Individually, the nine states present very different patterns of exit distribution from foster care. Illinois and California had the largest proportion of children remaining in care, which meant that they had the greatest share of long duration foster spells. Over one-third of the spells in Illinois and one-fifth of the spells in California had not been completed by the end of 1997. By contrast, in five states (Alabama, Michigan, Missouri, New Mexico, and Wisconsin) only one tenth or fewer of all spells remained open in 1997. To the extent that reunification is related to the length-of-stay in foster care, this is a very significant dynamic to remember during this analysis.
Reunification levels differ by state -- from 28.7 percent of exits in Alabama to 67.2 percent in California and 66.7 percent in Wisconsin. In the other six states, 45-60 percent of all foster care exits were family reunification. This state-to-state variation in the use of reunification to end episodes of foster care placement is substantial. Alabama is a clear outlier, with far lower proportions of reunification exits than any of the other states, but the reunification levels among the remaining eight states still vary widely, from 45 percent to 67 percent of observed exits.
Exits to relatives also differ widely among states. Alabama and Maryland reported the most family discharges (31.6% and 20.4%), while California and Illinois each reported less than 3 percent family exits.4 This is a difficult category to interpret, because it represents differences in foster care practice and in state data definitions. The child welfare agencies in Alabama and Maryland tended to use strategies that supported kinship placements outside of the formal foster care system, while California, Illinois and New York placed a substantial number of foster children in kinship arrangements within the foster care system. When exit to relatives and reunification are combined into a single category of "family exits," the variation between states is somewhat smaller than it is for either exit type considered alone. In part, the higher level of exit to relatives helps to explain why the reunification levels in Alabama and Maryland are very low. It should also be noted that children in care in California, New York, and Illinois (but in kin placements) may be similar to the exits to relative placements elsewhere.
Michigan has the highest rate of completed adoption with almost one-fifth (.193) of foster care spells ending with adoptive placement. Exits by adoption were least common in Alabama and Wisconsin (.054 and.065).
Few children exit foster care because they reach majority (or "age out"), with Maryland showing the highest proportion at .059.
Illinois (.126) and New York (.099) have the highest rates of runaway exits.
Overall, "other" exits represent almost 15 percent of all foster care spell endings. These are particularly common in Alabama (.306) and Missouri (.224). In both cases, most of the "other" exits are to unknown destinations. It is possible that many of these undefined exits are by reunification, but that the exit reason coded in the data system was represented in some unidentifiable way (such as by a change in legal status).
4. Other evidence suggests that a large share of the exits coded as "other" in Illinois (17.8% of exits) may actually be relative exits.