Kinship Care in the United States: A Systematic Review of Evidence-Based Research

Marc Winokur, Desiree Rozen, Stephen Thompson, Shawon Green, and Deborah Valentine Colorado State University

Final Report July 2005

Applied Research in Child Welfare Project Social Work Research Center School of Social Work College of Applied Human Sciences Colorado State University 222 W. Laurel Fort Collins, CO 80521

ACKNOWLEDGEMENTS

Project Director Deborah Valentine

Project Co-Chairs Jim Drendel Melisa Maling

Project Participants Adams County Arapahoe County Boulder County Broomfield County Denver County Douglas County El Paso County Jefferson County Larimer County Mesa County Pueblo County Weld County Colorado Department of Human Services Colorado Administrative Review Division Larimer Center for Mental Health

Technical Assistance Brian Cobb Jeff Gliner Brian Mattson James Thomas

Special Thanks David Gough Evidence for Policy and Practice Information and Coordinating Centre (EPPI-Centre), Social Science Research Unit, Institute of Education, University of London ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page ii

Kinship Care in the United States: A Systematic Review of Evidence-Based Research Executive Summary During the past 15 years, child welfare professionals have witnessed a rapid increase in the number of children removed the home and placed with relatives (Cuddeback, 2004). However, social work research has not kept pace with the exponential growth of kinship care as a placement option (Berrick & Barth, 1994; Dubowitz, 1994). Furthermore, much of the research on kinship care is anecdotal and conjectural, which does not allow for evidencebased decisions to be drawn from comparisons of children in out-of-home care (Goerge, Wulczyn, & Fanshel, 1994). To address this limitation, a systematic review of quantitative research on kinship care in the United States was conducted. For this study, the child welfare outcomes were permanency, behavior problems, mental health service utilization, reentry, adaptive behaviors, family relations, mental health problems, and educational attainment. The systematic review followed a rigorous vetting process informed by the What Works Clearinghouse (WWC) standards. After a comprehensive electronic and manual search of the kinship care literature, 1314 abstracts were reviewed, 190 articles and reports were acquired, and 74 eligible studies were assessed on the quality of their research designs and methods according to the WWC Study Design and Implementation Device. After data extraction using EPPI-Reviewer, 35 of the studies met the minimum inclusion criteria for the kinship care evidence base. Studies that did not qualify were deemed inadequate because of insufficient statistical reporting and non-equivalent group comparisons. In the final stage of the review, meta-analyses were conducted to calculate effect sizes for the outcomes represented by 23 of the studies in the evidence base, as the effect size estimates from the remaining12 studies were not directly comparable. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page iii

Overall, there are likely no harmful effects of kinship care on child welfare outcomes. According to the research, children in kinship care experience better outcomes in regard to behavior problems, reentry, adaptive behaviors, family relations, and mental health problems than do children in foster care. However, children placed with kin are less likely to achieve permanency and utilize mental health services. Although distinct trends were present for each outcome, the practical significance of these results is narrowed by the relatively small overall effect sizes. These findings also are undermined by numerous threats to the validity of studies in the kinship care evidence base. For example, the authors are somewhat unconfident that the outcome measures were properly defined, that children placed in kinship care were comparable to children placed in foster care, and that studies were free of events that happened concurrently with the intervention and may have confused its effects. Although these limitations complicate the interpretation of the results, several clear implications for social work practitioners and policymakers surfaced from the study. If the goal of kinship care is to lower reentry rates while improving the behavior, family relations, and mental health of children, then the evidence base is supportive. However, the findings do not support implementing kinship care solely to increase the permanency rates and service utilization of children in out-of-home placement. Finally, the systematic review answered some important questions about permanency outcomes for children in kinship care, but there still are major gaps in the literature regarding the safety and well-being of these children.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page iv

TABLE OF CONTENTS Introduction Topic Description Context Controversies Major Outcomes Research Synthesis

1 2 3 4 4

Search Strategy Search Results Keywording Criteria Data Extraction Meta-Analyses Effect Size Calculations

6 6 7 8 9 9

CREAD Ratings Overall Bivariate Results Permanency Behavior Problems Mental Health Service Utilization Reentry Adaptive Behaviors Family Relations Mental Health Problems Educational Attainment Overall Mulitvariate Results Subgroup Analyses

11 14 15 17 18 19 21 22 24 25 26 30

Conclusions Limitations Recommendations Future Research

31 33 34 34

Methodology

Findings

Discussion

References

36

Appendices Data Extraction Guidelines Studies in the Kinship Care Evidence Base Outcome Measures for Effect Size Calculations

51 65 71

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page v

Introduction This report describes a systematic review of evidence-based research on kinship care in the United States. The report provides a brief introduction to kinship care, details the search strategies used to locate the final literature set, and describes how the studies were coded on design and methodological criteria. In addition, findings, conclusions, limitations, and recommendations are presented for the studies that merited inclusion into the kinship care evidence base. Topic Description During the past 15 years, child welfare professionals have witnessed a rapid increase in the number of children removed from home and placed with relatives (Cuddeback, 2004). According to the most recent estimate from the Adoption and Foster Care Analysis and Reporting System (AFCARS), there were 121,030 children in the United States foster care system living with kin as of September 2003. The main reasons for the growth of this placement option include an influx of children into out-of-home care (Berrick, 1998), a persistent shortage in foster care homes (Berrick, 1998), and a shift in federal policy toward treating kin as appropriate caregivers with all of the legal rights and responsibilities of foster parents (Leos-Urbel, Bess, & Geen, 2002). Kinship care is the formal, informal, or private placement of children with family members rather than traditional foster parents. Formal kinship care is a legal arrangement in which the child welfare agency has custody of a child (Ayala-Quillen, 1998). Informal kinship care is when the child welfare agency facilitates the placement of a child but does not seek custody (Geen, 2000). Private kinship care is a voluntary arrangement between

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 1

the biological parents and family members without the involvement of the child welfare agency (Dubowitz, 1994). For a study to be included in the systematic review, kin caregivers could be certified or uncertified but the kinship care placement had to be formal. The rationale for this decision is twofold. Although the placement of children with kin is the essence of all kinship care interventions, children placed informally or privately often differ from children placed formally in the level of abuse and neglect precipitating the placement. Second, the meta-analyses required that all kinship care variations be combined into one group to allow for interpretable comparisons with the foster care group. Thus, the treatment group included children for whom kinship care was their first, last, only, or majority placement in out-of-home care. Context The following narrative incorporates kinship care research not eligible for inclusion in the evidence base (e.g., policy briefs, literature reviews) to provide a historical and policy context for this review. Originating in the form of extended families in the African-American community (Brooks, 2002; Danzy & Jackson, 1997), kinship care has existed for hundreds of years in the United States (Geen, 2000). As recently as 40 years ago, however, relatives were not eligible to become licensed foster parents or receive federal benefits (Leos-Urbel et al., 2002). This changed with the Miller v. Youakim Supreme Court case in 1979, which “determined that kin could not be excluded from the definition of foster parents and that under some conditions, kin might be eligible for foster care benefits” (Berrick & Barth, 1994, p. 1). The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 explicitly required states to give

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 2

preference to family members when placing a child in out-of-home care (Leos-Urbel et al., 2002). The most recent federal legislation, the Adoption and Safe Families Act of 1997, continues this federal commitment towards promoting and supporting kinship care (Ayala-Quillen, 1998). According to Geen and Berrick (2002), “historically, federal policy has been vague as to how state child welfare agencies should treat kinship care. Under this framework, many states have implemented policies that support greater use of kin” (p. 3). Most states first look to relatives (which often is broadly defined to include any adult with close family ties) when placing a child in out-of-home care (Geen, 2000). However, a great disparity still exists in state policies and practices regarding the assessment, selection, certification, and monitoring of kin caregivers (Ayala-Quillen, 1998; LeosUrbel et al., 2002). Social work research has not kept up with the exponential growth of kinship care as a placement option (Berrick & Barth, 1994; Dubowitz, 1994). Furthermore, much of the research supporting kinship care is anecdotal and conjectural, which does not allow for evidence-based decisions to be drawn from comparisons of children in out-of-home care (Goerge, Wulczyn, & Fanshel, 1994). Controversies Similar to other social work interventions, kinship care is faced with its fair share of controversial issues. The major controversy centers on the unequal financial support (Brooks, 2002) and service provision received by kinship caregivers as compared with traditional foster parents (Dubowitz, 1994; Roberts, 2001). The licensing and certification of kinship caregivers also is a source of much disagreement and dissatisfaction (Gibbs & Muller, 2000). One of the key debates is over the appropriate level of involvement for

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 3

biological parents prior to and after the removal of their children (Ayala-Quillen, 1998). Additionally, there is concern from all corners about the permanency (Leos-Urbel et al., 2002) and well-being (Geen & Berrick, 2002) of children in kinship care. Major Outcomes Most of the published research on kinship care within the past 15 years is grounded in caregiver and caseworker perceptions generated from surveys and interviews. The major outcomes from these mostly descriptive and qualitative studies are related to the mental and physical health, resource utilization, and social interactions of kin caregivers. Although these outcomes are very relevant for child welfare professionals, data on child outcomes is what drives the policy and practice of kinship care at the national, state, and local levels. Based on the existing quantitative research of kinship care, the child welfare outcomes for this systematic review were permanency, behavior problems, mental health service utilization, reentry, adaptive behaviors, family relations, mental health problems, and educational attainment. While there is great interest in the safety and well-being of children placed in kinship care, there is a dearth of experimental research on these outcomes (Gibbs & Muller, 2000). Research Synthesis A recent “substantive synthesis of research” by Cuddeback (2004) is presented to supplement the findings from this systematic review. Cuddeback did an admirable job of comprehensively reviewing the literature on kinship care and addressed many of the weaknesses of quantitative research on the topic. However, the synthesis lacked a systematic approach to including studies and did not employ meta-analytical techniques to quantify the presumed effects of kinship care on child welfare outcomes.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 4

Overall, Cuddeback confirms much of the conventional wisdom about kinship care. For example, he concludes that kinship caregivers are more likely to be AfricanAmerican, older, single, less educated, unemployed, and poor. Furthermore, kin caregivers report less daily activity, more health problems, higher levels of depression, and less marital satisfaction than do foster parents and non-custodial grandparents. Cuddeback is unequivocal in stating that kinship care families receive less training, services, and financial support than do foster care families. As for families-of-origin, Cuddeback notes that biological parents of children in kinship care are more likely than biological parents of children in foster care to physically neglect and abandon their children because of substance abuse problems. Cuddeback also argues that kinship birth families are less likely to receive family preservation services and other forms of assistance. As for child functioning, Cuddeback found inconclusive evidence that children in kinship care have greater problems than do children in foster care. Although children in kinship care are more likely to have repeated a grade or be enrolled in special education, there are no reported differences in educational attainment. Finally, Cuddeback concludes that children in kinship care have more stable placements but are reunified more slowly than are children in foster care.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 5

Methodology This systematic review followed a rigorous vetting process informed by the What Works Clearinghouse (WWC) standards. The following narrative provides a step-by-step description of the search, retrieval, coding, and meta-analytic procedures used to select and analyze the studies in the kinship care evidence base. Search Strategy In the first stage of the systematic review, a comprehensive search strategy was formulated. As the topic is in the social work domain, it was decided that the databases for Family and Society Studies Worldwide, Family Studies, Social Sciences Citation Index, Social Work Abstracts, and Sociological Abstracts would be accessed to identify literature on kinship care. The Educational Resource Information Center (ERIC), Master Abstracts International, and Dissertation Abstracts International databases also were accessed to cast a wider net for appropriate studies. The terms, “kinship care, kinship foster care, and kin care” were used to search these databases. The search was bounded to studies completed in the United States between 1992-2005, so that all kinship care interventions occurred during a similar policy and practice context. Search Results The initial electronic search yielded 1255 citations for journal articles, conference papers, evaluation reports, dissertations, book chapters, and policy papers on kinship care. During the second stage of the review process, abstracts for each of the 1255 references were read and analyzed according to the initial selection criteria. As a result, full-text copies for the 210 articles and reports that employed quantitative methods to study kinship care outcomes were sought. Due to out-of-print reports, unpublished

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 6

conference presentations, and extinct websites, 183 of the 210 documents were acquired (83% retrieval rate). To ensure that studies were not missed during the electronic search, a visual examination of the reference lists for each of the 183 articles and reports was conducted. After the manual search, 59 additional studies were identified for possible acquisition. After reading the abstracts for these 59 references, seven that met the inclusion criteria were obtained. As shown in the reference list, citations for all 190 studies were entered into Reference Manager 9, which is an interactive literature management software package. The citations were then uploaded to the EPPI-Reviewer online review database, which is a tool for storing and analyzing data for systematic reviews. EPPI-Reviewer is maintained by the Evidence for Policy and Practice Information and Coordinating Centre (EPPI-Centre) and can be accessed at http://eppi.ioe.ac.uk. Keywording Criteria During the third stage of the review, a “keywording” rubric was used to categorize each study by the type of intervention, intensity of data collection, type of research design, type of sample, and type of outcome. Specifically, a study had to investigate a formal kinship care intervention (i.e., study included cases in which there was state custody of the child and formal placement with either licensed or unlicensed kin caregivers) using a between-groups or within-subjects quantitative research design. The sample for an eligible study had to be drawn from children under the age of 18 placed in kinship care. Furthermore, the research had to include child, kin, or parent outcomes, although this criterion was changed later to include only studies with child outcomes. If there were multiple articles or reports from a single study, only the most recent and/or

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 7

complete document was eligible for inclusion. After being pilot tested for clarity and calibrated for interrater reliability, the instrument was revised before the 190 studies were keyworded in EPPI-Reviewer. Overall, 157 of the 190 studies (83%) were keyworded by two members of the research team to ensure consistency in the inclusion decision. Based on the aforementioned criteria, it was determined that 74 studies were qualified to be included in the next stage of the review process. The most common reasons for exclusion were the type of intervention (i.e., foster care only or no intervention), type of research design (i.e., descriptive or qualitative study), and the type of outcome (e.g., system). Data Extraction During the fourth stage of the review, the 74 eligible studies were assessed on the quality of their research designs and methods according to standards set forth in the WWC Study Design and Implementation Device (Study DIAD). As shown in Appendix A, a “data extraction” template was developed to provide guidelines for answering the eight composite questions posed by the Study DIAD. It should be noted that the original version of the Study DIAD was used in this systematic review. Although the new version is different from a procedural standpoint, the core standards that define the composite questions remain the same. Two researchers extracted data from each of the 74 studies and entered the results into EPPI-Reviewer. The researchers then came to consensus on the final coding for each study. Finally, the 74 studies were assessed according to the eight composite questions in the Study DIAD. As shown in Appendix B, 35 of the studies met the minimum criteria for all composite questions and were included in the kinship care evidence base. The studies

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 8

that did not qualify were deemed inadequate because of insufficient statistical reporting and non-equivalent group comparisons. Meta-Analyses The purpose of a meta-analysis is to aggregate studies that share common features (e.g., intervention, sample, outcome) by calculating an overall effect size based on the effect size for each study. Thus, a standardized effect size was computed using EPPIReviewer for the eight outcomes in the evidence base with at least three studies reporting appropriate data. Specifically, meta-analyses were conducted for permanency, behavior problems, mental health service utilization, reentry, adaptive behaviors, family relations, mental health problems, and educational attainment. However, only 23 of the 35 studies in the evidence base were represented in these meta-analyses. Studies that reported only multivariate results (e.g., odds ratios) were not included because the corresponding effect sizes would not be comparable to the bivariate results (e.g., raw percentages) from the other studies. Therefore, the results from the 12 multivariate studies are reported separately and then compared with the results from the 23 bivariate studies. The outcome measures used for effect size calculations and estimations in all 35 studies are shown in Appendix C. Effect Size Calculations Until effect sizes are considered, there is no sense of the magnitude of differences, regardless of statistical significance, between kinship care and foster care groups. The most common effect size statistics used in meta-analysis are Cohen’s d and Hedges’ g, both of which represent the strength of a relationship between an independent variable and a dependent variable in standard deviation units. In this study, Hedges’ g was

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 9

computed by dividing the difference between group means by the pooled population value of the standard deviation of the groups. An effect size may be positive or negative depending upon the direction of the difference between two groups. An alpha level of .05 was used to determine the statistical significance of the g values. Because effect sizes are estimates and not parameters, confidence intervals are provided in meta-analysis to quantify some of the uncertainty inherent in capturing the “true” effect of an intervention. Most researchers suggest the following scale for a g effect size: g < .20 is small, g = .50 is medium, and g > .80 is large. A small effect size is interpreted as having little consequence for child welfare outcomes, whereas medium and large effect sizes are interpreted as having real implications for social work practice and policy. The combined effect size for each outcome was computed as a weighted mean of the effect size for each study, with the weight being the inverse of the square of the standard error. Thus, a study is given greater weight for a larger sample size and more precise measurement, both of which reduce standard error. A t value also is computed to test the significance of the combined effect size by dividing the absolute value of the mean effect size by the standard error of the mean effect size. Specifically, this is a twotailed test of the null hypothesis that the combined effect size is not significantly different from zero. When calculating the combined effect size of an outcome, a Q statistic is computed to test for the heterogeneity of effect sizes for each study in the group. The Q statistic indicates whether the variability in effect sizes is due to sampling error or to some unmeasured variable(s), in which case the combined effect size is less reliable in its estimation of the population effect size.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 10

Findings The ratings from the WWC Cumulative Research Evidence Assessment Device (CREAD) provide an indication of the overall quality of the research designs and methods used by studies in the evidence base. The effect size results from the meta-analyses allow for a comparison of the directionality and statistical significance of the kinship care effect on selected child welfare outcomes. For studies included in the meta-analyses, the author and publication date are listed in the table and the title of the article or report is listed in the corresponding location in the figure. CREAD Ratings The 35 studies in the evidence base were evaluated on their construct, internal, external, and statistical conclusion validity as defined by the CREAD. As stated in the CREAD guidelines, the objective was to “provide an expression of the confidence with which a conclusion can be drawn about the existence of causal effects of an intervention based on an entire body of accumulated evidence.” The following ratings characterize the depth, breadth, and consistency of the quantitative research on kinship care. Composite Question #1: Construct Validity – Intervention The authors are confident that the intervention was properly defined in the evidence base. Specifically, 33 of 35 studies (94%) fully reflected commonly held or theoretically derived ideas about kinship care. The intervention was adequately described to allow for replication, and the implementation of kinship care was largely consistent with its defined characteristics. However, the description of kinship care implementation was lacking information on caregiver certification, service utilization, age at placement, and length of stay.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 11

Composite Question #2: Construct Validity – Outcomes The authors are somewhat unconfident that the outcome measures were properly defined in the evidence base. Specifically, 8 of 35 studies (23%) provided evidence of construct validity for the outcome measures considered in the systematic review. In the studies that did not merit a “yes” on this composite question, there was evidence that the outcome measures were properly aligned to the intervention. However, there was less evidence that the measures were reliable for assessing the effect of kinship care on child welfare outcomes. Composite Question #3: Internal Validity – Selection The authors are somewhat unconfident that children placed in kinship care were comparable to children placed in foster care for the studies in the evidence base. Because none of the 35 studies used random assignment, groups were not equivalent as defined by the WWC standards. However, 13 of the studies (37%) employed adequate equating procedures to make the groups comparable (e.g., use of covariates) and there were no indications of severe overall or differential attrition. Composite Question #4: Internal Validity – Contamination The authors are somewhat unconfident that the studies were free of events that happened concurrently with the intervention that may have confused its effects. Specifically, none of the 35 studies rendered contaminating events implausible through random assignment. However, there was no positive evidence of a contaminating event and no identified processes that were alternative explanations for the treatment effects in the evidence base.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 12

Composite Question #5: External Validity – Sampling The authors are confident that the kinship care intervention was tested for its effectiveness using targeted participants, settings, outcomes, and occasions. Although only 6 of the 35 studies (17%) merited a “yes” on this composite question, most aspects of the theoretical population and common variations of settings, classes of outcomes, and data collection occasions were represented in the evidence base. Specifically, children from all age groups and many different geographic locations were represented. The studies also employed appropriate permanency and well-being outcomes and used crosssectional and longitudinal measurement processes. Composite Question #6: External Validity – Testing within Subgroups The authors are somewhat confident that the intervention was tested for its effectiveness within a reasonable range of targeted participants, settings, outcomes, and occasions. For example, a few studies in the evidence base tested variations of the kinship care intervention (e.g., formal vs. informal placement). However, only 8 of the studies (23%) tested kinship care for its effectiveness on important subgroups. The majority of studies included subgroup analyses but did not disaggregate data by placement type. Composite Question #7: Statistical Conclusion Validity – Effect Size Estimation The authors are somewhat confident that the studies in the evidence base allow for a precise estimation of effect size. Specifically, 23 of 35 studies (66%) are based on statistical properties that allow for sufficiently precise estimate of effect sizes. However, we are less confident that the studies met all of the independence, distributional, and variance assumptions. We also are less confident that the studies utilized reliable and valid outcome measures to conduct appropriate quantitative analyses.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 13

Composite Question #8: Statistical Conclusion Validity – Completeness of Reporting The authors are confident that studies were not systematically excluded because of their results. The 35 studies clearly did not censor data at the outcome level because many of the findings were of no difference between kinship care and foster care groups. In addition, the literature search was effective in uncovering studies with limited availability, such as dissertations and unpublished reports. Overall Bivariate Results Bivariate data were used to calculate the relationship between kinship and foster care status on child welfare outcomes. As displayed in Table 1, effect sizes for the majority of outcomes were in the favor of children placed with kin (Note: positive effect sizes are in the favor of kinship care). Children in kinship care had better reported outcomes for behavior problems, reentry, adaptive behaviors, family relations, and mental health problems. However, children in foster care had better reported outcomes in permanency and mental health service utilization. There were other outcomes investigated by studies in the evidence base (e.g., maltreatment, employment) that were not included in the meta-analyses because there was only one study addressing each outcome.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 14

Table 1 Combined Effect Size Statistics ________________________________________________________________________ Test of Null Hypothesis Test of Heterogeneity ____________________ _________________________ Outcome Effect Sig. Q Value df p value (g) (p) ________________________________________________________________________ Permanency -.343 .001*** 457 7 .000*** Behavior Problems .321 .001*** 16.3 4 .002** 2.29 3 .514 Mental Health Services -.406 .001*** Reentry .108 .001*** 180 2 .000*** Adaptive Behaviors .365 .001*** 11.3 2 .004** Family Relations .253 .011* 42.7 2 .000*** Mental Health Problems .305 .001*** 6.73 2 .035* Educational Attainment -.045 .590 5.69 2 .058 _______________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

Permanency As measured by administrative databases, permanency was defined as adoption, reunification, or subsidized guardianship as opposed to remaining in out-of-home care. As displayed in Table 2, four studies (Barth, 2002; Berrick, 1999; Needell, 1996; Smith, 2003) found statistically significant differences in the favor of children placed in foster care. Two studies (McIntosh, 2002; Smith, 2002) reported nonsignificant differences in favor of children in foster care. One study (Testa, 1999) found statistically significant differences in favor of children in kinship care, while another (Testa, 2001) reported nonsignificant differences in favor of children in kinship care. The heterogeneity statistic (Q = 457) was significant for the permanency outcome. As displayed in Figure 1, the combined effect size for the permanency construct is -.343 with a lower and upper bound of -.354 and -.331 respectively. Thus, there is a statistically significant small to medium negative effect on permanency, as children in kinship care are less likely than children in foster care to be adopted, reunified, or in subsidized guardianship. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 15

Table 2 Effect Size Results for Permanency ______________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Barth (2002) Smith (2003)

25154 379 McIntosh (2002) 39 Testa (2001) 955 Smith (2002) 39 Needell (1996) 4090 Testa (1999) 2159 Berrick (1999) 15739 Combined

38430 878 54 955 36 6603 3167 32586

-.381 -.279 -.090 .039 -.395 -.143 .071 -.407

-.397 -.400 -.502 -.051 -.852 -.182 .016 -.426

-.365 -.158 .322 .129 .063 -.104 .126 -.388

* *

* * *

48554 82709

-.343

-.354

-.331

***

________________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

Figure 1 Forest Plot of Effect Sizes for Permanency

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 16

Behavior Problems As measured by case records and caregiver and teacher reports, this outcome was defined as internalizing (e.g., withdrawn, passive) and externalizing (e.g., aggressive, delinquent) problem behaviors. As displayed in Table 3, four studies (Benedict, 1996; Jones, 1998; Keller, 2001; Surbeck, 2000) found statistically significant differences in favor of children in kinship care. One study (Shore, 2002) reported nonsignificant differences in favor of children in foster care. The heterogeneity statistic was significant (Q = 16.3) for the behavior problems outcome. Table 3 Effect Size Results for Behavior Problems ________________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Benedict (1996) 86

Keller (2001) Shore (2002) Jones (1998) Surbeck (2000)

67 56 159 98

128 173 129 241 102

.557 .527 -.224 .316 .328

.278 .241 -.539 .115 .049

.835 .813 .090 .518 .607

* * * *

Combined

466

773

.321

.204

.439

***

________________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

As displayed in Figure 2, the combined effect size for the behavior problems construct is .321 with a lower and upper bound of .204 and .439 respectively. Thus, there is a statistically significant small to medium positive effect on behavior problems, as children in kinship care have fewer reported internalizing and externalizing problem behaviors than do children in foster care.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 17

Figure 2 Forest Plot of Effect Sizes for Behavior Problems

Mental Health Service Utilization As measured by child medical records and caregiver reports, mental health service utilization was defined as actual receipt of mental health services (e.g., visits to a provider). As displayed in Table 4, two studies (Carpenter, 2004; Clyman, 1998) found statistically significant differences in favor of children in foster care. Two other studies (Jenkins, 2002; Leslie, 2000) reported nonsignificant differences in favor of children in foster care. The heterogeneity statistic was not significant (Q = 2.29) for the mental health service utilization outcome. As displayed in Figure 3, the combined effect size for the mental health service utilization construct is -.406 with a lower and upper bound of -.544 and -.268 respectively. Thus, there is a statistically significant medium negative effect on mental health service utilization, as children in kinship care are less likely to receive mental health services than are children in foster care. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 18

Table 4 Effect Size Results for Mental Health Service Utilization ________________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Carpenter (2004) 394

Leslie (2000) 53 Jenkins (2002) 52 Clyman (1998) 41

149 243 83 48

-.441 -.259 -.337 -.632

-.632 -.557 -.686 -1.060

-.251 .039 .012 -.205

*

*

Combined 540 523 -.406 -.544 -.268 *** ________________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

Figure 3 Forest Plot of Effect Sizes for Mental Health Service Utilization

Reentry As measured by administrative databases, reentry was defined as a return to outof-home care after experiencing a permanent placement. As displayed in Table 5, all three studies (Berrick, 1999; Needell, 1996; Wulczyn) reported statistically significant findings in favor of children in kinship care. The heterogeneity statistic was significant (Q = 180) for the reentry outcome. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 19

Table 5 Effect Size Results for Reentry ________________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Needell (1996) 7703 8864 Berrick (1999) 22588 32586 Wulczyn (2002) 9727 15329

.221 .032 .201

.191 .015 .175

.252 .049 .226

* * *

Combined

.108

.095

.121

***

40018 56779

________________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

As displayed in Figure 4, the combined effect size for the reentry construct is .108 with a lower and upper bound of .095 and .121 respectively. Thus, there is a statistically significant small positive effect on reentry, as children in kinship care are less likely than children in foster care to reenter out-of-home care after adoption, reunification, or subsidized guardianship.

Figure 4 Forest Plot of Effect Sizes for Reentry

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 20

Adaptive Behaviors As measured by caregiver reports, adaptive behaviors were defined as the ability to perform competencies required for personal and social sufficiency. As displayed in Table 6, two studies (Belanger, 2001; Keller, 2001) found statistically significant differences in favor of children in kinship care. One study (Jones, 1998) reported nonsignificant differences in favor of children in kinship care. The heterogeneity statistic was significant (Q = 11.3) for the adaptive behaviors outcome.

Table 6 Effect Size Results for Adaptive Behaviors ________________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Keller (2001) Jones (1998)

57 107 Belanger (2001) 20

145 164 37

.397 .190 1.29

.088 -.054 .695

.706 .434 1.89

* *

Combined 184 346 .365 .183 .547 *** ______________________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

As displayed in Figure 5, the combined effect size for the adaptive behaviors construct is .365 with a lower and upper bound of .183 and .547 respectively. Thus, there is a statistically significant small to medium positive effect on adaptive behaviors, as children in kinship care have better reported competencies in communication, daily living, and socialization than do children in foster care.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 21

Figure 5 Forest Plot of Effect Sizes for Adaptive Behaviors

Family Relations As measured by caregiver reports, family relations were defined as the quality of relationships between caregivers and children in regard to problem-solving efforts, tolerance, long-term commitment, and conflicts. As displayed in Table 7, one study (Berrick, 1997) found statistically significant differences in favor of children in kinship care, while one study (Surbeck, 2000) reported nonsignificant differences in favor of children in kinship care. One study (Jenkins, 2002) reported nonsignificant differences in favor of children in foster care. The heterogeneity statistic was significant (Q = 42.7) for the family relations outcome.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 22

Table 7 Effect Size Results for Family Relations ________________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Berrick (1997) 28 Jenkins (2002) 76 Surbeck (2000) 98

33 105 102

2.20 -.168 .263

1.56 -.463 -.016

2.85 .128 .541

*

Combined

240

.253

.059

.446

**

202

________________________________________________________________________ *p < .05. ** p < .01. *** p < .001.

As displayed in Figure 6, the combined effect size for the family relations construct is .253 with a lower and upper bound of .059 and .446 respectively. Thus, there is a statistically significant small positive effect on family relations, as children in kinship care have better reported relationships with their caregivers than do children in foster care.

Figure 6 Forest Plot of Effect Sizes for Family Relations

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 23

Mental Health Problems As measured by case records and caregiver reports, mental health problems were defined as psychiatric illnesses or psychopathological conditions. As displayed in Table 8, two studies (Benedict, 1996; Tompkins, 2003) found statistically significant differences in favor of children in kinship care. One study (Belanger, 2001) reported nonsignificant differences in favor of children in kinship care. The heterogeneity statistic was significant (Q = 6.73) for the mental health problems outcome.

Table 8 Effect Size Results for Mental Health Problems ________________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Benedict (1996) 86 128 Tompkins (2003) 125856 191619 Belanger (2001) 21 38

.653 .305 .061

.373 .297 -.472

.933 .312 .594

* *

Combined

.305

.298

.312

***

125963 191785

________________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

As displayed in Figure 7, the combined effect size for the mental health problems construct is .305 with a lower and upper bound of .298 and .312 respectively. Thus, there is a statistically significant small positive effect on mental health problems, as children in kinship care have fewer reported psychiatric illnesses and psychopathological conditions than do children in foster care.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 24

Figure 7 Forest Plot of Effect Sizes for Mental Health Problems

Educational Attainment As measured by case records and administrative databases, educational attainment was defined as completion of high school. As displayed in Table 9, one study (Christopher, 1998) reported statistically significant differences in favor of children in foster care and one study (Benedict, 1996) reported nonsignificant findings in favor of children in foster care. One study (Prosser, 1997) found nonsignificant differences in favor of children in kinship care. The heterogeneity statistic was not significant (Q = 5.69) for the educational attainment outcome. Table 9 Effect Size Results for Educational Attainment ________________________________________________________________________ Study

Kin (n)

Foster (n)

Effect (g)

Lower (CI)

Upper (CI)

Sig. (p)

________________________________________________________________________ Benedict (1996) 86

24 Prosser (1997) 607

128 42 92

-.101 -.573 .089

-.375 -1.085 -.130

.172 -.062 .308

Combined

262

-.045

-.207

.118

Christopher (1998)

717

*

________________________________________________________________________ * p < .05. ** p < .01. *** p < .001.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 25

As displayed in Figure 8, the overall effect size for the educational attainment construct is a nonsignificant -.045 with a lower and upper bound of -.207 and .118 respectively. Thus, there is no effect on educational attainment, as children in kinship care do not differ significantly from children in foster care on school graduation rates. Figure 8 Forest Plot of Effect Sizes for Educational Attainment

Overall Multivariate Results Twelve of the 35 studies in the kinship care evidence base did not report the requisite bivariate data for effect size calculation using EPPI-Reviewer. These studies were retained in the systematic review because they reported multivariate statistics that can serve as estimates of the magnitude and direction of treatment effects. For example, these studies reported hazard, risk, and odds ratios, in addition to unstandardized coefficients and beta values. Multivariate results were reported to explore the relationship between kinship and foster care status on child welfare outcomes while controlling for demographic and placement characteristics. Additionally, the findings for studies that reported both bivariate and multivariate data were examined.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 26

The most confirmatory finding from the multivariate analyses was for the reentry outcome. Similar to the bivariate results, three studies found small to medium positive effects on reentry for children placed in kinship care. Specifically, Courtney (1995), Courtney, Piliavin, and Entner Wright (1997), and Wells and Guo (1999) found that children in kinship care were less likely to reenter out-of-home care after permanent placement than were children in foster care. As for other confirmatory findings, McMillen et al. (2004) found that children in kinship care were two and half times less likely to use current outpatient therapy than were children in foster care. Zima et al. (2000) found that children in therapeutic foster care homes were three times more likely to have a clinical behavior problem than were children living in kinship family homes. The most disconfirming finding from the multivariate analyses was for the permanency outcome. Courtney and Barth (1996) found that children in kinship care were twice as likely to be adopted or reunified than were children in foster care. Although reported by the same research team, the most convincing finding from the multivariate analyses was for the maltreatment construct. Two studies reported that children in kinship care were less likely to be maltreated than were children in foster care. Specifically, Zuravin, Benedict, and Somerfield (1993) found that foster care homes were almost three times more likely to maltreat than were kinship care homes, while Benedict, Zuravin, Somerfield, and Brandt (1996) found that placement in foster care increased the likelihood of association with maltreatment by over four times. These results are in agreement with the one study that reported bivariate maltreatment data, as Benedict, Zuravin, and Stallings (1996) also found that children in kinship care were less likely to be maltreated.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 27

Similar to the inconclusive results from the bivariate data, there was no difference in educational attainment between children in kinship and foster care. Specifically, Smithgall, Gladden, Howard, Goerge, and Courtney (2004) reported an odds ratio of 1.8 for the kinship care group and 1.7 for the foster care group on the likelihood of being old for grade. Furthermore, Shin (2003a) reported that placement in kinship care accounted for only 5.75% of the variance in reading achievement. There were several outcomes from the multivariate analyses that were not represented in the bivariate results including physical health problems and employment. For example, Carlson (2002) reported less adult risk of physical health problems for children in kinship care, while Goerge et al. (2002) reported that children in kinship care from Illinois were one and half times more likely to be employed than were children in foster care. There were seven studies in the evidence base that reported both bivariate and multivariate data for the same outcome measure in the same metric. These studies allowed for a comparison between bivariate and multivariate results to determine if the treatment effects were in the same direction and of similar magnitude. For permanency, Barth, Webster, and Lee (2002) reported a bivariate odds ratio of .22 and a multivariate odds ratio of .50, which indicates that children in kinship care were less likely to be adopted after four years of out-of-home placement. Smith (2003) reported a bivariate odds ratio of .52 and a multivariate hazard ratio of .28, which indicates that children in kinship care were less likely to be discharged within one year of termination of parental rights. Needell (1996) reported a bivariate risk ratio of 1.50 and a multivariate risk ratio of 1.27, which indicates that children in kinship care were less likely to be reunified or adopted.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 28

As for other outcomes, Carpenter, Berman, Clyman, Moore, and Xu (2004) reported a bivariate odds ratio of 2.6 and a multivariate odds ratio of 2.4, which indicates that children in foster care were more likely to receive mental health services. Clyman, Riley, Lewin, and Messer (1998) reported a bivariate odds ratio of 5.0 and a multivariate odds ratio of 12.8, which also indicates that children in foster care were more likely to receive mental health services. For behavior problems, Keller et al. (2001) reported a bivariate unstandardized coefficient of –6.1 and a multivariate unstandardized coefficient of –4.0, which indicates fewer reported behavior problems for children in kinship care. For adaptive behaviors, Keller et al. (2001) reported a bivariate unstandardized coefficient of 3.2 and a multivariate unstandardized coefficient of 2.8, which indicates greater reported competencies for children in kinship care. For reentry, Wulczyn, Hislop, and Harden (2002) reported a bivariate risk ratio of .56 and a multivariate risk ratio of .77, which indicates that children in kinship placements were less likely to reenter out-ofhome care. Needell (1996) reported a bivariate risk ratio of .72 and a multivariate risk ratio of .49, which also indicates that children in kinship care were less likely to reenter out-of-home care after achieving permanency. Overall, the effects were in the same direction and of similar magnitude for all seven studies, regardless of whether bivariate or multivariate data were analyzed. The effects were slightly attenuated in the multivariate analyses for some studies, as demographic and placement characteristics used as covariates in logistic regression or Cox proportional hazards models explained some of the variance in the outcome measures.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 29

Subgroup Analyses There were an insufficient number of studies that reported the necessary data to allow for subgroup analyses on all outcomes in this systematic review. Thus, effect sizes could not be calculated for socioeconomic status, gender, ethnicity, age at first placement, length of placement, caregiver certification, and the different definitions of kinship care (e.g., first placement, last placement). Although many of these demographic and placement characteristics were used as control or predictor variables in the studies that conducted multivariate analyses, the data were not sufficiently disaggregated by placement type to allow for effect size calculation. In addition, there were only a few studies that included an informal kinship care group, so a subgroup analysis on the type of kinship care placement was not possible at this time.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 30

Discussion Although this systematic review and other research syntheses have answered some important questions about kinship care, many new questions have been raised in the process (Cuddeback, 2004). The following discussion summarizes the findings from the review while providing a contextual framework for future conceptual and applied work in this area. From a practice and policy perspective, the implications are most relevant for formal kinship care placements, as data on informal and voluntary kinship care arrangements were not considered in this study. Conclusions The findings from the meta-analyses uncovered some of the underlying trends for the outcomes of children placed with kin. According to the research, children in kinship care experience better outcomes in regard to behavior problems, reentry, adaptive behaviors, family relations, and mental health problems than do children in foster care. However, children placed with kin are less likely to achieve permanency and utilize mental health services. Overall, there are likely no harmful effects of kinship care on child welfare outcomes because all of the effect sizes are small enough in magnitude to have limited practical consequence. The sparse results from the studies that conducted subgroup analyses also support this assertion, in that no child group was adversely impacted by kinship care. In addition, the results should be considered in light of the theoretical and historical support for the kinship care intervention and the relative costeffectiveness of kinship care implementation. Of course, any conclusions drawn from the review must be tempered by the pronounced methodological and design weaknesses of the studies in the kinship care evidence base.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 31

The seemingly negative effect of kinship care on permanency outcomes should be interpreted in the context that long-term kinship care arrangements often satisfy the definition of permanency. Thus, an undesirable outcome (i.e., remaining in care) might actually be desirable if the kinship care placement is considered to be safe and stable. The permanency rates for children in kinship care may be further underestimated because other permanency options (e.g., allocation of parental rights) were not considered by the studies in the evidence base. It is likely that children in foster care exhibit more behavioral and mental health problems and less adaptive behaviors before being placed than do children in kinship care. Thus, the lack of a baseline measurement of initial functioning undercuts the conclusion that children in foster care have lower levels of current functioning. Furthermore, caregiver reports may be biased because foster parents are less aware of potential behavioral and mental health issues, whereas relatives typically know what to expect from children placed in their care. The commonly held notion that foster parents are more “system involved” may explain the greater propensity for children in foster care to receive mental health services. The lower certification and licensure rate for kin caregivers is another factor in the unequal receipt of services for children in kinship care. However, the greater likelihood for children in foster care to utilize mental health services may have less to do with the type of placement and more to do with these children having a greater need for services. As for other interpretative caveats, the meta-analysis for reentry did not consider how age at placement or length of time to reentry might differentially impact children from kinship and foster care placements. It is understandable that kinship caregivers

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 32

report better relationships with their children than do foster parents, as that is the primary rationale for kinship care. The inconclusive findings for the educational attainment outcome may result from the notoriously unreliable measures of high school graduation. Limitations As evidenced by the small number of studies included in the evidence base, the major limitation encountered in this systematic review was the relatively weak standing of quantitative research on kinship care (Cuddeback, 2004). Specifically, the design and methodological flaws in the primarily ex post facto studies led to a lack of confidence regarding the construct validity of the outcomes, the comparability of groups, and the control of contaminating events. The studies also did not account for selection bias, which severely compromises the tenability of the findings from the systematic review. The reliability and validity of the meta-analyses were weakened by challenges confronted during the effect size calculations. Most notably, the heterogeneity statistic was significant for six of the eight outcomes, which indicates that the effect sizes were not always consistent within the same construct. Bivariate data were not reported by some authors, which restricted the meta-analysis of most outcomes to the bare minimum of three studies. Some studies analyzed a small sample of children (e.g., McIntosh, 2002), while others utilized a much larger data set of children in out-of-home care (e.g., Barth, 2002). As a result, studies with large sample sizes essentially eliminated the effects from studies with small sample sizes. Another limitation is the misalignment between the kinship care intervention and available outcomes, in that the fullest representation of its effects has yet to be measured (Cuddeback, 2004). When compared to traditional foster care, in which the relationship

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 33

between foster parents and the “system” is institutionalized, the effect of kinship care on child welfare outcomes may be more difficult to detect. For example, there is a lack of implementation fidelity within and across states in regard to kinship care implementation. This limitation provides support for researching kinship care only after it has been fully and consistently integrated into the fabric of child welfare policy and practice. Recommendations Although the implications of this study depend primarily on how individual counties and states interpret the results, several recommendations for social work practitioners and policymakers did emerge from the systematic review. If the goal of kinship care is to lower reentry rates while improving the behavior, family relations, and mental health of children, then the evidence base is supportive. However, the findings do not support implementing kinship care solely to increase the permanency rates and service utilization of children. Based on the work of other researchers in the field, the primary recommendation for practitioners is to provide appropriate levels of caseworker supervision and service delivery for kin caregivers (Geen, 2000). The main recommendation for policymakers is to consider more appropriate licensing standards (Geen, 2000), while providing adequate financial resources for both formal and informal kin caregivers (Hornby, Zeller, & Karraker, 1996). Future Research To address the major limitations of research on kinship care, Berrick and Barth (1994) call for studies that employ generalizable samples, equivalent groups, and repeated measurements. Cuddeback (2004) advocates for longitudinal designs to

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 34

investigate the outcomes of children over time, and the development of psychometrically sound instruments of family and child functioning that allow for more reliable comparisons across groups and studies. Additionally, Cuddeback (2004) believes that a greater emphasis should be placed on controlling and understanding selection bias through the use of emerging statistical models. There is a need to break out the effects of kinship care across important subgroups of target participants, settings, and intervention variations. For example, there are few studies with the appropriate data collection procedures to reliably measure the effect of kinship care on caregiver outcomes (Gibbs and Muller, 2000). The duration effect or the relationship between length of stay in out-of-home care and child welfare outcomes also should be explored in greater depth. In addition, research on informal and private kinship care arrangements should be a top priority for social work researchers. As for other topics, Testa (1992) calls for research on the financial implications of kin caregivers becoming certified, while Cuddeback (2004) recommends studies that examine the relationship between licensure and the provision of services to kin caregivers. Qualitative research that explores the underlying dynamics of kinship care is a natural outgrowth of this systematic review. Finally, for kinship care to remain a viable option in the social work repertoire, researchers must work more closely with stakeholders to design, implement, and disseminate innovative studies of the intervention.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 35

References Albert, V., Iaci, R., & Catlin, S. N. (2004). Facing time limits and kinship placements. Families in Society: Journal of Contemporary Social Services, 85(1), 63-70. Altshuler, S. J. (1998). Child well-being in kinship foster care: Similar to, or different from, non-related foster care? Children and Youth Services Review, 20, 369-388. Altshuler, S. J. (1999). The well-being of children in kinship foster care. In J. Gleeson & C. F. Hairston (Eds.), Kinship care: Improving practice through research (pp. 117-144). Washington, DC: Child Welfare League of America. Auslander, W. F., McMillen, J. C., Elze, D., Thompson, R., Jonson-Reid, M., & Stiffman, A. (2002). Mental health problems and sexual abuse among adolescents in foster care: Relationship to HIV risk behaviors and intentions. AIDS & Behavior, 6, 351-359. Ayala-Quillen, B. A. (1998). The Adoption and Safe Families Act of 1997 kinship care report: An analysis of key areas. Protecting Children, 14(3), 12-14. Barth, R. P. (1995). The reunification of very young children from foster care. Source, 5(1), 1-2, 15-16. Barth, R. P., Webster, D., & Lee, S. (2002). Adoption of American Indian children: Implications for implementing the Indian Child Welfare and Adoption and Safe Families Acts. Children and Youth Services Review, 24, 139-158. Beeman, S., & Boisen, L. (1999). Child welfare professionals' attitudes toward kinship foster care. Child Welfare, 78, 315-37. Beeman, S., Kim. H., & Bullerdick, S. K. (2000). Factors affecting placement of children in kinship and nonkinship foster care. Children and Youth Services Review, 22, 37-54. Belanger, S. A. (2002). The advisability of kinship foster placements: A comparison of adaptive behaviors and psychopathology of children in traditional and kinship foster care. Dissertation Abstracts International, 62(12), 5993. Benedict, M. I., Zuravin, S., Somerfield, M., & Brandt, D. (1996). The reported health and functioning of children maltreated while in family foster care. Child Abuse & Neglect, 20, 561-571. Benedict, M. I., Zuravin, S., & Stallings, R. Y. (1996). Adult functioning of children who lived in kin versus nonrelative family foster homes. Child Welfare, 75, 529-49.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 36

Berrick, J. D. (1997). Assessing quality of care in kinship and foster family care. Family Relations, 46, 273-280. Berrick, J. D. (1998). When children cannot remain home: Foster family care and kinship care. Future of Children, 8(1), 72-87. Berrick, J. D. (2002). Kinship care: Improving practice through research. Children and Youth Services Review, 24, 135-138. Berrick, J. D., & Barth, R. P. (1994). Research on kinship foster care: What do we know? Where do we go from here? Children and Youth Services Review, 16, 1-5. Berrick, J. D., Barth, R. P., & Needell, B. (1994). A comparison of kinship foster homes and foster family homes: Implications for kinship foster care as family preservation. Children and Youth Services Review, 16, 33-64. Berrick, J. D., Frasch, K., & Fox, A. (2000). Assessing children's experiences of out-ofhome care. Methodological Challenges and Social Work Research, 24, 119-127. Berrick, J. D., & Needell, B. (1999). Recent trends in kinship care: Public policy, payments, and outcomes for children. In P. A. Curtis, G. Dale, & J. C. Kendall (Eds.), The foster care crisis: Translating research into policy and practice (p. 6). Lincoln, NE: University of Nebraska Press. Bilaver, L. A., Jaudes, P. K., Koepke, D., & Goerge, R. M. (1999). The health of children in foster care. Social Service Review, 73, 401-417. Bonecutter, F. J. (1999). Defining best practice in kinship care through research and demonstration. In J. Gleeson & C. F. Hairston (Eds.), Kinship care: Improving practice through research (pp. 37-60). Washington, DC: Child Welfare League of America. Bonecutter, F. J., & Gleeson, J. P. (1997). Broadening our view: Lessons from kinship foster care. Journal of Multicultural Social Work, 5, 99-119. Brandon, P. D. (1995, April). An economic analysis of kin-provided child care. Paper presented at the annual meeting of the Population Association of America, San Francisco, CA. Brinig, M. F., & Nock, S. L. (2002). How much does legal status matter? Adoptions by kin caregivers. Family Law Quarterly, 36, 449-474. Brooks, D., & Barth, R. P. (1998). Characteristics and outcomes of drug-exposed and non drug-exposed children in kinship and non-relative foster care. Children and Youth Services Review, 20, 475-501.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 37

Brooks, S. L. (2002). Kinship and adoption. Adoption Quarterly, 5(3), 55-66. Brown, L. H. (2003). Intergenerational influences on perceptions of current relationships with grandparents. Journal of Intergenerational Relationships, 1(1), 95-112. Buehler, C., Orme, J. G., Post, J., & Patterson, D. A. (2000). The long-term correlates of family foster care. Children and Youth Services Review, 22, 595-625. Burnette, D. (1999). Custodial grandparents in Latino families: Patterns of service use and predictors of unmet needs. Social Work, 44(1), 22-34. Cariglia, N. (1999). “...in the best interest of the child.”: A study of the kinship care program of the Department of Social Services in Fitchburg, Massachusetts. Dissertation Abstracts International, 60(06), 1974. Carlson, G. L., Jr. (2002). An examination of selected variables affecting the outcomes for adult former foster children placed in Baltimore City: The impact of placement stability and type of out-of-home care. Dissertation Abstracts International, 63(01), 378. Carpenter, S. C., Berman, S., Clyman, R., Moore, L. A., & Xu, S. (2004) Are children in kinship care getting the mental health services they need? Pediatric Research, 55, 228A. Carpenter, S. C., & Clyman, R. B. (2004). The long-term emotional and physical wellbeing of women who have lived in kinship care. Children and Youth Services Review, 26, 673-686. Carpenter, S. C., Clyman, R. B., Davidson, A. J., & Steiner, J. F. (2001). The association of foster care or kinship care with adolescent sexual behavior and first pregnancy. Pediatriya, 108(3), E46. Carpenter, S. C., Clyman, R., Moore, L., Xu, S., & Berman, S. (2003). Are children in kinship care at greater risk than those in traditional foster care? Pediatric Research, 53, 193A-194A. Center for Nutrition Policy and Promotion. (1996). Health needs of young children in foster care. Family Economics and Nutrition Review, 9(2), 36-37. Chapman, M. V., Wall, A., & Barth, R. P. (2004). Children's voices: The perceptions of children in foster care. American Journal of Orthopsychiatry, 74, 293-304. Charon, S. L., & Nackerud, L. (1996). Kinship care in Walton County. Human Services in the Rural Environment, 20 (1), 14-18.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 38

Chipungu, S. S., Everett, J. E., Verdieck, M. J., & Jones, H. (1998). Children placed in foster care with relatives: A multi-state study. Executive Summary. Washington, DC: Administration on Children, Youth and Families. Christopher, J. A. (1998). Comparison of foster care teenagers in Kern County: High school graduates vs. dropouts. Masters Abstracts International, 36(06), 1500. Ciliberti, P. (1997). An innovative family preservation and support program in an African-American community: Analysis of six- and twelve-month follow-up data. Dissertation Abstracts International, 58(03), 1090. Clyman, R. B., & Harden, B. J. (2002). Infants in foster and kinship care. Infant Mental Health Journal, 23. 433-434. Clyman, R., Riley, A. W., Lewin, A. B., Messer, S. C., Palmer, B. A., & Altman, B. S. (1998). Service utilization by young children in out-of-home placement. (ERIC Document Reproduction Service No. ED433641) Cohon, D., Hines, L., Cooper, B. A., Packman, W., & Siggins, E. (2003). A preliminary study of an intervention with kin caregivers. Journal of Intergenerational Relationships, 1(3), 49-72. Cole, S. A. (2002). Security of attachment of infants in foster care. Journal of Reproductive and Infant Psychology, 20, 176. Cook, J. A., Boxer, A. M., Burke, J., Cohen, M. H., Weber, K., Shekarloo, P., Lubin, H., & Mock, L. O. (2004). Child care arrangements of children orphaned by HIV/AIDS: The importance of grandparents as kinship caregivers. Journal of HIV/AIDS & Social Services, 2(2), 5-20. Cordero, A. E. (2001). When reunification works: A family strengths perspective. Dissertation Abstracts International, 61(09), 3770. Courtney, M. E. (1994). Factors associated with the reunification of foster children with their families. Social Service Review, 68, 80-108. Courtney, M. E. (1995). Reentry to foster care of children returned to their families. Social Service Review, 69, 226-241. Courtney, M. E. (1996). Kinship foster care and children's welfare: The California experience. Focus, 17(3), 42-48. Courtney, M. E., & Barth, R. P. (1996). Pathways of older adolescents out of foster care: Implications for independent living services. Social Work, 41, 75-83.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 39

Courtney, M. E., & Needell, B. (1997). Outcomes of kinship care: Lessons from California. In J. D. Berrick, R. P. Barth, & N. Gilbert (Eds.), Child Welfare Research Review, Volume II (pp. 130-149). New York, NY, Columbia University Press. Courtney, M. E., Piliavin, I., & Entner Wright, B. R. (1997). Transitions from and returns to out-of-home care. Social Service Review, 71, 652-667. Cuddeback, G. S. (2004). Kinship and family foster care: A methodological substantive synthesis of research. Children and Youth Services Review, 26, 623-639. Cuddeback, G. S., & Orme, J. G. (2002). Training and services for kinship and nonkinship foster families. Child Welfare, 81, 879-909. Danzy, J., & Jackson, S. M. (1997). Family preservation and support services: A missed opportunity for kinship care. Child Welfare, 76(1), 31-44. Davis, I., Landsverk, J., Newton, R., & Ganger, W. (1996). Parental visiting and foster care reunification. Children and Youth Services Review, 18, 363-382. Denby, R. W. (2002). Kinship care: Improving practice through research. Families in Society: The Journal of Contemporary Human Services, 83, 103-105. Dubowitz, H. (1994). Kinship care: Suggestions for future research. Child Welfare, 73, 553-564. Dubowitz, H., Feigelman, S., Harrington, D., Starr, R., Jr., Zuravin, S., & Sawyer, R. (1994). Children in kinship care: How do they fare. Children and Youth Services Review, 16, 85-106. Dubowitz, H., Feigelman, S., & Zuravin, S. (1993). A profile of kinship care. Child Welfare, 72, 153-169. Dubowitz, H., Feigelman, S., Zuravin, S., Tepper, V., Davidson, N., & Lichenstein, R. (1992). The physical health of children in kinship care. American Journal of Diseases of Children, 146, 603-610. Dubowitz, H., & Sawyer, R. J. (1994). School behavior of children in kinship care. Child Abuse & Neglect, 18, 899-911. Dubowitz, H., Zuravin, S., Starr, R. H., Feigelman, S., & Harrington, D. (1993) Behavior problems of children in kinship care. Journal of Developmental and Behavioral Pediatrics, 14, 386-93. Early, T. J., & Mooney, D. D. (2002). Mental health service use for children in foster care in Illinois. Children and Family Research Center: Urbana-Champaign, IL. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 40

Ehrle, J., & Geen, R. (2002). Kin and non-kin foster care: Findings from a national survey. Children and Youth Services Review, 24, 15-35. Ehrle, J., Geen, R., & Clark, R. (2001). Children cared for by relatives: Who are they and how are they faring? (New Federalism: National Survey of America's Families, Series B, No. B-28). Washington, DC: Urban Institute. Epstein, M. H., Jayanthi, M., Dennis, K., Dennis, K. L., Hardy, R., Fueyo, V., Frankenberry, E., & McKelvey, J. (1998). Educational status of children who are receiving services in an urban family preservation and reunification setting. Journal of Emotional and Behavioral Disorders, 6, 162-169. Feigelman, S., Zuravin, S., Dubowitz, H., Harrington, D., Starr, R. H., Jr., & Tepper, V. (1995). Sources of health care and health needs among children in kinship care. Archives of Pediatrics & Adolescent Medicine, 149, 882-886. Fuller-Thomson, E., & Minkler, M. (2000). African American grandparents raising grandchildren: A national profile of demographic and health characteristics. Health and Social Work, 25, 109-118. Gaudin, J. M., Jr., & Sutphen, R. (1993). Foster care vs. extended family care for children of incarcerated mothers. Journal of Offender Rehabilitation, 19, 129-147. Gebel, T. J. (1996). Kinship care and non-relative family foster care: A comparison of caregiver attributes and attitudes. Child Welfare, 75, 5-18. Geen, R. (2000). In the interest of children: Rethinking federal and state policies affecting kinship care. Policy & Practice, 58(1), 19-27. Geen, R., & Berrick, J. D. (2002). Kinship care: An evolving service delivery option. Children and Youth Services Review, 24, 1-14. General Accounting Office. (1999). Foster care: Kinship care quality and permanency issues. Washington, DC: Author. Gennaro, S., York, R., & Dunphy, P. (1998). Vulnerable infants: Kinship care and health. Pediatric Nursing, 24, 119-125. Gibbons, C., & Jones, T. C. (2003). Kinship care: Health profile of grandparents raising their grandchildren. Journal of Family Social Work, 7(1), 1-14. Gibbs, P., & Muller, U. (2000). Kinship foster care moving to the mainstream: Controversy, policy, and outcomes. Adoption Quarterly, 4(2), 57-87.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 41

Gleeson, J. P. (1999). Kinship care as a child welfare service: What do we really know? In J. P. Gleeson & C. F. Hairston (Eds.), Kinship care: Improving practice through research (pp. 3-34). Washington, DC: Child Welfare League of America. Gleeson, J. P. (1999). Who decides? Predicting caseworkers' adoption and guardianship discussions with kinship caregivers. In J. P. Gleeson & C. F. Hairston (Eds.), Kinship care: Improving practice through research (pp. 61-84). Washington, DC: Child Welfare League of America. Gleeson, J. P., & Hairston, C. F. (1999). Future directions for research on kinship care. In J. P. Gleeson & C. F. Hairston (Eds.), Kinship care: Improving practice through research (pp. 281-313). Washington, DC: Child Welfare League of America. Gleeson, J. P., O’Donnell, J., & Bonecutter, F. J. (1997). Understanding the complexity of practice in kinship foster care. Child Welfare, 76, 801-826. Goerge, R. M., Bilaver, L., Lee, B. J., Needell, B., Brookhart, A., & Jackman, W. (2002). Employment outcomes for youth aging out of foster care. Chapin Hall Center for Children: Chicago, IL. Goerge, R., Wulczyn, F., & Fanshel, D. (1994). A foster care research agenda for the '90s. Child Welfare, 73, 525-549. Goodman, C. C., Potts, M., Pasztor, E. M., & Scorzo, D. (2004). Grandmothers as kinship caregivers: Private arrangements compared to public child welfare oversight. Children and Youth Services Review, 26, 287-305. Graff, J. C., Engle, V., & Pruett, J. (2003). Health disparities of custodial and noncustodial black grandparents caring for grandchildren. Gerontologist, 43, 39. Grant, R. (2000). The special needs of children in kinship care. Journal of Gerontological Social Work, 33(3), 17-33. Grinstead, L. N., Leder, S., Jensen, S., & Bond, L. (2003). Review of research on the health of caregiving grandparents. Journal of Advanced Nursing, 44, 318-326. Grogan-Kaylor, A. (2000). Who goes into kinship care? The relationship of child and family characteristics to placement into kinship foster care. Social Work Research, 24, 132-141. Grogan-Kaylor, A. (2001). The effect of initial placement into kinship foster care on reunification from foster care: A bivariate probit analysis. Journal of Social Service Research, 27(4), 1-31.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 42

Harden, A. W., Clark, R. L., & Maguire, K. (1997). Informal and formal kinship care: Volume II: Tables and Figures. Washington, DC: Department of Health and Human Services. Harden, A. W., Wulczyn, F., & Goerge, R. M. (1999). Adoption from foster care: The dynamics of the ASFA foster care population. Chicago, IL Chapin Hall Center for Children at the University of Chicago. Harden, B. J., Clyman, R. B., Kriebel, D. K., & Lyons, M. E. (2004). Kin and kin care: Parental attitudes and resources of foster and relative caregivers. Children and Youth Services Review, 26, 657-671. Harris, L., & Dollinger, S. C. (2002). Social and psychological well-being in custodial, traditional and secondary caregiver grandparents. Gerontologist, 42, 274-275. Harris, M. S. (1998). Factors that affect family reunification of African-American birth mothers and their children placed in kinship care. Dissertation Abstracts International, 58(08), 3308. Hawkins, C. A., & Bland, T. (2002). Program evaluation of the CREST project: Empirical support for kinship care as an effective approach to permanency planning. Child Welfare, 81, 271-292. Heflinger, C. A., Simpkins, C. G., & Combs-Orme, T. (2000). Using the CBCL to determine the clinical status of children in state custody. Children and Youth Services Review, 22, 55-73. Hornby, H., Zeller, D., & Karraker, D. (1996). Kinship care in America: What outcomes should policy seek? Child Welfare, 75, 397-418. Hunter, W. (1998). LONGSCAN Research Briefs, Volume 1, Summer 1998. National Center on Child Abuse and Neglect. (ERIC Document Reproduction Service No. ED460308) Iglehart, A. P. (1994). Kinship foster care: Placement, service, and outcome issues. Children and Youth Services Review, 15, 107-122. Iglehart, A. P. (1995). Readiness for independence: Comparison of foster care, kinship care, and non-foster care adolescents. Children and Youth Services Review, 17, 417-32. Jenkins, M. H. (2002). Quality of care study of 76 kinship and 105 non-kinship foster children. Dissertation Abstracts International, 63(04), 1557. Johnson, E. I., & Waldfogel, J. (2002). Parental incarceration: Recent trends and implications for child welfare. Social Service Review, 76, 460-479. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 43

Johnson-Garner, M., & Meyers, S. A. (2003). What factors contribute to the resilience of African-American children within kinship care? Child and Youth Care Forum, 32, 255-269. Jones, J. (1998). Functioning and adjustment of children in kinship care versus nonrelative foster family care placements. Dissertation Abstracts International, 59(09), 5089. Jones, M. R., & Hansen, C. (1996). Caregiving behaviors which predict adjustment of children raised by grandparents. (ERIC Document Reproduction Service No. ED409986) Keller, T. E., Wetherbee, K., Le Prohn, N. S., Payne, V., Sim, K., & Lamont, E. R. (2001). Competencies and problem behaviors of children in family foster care: Variations by kinship placement status and race. Children and Youth Services Review, 23, 915-940. Kelley, S. J., Whitley, D., Sipe, T. A., & Yorker, B. C. (2000). Psychological distress in grandmother kinship care providers: The role of resources, social support, and physical health. Child Abuse & Neglect, 24, 311-21. Kemp, S. P., & Bodonyi, J. M. (2002). Beyond termination: Length of stay and predictors of permanency for legally free children. Child Welfare, 81, 58-86. Kolomer, S. R., McCallion, P., & Janicki, M. P. (2002). African-American grandmother carers of children with disabilities: Predictors of depressive symptoms. Journal of Gerontological Social Work, 37(3/4), 45-63. Kontos, S., Howes, C., Shinn, M., & Galinsky, E. (1997). Children's experiences in family child care and relative care as a function of family income and ethnicity. Merrill - Palmer Quarterly, 43, 386-403. Landry-Meyer, L., Guzell, J., & Gerard, J. (2002). Grandparent caregivers: Social support and life satisfaction. Gerontologist, 42, 53. Landsverk, J., Davis, I., Ganger, W., Newton, R., & Johnson, I. (1996). Impact of child psychosocial functioning on reunification from out-of-home placement. Children and Youth Services Review, 18, 447-462. Leos-Urbel, J., Bess, R., & Geen, R. (2002). The evolution of federal and state policies for assessing and supporting kinship caregivers. Children and Youth Services Review, 24, 37-52. Leslie, L. K., Gordon, J. N., Ganger, W., & Gist, K. (2002). Developmental delay in young children in child welfare by initial placement type. Infant Mental Health Journal, 23, 496-516. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 44

Leslie, L. K., Landsverk, J., Ezzet-Lofstrom, Tschann, J. M., Slymen, D. J., & Garland, A. F. (2000). Children in foster care: Factors influencing outpatient mental health service use. Child Abuse & Neglect, 24, 465-476. Leslie, L. K., Landsverk, J., Horton, M. B., Ganger, W., & Newton, R. R. (2000). The heterogeneity of children and their experiences in kinship care. Child Welfare, 79, 315-34. Lewis, M., Giovannoni, J. M., & Leake, B. (1997). Ethnic variations in the two-year living arrangements of prenatally drug-exposed and comparison children placed at birth. Journal of Multicultural Social Work, 6, 17-40. Link, M. K. (1996). Permanency outcomes in kinship care: A study of children placed in kinship care in Erie County, New York. Child Welfare, 75, 509-28. Lipsey, M., & Wilson, D. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage Publications. Litrownik, A. J., Newton, R., Mitchell, B. E., & Richardson, K. K. (2003). Long-term follow-up of young children placed in foster care: Subsequent placements and exposure to family violence. Journal of Family Violence, 18(1), 19-28. Magruder, J. (1994). Characteristics of relative and non-relative adoptions by California public adoption agencies. Children and Youth Services Review, 16, 123-131. Mallon, G. P., Aledort, N., & Ferrera, M. (2002). There's no place like home: Achieving safety, permanency, and well-being for lesbian and gay adolescents in out-ofhome care settings. Child Welfare, 81, 407-439. Martin, M. H., Barbee, A. P., Antle, B. F., & Sar, B. (2002). Expedited permanency planning: Evaluation of the Kentucky adoptions opportunities project. Child Welfare, 81, 203-224. Mcintosh, M. M. (2002). Barriers to reunification in the child welfare system: An analysis of kinship and non-kinship placements. Masters Abstracts International, 40(06), 1423. McLean, B., & Thomas, R. (1996). Informal and formal kinship care populations: A study in contrasts. Child Welfare, 75, 489-505. McMillen, J. C., Scott, L. D., Zima, B. T., Ollie, M. T., Munson, M. R., & Spitznagel, E. (2004). Use of mental health services among older youths in foster care. Psychiatric Services, 55, 811-817. McNichol, T. (1999). The impact of drug-exposed children on family foster care. Child Welfare, 78, 184-96. ________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 45

McNichol, T., & Tash, C. (2001). Parental substance abuse and the development of children in family foster care. Child Welfare, 80, 239-56. Metzger, J. W. (1997). The role of social support in mediating the well-being of children placed in kinship foster care and traditional foster care. Dissertation Abstracts International, 58(06), 2394. Mills, C. S., & Usher, D. (1996). A kinship care case management approach. Child Welfare, 75, 600-618. Miltenberger, P., Hayslip, B., Jr., Harris, B., & Baird, A. (2002). Perceptions of custodial grandparents. Gerontologist, 42, 274. Musil, C. M. (2000). Health of grandmothers as caregivers: A ten month follow-up. Journal of Women and Aging, 12, 129-145. Needell, B. (1996). Placement stability and permanence for children entering foster care as infants. Dissertation Abstracts International, 58(02), 587. Nichols, C. A. (2002). Providing foster care: A study of the challenges for regular and kinship foster parents and the child welfare community. Dissertation Abstracts International, 63(05), 1998. O'Donnell, J. M. (2001). Paternal involvement in kinship foster care services in one father and multiple father families. Child Welfare, 80, 453-79. Patton, N. (2003). The effects of parental drug use: Children in kinship care: A review of the literature. St Kilda South, Victoria: Mirabel Foundation. Petras, D. D. (1999). The effect of caregiver preparation and sense of control on adaptation of kinship caregivers. In J. P. Gleeson & C. F. Hairston (Eds.), Kinship care: Improving practice through research (pp. 233-256). Washington, DC: Child Welfare League of America. Potter, C. C., & Klein-Rothschild, S. (2002). Getting home on time: Predicting timely permanence for young children. Child Welfare, 81, 123-150. Prosser, W. R. (1997). Family structure, substitute care, and educational achievement (Discussion Paper No. 1140-97). Madison, WI: Institute for Research on Poverty at the University of Wisconsin-Madison. Rhodes, J., Haight, W. L., & Briggs, E. C. (1999). The influence of mentoring on the peer relationships of foster youth in relative and nonrelative care. Journal of Research on Adolescence, 9, 185-201.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 46

Rittner, B. (2002). The use of risk assessment instruments in child protective services case planning and closures. Children and Youth Services Review, 24, 189-207. Rittner, B., & Dozier, C. D. (2000). Effects of court-ordered substance abuse treatment in child protective services cases. Social Work, 45, 131-140. Roberts, D. E. (2001). Kinship care and the price of state support for children. Chicago Kent Law Review, 76, 1619-1642. Rozario, P. (2003). Social structural impacts on family caregiver well-being: Macro-level influences on micro-level outcomes. Dissertation Abstracts International, 63(10), 3733. Rubin, D. M., Alessandrini, E. A., Feudtner, C., Mandell, D. S., Localio, A. R., & Hadley, T. (2004). Placement stability and mental health costs for children in foster care. Pediatrics, 113, 1336-1341. Ruiz, D. S., Zhu, C. W., & Crowther, M. R. (2003). Not on their own again: Psychological, social, and health characteristics of custodial African American grandmothers. Journal of Women & Aging, 15, 167-184. Sainer, J. S., Hayslip, B., Jr., & Cantor, M. (2003). Strengthening grandparents and other kinship care families through innovations in research and practice. Gerontologist, 43, 290-291. Sands, R. G., & Goldberg-Glen, R. S. (2000). Factors associated with stress among grandparents raising their grandchildren. Family Relations, 49(1), 97-105. Sands, R. G., & Goldberg-Glen, R. S. (2000). Grandparent caregivers' perception of the stress of surrogate parenting. Journal of Social Service Research, 26, 77-95. Sawyer, R. J., & Dubowitz, H. (1994). School performance of children in kinship care. Child Abuse & Neglect, 18, 587-97. Scannapieco, M., & Hegar, R. L. (1996). A nontraditional assessment framework for formal kinship homes. Child Welfare, 75, 567-582. Scannapieco, M., & Hegar, R. L. (2002). Kinship care providers: Designing an array of supportive services. Child and Adolescent Social Work Journal, 19, 315-327. Scannapieco, M., Hegar, R. L., & McAlpine, C. (1997). Kinship care and foster care: A comparison of characteristics and outcomes. Families in Society, 78, 480-488. Shin, H. S. (2003a). Building evidence to promote educational competence of youth in foster care. Child Welfare, 82, 615-632.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 47

Shin, H. S. (2003b). The well-being and reading achievement of older youth in out-ofhome care. Dissertation Abstracts International, 63(11), 4096. Shlonsky, A., Webster, D., & Needell, B. (2003). The ties that bind: A cross-sectional analysis of siblings in foster care. Journal of Social Service Research, 29, 27-52. Shore, N., Sim, K. E., Le Prohn, N. S., & Keller, T. E. (2002). Foster parent and teacher assessments of youth in kinship and non-kinship foster care placements: Are behaviors perceived differently across settings? Children and Youth Services Review, 24, 109-134. Smith, B. D. (2003). After parental rights are terminated: Factors associated with exiting foster care. Children and Youth Services Review, 25, 965-985. Smith, C. J., Rudolph, C., & Swords, P. (2002). Kinship care: Issues in permanency planning. Children and Youth Services Review, 24, 175-188. Smithgall, C., Gladden, R. M., Howard, E., Goerge, R., & Courtney, M. (2004). Educational experiences of children in out-of-home care. Chicago, IL: Chapin Hall Center for Children at the University of Chicago. Stolley, K. S. (1993). Statistics on adoption in the United States. Future of Children, 3(1), 26-42. Strozier, A. L., Elrod, B., Beiler, P., Smith, A., & Carter, K. (2004). Developing a network of support for relative caregivers. Children and Youth Services Review, 26, 641-656. Surbeck, B. (2000). The social environment's impact on the functioning of children in kinship and nonkinship care. Dissertation Abstracts International, 61(05), 2042. Sykes, J., Sinclair, I., Gibbs, I., & Wilson, K. (2002). Kinship and stranger foster carers: How do they compare? Adoption and Fostering, 26(2), 38-48. Szinovacz, M. E., DeViney, S., & Atkinson, M. P. (1999). Effects of surrogate parenting on grandparents' well-being. Journals of Gerontology. 54, S376-S388. Talbot, C., & Williams, M. (2003). Kinship care. Family Law, 33, 502-507. Terling-Watt, T. (2001). Permanency in kinship care: An exploration of disruption rates and factors associated with placement disruption. Children and Youth Services Review, 23, 111-126. Testa, M. F. (1992). Conditions of risk for substitute care. Children and Youth Services Review, 14, 27-36.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 48

Testa, M. F. (2001). Kinship care and permanency. Journal of Social Service Research, 28(1), 25-43. Testa, M. F. (2002). Subsidized guardianship: Testing an idea whose time has finally come. Social Work Research, 26, 145-158. Testa, M. F., & Rolock, N. (1999). Professional foster care: A future worth pursuing? Child Welfare, 78, 108-124. Testa, M. F., & Slack, K. S. (2002). The gift of kinship foster care. Children and Youth Services Review, 24, 79-108. Testa, M. F., Shook, K. L., Cohen, L. S., & Woods, M. G. (1996). Permanency planning options for children in formal kinship care. Child Welfare, 75, 451-70. Thorpe, R. (2002, September). Examining the evidence in out-of-home care. What Works? Evidence based practice in child and family services. Paper presented at the annual meeting of ACWA, Bondi Beach, New South Wales. Tompkins, W., Jr. (2003). Mental and physical health and educational performance outcomes among children in three out-of-home placement types. Dissertation Abstracts International, 63(12), 4477. Trafton, P. (1998). Parenting grandchildren: effects on developmental tasks of later adulthood. Progress: Family Systems Research and Therapy, 7, 79-88. Vogel, C. A. (1999). Using administrative databases to examine factors affecting length of stay in substitute care. Children and Youth Services Review, 21, 677-690. Waldfogel, J. (2000). Child welfare research: How adequate are the data? Children and Youth Services Review, 22, 705-741. Webster, D., Barth, R. P., & Needell, B. (2000). Placement stability for children in outof-home care: A longitudinal analysis. Child Welfare, 79, 614-32. Wells, K., & Guo, S. (1999). Reunification and reentry of foster children. Children and Youth Services Review, 21, 273-294. Whitley, D. M., Kelley, S. J., & Sipe, T. A. (2001). Grandmothers raising grandchildren: Are they at increased risk of health problems? Health & Social Work, 26, 105114. Wilson, L., & Conroy, J. (1999). Satisfaction of children in out-of-home care. Child Welfare, 78, 53-69.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 49

Wulczyn, F. (2003). Closing the gap: Are changing exit patterns reducing the time African American children spend in foster care relative to Caucasian children? Children and Youth Services Review, 25, 431-462. Wulczyn, F. H., & Goerge, R. M. (1992). Foster care in New York and Illinois: The challenge of rapid change. Social Service Review, 66, 278-294. Wulczyn, F. H., Hislop, K. B., & Goerge, R. M. (2000). An update from the Multistate Foster Care Data Archive: Foster care dynamics 1983-1998. Alabama, California, Illinois, Iowa, Maryland, Michigan, Missouri, New Jersey, New York, North Carolina, Ohio, and Wisconsin. Chicago, IL: Chapin Hall Center for Children at the University of Chicago. Wulczyn, F., Hislop, K. B., & Harden, B. J. (2002). The placement of infants in foster care. Infant Mental Health Journal, 23, 454-475. Zima, B. T., Bussing, R., Freeman, S., Yang, X., Belin, T. R., & Forness, S. R. (2000). Behavior problems, academic skill delays and school failure among school-aged children in foster care: Their relationship to placement characteristics. Journal of Child and Family Studies, 9, 87-103. Zima, B. T., Bussing, R., Yang, X., & Belin, T. R. (2000). Help-seeking steps and service use for children in foster care. Journal of Behavioral Health Services & Research, 27, 271-285. Zlotnick, C., Robertson, M. J., & Wright, M. A. (1999). The impact of childhood foster care and other out-of-home placement on homeless women and their children. Child Abuse & Neglect, 23, 1057-1068. Zuravin, S. J., Benedict, M., & Somerfield, M. (1993). Child maltreatment in family foster care. American Journal of Orthopsychiatry, 63, 589-596.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 50

Appendix A Data Extraction Guidelines Section A: Rationale and research question(s) or hypotheses of the study A.1 What are the broad aims of the study?

A.1.1 Explicitly stated (please specify)

Please write in authors’ description if there is one. Look for a purpose statement or a problem statement. Do not give specific questions about the research questions or hypotheses here.

A.1.2 Implicit (please specify)

A.2 What are the study research questions and/or hypotheses?

A.2.1 Yes, questions or hypotheses are given (please specify all)

Research questions or hypotheses operationalize the aims of the study. Please write in authors’ exact questions or hypotheses or both.

A.3 With what level of clarity and thoroughness was the study informed by, or linked to an existing body of empirical and/or theoretical research? Look at the introduction and examine the logic and flow of the rationale of the study. Check how recent and extensive the quantity and quality of the references are, and how compelling the case is made for examining the research questions and hypotheses for this study.

A.1.3 Not stated/unclear (please specify)

A.2.2 Yes, questions or hypotheses are given, but they are unclear (please describe) A.2.3 No, questions or hypotheses are not given by the authors of the study

A.3.1 High quality – probably reflective of the best evidence criterion A.3.2 Medium quality – clearly satisfactory but lacking on one or more dimensions of clarity, thoroughness, and quality of literature cited as the rationale for the study A.3.3 Low quality – achieves minimum standards, but clearly lacking on most dimensions of clarity, thoroughness, and literature cited as the rationale for the study A.3.4 Unacceptable – does not meet even the most fundamental standards of clarity, thoroughness, and literature cited as the rationale for the study

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 51

Section B: Description and DIAD Rating of the Intervention(s) This section assesses information about the nature of the intervention(s) and ends with a judgment of DIAD Composite Question #1 about the construct validity of the intervention description(s). B.1 How was placement in kinship care defined for participants in this study?

B.1.1 Kinship care was only out-of-home placement B.1.2. Kinship care was first out-of-home placement B.1.3 Kinship care was last out-of-home placement B.1.4 Kinship care was majority of out-of-home placement (please specify other placement type) B.1.5 Other (please specify) B.1.6 Unlcear/Not Stated

B.2 How were children placed into kinship care in this study?

B.2.1 Formal placement with kin certification B.2.2 Formal placement without kin certification B.2.3 Informal placement with state involvement B.2.4. Other (please specify) B.2.5. Unclear/Not Stated

B.3 What is the mean length of placement in kinship care for participants in this study?

B.3.1 Please specify B.3.2 Unclear/Not Stated

B.4 What services/training were caregivers receiving during the kinship placement in this study?

B.4.1 Please specify B.4.2 Unclear/Not Stated

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 52

B.5 How would you rate the alignment of the intervention to commonly-held ideas of the approach?

What level of description did the author(s) of the study provide such that the intervention or approach that was implemented in the study adhered to commonly-held or theoretically derived ideas of what the intervention or approach should be?

B.5.1 Yes - The intervention was adequately described and it fully reflected commonly-held or theoretically derived ideas about what the intervention should be. B.5.2 Maybe yes - At a minimum the intervention was adequately described, and it at least somewhat reflected commonly-held or theoretically derived ideas about what the intervention should be. B.5.3 Maybe no - The intervention was described only as a member of broader classes (across which significant variation in content can be expected). B.5.4 No - It is unclear what the intervention was, OR the intervention did not reflect commonly-held or theoretical ideas about what it should be.

B.6 How would you rate the implementation and replicability of the intervention?

What level of description did the author(s) of the study provide such that the intervention that was implemented in the study would be replicable by others and was implemented consistently with its described characteristics?

B.6.1 Yes - The intervention was sufficiently described at a level which would allow relatively easy and thorough replication by other implementers, and the description of the implementation of the intervention was fully consistent with its defined characteristics. B.6.2 Maybe yes - The intervention was adequately described to allow replication of the most essential elements by other implementers, and the description of the implementation was largely consistent with its defined characteristics. B.6.3 Maybe no - The authors omit important descriptive information concerning the essential elements of the intervention such that its replication would be impossible, OR it is plausible that the implementation of the intervention may well have been inconsistent with its defined characteristics. B.6.4 No -The authors of the study omit important descriptive information concerning the essential elements of the intervention such that its replication would be impossible, AND it is plausible that the implementation of the intervention may well have been inconsistent with its defined characteristics.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 53

Section C: Description and DIAD Rating of the Outcome(s) This section assesses information about the nature of the outcome(s) of the study and ends with DIAD Composite Question #2 about the construct validity of the outcome(s) description(s). C.1 For whom were the outcomes of the intervention being researched in the study?

C.1.1 Child outcomes C.1.2 Kin outcomes

If there was more than one outcome, answer the following questions for each.

C.1.3 Parent outcomes

C.2. What is the conceptual name for the outcome construct that is the focus of this study?

C.2.1. Please specify

C.3. What is the measure that will be used in the calculation of the effect size for this study?

C.3.1. Please specify

C.4 Did the study identify the actual name(s) of the instrument used to measure the outcome(s) selected for this study?

C.4.1 Yes (please specify)

C.5 Does the author describe any methods that have been used to address the validity of data collection tools?

C.5.1 Yes (please specify)

For example, the direct mention of criterion related or construct validity

C.5.3 Unclear (please specify)

C.6 How would you rate the adequacy with which the validity of the outcome measure(s) were defined?

C.6.1 Yes - The study provided adequate evidence that the outcome measure was properly defined and was appropriate for the context of the study.

With what level of precision and clarity did the author(s) of the study describe the outcome measure(s) used in the study? Was there evidence that the outcome measure was aligned to the intervention?

C.4.2 No - not stated by author(s)

C.5.2 No – not stated by author(s)

C.6.2 Maybe yes - Although the study did not present adequate evidence that the outcome measure was properly defined, the measure did appear to be appropriate to the content of the outcome and the context of the study. C.6.3 Maybe no - The outcome and/or the measure used to assess the outcome were only described conceptually as a member of a broader class of outcomes/measures about which significant variation exists as to their specific content. C.6.4 No – It is unclear what the outcome is and how it was measured.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 54

Section D: Description and DIAD Rating of the Selection Procedures This section assesses information about the procedures used to select participants for the study, including issues associated with assignment of participants to groups, attrition, and pre- and postequating of participants across groups. This section ends with an assessment of DIAD Composite Question #3 about the internal validity of the selection procedures. D.1 Which of the following design purposes was the focus of this study?

D.1.1 To compare formal kinship care with nonfamily foster care D.1.2 To compare formal kinship care with other out-of-home placements (please specify) D.1.3 To compare formal kinship care with general population D.1.4 To compare formal kinship care with informal kinship care D.1.5 To compare single group in formal kinship care (i.e., within-group design)

D.2 Was any information provided on the sampling frame for this study?

D.2.1 Yes (please specify) D.2.2 Not Stated

How did the researchers put together the list of sampling units (i.e. children, kin, parents) from which they drew the actual sample for this study?

D.2.3 Unclear (please specify)

For example, was there random selection of participants from a larger sampling frame? D.3 What was the unit of allocation into each intervention and control/ comparison group?

D.3.1 Children D.3.2 Kin D.3.3 Parents D.3.4 Unclear/Not Stated

D.4 What was the attrition rate?

D.4.1 Reported for study population as a whole (please specify)

If not reported, subtract the final sample size (time of measurement) from the initial sample size (time of assignment) and divide by initial sample size to determine attrition rate.

D.4.2 Reported for one/some group(s) (please specify) D.4.3 Reported for all groups (please specify) D.4.4 Unclear/Not stated

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 55

D.5 Was any information provided on those who dropped out of the study?

D.5.1 Yes (please specify) D.5.2 Not Stated D.5.3 Unclear (please specify) D.5.4 Not applicable

D.6 How would you rate the adequacy with which participants in the comparison or alternative treatment group(s) were made comparable to those in the treatment group?

D.6.1 Yes – Participants were randomly assigned to conditions, and there does not appear to have been any serious differential attrition within groups or severe overall attrition across groups or within subjects when participants serve as their own controls.

This is the fundamental issue of internal validity of selection of participants.

D.6.2 Maybe yes - EITHER randomized assignment was used but there appears to have been serious differential attrition within groups or serious overall attrition across groups or within subjects, OR although random assignment was not used, there does not appear to have been serious attrition problems within or across groups or within subjects and reasonable attempts were made to make the groups comparable (i.e. matched sampling, use of a covariate). D.6.3 Maybe no - Randomized assignment was not used and despite some steps taken to make the groups comparable, they do not appear to have been adequate. D.6.4 No - It is unlikely or unknown if the participants in the groups are comparable.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 56

Section E: Description and DIAD Rating of Contamination This section assesses the steps used by the study’s authors to minimize the probability that events alternative to the intervention in the research context could be responsible for the effects measured in the study. This section ends with an assessment of DIAD Composite Question #4 about the internal validity of controls over alternative events. E.1 What is the research design associated with this study?

E.1.1 Pretest-posttest control group design E.1.2 Posttest only control group design E.1.3 Pretest-posttest non-equivalent comparison group design E.1.4 Posttest only non-equivalent comparison group design E.1.5 Pretest-posttest single group design E.1.6 Posttest only single group design E.1.7 Other (e.g., regression discontinuity, interrupted time series)

E.2 Was outcome measurement done in such a way that those conducting the measurement were unaware of group membership (in the reviewer's judgment)?

E.2.1 Yes E.2.2 No E.2.3 Unclear

Were those assessing the outcomes unaware whether the participant had been in the control/comparison or intervention group?

E.3 How did the different groups compare to one another?

E.2.4 Not stated E.2.5 Not applicable (within-subjects design) E.3.1 Equivalent E.3.2 Non-equivalent

Were the groups equivalent at baseline or was incomparability addressed by the study authors? Groups are likely to be equivalent if they were drawn from the same sample and have similar demographic variables and pre-test outcome measures. Groups are likely to be non-equivalent if they were drawn from an external population or archival/historical sample.

E.3.3 Unclear E.3.4 Not stated E.3.5 Not applicable (within-subjects design)

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 57

E.4 How would you rate the adequacy with which the study controlled events that happened concurrently with the intervention or approach that might have confused its effect(s)?

Specifically, was there evidence of a changed expectancy/novelty/disruption, a local history event, or any other intervention contaminants?

E.4.1 Yes - Concurrent processes and events that might be alternative explanations to a treatment effect have been ruled out, either explicitly or implicitly. E.4.2 Maybe yes - There were no identified processes or events that could be alternative explanations for a treatment effect, but some alternative explanations cannot be explicitly ruled out either because there was some evidence that alternative explanations might exist, or because no attention was given to ruling out an alternative explanation and it is reasonable to expect that one or more alternative explanations might exist. E.4.3 No - Identifiable processes or events that are described to be occurring simultaneously with the treatment or approach may have caused the observed effect.

Section F: Description and DIAD Rating of Sampling External Validity This section assesses how representative the actual participants, settings, outcomes, and data collection activities were to the theoretical population, school settings, typical measures, and appropriate measurement processes. This section ends with an assessment of DIAD Composite Question #5 about the external validity of sampling of participants, settings, outcomes, and measurement occasions. F.1 Which method(s) was/were used to collect the data for this study (if this is a secondary analysis of an existing dataset, please describe the method(s) used in the original data collection process)?

Please indicate all that apply and give further detail where possible.

F.1.1 Case records (state/county) F.1.2 Survey/Questionnaire F.1.3 Interview F.1.4 Standardized assessment F.1.5 School records (e.g., grades, graduation) F.1.6 Observations F.1.7 Other (please specify) F.1.8 Unclear/Not stated

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 58

F.2 What was the timing of measurement for the treatment and comparison groups in the study?

F.2.1 Cross-sectional: Data were collected at the same point in time for both groups F.2.2 Longitudinal: Data were collected over the same period of time for both groups F.2.3 Cohort Comparison/Historical Control: Data were collected at a different point in time for both groups F.2.4 Unclear/Not Stated

F.3 How long a period before the intervention was the final pre-intervention measurement made?

F.3.1 Please specify F.3.2 None/Not Applicable F.3.3 Unclear/Not stated

F.4 How long a period after the intervention was the first post-intervention measurement made?

F.4.1 Please specify F.4.2 None/Not Applicable F.4.3 Unclear/Not stated

F.5 How would you rate the adequacy with which the actual sample, setting, outcome(s), and measurement processes reflected the theoretical population and typical norms for settings, outcomes, and measurement processes?

F.5.1 Yes - The actual sample generalizes well to the theoretical population and the setting, outcome(s) and measurement processes generalize well to common variations in settings, classes of outcome(s), and processes and timing of data collection. F.5.2 Maybe yes - Most aspects of the theoretical population and common variations of settings, classes of outcomes, and data collection processes and timing are represented in the study. F.5.3 Maybe no - Although some important characteristics of the theoretical population and typical settings, outcomes, and data collection processes and timing are represented by the study, many important characteristics are not. F.5.4 No - The actual sample does not adequately reflect any characteristics of the theoretical population, and the setting, outcomes, and data collection timing and processes have characteristics that are not within the boundaries of accepted and typical practice.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 59

Section G: Description and DIAD Rating of Testing within Sub-groups External Validity This section assesses how broadly the intervention was tested across important sub-groups of students, and across substantive variations within the intervention as a whole. This section ends with an assessment of DIAD Composite Question #6 about the external validity of testing within sub-groups. G.1 What is the mean age of the participants in this study?

G.1.1 Children (please specify age at placement) G.1.2. Kin (please specify) G.1.3. Parents (please specify) G.1.4 Unclear/Not Stated

G.2 What is the stated percentage(s) for each of the ethnic classifications of the participants in this study?

G.2.1 White/European G.2.2 Black/African American G.2.3 Native American/American Indian

If available, use percentage calculations by the authors of the study. If feasible and not reported by the authors, calculate these percentages yourself, but specify that you calculated these percentages and not the study authors

G.2.4 Asian

G.3 What is the predominant gender of the participants of this study?

G.3.1 Please specify

G.2.5 Hispanic/Latino G.2.6 Unclear/Not Stated

G.3.2 Unclear/Not Stated State in terms of proportion of females in study

G.4 What is the socio-economic composition of the participants in this study?

G.4.1 Low income (please specify) G.4.2. Middle income (please specify)

Eligibility for AFDC is a common indicator of SES for children in kinship care

G.4.3. High income (please specify) G.4.4 Unclear/Not Stated

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 60

G.5 How broadly was the intervention analyzed across important sub-groups of students, and across substantive variations within the intervention as a whole?

G.5.1 Yes - The analyses in the study examined the effect(s) of the intervention across important subgroups of students AND included separate analyses of key sub-components of the intervention for differential effectiveness on those different subgroups of students G.5.2 Maybe yes - Some sub-group analyses were conducted AND some estimates were made exploring differential effects of different intervention components. G.5.3 Maybe no - Some sub-group analyses were conducted OR some estimates were made exploring differential effects of different intervention components. However, significant sub-groups were omitted from the analyses, and no separate effects of different intervention components by sub-groups were explored. G.5.4 No - Only main effects for the intervention as a whole were reported with no sub-group or intervention component analyses.

Section H: Description and DIAD Rating of the Statistical Validity of the Data This section assesses thoroughness with which the statistical properties of the data were reported, including how well the data satisfied important assumptions underlying the analytic techniques that were used, and how reliable the instruments were for measuring the outcome(s) of the study. This section ends with an assessment of DIAD Composite Question #7 about statistical validity of measurement. H.1 Which statistical techniques were used to analyze the data?

H.1.1 Please specify H.1.2 Unclear (please specify) H.1.3 Not stated

H.2 Does the author describe any technical information about the reliability of instruments used to measure the outcome(s) of the study (e.g. internal consistency, test – retest, interrater reliability)?

H.2.1 Yes (please specify) H.2.2 No H.2.3 Unclear (please specify)

Where more than one approach was used, please provide details for each.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 61

H.3 What is the unit of data analysis

Were the results reported according to the unit of allocation? For example, if individuals were allocated to different groups, results from individuals should be analyzed and reported.

H.3.1 Same as unit of allocation (please specify) H.3.2 Different from unit of allocation (please specify) H.3.3 Unclear (please specify) H.3.4 Not stated

H.4 Are there any obvious shortcomings in the numerical reporting in this study?

H.4.1 Yes (please specify) H.4.2 No H.4.3. Unclear/Not Stated

H.5 How thoroughly were the assumptions underlying the statistical analyses for the study reported and how reliable does the outcome measurement instrumentation appear to be?

H.5.1 Yes - Analyses were conducted and reported for all important assumptions underlying the statistics for the study and important instrument reliability estimates are given. H.5.2 Maybe yes - Some reporting of the statistical properties of the data are provided as are some estimates of instrument reliability.

Most importantly, are students statistically independent (i.e., the outcomes for some participants in a group are unrelated to the outcomes of others in that group)?

H.5.3 Maybe no - Key information about how well the data met important assumptions the statistical analyses are omitted OR no reliability information is given for the instruments used in the study. H.5.4 No - Neither reliability information nor descriptions of how well the data met important assumptions underlying the statistics used in the study were reported.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 62

Section I: Description and DIAD Rating of the Effect Size Calculation This section assesses the adequacy with which the study reports effect size estimates or the data needed to calculate effect sizes. This section ends with an assessment of DIAD Composite Question #8 about completeness of data reporting. I.1 How many participants were in this study?

I.1.1 Please specify

State the final number of participants who were included in the analyses. For multi-group studies, state total number of experimental and control participants.

I.1.2 Unclear

I.2 Was the sample large enough for sufficiently precise estimates of effects?

I.2.1 Yes

I.1.3 Not Stated

I.2.2 No I.2.3 Unclear I.3 What is/are the effect size(s) for the intervention in this study?

I.3.1 Reported by the author (please specify) I.3.2 Calculated by the reviewer (please specify) I.3.3 Unclear (please specify)

Specify effect size(s) for each outcome in the study.

I.3.4 Not Stated I.3.5 Not applicable

I.4 How adequately were the results of statistical tests reported such that effect sizes for all important outcomes in the study are available?

I.4.1 Yes - All important outcomes either have effect sizes reported by the authors or provide data to allow precise calculation of effect sizes. I.4.2 Maybe yes - Sufficient statistical information was reported to allow, at a minimum, imprecise effect sizes to be calculated for most measured outcomes. I.4.3 Maybe no - For most outcomes, effect sizes were not reported, nor is there adequate statistical information to allow effect sizes to be calculated with precision for most outcomes. I.4.4 No - Neither sample sizes nor effect sizes were reported, OR insufficient data were provided to allow those effect sizes to be calculated.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 63

Section J: Inclusionary Criteria This section assesses whether the study meets all DIAD standards for inclusion in the systematic review. J.1 Does the study meet all DIAD standards for inclusion in the systematic review?

J.1.1 Yes J.1.2 No

J.2 If the study does not meet all DIAD standards for inclusion, please indicate which standard(s) was not met.

J.2.1 Intervention specification (Section B) J.2.2 Outcome specification (Section C) J.2.3 Group comparison (Section D) J.2.4 Contamination (Section E) J.2.5 Sampling Validity (Section F) J.2.6 Statistical Validity (Section H) J.2.7 Effect Size Calculation (Section I)

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 64

Appendix B Studies in the Kinship Care Evidence Base Author and Publication Date

Barth, Webster, & Lee (2002)

Belanger (2001)

Benedict, Zuravin, Somerfield, & Brandt (1996)

Benedict, Zuravin, & Stallings (1996)

Berrick (1997)

Berrick & Needell (1999)

Sample

Treatment

Children less than 6 years old who entered out-of-home care in California for the first time between Jan. 1988 and Dec. 1992

Kinship care was majority of out-of-home placement

Children in NYC referred to the Jewish Child Care Association for placement from 1999 to 2001

Kinship care was first placement

Children from all substantiated maltreatment reports in 1984-1988 for the Baltimore City Department of Social Services

Kinship care was only outof-home placement

Outcomes

Results

Permanency

Children in kinship care were less likely to be reunified, adopted, or in legal guardianship

Adaptive Behaviors Mental Health Problems

Maltreatment

Behavior Problems

Children who lived in family foster homes licensed by the Baltimore City Department of Social Services between 1984 and 1988

Kinship care was majority of out-of-home placement

Foster care homes in Santa Clara, California with children ages 5-12 in care as of Dec. 1995

Not Stated

All children in California who entered out-of-home care for the first time between 1989-1991

Kinship care was majority of out-of-home placement

Mental Health Problems Maltreatment Educational Attainment

Family relations

Permanency Reentry

Children in kinship care had greater reported adaptive behaviors No difference in reported mental health problems between children in kinship and foster care

Children in kinship care were less likely to be maltreated

Children in kinship care had fewer reported behavior problems, mental health problems and substantiated maltreatment No difference in high school graduation rates between children in kinship and foster care Children in kinship care had higher reported family relations Children in kinship care were less likely to be in a permanent placement and less likely to reenter care after reunification

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 65

Author and Publication Date

Sample

Adults formerly in out-of-home care through the Baltimore City Department of Social Services Foster Care Division

Treatment

Outcomes

Christopher (1998)

Clyman, Riley, Lewin, Messer, Palmer & Altman (1998)

Courtney (1995)

Courtney & Barth (1996)

Adults formerly placed in kinship care had fewer health risks

Kinship care was any placement during out-ofhome care

Adult Outcomes (behavior risk, economic risk, health risk, education risk)

Children from the National Survey of America's Families

Not Stated

Mental Health Service Use

Children in kinship care had lower rates of mental health service utilization

Closed foster care cases from the Kern County Department of Human Services (CA) for children who emancipated in 1995 and 1996

Kinship care was last outof-home placement

Educational Attainment

Children in kinship care had lower high school graduation rates

Mental Health Service Use

Children in kinship care had lower rates of mental health service utilization

Carlson (2002) Carpenter, Berman, Clyman, Moore, & Xu (2004)

Results

All families in a large suburban eastern county who had children under the age of six legally placed with them for at least three months Children discharged from a first episode in the foster care system in California between Jan. 1and June 30, 1988 All foster children from California who experienced a final discharge between July 1, 1991 and Dec. 31, 1992

Not Stated

Developmental Service Use Physician Service Use

No difference between adults formerly in kinship and foster care on behavior risk, economic risk, and education risk

No difference between children in kinship and foster care on rates of developmental and physician service use

Kinship care was last outof-home placement

Reentry

Children in kinship care were less likely to reenter care after reunification

Kinship care was last outof-home placement

Permanency

Children in kinship care were more likely to be adopted or reunified

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 66

Author and Publication Date

Courtney, Piliavin, & Entner Wright (1997)

Goerge, Bilaver, Lee, Needell, Brookhart, & Jackman (2002)

Sample

Treatment

Children placed in out-of-home care by California county child welfare departments during 1988

Kinship care was first outof-home placement for permanency and last for reentry

Children in California, Illinois, and South Carolina from the Multi-State Foster Care Data Archive

Kinship care was last outof-home placement

Every foster family under the supervision of two voluntary, sectarian foster care/adoption agencies in New York City

Outcomes

Permanency Reentry

Employment status Earnings

Physician Service Use Not Stated

Mental Health Service Use Family Relations

Jenkins (2002)

Children in out-ofhome placements in Erie County, NY

Behavior Problems Not Stated Adaptive Behaviors

Jones (1998)

Keller, Wetherbee, Prohn, Payne, Sim, & Lamont (2001)

Children in the Casey Family Program between Jan. 1994 and June 1997

Behavior Problems Not Stated

Adaptive Behaviors

Results

No difference between children in kinship and foster care on likelihood of reunification Children in kinship care were less likely to reenter care after reunification Children in kinship care had higher rates of employment within two years of exit from care Children in kinship care had higher reported earnings during first two years after emancipation Children in kinship care had lower rates of physician service use No difference between children in kinship and foster care on reported mental health service use and family relations Children in kinship care had fewer reported behavior problems No difference between children in kinship and foster care on reported adaptive behaviors Children in kinship care had fewer reported behavior problems and greater reported adaptive behaviors

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 67

Author and Publication Date

Sample

Treatment

Leslie, Landsverk, EzzetLofstrom, Tschann, Slymen, Garland (2000)

Children ages 0-16 in San Diego County removed from their homes between May 1990 and October 1991

Kinship care was only outof-home placement

McIntosh (2002) McMillen, Scott, Zima, Ollie, Munson, & Spitznagel (2004)

Needell (1996)

Prosser (1997)

Shin (2003a)

Shore, Sim, Prohn, & Keller (2002)

Children in out-ofhome care from Los Angeles County in the Department of Children and Family Services Database from December, 2000 Children in eight counties from four of the seven administrative regions of the Missouri Children's Division

Outcomes

Results

Mental health service use

No difference between children in kinship and foster care on rates of mental health service utilization

Not Stated

Permanency

No difference between children in kinship and foster care on likelihood of reunification

Not Stated

Mental Health Service Use

Children in kinship care had lower rates of mental health service utilization Children in kinship care were less likely to be reunified or adopted and less likely to reenter care after reunification

Children in foster care between 1988 and 1994 from the California Foster Care Information System

Kinship care was majority of out-of-home placement

Children from the National Longitudinal Survey of Youth, which is a nationally representative sample of 12,000 people born between 1957-1964

Kinship care was at least a 4 month spell before age 18

Educational Attainment

No difference between children in kinship and foster care on high school graduation rates

Not Stated

Educational Achievement

Placement in kinship care accounted for 5.75% of the variance in reading achievement

Not Stated

Behavior Problems

Youth in Illinois substitute care between the ages of 16.5 and 17.5 years as of December 1998 Children served by Casey Family Program between Jan. 1994 and June 1997

Permanency Reentry

No difference between children in kinship and foster care on reported behavior problems

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 68

Author and Publication Date

Smith (2003)

Smith, Rudolph, & Swords (2002)

Smithgall, Gladden, Howard, Goerge, & Courtney (2004)

Sample

Treatment

Outcomes

Results

Children in the AFCARS database who became eligible for adoption in October 1997

Kinship care was last outof-home placement

Permanency

Children in kinship care were less likely to be discharged from care

All children placed with a relative in a middle size county in New York State between October 1993 and April 1994

Kinship care was first outof-home placement

Permanency

No difference between children in kinship and foster care on likelihood of permanent placement

Educational Achievement

No difference between children in kinship and foster care on ITBS reading test and likelihood for being old for grade

Children in out-ofhome placement in 2000 from Chapin Hall's Integrated Database on Child and Family Services

Children adjudicated dependent through Children and Youth Services of Delaware County and in out-ofhome care in June 1997

Kinship care was majority of out-of-home placement

Behavior Problems Not Stated

Testa (2001)

Testa & Rolock (1999)

Tompkins (2003)

Children from the 1994 National Study of Protective, Preventive, and Reunification Services

Educational Attainment Family Relations

Surbeck (2000) Children in kinship and non-kinship foster placements in Cook County, Illinois that began between July 1991 and June 1995 Homes in Cook County, Illinois that received children from Dec. 1, 1994 to Sept. 30, 1996

Old for Grade

Kinship care was first out-of-home placement for 61% of children

Not Stated

Permanency

Permanency

Mental Health Problems Not Stated Physical Health Problems

Children in kinship care had fewer reported behavior problems No difference between children in kinship and foster care on reported academic performance and family relations No difference between children in kinship and foster care on likelihood of reunification, adoption, or subsidized guardianship Children in kinship care were more likely to be reunified, adopted, or in subsidized guardianship Children in kinship care had fewer reported mental health problems and physical health problems

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 69

Author and Publication Date

Wells & Guo (1999)

Wulczyn, Hislop, & Harden (2002)

Zima, Bussing, Freeman, Yang, Belin, & Forness (2000)

Zuravin, Benedict, & Somerfield (1993)

Sample

Treatment

Children placed in out-of-home care by the Cuyahoga County Department of Children and Families Services at any time in 1992 or 1993

Kinship care was first outof-home placement for permanency and last for reentry

Children from the Multistate Foster Care Data Archive placed for the first time between Jan. 1990 and Dec. 1997 Children ages 6-12 years living in out-ofhome placement from three of the eight Los Angeles counties between July 1996 and March 1998 Children from Baltimore City Department of Social Services supervised foster homes from Jan. 1984 to Dec. 1988

Outcomes

Results

Permanency

No difference between children in kinship and foster care on likelihood of reunification

Reentry

Children in kinship care were less likely to reenter care after reunification

Kinship care was majority of out-of-home placement

Reentry

Children in kinship care were less likely to reenter care after a completed first spell

Not Stated

Behavior Problems

Children in kinship care had fewer reported behavior problems

Not Stated

Maltreatment

Children in kinship care were less likely to be maltreated

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 70

Appendix C Outcome Measures for Effect Size Calculations Bivariate Analyses Barth, Webster, & Lee (2002): the percentage of children who achieved permanency or were still in care at six years after entry into out-of-home placement was used to calculate effect size. The permanency outcome included reunification, adoption, subsidized guardianship, and an “other” category. Belanger (2001): the mean total score on the Devereaux Scales of Mental Disorders (DSMD) and the mean composite score on the Vineland Scales of Adaptive Behavior (VABS) were used to calculate effect size. Benedict, Zuravin, & Stallings (1996): the percentage of children with behavior problems, mental health problems, and substantiated maltreatment reports was used to calculate effect size. The percentage of children who graduated high school (including GED completion) also was used to calculate effect size. Berrick (1997): the mean score on the Index of Family Relations (IFR) was used to calculate effect size. Berrick & Needell (1999): the percentage of children who achieved permanency or remained in care at four years after out-of-home placement was used to calculate effect size. The permanency outcome included reunification, adoption, subsidized guardianship, and an “other” category. The percentage of children who reentered out-of-home care or remained reunified after 6 to 36 months also was used to calculate effect size. Carpenter, Berman, Clyman, Moore, & Xu (2004): the percentage of children using mental health services was used to calculate effect size.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 71

Christopher (1998): the percentage of children who graduated high school was used to calculate effect size. Clyman, Riley, Lewin, & Messer (1998): the percentage of children receiving mental health services was used to calculate effect size. Jenkins (2002): the percentage of children receiving good or poor mental health care and the percentage of children having good or poor family relations were used to calculate effect size. Jones (1998): the mean score on the total problems scale of the Child Behavior Checklist (CBCL) and the mean composite score on the VABS were used to calculate effect size. Keller et al. (2001): the mean score on the total problems scale and total competence scale of the CBCL were used to calculate effect size. Leslie et al. (2000): the mean number of outpatient mental health visits was used to calculate effect size. McIntosh (2002): the percentage of children who achieved permanency or remained in care at one to five years after entry into out-of-home placement was used to calculate effect size. The permanency outcome only included reunification. Needell (1996): the percentage of children who achieved permanency or remained in care at four years after a first spell in out-of-home placement was used to calculate effect size. The permanency outcome included reunification and adoption. The percentage of children who reentered out-of-home care or remained reunified after 6 and 36 months also was used to calculate effect size. Prosser (1997): the percentage of children who graduated from high school was used to calculate effect size.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 72

Shore, Sim, Le Prohn, & Keller (2002): the mean score on the total problems scale of the Teacher Report Form (TRF) was used to calculate effect size. Smith (2003): the percentage of children discharged from or remaining in out-of-home care after one year was used to calculate effect size. Smith, Rudolph, & Swords (2002): the percentage of children who achieved permanency or remained in care at two or three years after entry into out-of-home placement was used to calculate effect size. The permanency outcome included reunification, adoption, relative custody, and long-term foster care. Surbeck (2000): the mean composite score on the child behavior scale, the mean score on the family relations scale, and the mean score on the academic performance scale of the Child Well-Being Scales were used to calculate effect size. Testa (2001): the percentage of children who achieved permanency or remained in care at four to eight years after entry into out-of-home placement was used to calculate effect size. The permanency outcome included reunification, adoption, and subsidized guardianship. Testa and Rolock (1999): the percentage of children who achieved permanency or remained in care at one to three years after entry into out-of-home placement was used to calculate effect size. The permanency outcome included reunification, adoption, and subsidized guardianship. Tompkins (2003): the percentage of children with mental health problems was used to calculate effect size. Wulczyn, Hislop, & Harden (2002): the percentage of children who reentered out-ofhome care or remained discharged after a first spell was used to calculate effect size.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 73

Multivariate Analyses Benedict, Zuravin, Somerfield, & Brandt (1996): the odds ratio of association with maltreatment while in out-of-home care was used as an effect size estimate. Carlson (2002): the ordinary least squares regression coefficients for behavior risk index, economic risk index, health risk index, and education risk index were used as effect size estimates. Courtney (1995): the parameter value for reentry to out-of-home care for reunified children was used as an effect size estimate. Courtney & Barth (1996): the odds ratio of returning home or being adopted over unsuccessful discharge from out-of-home care was used as an effect size estimate. Courtney, Piliavin, & Entner Wright (1997): the parameter value for reunification within 4 years of placement into out-of-home care was used as an effect size estimate. The parameter value for reentry to out-of-home care within 2 years after reunification also was used as an effect size estimate. Goerge et al. (2002): the odds ratio of likelihood of employment during the first two post-exit years was used as an effect size estimate. The parameter value for estimated wages during the first two years after age 18 also was used as an effect size estimate. McMillen et al. (2004): the odds ratio of current outpatient therapy utilization was used as an effect size estimate. Shin (2003a): the beta value for the relationship between kinship care placement and reading achievement was used as an effect size estimate.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 74

Smithgall, Gladden, Howard, Goerge, & Courtney (2004): the mean adjusted score on the Iowa Test of Basic Skills (ITBS) reading scale and the odds ratio for being old for grade were used as effect size estimates. Wells and Guo (1999): the risk ratio of reunification was used as an effect size estimate. The risk ratio of reentry to out-of-home care for children reunified within 24 months also was used as an effect size estimate. Zima, Bussing, Yang, & Belin (2000): the odds ratio of having clinical behavior problems was used as an effect size estimate. Zuravin, Benedict, & Somerfield (1993): the odds ratio of children being maltreated in out-of-home care was used as an effect size estimate.

________________________________________________________________________________________________ Social Work Research Center Kinship Care Systematic Review Page 75