Symposium
Chair: Lori A. Roggman
Discussant: Richard N. Roberts
Relationship-level interventions involving infants typically focuson promoting appropriate parent-infant interactions. One general programstrategy used to support parent-infant interactions is home visiting. Homevisiting programs vary in their reasons for selecting a home visitingstrategy in the first place and vary in the objectives they intend to meeton their home visits. In addition, individual home visitors within asingle program may implement visit strategies in different ways that varyin effectiveness. Therefore, evaluating home visits requires more thansimply comparing a group receiving home visits to a group receiving otherkinds of services; it requires an examination of the processes that occurduring home visits. Although the effectiveness of home visiting has beenquestioned, few studies have examined the actual process of home visiting.The studies presented here have examined home visiting processes withinvarious intervention programs serving infants. The first two papers present a methodology for evaluating homevisits in an Early Head Start home visiting program that emphasizesrelationships: parent-infant relationships as a target of intervention andstaff-parent relationships as a strategy of intervention. Methods forevaluating this home visiting program needed to be developed to beappropriate for the program. Researchers and program staff collaborated toidentify assessment objectives, to develop appropriate measures, and tointerpret the data for program improvement. This process and the initialresults will be presented in the first paper. The second paper extends thefirst by examining not only how home visits work but also what works bestfor whom, specifically how parents' mental health and use of socialresources interact with home visit quality to affect family improvement.The third paper examines a home-visiting intervention,the Mutual CompetenceModel, that emphasizes parent-child relationships as the target ofintervention and staff-parent relationships as the strategy of change forfamilies at risk because of low SES status. The primary focus of this paperwill be on staff training, verification of treatment implementation, andthe impacts of participation. These issues are critical to increasing ourknowledge of why and when home visit interventions are or are noteffective. The final paper in this symposium presents a multi-dimensionalassessment of home visit process in an Early Head Start program with twodifferent kinds of home visitors, Family Development Specialists andInfant-Toddler Development Specialists. By using a standardized assessmentform (the Home Visit Observation Form), the researchers were able todescribe commonalities and distinctions between the two kinds of homevisits as a first step toward identifying effective home visitinterventions. By developing methods for assessing the process, content, andquality of home visits, these studies have described methods that allowearly intervention research to move beyond simple program level comparisonstoward process evaluations of relationship-level interventions involvinginfants and their parents. The discussant has been asked to consider theappropriateness of these methods for studying the effectiveness of homevisiting in early intervention programs aimed at promoting infantdevelopment.
Details of individual items:
paper
Facilitating infant development through strengthening parents'roles has been the focus of the Mid-Iowa Community Action, Inc. (MICA)Early Head Start, which serves 75 families in a rural area of Central Iowa.MICA's intervention strategies are carried out through a unique home-basedmodel that utilizes the knowledge and skills of two developmental resourcepersons: a Family Development Specialist to work with family concerns anda Infant-Toddler Development Specialist to focus on the development of thechild(ren) within the context of their families. This presentation will feature data collected via nonparticipantobservations of home visits. The Home Visit Observation Form (HVOF) wasdeveloped via collaborative efforts between research and program teammembers. We felt it was essential to involve program staff in thedevelopment of our observation protocol for this project. First, becausewe were documenting their behavior, we needed their support and involvingthem in the process increased their investment. Second, while this programhas a fairly well articulated program philosophy, specific interventionstrategies and behavioral indicators of their approach are not clearlydocumented. Using focused discussions, research team members provided basicobservation categories that had been employed previously. The program teammembers were asked to identify their a) goals for intervention, b) thestrategies employed, and c) what we would observe. Research team membersused this information to develop initial subcategories. After pilot codingvideotaped home visits, the research team and program team met a secondtime to refine the instrument. The HVOF facilitates observation of ongoing activities in each ofthree main categories: a) primary interactors (e.g., parent-child,parent-interventionist, etc.), b) content of the interaction (e.g., childdevelopment, family functioning, etc.), and c) nature of theinterventionist's role (e.g., asking or giving information, supportingparent-child interaction, etc.) every 30 seconds. Data are collapsedacross intervals to indicate the percent of time each subcategory ofbehavior occurred during home visits. Findings to date indicate that the content and process of homevisits are related to the home interventionist's role. Infant-ToddlerDevelopment Specialists (ITDSs) are more likely to address childdevelopment issues while Family Development Specialists (FDSs) focus onfamily issues such as relationships, employment, or education. Both theITDSs and FDSs primarily support interaction with parents or other adults(over 50% of the time). The amount of time devoted to supportingparent-child interaction by modeling or coaching parent-child interactionsis limited. However, the percentages of time devoted to various activitiesvary widely between families, as well as between interventionists. We have taken a first step toward being able to identify whatintervention strategies work for whom by clearly describing theinterventions. Next, we can move to looking at how closely the actualinterventions match program goals and how they are related to familycharacteristics and outcomes realized. Only then can we make specificrecommendations about interventions, consider the integrity with which theyare implemented, and make generalizations about their effectiveness.
paper
Relationship-level interventions involving infants typically focuson promoting appropriate parent-infant interactions. One general programstrategy used to support parent-infant interactions is home visiting. Homevisiting programs vary in their reasons for selecting a home visitingstrategy in the first place and vary in the objectives they intend to meeton their home visits. In addition, individual home visitors within asingle program may implement visit strategies in different ways that varyin effectiveness. Therefore, evaluating home visits requires more thansimply comparing a group receiving home visits to a group receiving otherkinds of services; it requires an examination of the processes that occurduring home visits. Although the effectiveness of home visiting has beenquestioned, few studies have examined the actual process of home visiting.The studies presented here have examined home visiting processes withinvarious intervention programs serving infants. The first two papers present a methodology for evaluating homevisits in an Early Head Start home visiting program that emphasizesrelationships: parent-infant relationships as a target of intervention andstaff-parent relationships as a strategy of intervention. Methods forevaluating this home visiting program needed to be developed to beappropriate for the program. Researchers and program staff collaborated toidentify assessment objectives, to develop appropriate measures, and tointerpret the data for program improvement. This process and the initialresults will be presented in the first paper. The second paper extends thefirst by examining not only how home visits work but also what works bestfor whom, specifically how parents' mental health and use of socialresources interact with home visit quality to affect family improvement.The third paper examines a home-visiting intervention,the Mutual CompetenceModel, that emphasizes parent-child relationships as the target ofintervention and staff-parent relationships as the strategy of change forfamilies at risk because of low SES status. The primary focus of this paperwill be on staff training, verification of treatment implementation, andthe impacts of participation. These issues are critical to increasing ourknowledge of why and when home visit interventions are or are noteffective. The final paper in this symposium presents a multi-dimensionalassessment of home visit process in an Early Head Start program with twodifferent kinds of home visitors, Family Development Specialists andInfant-Toddler Development Specialists. By using a standardized assessmentform (the Home Visit Observation Form), the researchers were able todescribe commonalities and distinctions between the two kinds of homevisits as a first step toward identifying effective home visitinterventions. By developing methods for assessing the process, content, andquality of home visits, these studies have described methods that allowearly intervention research to move beyond simple program level comparisonstoward process evaluations of relationship-level interventions involvinginfants and their parents. The discussant has been asked to consider theappropriateness of these methods for studying the effectiveness of homevisiting in early intervention programs aimed at promoting infantdevelopment.
paper
Home visiting programs vary in their objectives or 'theory ofchange' and their effectiveness in helping families to meet thoseobjectives. Within home visiting programs themselves, some families benefitmore than others (Gomby, Culross, & Behrman, 1999).Families enrolled in early intervention programs vary in their mentalhealth and their use of social resources. A better understanding of therelation between these family characteristics and the relevant indicatorsof home visit quality will help predict those families who will benefitmost from a particular home visiting intervention. To explore the association between effective home visiting andvarious family characteristics, we interviewed mothers in a home-basedEarly Head Start program (N 92), interviewed program staff about familyimprovement, and rated the quality of videotaped home visits. Mothers wereinterviewed about symptoms of depression (CES-D), use of social support(F-Copes), and basic demographics when they first entered the program. Thehome visitors were asked after the second year of the program to rate theextent of improvement for each family. Improvement was rated by a 16item-scale that asked about specific domains of functioning. The scaleranged from 1 for 'inadequate progress' to 5 for 'outstanding, a truesuccess story.' Mothers' engagement during the home visit and home visitorfacilitation of positive mother-child interactions were rated byresearchers from videotaped home visit observations. These two globalrating scales were used to assess home visiting quality through theeffective implementation of the programs' objectives of increased positiveparent-infant play interactions and responsive parenting. Mothers'engagement during the home visits ranged from 1 (unengaged) to 6 (highlyengaged). Home visitor facilitation ranged from 1 representing nofacilitation or overly intrusive and directive behavior to 5 representingeffective facilitation and responsiveness to the cues of both parent andchild. Variations in maternal engagement and home visitor facilitationsuggest that home visit quality may vary by family and home visitor. Home visitor facilitation and family improvement scores wererelated to maternal characteristics such as depression and use of socialsupport. Briefly, mothers who were depressed, compared with those who werenot, were less likely to have home visits where the home visitor was ableto facilitate mother-child interaction and were less likely to use socialsupport. Also, mothers who turned to family and friends as well as thosewho sought spiritual support when problems arose received higherimprovement ratings than those mothers who either did not use socialsupport or used other types of social support. Additional results related to specific types of social support aswell as to interaction patterns among home visiting quality variables andmaternal characteristics will be presented in detail. Examination of thesematernal characteristics inrelation to home visitor facilitation ofparent-child interaction and to family improvement offers an importantsource of data for understanding the home visiting process and foranswering the question, 'What works best for whom?'
paper
Home visiting interventions have come under increased scrutiny interms of their effectiveness (Gomby, Culross, & Behrman, 1999) and researchis generally equivocal with specific intervention approaches showingdifferent outcomes at different times and in different places. Homevisiting, however, is not a treatment but a location for intervention.Unfortunately both politicians and researchers do not always providesufficient attention to this fact. Given the equivocal research andpotential impact on services, it becomes imperative that we betterunderstand the treatment(s) that is(are) home visiting. The Promoting Resilient Outcomes (PRO) project has developed a homevisitor intervention, the Mutual Competence Model, for environmentallyat-risk children, zero to two years of age, and their families.Community-based programs have been involved in the project and receivetraining and are involved in a comparative evaluation of the intervention.The primary goal of the intervention is an increase in secure attachmentand positive parent-child interactions. The model views the relationshipas the agent of change at all levels; parent/child, home visitor/parent,supervisor/home visitor, etc. The basic premise is that the mutualcompetency of the parent/infant relationship should become the focal aspectof intervention, even when the intervention includes other agendas. Theadoption of this model by program sites was a departure from approachesthat had been used in the past. Participating sites were involved inintensive training activities for 18 months. The focus of this presentation will not be on the model per se,although information on the model will be presented to provide perspectiveto the audience, but on the process of the programs moving their practicestoward the new model and to our understanding of the process asresearchers. In an attempt to better understand process activities, avariety of data were collected. These included: videotaped trainingsessions; a process focused questionnaire following each session;assessment of model philosophy and application; and videotaped homevisitation sessions. A formal qualitative analysis of individualinterviews with program staff was completed. In sum, a variety ofopportunities were taken to better understand the process of training. Although it is impossible to present all of this data, thispresentation will provide an overview of the process involved in programsimplementing the model. Most staff needed approximately one year oftraining before the basic concept and activities central to the model hadinfiltrated into their daily practice. Even after 18 months, concernsstill existed regarding the regular and ongoing implementation of themodel. This raises multiple questions both from a training and a researchperspective. Quantitative data collected on child and family outcomes willbe used to augment the qualitative information presented. This presentation will address not only the results of ourtreatment verification but also discuss practice issues learned as aresult of engaging in these various activities. Treatment informationcombined with results can provide researchers with more informationregarding the phenomenon being studied. Hopefully, this presentation willprovide information that will facilitate this process.