The assessment of the timeliness of reunification addresses national child welfare outcome 4: reduce time in foster care to reunification without increasing reentry. The wording of this outcome is intended to ensure that reunifications are not viewed as timely if they are not permanent; that is, if the child reenters foster care fairly soon after being reunified.

Table III-1 presents summary data regarding State performance in 2005 on key measures assessing both the timeliness and permanency of reunification. The table includes one of the measures originally developed for the Child Welfare Outcomes Reports (measure 4.1), as well as the four new measures developed for use in the second round of the CFSRs.⁴³ The calculations of three of the new measures shown in the table (C1.1, C1.2, and C1.3) incorporate the “less than 8 days exclusion” and the “trial home visit adjustment.” These are described in detail in appendix B.

Table III-1. Summary of State performance in 2005 on measures pertaining to child welfare outcome 4: Reduce time in foster care to reunification without increasing reentry

* Appendix B provides complete definitions of each measure.
** For these measures, lower percentages are associated with higher performance and higher percentages are associated with lower performance.

Impact of the less than 8 days exclusion and the trial home visit adjustment on State performance regarding reunifications occurring in less than 12 months

Outcome measures 4.1 and C1.1 assess the percentage of reunifications occurring in less than 12 months from a child’s entry into foster care. The key difference between the measures is that the calculation of measure C1.1 excludes children who were discharged from foster care in less than 8 days and incorporates the trial home visit adjustment, which takes into account children who are physically reunified with their families for a period of time prior to reunification but are not officially discharged from foster care. The calculation of measure 4.1 does not involve these adjustments. The data presented in table III-1 indicate that these adjustments resulted in minimal differences in percentages for the national medians and percentiles. For example, the median across States for measure 4.1 was 71.0 percent, while the median for measure C1.1 was 70.4 percent.

However, there were 16 States whose percentages for measures 4.1 and C1.1 were quite different. For six of these States (Arizona, Hawaii, Illinois, Nevada, New Mexico, and Washington), in 2005, the percentage of reunifications occurring in less than 12 months was considerably lower when the adjustments were applied (measure C1.1) than when the adjustments were not applied (measure 4.1). The direction of this difference indicates that these States had relatively high percentages of children discharged from foster care in less than 8 days.

In comparison, there were 10 States (Alaska, Kansas, Maryland, Missouri, Nebraska, North Carolina, Oklahoma, Oregon, Puerto Rico, and Texas) in which the percentage of reunifications occurring in less than 12 months was substantially higher when the adjustments were applied to the measure (measure C1.1) than when the adjustments were not applied (measure 4.1). The direction of this difference suggests that a substantial percentage of children in these States were in a trial home visit for more than 30 days at the time of their discharge from foster care to reunification. These States may engage in the practice of physically reunifying children for a period of time prior to actual discharge from foster care.

Performance with regard to the timeliness of reunification

Timeliness of reunification is assessed through a number of perspectives in order to provide a comprehensive picture of performance with regard to this outcome. Three of the measures of timeliness of reunification focus on children’s length of stay in foster care at the time that they are discharged from foster care to reunification (measures 4.1, C1.1, and C1.2). The 2005 data shown in table III-1 suggest that, in many States, a majority of children discharged to reunification were reunified in a timely manner. Across States, the median length of stay of children reunified was 6.8 months (measure C1.2), and the median percentage of reunifications occurring in less than 12 months was about 70 percent (measure 4.1 and C1.1).

In contrast to these measures, one of the measures of timeliness provides a longitudinal analysis of children from the time of entry into foster care (i.e., an entry cohort analysis). This measure follows children entering foster care for 12 months to determine the percentage that was reunified within this time period. As shown in table III-1, the median across States of children entering foster care who were reunified in less than 12 months was only 38.7 percent and the 75th percentile was only 48.3 percent (measure C1.3). Thus, in at least 75 percent of the States, the majority of children who entered foster care in the last 6 months of 2004 were still in foster care 12 months later.

Measure C1.4 assesses the permanency of reunification by following children who were discharged from foster care to reunification for a 12-month period and identifying the percentage of those children who reentered foster care within that time period.⁴⁴ The data for measure C1.4 indicate that reentry into foster care is an area needing improvement for many States. The median across States for this measure is close to 15 percent, and in five States (Delaware, Indiana, Massachusetts, Oregon, and Pennsylvania), approximately 25 percent or more of the children exiting foster care in 2004 reentered foster care in less than 12 months.

The following are additional key findings with regard to the timeliness of reunification for 2005:

• A moderate correlation was found between timeliness of reunification and reentry. Many States with a relatively high percentage of children reunified in less than 12 months (measure C1.1) also had a relatively high percentage of children reentering foster care in less than 12 months (measure C1.4) (Pearson’s r = 0.45). The relationship between reunification and reentry is even higher (Pearson’s r = 0.54) when the reunification measure does not exclude children who were discharged from foster care in less than 8 days (measure 4.1). The fact that the correlation is higher when the measure includes children in foster care for less than 8 days suggests that many children who are discharged from foster care in less than 8 days may be reentering foster care within the year. 

• Moderate correlations were found between the timeliness of reunification measures and foster care entry rates. Many States with relatively high foster care entry rates also had relatively high percentages of reunifications occurring in less than 12 months (Pearson's r = 0.39) and relatively low median lengths of stay (Pearson's r = -0.41). More investigation is needed to understand the reasons for these relationships.

• Moderate correlations were found between the measure of reentry into foster care and the age of children entering foster care in a State. Many States with a relatively high percentage of children entering foster care who were age 13 or older at the time of entry also had a relatively high percentage of children reentering foster care (measure C1.4) (Pearson’s r = 0.48). Similarly, many States with a relatively high percentage of children entering foster care at age 12 or younger also had a relatively low percentage of children reentering foster care (Pearson’s r = -0.421). Compared to younger children, older children may present very different challenges to States with regard to meeting their short- and long-term needs and addressing the needs of their families.

CHANGES IN STATE PERFORMANCE WITH REGARD TO ACHIEVING TIMELY REUNIFICATIONS

Table III-2 provides findings with regard to changes in performance on the measures relevant to child welfare outcome 4: reduce time in foster care to reunification without increasing reentry. The change in State performance over time was assessed by calculating a percent change in performance.⁴⁵ Using a percent change calculation permits an assessment of the extent of change occurring over time by taking into account the size of the percentages being compared. For example, a change from 6 percent in 2002 to 7 percent in 2005 represents a change of only 1 percentage point; however, it reflects a +16.7 percent change. In contrast, a change from 75 percent to 80 percent represents a change of 5 percentage points, but only a +6.7 percent change.  Consistent with the Department's approach in prior Child Welfare Outcomes Reports, a percent change of 5.0 or greater was used as an indicator that meaningful change occurred. That is, for purposes of the analyses presented in this chapter, if the percent change in performance from 2002 to 2005 was not at least 5.0 in either direction (i.e., positive or negative), a determination was made that there was “no change” in performance.

Change in performance for measure 4.1 is assessed from 2002 to 2005. Change in performance on the other measures is assessed from 2004 to 2005 because data are not available for these measures in prior years.

Table III-2. Number and percentage of States exhibiting improvement, decline, or no change in performance over time on measures pertaining to outcome 4: Reduce time in foster care to reunification without increasing reentry

As shown in table III-2, there are no national patterns of improvement or decline in performance on these measures. Regardless of the measure, some States exhibited improvement, some declined in performance, and some exhibited no change. States were least likely to exhibit a change in performance on outcome measure C1.1.

TIMELINESS OF ADOPTIONS

The assessment of timeliness of adoptions addresses national child welfare outcome 5: reduce time in foster care to adoption. Table III-3 presents summary data regarding State performance in 2005 on the measures relevant to this outcome. The table includes one of the measures originally developed for the Child Welfare Outcomes Reports (measure 5.1) as well as five new measures pertaining to adoption recently developed for use in the second round of the CFSRs (measures C2.1, C2.2, C2.3, C2.4, and C2.5).⁴⁶

Table III-3. Summary of State performance in 2005 on the measures pertaining to achieving adoptions in a timely manner

* Chapter I and appendix B provide complete definitions of each measure.

Outcome measures 5.1, C2.1, and C2.2 focus on the length of time in foster care of children who are discharged to adoption. Performance on these measures in 2005 suggests that achieving timely adoptions is a challenge for all but a few States. As shown in table III-3, the percentage of adoptions that occur in less than 12 months from the child’s entry into foster care is quite low in almost all States (measure 5.1, median = 3.6 percent). Three States, however, had relatively large percentages of adoptions occurring in less than 12 months. These were Utah (34 percent), Iowa (32 percent), and Wyoming (20.6 percent).

The data in table III-3 indicate that the percentage of adoptions occurring in less than 24 months from the child’s entry into foster care also is fairly low (measure C2.1, median = 29.1 percent), although higher than the percentage for measure 5.1. There were only three States, for example, in which at least 50 percent of adoptions occurred in less than 24 months of the child’s entry into foster care (Colorado, Iowa, and Utah). In addition, across States, the median length of stay in foster care for children adopted is 31.5 months (measure C2.2). In about 25 percent of the States, the median length of stay of children adopted is at least 3 years.

Two measures reported in table III-3 (measures C2.3 and C2.4) assess the timeliness of adoptions using a longitudinal analysis. The measures focus on children who were in foster care for at least 17 months on the first day of the fiscal year and follow them to determine: (1) the percentage that are adopted by the end of the fiscal year (measure C2.3) and (2) the percentage who became legally free for adoption in the first 6 months of the fiscal year. The data presented in table III-3 suggest that, across States, only a small percentage of children in foster care for 17 months or longer at the start of 2005 were adopted by the end of that year (measure C2.3, median = 20.9 percent).⁴⁷ The data also indicate that, across States, an even smaller percentage of children in foster care for 17 months or longer on the first day of 2005 became legally free for adoption within the first 6 months of that year (measure C2.4, median = 8.8 percent). As shown in the range for this measure, in the highest performing State, only 23.7 percent of the children in foster care for 17 months or longer became legally free for adoption in the first 6 months of 2005.

Measure C2.5 follows children for a 12-month period from the time that they became legally free for adoption to identify the percentage of children whose adoptions were finalized during that time period. As shown in table III-3, the range in performance on this measure across States is extensive (16.4 to 87.7 percent). For example, in three States (Maine, South Dakota, and West Virginia), less than 25 percent of the children who were legally free for adoption were adopted less than 12 months after becoming legally free for adoption. However, in three States (District of Columbia, New Hampshire, and Utah), 80 percent or more of the children who were legally free for adoption were adopted in less than 12 months after becoming legally free. An additional key finding with regard to the timeliness of adoptions for 2005 was a moderate correlation between performance on the measure of adoptions occurring in less than 24 months (measure C2.1) and performance on the measure of reunifications occurring in less than 12 months (measure C1.1) (Pearson’s r = 0.51). Many States with a relatively high percentage of children adopted in less than 24 months also had a relatively high percentage of children reunified in less than 12 months. This suggests that some States are more successful than others in achieving timely permanency for children in foster care, regardless of whether it is through reunification or adoption.

CHANGE IN PERFORMANCE WITH REGARD TO TIMELINESS OF ADOPTIONS

Table III-4 provides key findings with regard to changes in performance on the measures assessing timely adoptions. Change in performance for the original outcome measure 5.1b (adoptions occurring in less than 24 months) is assessed from 2002 to 2005. Change in performance on the new outcome measures is assessed from 2004 to 2005 because data are not available for these measures in prior years.

Table III-4. Number and percentage of States exhibiting improvement, decline, or no change in performance on measures pertaining to outcome 5: Reduce time in foster care to adoption

The data presented in table III-4 reveal a national pattern of change with regard to the percentage of adoptions occurring in less than 24 months. For measure 5.1b, 63 percent of the States (N= 33) demonstrated improved performance from 2002 to 2005.⁴⁸ There are no noteworthy patterns of change in performance for the other measures.

SUMMARY OF FINDINGS REGARDING ACHIEVING REUNIFICATIONS AND ADOPTIONS IN A TIMELY MANNER

Achieving timely reunifications

For 2005, the measures that assess the timeliness of reunification from the point of discharge from foster care suggest that, in most States, many children who exit foster care to reunification are reunified in a timely manner. However, the measure that assesses timeliness of reunification from the perspective of children entering foster care (i.e., an entry cohort measure), indicates that, in most States, the majority of children entering foster care in the last 6 months of 2004 were still in foster care 12 months later. Although not all of these children will eventually exit foster care to reunification, many of them will be reunified, and they will have been in foster care for a year or longer.

The data for 2005 also suggest that many States experience challenges in ensuring that children who are reunified do not reenter foster care within the year. A consistent finding of the data analyses for the Child Welfare Outcomes Reports is that many States that have a high percentage of reunifications occurring in less than 12 months from the child’s entry into foster care also have a high percentage of children who reenter foster care in less than 12 months from the time of reunification. This is an important finding because it raises the possibility that, at the time of reunification, not all of the problems that resulted in the child’s initial entry into foster care were adequately resolved, or that new problems arose at the point of reunification that were not sufficiently addressed by the agency.

An additional consistent finding is that many States with a relatively high percentage of foster care reentries also had a relatively high percentage of children entering foster care who were adolescents (age 13 or older). The challenges that these children present to State child welfare systems with regard to meeting the reunification needs of the children and their families are quite different than those encountered in working with younger children and their families.

There was no clear pattern nationally with regard to a change in performance on most of the measures pertaining to reunification. Although many States exhibited improved performance on the measures, many other States exhibited declines or no change in performance.

Achieving timely adoptions

The data for 2005 with regard to timely adoptions indicate that achieving adoptions in a timely manner remains a significant challenge for almost all States. The exceptions to this pattern are Colorado, Iowa, and Utah where approximately 50 percent or more of the adoptions finalized in 2005 occurred less than 24 months from the time of the child’s entry into foster care. A clear area for further investigation is why these States are so much more successful than other States in achieving timely adoptions.

The challenges in achieving adoptions in a timely manner are apparent from the data for all of the relevant measures. However, 63 percent of States exhibited an improvement in performance in the percentage of adoptions occurring in less than 24 months from the child’s entry into foster care. This indicates that, despite the challenges, many States are making concerted efforts to expedite the adoption process for children.

⁴² For the Child Welfare Outcomes Reports and the CFSRs, children are considered to be reunified if the discharge reason provided to AFCARS is either (1) reunified with parent or primary caretaker or (2) living with other relatives. Back

⁴³ Four measures comprise composite 1: timeliness and permanency of reunifications. More information regarding this composite is provided in appendix B. Back

⁴⁴ As noted in chapter I, in prior Child Welfare Outcomes Reports, reentry into foster care was measured by the percentage of children entering foster care who were reentering in less than 12 months. This measure was intended as a proxy measure of reentry because, at the time that the measures were developed for the Child Welfare Outcome Report, it was not possible to track children in AFCARS across fiscal years. The effectiveness of the original measure 4.2 as a proxy measure is supported by the high correlation between performance on measure 4.2 and new measure C1.4 (Pearson’s r = 0.872). Measure C1.4 will now be reported as the key measure for reentry because it is more precise than the proxy measure. Back

⁴⁵ Percent change is calculated by subtracting “old” data from “new” data, dividing by old data and multiplying by 100. For example, if maltreatment recurrence was 9.2 in 2002 and 7.6 in 2005, the percentage change would be [(7.6-9.2)/9.2] x 100 = -17.4 percent change. Back

⁴⁶ These five new measures comprise composite 2: timeliness of adoptions. More information on this composite is provided in appendix B. Back

⁴⁷ Measures C2.3 and C2.4 exclude children retrospectively from the denominator if the children exited from foster care during the specified time period to either reunification or guardianship. Back

⁴⁸ Original Child Welfare Outcomes measure 5.1 incorporates a range of time periods. The 24-month time period was chosen for this analysis because data are available for 2002, 2003, 2004, and 2005. This time period also is captured in measure C2.1. Back