a randomized controlled trial. Interna- Hispanic Community Health Study/Study of Latinos. Annals of tional Journal of Nursing Studies, 52(1), 75–87. Epidemiology, 24(11), 822–830. Murphey, D., Guzman, L., & Torres, A. (2014). America’s Hispanic Children: Gaining ground, looking forward. 1 3

Journal

Maternal and Child Health JournalSpringer Journals

Published: May 31, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off