The Quality Standards Assessment: Connecting Quality Care with Youth Outcomes

Background. Some policy advocates and researchers are critical of children’s residential care (RC) and argue that it should be reduced or eliminated as it is expensive and produces poor outcomes. However, emerging evidence indicates that the quality-of-care children receive in RC affects their outcomes. Thus, a proactive, research-based approach is needed to define what comprises quality RC and establish quality practice standards to evaluate care and inform quality improvement. To that end, Florida established a statewide accountability system based on empirically driven and measurable quality standards. Evidence from statewide studies demonstrated that RC programs that scored higher on the quality standards reported fewer critical incidents (e.g., child runaway episodes, hospitalizations, law enforcement intervention) and staff turnover. Despite these promising results, the question of the extent to which RC programs’ performance on the standards predicts youth outcomes remains. This study addresses this gap by examining associations between quality standards and child outcomes to determine which practices standards are most strongly associated with improvement.

Method: A total of 555 children from 103 distinct RC programs in Florida were identified using a centralized child welfare database. For each child, the child and program staff were asked to complete a Strengths and Difficulties Questionnaire (SDQ) over three phases including an initial assessment (Time 1) and two follow-ups (60 days, Time 2 and 90 days, Time 3). Children with an initial and at least one follow-up SDQ were included in the analytic sample (n = 59) clustered in 18 RC programs. Quality of care in RC programs was measured using the Quality Standards Assessment, a validated measure of RC programs’ performance on a set of 59 quality standards across eight core practice domains. Using confirmatory factor analysis (CFA), quality standards were identified and subsequently added to stepwise regression models controlling for SDQ scores and child demographics and time spent in RC.

Results: In total, higher scores on 12 standards from eight domains were associated with lower emotional symptoms (β=0.61, p=0.021), conduct problems (β=0.39, p=0.001), prosocial problems (β=0.51, p=0.034) and total difficulties (β=1.25, p=0.026) at follow-up with data suggesting children placed in programs with higher performance ratings experienced significantly greater improvements. Results also showed that younger children demonstrated better outcomes (β=12.31, p=0.011) and this was moderated by the time (weeks) spent in the RC program.

Conclusion and Implications: Findings from this pilot study provide preliminary evidence suggesting that quality practice standards are associated with improvement in children’s symptoms. Specifically, higher scores on certain practice standards may be associated with improvements in children’s emotional symptoms, conduct problems, prosocial problems, and total difficulties. This study yielded findings to suggest there is value in further research identifying quality practices and how they impact child outcomes in RC. Future analyses aimed at collecting child outcomes data will require careful consideration of strategies to mitigate attrition and increase participation, in addition to utilizing a more rigorous study design to produce more robust findings.