Improving relationship quality in group care settings: The impact of implementing the CARE model

May combine two claims.

Children's positive relationships with caregivers predict improved behavioral and emotional functioning and are associated with fewer iatrogenic effects of residential placement.

“research conducted in group care settings has found that children's positive relationships with caregivers predicts improved behavioral and emotional functioning, and is associated with fewer iatrogenic effects of residential placement”

• cites Duppong-Hurley, Lambert, Gross, Thompson, & Farmer, 2017 sourceNot checked — no full text of the source
• cites Farmer, Murray, Ballentine, Rauktis, & Burns, 2017 sourceNot checked — no full text of the source
• cites Southerland, Mustillo, Farmer, Stambaugh, & Murray, 2009 sourcePartly supported

  The Discussion (page unavailable) reports that better relationships between youth and treatment parents in therapeutic foster care were associated with better emotional and behavioral functioning. But this was a cross-sectional study, so it did not establish prediction or causation. Our copy does not report findings from group care settings or an association with fewer iatrogenic effects of residential placement. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim. May combine two claims.

CARE is based on the philosophy that enhancing the quality of relationships between children and direct-care staff across the organization and fostering more positive daily interactions contributes to children's recovery and growth.

“CARE is based on the philosophy that enhancing the quality of relationships between children and direct-care staff across the organization and fostering more positive daily interactions contributes to children's recovery and growth”

• cites Trieschman, Whittaker, & Brendtro, 1969Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Whittaker et al., 2016 sourceCould not verify

  The References section lists a chapter on the CARE model, but does not describe CARE’s philosophy. Searches for CARE, staff–child relationships, daily interactions, and recovery found no passage in our copy that establishes the attributed claim; no page number is available for the reference passage. (LLM, conf=0.72)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim.

The stepped-wedge design is a quasi-experimental design commonly used to evaluate organization-wide interventions.

“The stepped-wedge is a quasi-experimental design commonly used to evaluate organization-wide interventions”

• cites Handley, Lyles, McCulloch, & Cattamanchi, 2018 sourceSupported
• cites Highfield et al., 2015 sourcePartly supported

  Methods > Trial design describes this study as a non-randomized stepped-wedge evaluation in which clinics implement the intervention at staggered times. That supports the design as an example for evaluating a clinic-level intervention, but the paper does not establish that stepped-wedge designs are commonly used for organization-wide interventions or that they are necessarily quasi-experimental; it also discusses randomized stepped-wedge trials. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S2 (checkable).

Rich relational environments that stimulate positive social interaction among children and adults and foster a sense of belonging are essential for healthy growth and thriving in any child service setting.

“Perhaps equally important, agencies must also focus on creating rich relational environments that stimulate positive social interaction among children and adults and foster a sense of belonging, factors that are essential for healthy growth and thriving in any child service setting (Allen & Tan, 2016;Baumeister & Leary, 1995;Feeney & Collins, 2015;Holt-Lunstad, Robles, & Sbarra, 2017;Pianta, Hamre”

• cites Allen & Tan, 2016Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Baumeister & Leary, 1995 sourceNot checked — no full text of the source
• cites Feeney & Collins, 2015 sourcePartly supported

  The ‘Relational Support Functions as Predictors of Thriving’ section says that well-functioning close relationships are fundamental to thriving, and ‘Elaboration on Support Functions’ describes a network of responsive relationships as an optimal environment. These passages support the general importance of relationships, but do not establish that agencies must create such environments or that they are essential in every child service setting. Page numbers were not available in our copy. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim.

The model is rooted in the core idea that a critical pathway for healing and growth among children in group care is through experiencing a consistent pattern of positive, reciprocal interactions with supportive adults.

“The model is rooted in the core idea that a critical pathway for healing and growth to occur among children in group care is through experiencing a consistent pattern of positive, reciprocal interactions with supportive adults”

• cites Bronfenbrenner & Morris, 2006Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Li & Julian, 2012 sourceNot checked — no full text of the source
• cites Schuengel & van IJzendoorn, 2001 sourcePartly supported

  In “Patient-staff relationships,” the review says stable relationships and secure-base support are important in institutional care. On p. 313, it says a secure patient-staff attachment may offer a corrective experience. But it treats that benefit as conditional and notes difficulties and risks; it does not establish consistent, positive, reciprocal interactions as a critical pathway to healing and growth for children in group care generally. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim.

CARE is an organizational intervention that provides ongoing consultation, training and technical assistance to help group care agencies create a more therapeutic environment.

“CARE is an organizational intervention that provides ongoing consultation, training and technical assistance to help group care agencies create a more therapeutic environment”

• cites Holden, 2009Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Holden, Anglin, Nunno, & Izzo, 2014Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Holden & Sellers, 2019 sourceSupported

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Achieving the goal of improving outcomes for children requires sustained commitment, agency-wide education, and critical reflection about existing beliefs and norms, all within the context of a coherent program model centered around the best interests of children.

“As highlighted by Anglin (2011) and others (Bryson et al., 2017;Whittaker et al., 2016), achieving this goal requires sustained commitment, agency-wide education, and critical reflection about existing beliefs and norms, all within the context of a coherent program model centered around the best interests of children.”

• cites Anglin (2011)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Bryson et al., 2017 sourceSupported
• cites Whittaker et al., 2016 sourcePartly supported

  The Basic principles section calls for a child-development-focused program that is carefully implemented and continually evaluated. The concluding research agenda identifies targeted training and critical evaluation as ways to support residential-care staff. Neither passage specifies sustained commitment, agency-wide education, or reflection on existing beliefs and norms as requirements for improving outcomes. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

May combine two claims.

The CARE effect did not differ by age, gender, race, length of stay, DSS referral, or problem behavior.

“but did not differ by age, gender, race, length of stay, DSS referral, or problem behavior.”

May combine two claims.

Post-hoc analyses indicated that the proportion of children in Cohort 1 with 2 or more placements steadily decreased over time, while in Cohort 2, the proportion in each placement category varied over time but did not follow a consistent pattern.

“Post-hoc analyses (not shown) indicated that the proportion of children in Cohort 1 with 2 or more placements steadily decreased over time, while in Cohort 2, the proportion in each placement category varied over time but did not follow a consistent pattern.”

May combine two claims.

The mean YPRQ score was 3.0 during the Pre-CARE period and ranged from 3.25 to 3.40 during CARE implementation.

“YPRQ estimates at each year of CARE implementation, derived from the Design Group variable, showed that the mean YPRQ score was 3.0 during the Pre-CARE period and during CARE implementation, scores ranged from 3.25 to 3.40 (see Fig. 5).”

Probably not a substantial claim: may not pass S3 (load-bearing).

The results of the IPPA were comparable with those of the YPRQ unless otherwise stated.

“Unless otherwise stated, the results of the IPPA were comparable with those of the YPRQ.”

May combine two claims.

Older children and children with higher SDQ Total Difficulties scores had lower YPRQ scores, on average.

“Older children and children with higher SDQ Total Difficulties scores had lower YPRQ scores, on average.”

May combine two claims.

The test of cohort equivalence at the start of the study in 2010 was not statistically significant, nor was the test of baseline equivalence between Cohort 1 in 2010 and Cohort 2 in 2011.

“The test of cohort equivalence at the start of the study in 2010 was not statistically significant, nor was the test of baseline equivalence between Cohort 1 in 2010 and Cohort 2 in 2011.”

May combine two claims.

With Site A included, the CARE effect at Year 1 was different from zero, but the CARE effects at Years 2 and 3 were not.

“Tests of the CARE effect with Site A included in the analysis indicate that, based on null hypothesis testing at α = 0.05, the CARE effect at Year 1 was different from zero, but the CARE effects at Years 2 and 3 were not.”

May combine two claims.

Over the five-year study period, the composition of the sample remained consistent with regard to age, gender, race, SDQ score, referral source, and length of stay.

“Over the five-year study period, the composition of the sample remained consistent with regard to age, gender, race, SDQ score, referral source, and length of stay.”

May combine two claims.

Only number of previous placements reached statistical significance and an interaction between cohort and year was found.

“Only number of previous placements reached statistical significance and an interaction between cohort and year was found.”

May combine two claims.

For the IPPA, the Cohort Equivalence and Cohort 2 Stability tests were not statistically significant, while the Baseline Equivalence test was statistically significant (mean difference = -0.53, 95% CI [-1.04, -0.01], p = .045).

“For the three pre-CARE contrasts, the Cohort Equivalence and Cohort 2 Stability tests were not statistically significant, as expected, while the Baseline Equivalence test was statistically significant (mean difference = -0.53,95% CI [-1.04, -0.01], p = .045).”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Although the CARE implementation model provided a standard progression of activities, implementation unfolded in unique ways at each agency, and many developed their own innovative strategies to apply the CARE principles.

“Although the CARE implementation model provided a standard progression of activities, implementation unfolded in unique ways at each agency, and many developed their own innovative strategies to apply the CARE principles.”

May combine two claims.

For the IPPA, Years 1 and 2 CARE effects were not statistically significant, while Year 3 was statistically significant (mean difference = 0.34, 95% CI [0.02, 0.66], p = .038).

“Years 1 and 2 were not statistically 32 [0.07, 0.57], p = .013Random Effects Site*Time 0.012 0.020 [0.000, 0.338] Site*Time*Cottage 0.123 0.032 [0.074, 0.207] Residual 0.346 0.024 [0.302, 0.396] significant, while Year 3 was statistically significant (mean difference = 0.34, 95% CI [0.02, 0.66], p = .038)”

May combine two claims.

Site A had a positive influence score of 0.15; when Site A was removed, the estimate of the Year 3 CARE effect increased from 0.17 to 0.32.

“As can be seen in Fig. 6, Site A had a positive influence score of 0.15; that is, when Site A was removed the estimate of the Year 3 CARE effect increased from 0.17 to 0.32.”

May combine two claims.

Compared to Cohort 1, children in Cohort 2 were more likely to be older, male, non-white, and referred through DSS.

“compared to Cohort 1, children in Cohort 2 were more likely to be older, male, non-white, and referred through DSS.”

May combine two claims.

Intraclass correlations were 0.32 at the cottage level and 0.05 at the site level.

“Intraclass correlations, computed with intercept-only models, were 0.32 at the cottage level and 0.05 at the site level.”

May combine two claims.

The increase in YPRQ peaked at 0.40 at the end of the first year and decreased slightly in subsequent years.

“The increase peaked at 0.40 at the end of the first year and decreased slightly in subsequent years.”

May combine two claims.

One agency (Site A) had an average Pre-CARE YPRQ score of 4.1, while most agencies ranged from 2.5 to 3.5.

“although Pre-CARE scores for most agencies ranged from 2.5 to 3.5, one agency had an average score of 4.1 (see Fig. 4).”

May combine two claims.

Most agencies showed improvement over time, some agencies did not, and one agency (Site A) showed a substantial decline in relationship quality between baseline and Year 3.

“Lastly, although most agencies showed improvement over time, some agencies did not, and one agency (Site A) showed a substantial decline in relationship quality between baseline and Year 3.”

May combine two claims.

For Year 1, t(39.5) = 2.26, p = 0.030, 95% CI [0.03, 0.56]; for Year 2, t(40.1)= 1.13, p = 0.265, 95% CI [-0.12, 0.43]; for Year 3, t(38.5)= 1.19, p = 0.243, 95% CI [-0.12, 0.46].

“For Year 1, t(39.5) = 2.26, p = 0.030, 95% CI [0.03, 0.56]; for Year 2, t(40.1)= 1.13, p = 0.265, 95% CI [-0.12, 0.43]; for Year 3, t(38.5)= 1.19, p = 0.243, 95% CI [-0.12, 0.46].”

May combine two claims.

The pattern of effects for the covariates differed somewhat from that observed for the YPRQ; however, the results of the six planned contrasts were similar to what we found with the YPRQ, though not as strong.

“The pattern of effects for the covariates differed somewhat from that observed for the YPRQ; however, the results of the six planned contrasts were similar to what we found with the YPRQ, though not as strong.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Across the full sample of 733 surveys, the unadjusted YPRQ mean and median were both 3.3, indicating a symmetric distribution located at the center of the 5-point Likert scale.

“Across the full sample of 733 surveys, the unadjusted YPRQ mean and median were both 3.3, indicating a symmetric distribution located at the center of the 5-point Likert scale.”

May combine two claims.

After controlling for several important variables, the perceived quality of child-adult relationships within agencies increased significantly after CARE implementation began, with improvements observed at the end of each year of CARE implementation.

“With regard to the first research question, results indicated that, after controlling for several important variables, the perceived quality of child-adult relationships within agencies increased significantly after CARE implementation began, with improvements observed at the end of each year of CARE implementation.”

• AssumptionRests on an assumption: The CARE model was implemented with sufficient fidelity and uniformity across agencies to attribute observed relationship quality improvements to CARE.

Probably not a substantial claim: may not pass S3 (load-bearing).

At baseline, Cohort 1 and Cohort 2 differed significantly in the proportion of males (F(1,183)=5.72, p=.018).

“| Male | 55.8% | 41.1% | 73.3% | F(1, 183) = 5.72 | F(4, 184) = 0.47 | F(3, 183) = 0.25 |”

• Also stated in: § Declaration of Competing Interests · plain

  “| | | | p = .018 | p = .761 | p = .864 |”

Probably not a substantial claim: may not pass S3 (load-bearing).

There was a significant cohort by year interaction for previous placements (F(9,289)=2.47, p=.010).

“| Previous Placements | | | | F(3, 287) = 5.94 | F(12, 302) = 1.72 | F(9, 289) = 2.47 |”

• Also stated in: § Declaration of Competing Interests · plain

  “| | | | | p = .001 | p = .063 | p = .010 |”

Probably not a substantial claim: may not pass S3 (load-bearing).

At baseline, Cohort 1 and Cohort 2 differed significantly in age (F(1,45)=10.92, p=.002).

“| Age (years) | 14.8 (2.2) | 14.5 (2.2) | 15.2 (2.1) | F(1, 45) = 10.92 | F(4, 45) = 0.49 | F(3, 45) = 0.23 |”

• Also stated in: § Declaration of Competing Interests · plain

  “| | | | p = .002 | p = .743 | p = .876 |”

Probably not a substantial claim: may not pass S3 (load-bearing).

At baseline, Cohort 1 and Cohort 2 differed significantly in the proportion with DSS custody (F(1,183)=14.53, p<.001).

“| DSS Custody | 40.0% | 30.0% | 52.0% | F(1, 183) = 14.53 | F(4, 185) = 0.97 | F(3, 182) = 0.81 |”

• Also stated in: § Declaration of Competing Interests · plain

  “| | | | p < .001 | p = .424 | p = .489 |”

Probably not a substantial claim: may not pass S3 (load-bearing).

At baseline, Cohort 1 and Cohort 2 differed significantly in the proportion of non-White youth (F(1,178)=14.49, p<.001).

“| Non-White | 27.9% | 18.9% | 38.7% | F(1, 178) = 14.49 | F(4, 174) = 1.32 | F(3, 178) = 0.12 |”

• Also stated in: § Declaration of Competing Interests · plain

  “| | | | p < .001 | p = .264 | p = .950 |”

Probably not a substantial claim: may not pass S3 (load-bearing).

At baseline, Cohort 1 and Cohort 2 differed significantly in previous placements (F(3,287)=5.94, p=.001).

“| Previous Placements | | | | F(3, 287) = 5.94 | F(12, 302) = 1.72 | F(9, 289) = 2.47 |”

• Also stated in: § Declaration of Competing Interests · plain

  “| | | | | p = .001 | p = .063 | p = .010 |”

Probably not a substantial claim: may not pass S3 (load-bearing).

The YPRQ was designed such that most items asked respondents to report on their experience within the context of specific scenarios commonly experienced in residential settings.

“The YPRQ was designed such that most items asked respondents to report on their experience within the context of specific scenarios that are commonly experienced in residential settings”

Classified as probably not a substantial claim.

Steps were taken to monitor and prevent sharing of CARE-related information across cohorts.

“Given the potential for contamination across cohorts, we took steps to monitor and prevent sharing of CARE-related information.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The YPRQ included 33 items generated based on a review of the scientific literature and focus groups with children from non-study agencies.

“Developed for the current study, the YPRQ included 33 items that were generated based on a review of the scientific literature and focus groups with children from non-study agencies.”

Classified as probably not a substantial claim.

The Youth Perceptions of Relationship Quality (YPRQ) was designed to be salient to the unique conditions of children living in group care settings.

“The Youth Perceptions of Relationship Quality (YPRQ) was our primary measure of relationship quality, as it was designed to be salient to the unique conditions of children living in group care settings.”

Probably not a substantial claim: may not pass S3 (load-bearing).

All assessment procedures were approved by the Cornell University Institutional Review Board.

“All assessment procedures were approved by the Cornell University Institutional Review Board.”

Classified as probably not a substantial claim.

CARE is an organization-wide program model that involves a range of structural change and staff-development activities.

“Children and Residential Experiences (CARE), an organization-wide program model that involves a range of structural change and staff-development activities”

May combine two claims.

CARE was implemented and assessed at multiple agencies across the state, which helped eliminate selection bias as a concern.

“as did the fact that CARE was implemented and assessed at multiple agencies across the state.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

CARE implementation lasted three years and involved the application of six evidence-informed principles throughout the organization.

“CARE implementation lasted three years and involved the application of six evidence-informed principles throughout the organization”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Although no factors were identified empirically, the YPRQ was designed to reflect constructs pertaining to children's relational experiences with staff.

“Although no factors were identified empirically, the measure was designed to reflect the following constructs pertaining to children's relational experiences with staff:”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The Inventory of Parent and Peer Attachment (IPPA) was included as a secondary measure of relationship quality and as a source of construct validity for the YPRQ.

“The Inventory of Parent and Peer Attachment (IPPA; Armsden & Greenberg, 1987) was included as a secondary measure of relationship quality, as well as a source of construct validity for the YPRQ.”

• cites Armsden & Greenberg, 1987 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

CARE implementation is an intensive and complex organizational process that calls for substantive changes in role expectations, theoretical perspective, organizational priorities and norms.

“Ultimately, CARE implementation is an intensive and complex organizational process that calls for substantive changes in role expectations, theoretical perspective, organizational priorities and norms.”

May combine two claims.

Most other empirically-supported models are based on studies at a single agency, making it difficult to know the extent to which findings were influenced by unique conditions at those locations.

“Most other empirically-supported models are based on studies at a single agency (Elwyn, Esaki, and Smith, 2015;Hodgdon et al., 2013;McCall et al., 2010;McCurdy & McIntyre, 2004;Morral, McCaffrey, & Ridgeway, 2004;Rivard, Bloom, McCorkle, & Abramovitz, 2005), making it difficult to know the extent to which findings were influenced by unique conditions at those locations.”

• cites Elwyn, Esaki, and Smith, 2015 sourceNot checked — no full text of the source
• cites Hodgdon et al., 2013 sourceNot checked — no full text of the source
• cites McCall et al., 2010 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Our results say little about the mechanism of change responsible for the improvements in relational quality.

“Our results say little about the mechanism of change responsible for the improvements in relational quality.”

Probably not a substantial claim: may not pass S2 (checkable).

Use of a coherent program model such as CARE helps to accomplish the goal of creating rich relational environments by offering a sound philosophy that creates congruence across the entire agency and aligns everyone's efforts around creating a consistent, coherent experience for children.

“Use of a coherent program model such as CARE helps to accomplish this challenging goal by offering a sound philosophy that creates congruence across the entire agency, and aligns everyone's efforts around creating a consistent, coherent experience for children (Anglin, 2002).”

• cites Anglin, 2002Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

Our assessment of agency-level factors was limited.

“Our assessment of these factors was limited.”

Probably not a substantial claim: may not pass S3 (load-bearing).

We have no evidence about how this transition may have occurred in our agencies.

“We have no evidence about how this transition may have occurred in our agencies, however.”

Classified as probably not a substantial claim.

The statistical model included all variables as fixed effects covariates, regardless of statistical significance, to account for baseline differences across cohort and time period.

“To account for these baseline differences across cohort and time period, our statistical model included all variables as fixed effects covariates, regardless of statistical significance.”

Probably not a substantial claim: may not pass S3 (load-bearing).

These children's history of disrupted relationships with past caregivers may have made them particularly responsive to the enhancements made by agencies as they put the CARE principles into practice.

“Given that placement disruption can influence children's relationships with care providers (Rauktis et al., 2005), this finding may suggest that these children's history of disrupted relationships with past caregivers made them particularly responsive to the enhancements made by agencies as they put the CARE principles into practice.”

• cites Rauktis et al., 2005Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

Because the current study is based on direct reports from children, it provides an important complement to the existing studies which rely on administrative data and staff report.

“Because the current study is based on direct reports from children, it provides an important complement to the existing studies which rely on administrative data and staff report.”

Classified as probably not a substantial claim.

The YPRQ was designed to capture key aspects of relationships between children and caregivers that are associated with engagement, growth, and wellbeing.

“It was designed to capture key aspects of relationships between children and caregivers that are associated with engagement, growth, and wellbeing.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

CARE may have helped staff become more capable and purposeful about meeting the unique relational needs of these children.

“It is also possible that CARE may have helped staff become more capable and purposeful about meeting the unique relational needs of these children.”

May combine two claims.

The improved relationship quality scores may suggest that CARE consultants' efforts succeeded in changing how staff approached their work with children, reflecting more trauma sensitive responses, greater autonomy support and flexibility, competence-building, and prioritizing relationship quality over punishment and power struggles.

“Although our study did not examine these implementation processes directly, the improved relationship quality scores we observed may suggest that their efforts succeeded in changing how staff approached their work with children, reflecting more trauma sensitive responses, greater autonomy support and flexibility, competencebuilding, and prioritizing relationship quality over punishment and power”

• AssumptionRests on an assumption: The CARE model was implemented with sufficient fidelity and uniformity across agencies to attribute observed relationship quality improvements to CARE.

Probably not a substantial claim: may not pass S3 (load-bearing).

Accomplishing these changes requires a multi-year effort by administration and leadership to support and sustain the process.

“Accomplishing these changes requires a multi-year effort by administration and leadership to support and sustain the process”

• cites Anglin, 2002Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

We were unable to test for moderation with OSC because there were too few Cohort 2 agencies with positive OSC profiles at the start of the study.

“We were unable to test for moderation with OSC because there were too few Cohort 2 agencies with positive OSC profiles at the start of the study.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

Leadership changes at the agency may have contributed to the decline in Site A.

“Leadership changes at the agency may have contributed as well.”

Probably not a substantial claim: may not pass S3 (load-bearing).

It is unclear whether the OSC assesses dimensions of organizational functioning that are most salient for these agencies.

“It is unclear whether it assesses dimensions of organizational functioning that are most salient for these agencies.”

May combine two claims.

The stepped-wedge design allowed us to demonstrate that observed changes were not likely due to a secular trend affecting agencies throughout the region, because scores in Cohort 2 agencies remained stable during the Pre-CARE period.

“Specifically, the stepped-wedge design allowed us to demonstrate that observed changes were not likely due to a secular trend affecting agencies throughout the region (i.e., scores in Cohort 2 agencies remained stable during the Pre-CARE period).”

• AssumptionRests on an assumption: The CARE model was implemented with sufficient fidelity and uniformity across agencies to attribute observed relationship quality improvements to CARE.

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

This need for professional learning and support is especially acute in group care settings.

“This need is especially acute in group care settings.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based). May combine two claims.

The change may have been more troublesome when shifting from a longstanding program model, especially if aspects of the new model were perceived as incompatible, generated resistance or conflict among staff, or if the process did not result in a coherent milieu across the agency.

“The change may have been more troublesome, however, when shifting from a longstanding program model, especially if aspects of the new model were perceived as incompatible, generated resistance or conflict among staff, or if the process did not result in a coherent milieu across the agency.”

Classified as probably not a substantial claim.

Although differences between cohorts were accounted for statistically, those agencies may have differed on other variables that were not measured.

“Furthermore, although differences between cohorts were accounted for statistically, those agencies may have differed on other variables that were not measured.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The decline in Site A may have been related to difficulties transitioning to CARE from their pre-existing program model.

“However, we must also consider the possibility that the decline was related to difficulties transitioning to CARE from their pre-existing program model.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

It is typical to find that agencies face many challenges when adjusting their policies and practices to adapt to a new philosophy of service.

“It is typical to find that agencies face many challenges when adjusting their policies and practices to adapt to a new philosophy of service.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The OSC, although validated in some child service settings, has not been validated in group care settings like the ones represented in this study.

“The OSC, although validated in some child service settings, has not been validated in group care settings like the ones represented in this study.”

May combine two claims.

The YPRQ score was highly correlated with the modified IPPA score (r = 0.84, p < .0001), providing a good indicator of construct validity.

“The YPRQ score was highly correlated with the modified IPPA score (r = 0.84, p < .0001),providing a good indicator of construct validity.”

May combine two claims.

The YPRQ showed high internal reliability and its validity was demonstrated by its high correlations with established measures of social and emotional functioning and relationship quality.

“It showed high internal reliability and its validity was demonstrated by its high correlations with established measures of social and emotional functioning and relationship quality.”

May combine two claims.

These findings demonstrate the disproportionate influence of Site A on estimates of the CARE effect, which is large enough to affect interpretation of the results.

“These findings demonstrate the disproportionate influence of Site A on estimates of the CARE effect, which is large enough to affect interpretation of the results.”

Classified as probably not a substantial claim.

CARE helps agencies accomplish this goal by engaging leadership, management and staff throughout the organization in a systematic effort to learn and apply a set of scientifically grounded principles.

“CARE helps agencies accomplish this goal by engaging leadership, management and staff throughout the organization in a systematic effort to learn and apply a set of scientifically grounded principles”

May combine two claims.

CARE represents a promising approach for elevating the quality of group care and thus strengthening the continuum of care within child welfare.

“CARE represents a promising approach for elevating the quality of group care and thus strengthening the continuum of care within child welfare.”

• AssumptionRests on an assumption: The CARE model was implemented with sufficient fidelity and uniformity across agencies to attribute observed relationship quality improvements to CARE.