Parent mentoring and child anticipatory guidance with Latino and African American families.
Abstract: Poor health and developmental outcomes for children are linked to scarcity of economic resources, various barriers in the delivery of health services, and inadequate parenting. To mitigate such adverse effects and address the needs of 50 high-risk, low-income Latino and African American families receiving well-baby care at an urban primary care health center, a collaborative team from the social work, nursing, and education fields piloted a preventive two-year parent mentoring project. The intervention was theoretically anchored in the transactional model of child development. The mentoring practices used an activity-based approach for strengthening child anticipatory guidance and meeting family needs. Thirty-five intervention families completed the project. Compared with a matched community sample, intervention families showed positive statistically significant changes in parent and child outcomes. The discussion addresses the practical benefits of the intervention, limitations of the evaluation design, and implications for collaborative multidisciplinary practice.

KEY WORDS: clinical social work intervention; early childhood guidance; high-risk families; infant development; parent mentoring
Article Type: Report
Subject: Child development (Social aspects)
Child development (Economic aspects)
Parenting (Social aspects)
Parenting (Health aspects)
Child health services (Management)
Hispanic Americans (Social aspects)
Hispanic Americans (Economic aspects)
African Americans (Economic aspects)
African Americans (Social aspects)
Children (Health aspects)
Children (Economic aspects)
Children (Social aspects)
Author: Farber, Michaela L.Z.
Pub Date: 08/01/2009
Publication: Name: Health and Social Work Publisher: National Association of Social Workers Audience: Academic; Professional Format: Magazine/Journal Subject: Health; Sociology and social work Copyright: COPYRIGHT 2009 National Association of Social Workers ISSN: 0360-7283
Issue: Date: August, 2009 Source Volume: 34 Source Issue: 3
Topic: Event Code: 290 Public affairs; 200 Management dynamics Computer Subject: Company business management
Product: Product Code: 8000187 Maternal & Child Health Care; 9105264 Maternal & Child Health Programs; E123400 Hispanic Americans NAICS Code: 621999 All Other Miscellaneous Ambulatory Health Care Services; 92312 Administration of Public Health Programs SIC Code: 8099 Health and allied services, not elsewhere classified
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 204610806
Full Text: Early childhood constitutes a relatively short period of human development, yet it is disproportionately important in setting the stage for optimal health and development across the life span (Starfield, 2004). Specifically, a child's health status, parental child-rearing choices around early developmental milestones, and the family environment set the stage for child well-being, create long-term developmental consequences for school-based learning, and affect risks for developing chronic diseases (Borkowski & Weaver, 2006). The centrality of the family environment in early child development forms the conceptual base for strengthening young children by providing their families with anticipatory child guidance and promoting family-centered care in the delivery of well-baby health services (American Academy of Pediatrics, 2003). Recently, however, the U.S. Preventive Service Task Force found that the amount of time required for delivering all of the needed child well-health preventive services, including necessary support and guidance for families, is not feasible during routine well-baby visits (Yarnall, Pollack, Ostbye, Krause, & Michener, 2003). Research has also revealed that a large proportion of young children living in poverty receive inadequate preventive health care, with the most frequent barriers being providers' lack of comprehension of family needs and lack of sufficient time for service (Zuckerman, Stevens, Inkelas, & Halfon, 2004).

Although family-centered interventions have long been central in social services provision (Kilpatrick & Holland, 2006), developmentally sensitive interventions for families of very young children are relatively new and require extensive multidisciplinary collaboration (Applegate & Shapiro, 2005).Thus, a team of professionals from the social work, nursing, and education fields created and implemented a two-year preventive intervention project of parent mentoring designed to strengthen the anticipatory child guidance of 50 Latino and African American families at an urban primary health care center in Washington, DC. This article presents the pilot intervention, the evaluation process, findings for parents' and children's outcomes, and lessons learned.

CONCEPTUAL FRAMEWORK

Four conceptual domains affect infant development (Sameroff, McDonough, & Rosenblum, 2004) and frame parent mentoring during early childhood.

Caregiving

Infant development is a product of continuous dynamic transactions between developmental processes and caregiving experiences. Indeed, recent studies on brain development confirm that infants' dependence is so pronounced that early inadequate or harmful parenting practices yield persistent negative developmental effects (DeBellis, 2005; Osofsky, 2004). To mitigate these risks, current research suggests that preventive interventions need to be directed at infants, parents, and family home environments (Borkowski & Weaver, 2006).

Social Environmental Resources

Within this transactional model of child development (Sameroff & Chandler, 1975), poor health outcomes are linked to a paucity of family economic resources, gaps in parental education and child development knowledge, and barriers in access to health care (Borkowski &Weaver, 2006). Research has also found that children's well-being and early school readiness are compromised by parents' depression, substance abuse, domestic discord, harsh child-rearing practices, and child neglect (Bradley & Corwyn, 2002). Such risks are particularly prevalent in poor families, and they pose even greater risks in poor immigrant families (McLanahan, 2005). Poverty is also the most significant factor associated with low rates of vaccination and receipt of basic medical care (Rodewalt & Santoli, 2001).

Well-Baby Care

Routine well-baby health care represents a universal opportunity for early engagement of parents in monitoring infant health and growth and learning through anticipatory child guidance. The well-baby care contacts between health care providers and parents serve as focal contact points during which providers and parents share a focus on the well-being of the infant. In turn, this mutual focus creates a therapeutic leverage for developing a working alliance between parents and providers (Brazelton & Greenspan, 2000).

Parent Guidance

As a major life change for adults, the transition to parenthood requires adequate life skills and resilience (Bornstein, 2002). Review of the research reveals a striking consensus on major parental prebirth factors that exert a negative influence on postnatal parenting development and create risks for child development (Borkowski &Weaver, 2006). Parents whose mental health is compromised or who lack persistence or sensitivity in their approach to general problem solving tend to have incompetent self-regulation and insufficient empathy to sustain relationships. Others have inadequate basic education for meeting their own or their family's needs. Such parents tend to show a lack of resilience in meeting life's demands, have unrealistic expectations and rigidity toward children's behaviors, manage children's developmental needs poorly, and often maltreat their children (Landy & Menna, 2006).

PARENT MENTORING INTERVENTION

In this project, Brazelton's (1992) Touchpoints approach served as the educational practice model for mentoring and strengthening parents' anticipatory guidance during routine well-baby health care. The parent mentoring was carried out by two bilingual (in Spanish) college-educated (in early childhood development) parent coaches. To ensure delivery of a quality intervention, the entire project team (social work clinical research director, nursing health center director, administrative education director, parent coaches, coaches' supervisor, project manager, and data manager) was trained in the promotion of the transactional approach to parent-child interaction and guidance through video-based training delivered by an on-site Keys to Caregiving consultant (Barnard, 1994). This week-long training included guided viewing of six instructional videos (addressing infant state, behavior, cues, temperament, feeding interaction, and provider-parent communication), completion of study guides, and engagement in role play using five parent handouts. The videos demonstrated various preferred parent provider and infant transactional behaviors. The study guides provided scaffolded information about parent-child interaction, guidance for facilitating self-reflective practice in mentoring parents, and questions for assessing comprehension of the video vignettes. The parent handouts were used to demonstrate preferred parent-child interactions at different stages of infant development. In addition, to ensure fidelity to Brazelton's developmental method of teaching parents about infant development and ensure reliability of infant observations during the delivery of well-baby health care and parent mentoring, all project team members participated in an intensive three-day training at the Brazelton Institute in Boston or locally. These training experiences ensured the buy-in of all project personnel and the relevant educational knowledge and skill sets needed across disciplines.

The parent mentoring intervention was implemented with individual families through a systematized protocol. In implementing this protocol, each parent coach first met with a family during a planned health center visit. A two-hour home visit followed two weeks later. There were multiple phone contacts throughout. Visits started at birth and continued until the child reached 18 months of age. The well-baby visits at the health center followed the requirements set by the Medicaid Early and Periodic Screening, Diagnosis, and Treatment (EPSDT) protocol (Centers for Medicare and Medicaid Services, 2003).This protocol is based on infant developmental progression, with designated time periods serving as critical anticipatory guidance and assessment points at which infant health, development, and caregiving are reviewed for progress, needs, and treatment planning.

First, during the EPSDT health center visit, parent coaches participated in the child health and developmental assessment and observed parent-health provider communication and interaction. Parent coaches used the knowledge gained in their subsequent mentoring during home visits. Overall, coaches structured their mentoring contacts to meet three goals: (1) to strengthen the parent-health provider relationship and communication about the focus child's development and health, (2) to educate parents about the importance of nonharsh parent-child interactions and age-appropriate child behaviors, and (3) to increase families' knowledge and use of community resources to meet their needs. To accomplish these goals, the project first manualized all EPSDT planned contacts between coaches and families on the basis of normative developmental expectations for infant physical and socioemotional functioning. The planned contacts had separate activity components for well-health visits at the health center and follow-up visits at home. Typically, the planned contacts had an associated educationally planned activity to enhance parent skills in infant stimulation or behavior management, promote positive parent-child interaction and attachment, give parents sufficient time to ask questions about their child or related family needs or issues, and allow parents to explore feelings about recommendations made for their child's health or development. The components of these contacts were often sequenced into child health or development training, home safety, and parenting practices.

Second, in all planned contacts from four to 18 months, coaches used the Ages and Stages Questionnaire (ASQ) (Bricker & Squires, 2002), an interactive activity-based system to help parents evaluate, understand, and anticipate their child's unfolding growth and socioemotional competencies as well as explore options for intervention. Within this ASQ activity-based system, coaches used pencil-and-paper questionnaires to help parents assess their infant's competencies. They also used videotaped vignettes to guide parents in learning about developmental benchmarks and the meaning of infant normative behaviors, explore various beliefs and common myths about child-rearing practices, teach parents how to anticipate the progression in their infant's physical and socioemotional competencies, and model for and practice with parents sensitive interaction with their child (for examples, see Squires & Bricker, 2007).

Third, to increase parental sensitivity to infants' emotional regulatory needs, coaches used culturally skilled dialogue in providing developmental guidance and information in a respectful, sensitive, collaborative, and nonjudgmental manner. Starting at enrollment, coaches asked about parents' comfort level in speaking, reading, and writing English. Coaches followed the parents' lead in using English, Spanish, or both during visits and with informational materials. Coaches initiated parent dialogue in a way that encouraged parents to identify what they considered important for their child's care. Before addressing any other concerns, coaches asked parents what questions or concerns they had about their child's health, development, or needs, thus allowing parents to lead the conversation in their preferred direction. Coaches used natural opportunities that presented themselves during contacts to reflect and comment on parents' competencies and strengths. Coaches sought to validate parents' feelings by using reflective language, offering compliments on care or attunement, and helping parents interpret the meaning of infant behavior and utterances (for dialoguing, see Barrera & Corso, 2003).

To facilitate use of community resources, coaches helped parents to secure instrumental resources (toys; basic equipment such as cribs, high chairs, and car seats) by identifying low-cost stores, securing donations through local charities and foundations, and directly showing parents how to find these resources. Coaches also provided information on accessing legal and immigration services and using available resources for employment, education, health insurance, housing, and public transportation. When necessary, coaches provided referrals to appropriate support services for parental coping with substance abuse, mental health issues, and domestic violence. Although coaches followed the EPSDT schedule for regularly planned mentoring, they also provided parents with emotional support through frequent follow-up phone calls. To support family diversity of lifestyle and choices, coaches were expected to be flexible in finding times convenient to parents for visiting, to engage in conversations relative to parents' preferred issues, and to link these conversations to the jointly established goals for promoting adequate care and healthy development.

To ensure fidelity in providing a quality intervention, monitoring the developmental needs of focus children, providing support to coaches, and coordinating information, the project included weekly individual supervision of coaches, meetings with the entire project staff, and meetings of the directors from the three disciplines to review developmental progress. All parent contacts were documented in a confidential case record accessible only to selected staff. For all contacts, coaches completed a structured form that tracked time, location, purpose, objectives, people involved, activity or accomplishments, and coaches' perceptions of strengths and barriers present. This information was entered into a password-protected database and regularly reviewed by directors. For health care quality assurance, the nursing director also reviewed medical records and coaches' reports monthly to monitor children's progress. Each quarter, selected staff made random calls to spot-check families' satisfaction with services. Every six months, families completed an anonymous satisfaction questionnaire. These internal reviews served to ensure consistency, maintain quality of service delivery, and triangulate coaches' and parents' reports with health center staff observations in the child's medical record.

In addition, different team members coordinated various project-related tasks. For example, the social work director grounded the intervention's clinical practices in theory and research, designed and implemented the evaluation, and guided coaches' family-centered mentoring practices in home visits and use of community resources. The nursing director supervised coaches' well-child health center visits and provided education about child health and illness. The education director worked with an advisory body, monitored administrative reporting, and facilitated coaches' practices in mentoring parents who had cognitive impairments.

To ensure the project's saliency to local community needs, quarterly meetings with a community advisory group provided further input and periodic oversight. This advisory group was composed of health, education, and social services professionals and parents of young children. Overall, the project used parent mentoring to provide parents with anticipatory guidance in child health and development. For parents, the project outcomes focused on strengthening adequacy of family needs and resources, increasing knowledge of nurturing practices, and promoting personal resilience. For children, outcomes focused on ensuring timely child immunization and age-appropriate developmental and language progress.

PROJECT EVALUATION METHOD

To assess intervention benefits, outcome data were collected for two research questions: (1) Do focus children evidence positive changes in child development-related outcomes? (2) Do intervention families evidence positive changes in parent-related outcomes?

Participants

The project enrolled 50 families, with 30 of the focus children (60 percent) enrolled during their mothers' last month of pregnancy and the remainder enrolled during their first month of age. Thirty-five families (70 percent) were fully served over the designated evaluation period from enrollment to when the focus child turned 16 to 18 months of age. Eleven dropped out by the end of the first year because their home was too far from the primary health center. Four moved out of state. There were no statistically significant differences in baseline characteristics between fully served families and those who left.

The project enrolled an equal number of boys and girls with a medically normal birth (that is, there was no special medical intervention following birth, and both mother and baby were discharged within three days). Seventeen focus infants (34 percent) were only children; 18 (36 percent) had one older sibling under 18 years of age, and 15 (30 percent) had two or more siblings. Of the 33 infants with siblings (66 percent), 17 (52 percent) also had half siblings residing elsewhere.

At enrollment, the mother's average age was 23 years (SD = 5.6), and the partner or spouse's average age was 25 years (SD = 6.9). Nine mothers (18 percent) were legally married with a spouse living at home; 20 (40 percent) occasionally lived with a spouse or partner after having legally separated or divorced, and 21 (42 percent) had never been legally married but occasionally lived with a partner. Twenty-eight infants (56 percent) had regular daily or weekly contact with their biological father.

Nineteen families (38 percent) had close relatives residing nearby, 10 (20 percent) had family relatives living elsewhere in the United States, and 21 (42 percent) had relatives living abroad. Thirty-five families (70 percent) had immigrated from South or Central America or Puerto Rico. In 25 of these Latino families (71 percent), both parents were immigrants. Ten mothers (29 percent) had been born in the United States of immigrant parents. Although most of the Latino immigrant families preferred to speak their native Spanish, about half were able to converse and write in basic English. The remaining 15 families (30 percent) in the project were U.S.-born African Americans.

Although family income level made the focus children eligible for Medicaid, half did not have health insurance at enrollment. In most families, parent education was minimal. Thirty mothers (60 percent) had not completed high school; the rest were in the process of obtaining a GED. Twenty-four biological fathers (86 percent) did not have a high school diploma.

Design and Sampling

The project evaluation used a quasi-experimental nonequivalent group design with a static comparison group. The one-group pre-post evaluation compared targeted parenting and child outcomes of intervention families with a comparison group of 30 matched families. Self-selected intervention families were enrolled on a first-come first-served basis by referrals from hospital clinics. The comparison families were recruited from the health center catchment area through invitational fliers and provided with information about the evaluative survey and project enrollment availability. To create a comparative sample, matching criteria for comparison and intervention families included project recruitment criteria: focus child's medically normal birth status and age and mother's age (17 years or older at the time of child's birth), education (less than high school), and race or ethnicity.

The comparison family could not be receiving any parent mentoring. These criteria stemmed from early prevention research (Borkowski & Weaver, 2006) .The project was approved by the participating university and hospital Internal Review Boards. All intervention and comparison families completed informed consent forms prior to participation. As a reward for completing evaluation assessments, all families received toys and books for their focus child, calendars, and grocery gift certificates.

Data Collection

All family data were collected through structured interviews by parent coaches at enrollment baseline and at exit. Some data were also collected when the child turned 12 months old. Mothers were the primary informants; however, both parents sometimes negotiated the one answer they wished to provide during evaluation. All reasonable efforts were made to interview the same individuals at each assessment point and in their preferred language.

Data Analysis

The Statistical Package for the Social Sciences (SPSS 13.0) was used for all analyses. Measures of central tendency and percentages described participants. Paired t tests were used to evaluate the mean outcome change in the intervention group from pre- to posttest. Independent t tests were used to compare the mean change between the intervention and comparison groups at posttest.

Outcome Variables

Parent and child outcomes were based on previous research noting the negative impact poverty and poor parenting practices exert on child development and on scale appropriateness, brevity, and ease of completion (DelCarmen-Wiggins & Carter, 2004). All instruments had appropriately developed Spanish versions (Chavez & Canino, 2005).

The three parent variables were as follows: (1) Adequacy of family needs and resources was measured by the Family Resource Scale (Dunst & Leet, 1987), which has research-adequate criterion validity and consistency. The five-point scale responses were aggregated into usually inadequate (rated as never, rarely, or sometimes) and adequate (rated as usually or most of the time). (2) Parenting knowledge of nurturing practices and childrearing beliefs was measured by the Adult-Adolescent Parenting Inventory (AAPI-2) (Bavolek & Keene, 1999), which has research-adequate construct validity and consistency. (3) Personal resilience was measured by an adapted brief version of the Resiliency Attitude Scale (Briscoe & Harris, 1994). This five-point, 30-item adapted scale has basic content validity achieved through expert review and pretesting. All measures had good research reliability (Cronbach's [alpha] = .75 or better) during administrations.

The three child variables were as follows: (1) Completion of infant immunizations was measured through the health center's standard medical records documenting compliance with the EPSDT immunization schedule. (2) Achievement of age-appropriate developmental milestones was measured with the ASQ (Bricker & Squires, 2002). (3) Emerging language competency was measured by parent reports using the MacArthur Communicative Development Inventories (CDI) (Level I, Short Form); the brief-screening CDI is used widely to assess children's normative vocabulary accumulation in understanding and producing words (Fenson et al., 2000). The screening derives vocabulary percentile achievement scores that are normed for age and gender. The ASQ system and CDI short form have established adequate research validity and reliability and current alphas of .70 or above.

FINDINGS

Parent Outcomes

Baseline reports on adequacy of basic instrumental, parenting, and interpersonal needs and resources reflected family struggles with poverty. At baseline, the average total scale score for the majority of intervention families (87 percent) reflected having usually inadequate resources. By project exit, as Table 1 shows, the positive mean changes and robust effects that occurred strengthened the overall resources of these families.

Baseline findings about nurturing and childrearing beliefs revealed that some parents evidenced abusive beliefs and practices on the AAPI-2 scale (see Table 2).At exit, five of the completing families did not complete these data due to family illness or scheduling conflicts. Thirty completing parents provided information about their nurturing and evidenced a statistically significant positive mean improvement in parenting practices, showing a modest positive effect (see Table 2). Because the project actually collected AAPI-2 data at three time points (baseline, 12 months, and exit), a repeated-measures analysis of variance general linear model was used to compare the mean total scores across these time points. The findings revealed that the three mean AAPI-2 scores were statistically significantly different from each other [baseline [M.sub.1] = 1.57, SD = 1.40; 12-month [M.sub.2] = 0.97, SD = 1.00; exit [M.sub.3] = 0.63, SD = 0.76; between-subjects effect: F(2, 29) = 42.31, p = .000, partial [[eta].sup.2] = .59; multivariate tests: Pilai's trace = .39, p = .001, partial [[eta].sup.2] = .39]. Pairwise comparisons specifically revealed that the [M.sub.1] AAPI-2 failure score was statistically significantly different from the [M.sub.2] (p = .001) and [M.sub.3] (p = .003) failure scores. The [M.sub.2], however, did not significantly differ from the [M.sub.3] (p = .230).These findings suggest that parents' positive and powerful impact on their child's life was most evidenced by the first year, a finding that is not surprising given the short time (four to six months) between the second assessment point and exit.

Of the 35 mothers who completed the project, 28 (80 percent) provided information on their personal resilience. At enrollment, these mothers did not significantly differ in their demographics or mean outcomes from other completing mothers in the project. Twenty-four had a total baseline score that was less than 60 percent of the total possible (highest) score. On the basis of previous studies reported by Biscoe and Harris (1996), such results suggest that these mothers were having many life-skills-related difficulties in their personal resilience. At exit, these mothers evidenced a moderate to strong positive effect on their resilience [M1 = 102.2, SD = 13.3;[M.sub.2] = 108.5, SD = 11.0; AM = 5.0, paired t(27) = -11.1,p = .000, [[eta].sup.2] = .85]. Seven mothers did not complete an exit resiliency measure due to scheduling conflicts; however, they did not significantly differ in their baseline measures from the remainder.

As shown in Table 3, independent t-test analyses were used to further compare all mean parent outcomes of intervention and matched comparison families at exit. These findings suggest beneficial significant changes for intervention families. Notably, the mean levels of parent outcomes for comparison families did not differ significantly from baseline levels of intervention families.

Child Outcomes

At exit from the project, infants were 16 months old on average (SD = 1.9). All 35 intervention children who completed the project and 11 who did not were appropriately immunized on the basis of the EPSDT schedule by project exit. No information was available for the four children whose families moved away. To ensure child immunizations, coaches averaged four to five additional phone calls (SD = 5) beyond the planned EPSDT contacts. They also encountered various barriers in assisting parents with children's immunizations. For example, to address parents' fears or myths about immunizations, coaches used several strategies to provide accurate information and guidance. First, they explored and listened to parents' stories of their own upbringing and tales about immunizations. Second, they provided emotional support when parents expressed their concerns. Third, they reflected on positive and challenging aspects of the current state of knowledge on child immunizations, helped parents synthesize and comprehend the published health information (for example, information available through the Centers for Disease Control and Prevention or the American Academy of Pediatrics), and explored consequences of different decisions. To overcome transportation barriers during health center visits, coaches taught parents how to navigate the local metro train and bus connections and provided tokens on the basis of need.

In addition, at enrollment all families received a bilingual monthly calendar (developed by the project) that described parent-guiding activities for promoting infant development during the first year and also included childhood immunization requirements through adolescence. The calendar offered note space and phone numbers for the parent coaches, the health center, and other community resources. Coaches used these calendars for setting appointments with families and in commenting on children's overall progress. At exit, all parents reported positively on the calendars' usefulness in scheduling health visits, keeping notes, and sharing specific information about their child with other family members. Comparing immunization between intervention and comparison children revealed that 25 percent of the parents in the sample of 30 matched children could not provide evidence of their child's immunizations. These comparison families were also subsequently assisted in having their children appropriately immunized.

Almost three-quarters of the intervention children (74 percent) completed all six ASQ developmental assessments by project exit, and no child received fewer than four assessments. The coaches, families, and health care staff viewed the ASQ assessments as beneficial in fostering parents' understanding of their child's developmental and socioemotional needs. The coaches particularly liked the way the ASQ system helped parents concretely demonstrate their child's developmental progress. In addition, three children were identified early by the ASQ system as being at risk for delayed development. This identification fostered a prompt pediatric assessment and referral for early intervention. Of the 11 intervention children whose families dropped out, six completed at least the first two ASQ assessments at four and six months.

To examine children's language development on the CDI, the evaluation took into account whether the child was in home- or center-based child care at project exit (Borkowski &Weaver, 2006).Twenty Latino intervention children (57 percent) were in home-based child care with their own mother; the remaining 15 (seven Latino, eight African-American) children (43 percent) were in center-based care. Of the intervention children in home-based child care, 12 scored within the normative range for age and gender in vocabulary understanding and production. Three children scored below the normative range and were thus identified as at risk for delayed language development. These children were also identified as being at risk by the ASQ developmental assessments and referred for early intervention. With parent permission, CDI vocabulary assessments were also obtained for intervention children in center-based child care. The results for these children revealed that their normative vocabulary understanding and production did not significantly differ from those of intervention children in home-based care.

The matched comparison sample children were all in home-based child care. Therefore, only 20 Latino children were matched with and compared with 20 intervention children in home-based child care. Five comparison children scored below the normative range for age and gender in understanding words, and 12 scored below the normative range in producing words. Five comparison children also scored below the normative range in both. On average, intervention children evidenced a statistically significant mean difference in knowing nine more words and producing 10 more words than the matched comparison children (see Table 4).

DISCUSSION

The nature of infant development, the findings from prevention research about the pervasive effects of inadequate parenting in the context of poverty, the lack of available time during well-health care for engaging in anticipatory child guidance, and a local need for service constituted the basis for developing and implementing parent mentoring with the high-risk Latino immigrant and African American families in this study. Two bilingual parent coaches individually mentored parents during their well-baby health care and home visits. The intervention followed children's developmental progression during their first 18 months. The mentoring child guidance focused on helping parents to recognize and comprehend their children's needs and competencies, manage distress or unwanted behaviors, and anticipate children's future growth. Coaches also provided concrete assistance in accessing community resources and guided parents in effective communication of their perceptions and wishes to health providers. All mentoring activities promoted the personal resilience needed for effective and sensitive parenting. Coaching practices included educating, counseling, obtaining resources, and emotionally supporting parents in child guidance and meeting family needs. Coaches relied on the ASQ activity-based developmental assessment system in fostering and facilitating parents' knowledge and child-rearing skills. They also used developmentally based toys, books, and calendars to enrich families' resources and promote sensitive, nurturing parent-child interaction. The ASQ system became instrumental in facilitating early identification of children's suspected developmental delays. The overall project was supported by input from a community advisory body. In evaluation of outcomes, the families who completed the project evidenced increased family resources, stronger nurturing and sensitivity in meeting their child's developmental needs, and better personal resilience. Their focus children were appropriately immunized and demonstrated age-appropriate development and language vocabulary. For children who were identified as being at risk for developmental delay, the intervention process ensured early recognition and timely service delivery. In contrast, the matched comparison sample of families evidenced parent outcomes that were statistically similar to the baseline levels of intervention families, suggesting no gains. The matched comparison Latino children evidenced lower levels of vocabulary than intervention children, and a substantial proportion of these children were not immunized.

The project results are in line with findings in other preventive health-related home visiting trials (Olds et al., 2004). That is, the project showed that to be successful in mentoring, parent coaches need to have strong knowledge of and appreciation for parent-child interaction and to use a mentoring approach that is collaborative, family centered, and culturally sensitive. Their approach also should incorporate reflective communication along with hands-on activities to impart child guidance knowledge and assist in obtaining resources. Only then is mentoring likely to make a positive impact (Weatherston, 2000).

This project yielded several important lessons for practice. First, the multidisciplinary collaborative approach to project execution heralds the necessary knowledge and oversight for prevention practices in early childhood development. Second, the low-cost use of the ASQ system provides assessments that easily convey information to parents and social services, health, and education providers; systematizes parent-mentor contacts; concretely illustrates child development through anticipatory guidance; and easily links assessment to intervention options in a meaningful way. Third, the timing of the project during universally present well-baby health care services leverages therapeutic goals and benefits. These lessons further imply that work with families of very young children requires that social workers develop the capacity to integrate specialized knowledge of child development, healthy parenting practices, and prevention research in a context of delivering services through multidisciplinary teamwork.

The results of the project are, however, subject to methodological limitations. The quasi-experimental nonequivalent group design precludes a determination that the intervention can be solely credited with the observed beneficial impacts (Rubin, 2008).The use of a static comparison group was a response to funding limitations and is one of the internal validity biases that compromise the causal inference of intervention benefits. The results of outcomes derived from matched comparisons and the triangulation of coaches' reports, ASQ assessments, and parental satisfaction with the documentation of child development and health in medical records, however, support the promising nature of the project's positive short-term findings.

Recommendations for future research include using a randomized experimental design for replicating the effectiveness of parent mentoring and exploring intervention effects with different family variables (culture, caregiver's gender, other family demographics), different caregivers (grandparents, foster parents), and different levels of parent coaches' training and experience. Research also needs to examine the long-term impact of this and other such short-term interventions with young children. Indeed, engaging in more multidisciplinary practice research is a necessity in this era of increasing complexity of knowledge and practice issues.

Original manuscript received June 19, 2006

Final revision received April 7, 2008

Accepted July 17, 2008

REFERENCES

American Academy of Pediatrics. (2003). Policy statement: Family-centered care and the pediatrician's role. Pediatrics, 112, 691-696.

Applegate, J. S., & Shapiro, J. (2005). Neurobiology for clinical social work. New York: W.W. Norton.

Barnard, K. (1994). Keys to care giving--NCAST AVENUW program. Retrieved April 28, 2009, from http://www.ncast.org/index.cfin?category=18

Barrera, I., & Corso, R. M. (2003). Skilled dialogue: Strategies for responding to cultural diversity in early childhood. Baltimore: Patti H. Brookes.

Bavolek, S.J., & Keene, R. (1999). Adult-Adolescent Parenting Inventory. Eau Claire, WI: Family Development Resources.

Biscoe, B., & Harris, B. (1994). Resiliency Attitude Scale. Oklahoma City, OK: Eagle Ridge Institute.

Borkowski, J. G., & Weaver, C. M. (2006). Prevention. Baltimore: Paul H. Brookes.

Bornstein, M. H. (Ed.). (2002). Handbook of parenting, Vol. 4: Social condition and applied parenting. Mahwah, NJ: Lawrence Erlbaum.

Bradley, R. H., & Corwyn, R. E (2002). Socioeconomic status and child development. Annual Review of Psychology, 53, 371-399.

Brazelton, T. B. (1992). Touchpoints : The essential reference--Your child's emotional and behavioral development. Cambridge, MA: Perseus Press.

Brazelton, T. B., & Greenspan, S. I. (2000). The irreducible needs of children: What every child must have to grow, learn, and flourish. Perseus Press.

Bricker, D., & Squires, J. (2002). Ages & Stages Questionnaires (ASQ). Baltimore: Paul H. Brookes.

Briscoe, B., & Harris, B. (1994). Resiliency Attitude Scale (RAS) manual. Oklahoma City, OK: Eagle Ridge Institute.

Centers for Medicare and Medicaid Services. (2003). Early and periodic screening, diagnosis, and treatment (EPSD T). Retrieved April 28, 2009, from http://www.cms.hhs.gov/MedicaidEarlyPeriodicScrn/

Chavez, L. M., & Canino, G. (2005). PN-54--Toolkit on translating and adapting instruments. Retrieved April 28, 2009, from http://www.tecathsri.org/products_fist.asp

DeBellis, M. D. (2005). The psychobiology of neglect. Child Maltreatment, 10, 150-172.

DelCarmen-Wiggins, R., &Carter, A. (Eds.). (2004). Handbook of infant, toddler, and preschool mental health assessment. Oxford, England: Oxford University Press.

Dunst, C., & Leet, H. (1987). Measuring the adequacy of resources in households with young children. Child: Care, Health, and Development, 13, 111-125.

Fenson, L., Pethick, S., Renda, C., Cox, J. L., Dale, P. S., & Reznick, J. S. (2000). Short-form versions of the MacArthur Communicative Development Inventories. Applied Psycholinguistics, 21, 95-115.

Green, S. B., & Salkind, N.J. (2003). Using SPSS for Windows and Macintosh. Englewood Cliffs, NJ: Prentice Hall.

Kilpatrick, A., & Holland, T. P. (2006). Working with families: An integrative model by level of need (4th ed.). Boston: Allyn & Bacon.

Landy, S., & Menna, R. (2006). Early intervention with multi-risk families. Baltimore: Paul H. Brookes.

McLanahan, S. (Ed.). (2005). School readiness: Closing racial and ethnic gaps [Electronic version]. Future of Children, 15(1). Retrieved April 28, 2009, from http://www.futureofchildren.org/pubs-info2825/ pubs-info_show.htm?doc_id=255946

Olds, D., Robinson, J, Pettit, L., Luckey, D., Holmberg, J., Ng, R., et al. (2004). Effects of home visits by paraprofessionals and by nurses: Age 4 follow-up results of a randomized trial. Pediatrics, 114, 1560-1568.

Osofsky, J. D. (2004). Young children and trauma. New York: Guilford Press.

Rodewalt, L., & Santoli, J. (2001).The challenge of vaccinating vulnerable children. Journal of Pediatrics, 139, 613-615.

Rubin, A. (2008). Practitioner's guide to using research for evidence-based practice. New York: John Wiley & Sons.

Sameroff, A.J., & Chandler, M.J. (1975). Reproductive risk and the continuum of caretaking causality. In F. Horowitz, M. Hetherington, S. Scarr-Salapatek, & G.

Siegel (Eds.), Review of child development research (Vol. 4, pp. 187-244). Chicago: University of Chicago Press.

Sameroff, A. J., McDonough, S. C., & Rosenblum, K. U (Eds.). (2004). Treating parent-infant relationship problems. New York: Guilford Press.

Squires, J., & Bricker, D. (2007). An activity-based approach to developing young children's social emotional competence. Baltimore: Paul H. Brookes.

Starfield, B. (2004). U.S. child health: What's amiss, and what should be done about it? Health Affairs, 23, 165-170.

Weatherston, D. (2000). The infant mental health specialist. Zero to Three, 2I(2), 46-83.

Yarnall, K. S., Pollack, K. I., Ostbye, T., Krause, K. M., & Michener, J. L. (2003). Primary care: Is there enough time for prevention? American Journal of Public Health, 93, 635-641.

Zuckerman, B., Stevens, G. D., Inkelas, M., & Halfon, N. (2004). Prevalence and correlates of high-quality basic pediatric preventive care. Pediatrics, 114, 1522-1528.

Michaela L. Z. Farber, PhD, BCD, LCSW-C, is assistant professor, National Catholic School of Social Service, Catholic University of America, Shahan Hall, 620 Michigan Avenue NE, Washington, DC 20064; e-mail: farber@cua.edu. The research reported in this article was supported by a collaborative partnership between the National Catholic School of Social Service, the National Research Center for Child and Family Services, Catholic University of America, the Lt. Joseph P. Kennedy Institute of Catholic Community Services, and Perry Family Health Center of Providence Hospital in Washington, DC. Funding was provided through grant R305 TO 10754- 02 from the U.S. Department of Education, Office of Educational Research and Improvement. On behalf of the project partnership, the author expresses gratitude to the participating families and collaborative partners. An earlier version of this article was presented at the Council on Social Work Education conference, February 2005, New York.
Table 1: Mean Change in Adequacy of Needs and Resources of
Intervention Families (N = 35)

                                        Pretest       Posttest
Types of Needs and Resources            M (SD)         M (SD)

Basic family needs
     Food for a week, housing,
     money to pay utilities or
     bills, money for family needs,
     clothes for family, furniture,
     heat, hot water or plumbing,
     transportation, telephone
     access, medical care or dental
     care for family                   46 (9.7)     52 (a) (5.8)

Parenting needs
     Time for family to be
     together, time to be with
     children, babysitting, regular
     child care, money for
     children's needs,
     age-appropriate toys,
     information about child
     rearing or parenting              19 (6.2)     28 (b) (3.3)

Interpersonal needs
     Time to sleep and rest, time
     for self, time to be with
     partner, time to be with close
     friends, time to get emotional
     support, time to socialize,
     resources to keep in shape and
     look nice, resources to
     participate in the community
     money to buy things for self,
     money to save for family,
     money for family entertainment    31 (8.6)     37 (c) (6.7)

Total adequacy of needs                97 (21.1)   118 (d) (15.6)
       and resources

Note: [[eta].sup.2]= effect size per the 0-1 range (Green &
Salkind, 2003).

(a) t(34) = 5.8, p = .000, [[eta].sup.2] = .52.

(b) t(34) = 9.8, p = .000, [[eta].sup.2] = .76.

(c) t(34) = 5.2, P = .000, [[eta].sup.2] = .47.

(d) t(34) = 9.1, P = .000, [[eta].sup.2] = .73.

Table 2: Nurturing and Child-rearing Beliefs and Parenting Practices
of Intervention Families on the Adult-Adolescent Parenting Inventory
(AAPI-2)

                                    Mothers with Abusive Practices
AAAI-2 Scale Factor                   at Enrollment (%) (N = 50)

Inappropriate expectations of                  --
  children's developmental needs

Lack of empathic awareness and                 50
  understanding of children's
  needs

Strong beliefs in the use and                  25
  value of corporal punishment
  in discipline

Engagement in parent-child                     40
  role reversal, or promoting
  inappropriate child behaviors

Oppressing children's need for                 24
  appropriate developmental
  independence

                        Mothers Experiencing
                         Factor Failures at     Mothers Experiencing
Number of AAPI-2         Enrollment (%) (N =     Factor Failures at
Scale Factor Failures          50) (a)         Exit (%) (n = 30) (b,c)

0                                 3                      75
1                                20                      21
2                                30                       4
3                                 7                       0
4                                11                       0
5                                 2                       0

Note: Factor failure on the AAPI-2 scale reflects having parenting
difficulties.

(a) [M.sub.1] (enrollment factor failure) = 1.57 (SD= 1.40).

(b) [M.sub.1] (exit factor failure) = 0.63 (SD = 0.76).

(c) Paired t(29) = 3.62, p = .001, [[eta].sup.2] = .29.

Table 3: Mean Differences in Parents' Selected Outcomes between
Intervention and Matched Comparison Families at Posttest

                                    Intervention         Comparison
                                       Family              Family
Parents' Selected Outcome              M (SD)              M (SD)

Total basic needs score            52.4 (7.8) (b)      44.0 (9.0) (c)
Total parenting needs score        28.0 (3.4) (b)      20.2 (7.2) (c)
Total interpersonal needs score    28.0 (5.0) (b)      25.5 (5.8) (c)
Total needs and resources score   117.8 (15.9) (b)     96.1 (20.3) (c)
Total AAPI-2 score failures         0.63 (0.76) (c)     1.50 (1.10) (c)
Total resilience score            108.5 (11.0) (d)    101.2 (11.2) (c)

Parents' Selected Outcome          t    df    p     [[eta].sup.2] (a)

Total basic needs score           3.4   63   .001          .15
Total parenting needs score       5.3   63   .000          .30
Total interpersonal needs score   2.3   63   .025          .08
Total needs and resources score   4.6   63   .000          .25
Total AAPI-2 score failures       3.5   58   .001          .17
Total resilience score            2.3   46   .026          .08

Note: AAPI-2=Adult-Adolescent Parenting Inventory.

(a) Effects for independent samples: [[eta].sup.2] = [t.sup.2]/
[t.sup.2] + ([N.sub.1], + [N.sub.2]), in which [[eta].sup.2]s of
.01, .06, and .14 are Interpreted as small, medium, and large
effect sizes, respectively (Green & Salkind, 2003.)

(b) n = 35.

(c) n = 30.

(d) n = 28.

Table 4: Mean Differences in Vocabulary Achievement of Intervention
and Matched Comparison Latino Children on the MacArthur Communicative
Development Inventories Short Form at Program Exit (N = 20)

                           Intervention   Comparison
                              Child         Child

Child has the ability to      M (SD)        M (SD)      t

Understand the word         89 (11.6)     79 (12.5)    2.5
Say the word                 83 (9.6)     73 (12.2)    3.0

Child has the ability to   df    p     [[eta].sup.2] (a)

Understand the word        38   .015          .14
Say the word               38   .006          .19

(a) Effects for independent samples: [[eta].sup.2] = [t.sup.2]/
[t.sup.2] + ([N.sub.1], + [N.sub.2] -2), in which [[eta].sup.2]s
of .01, .06, and .14 are by convention interpreted as small,
medium, and large effect sizes, respectively (Green & Salkind,
2003).
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.