Orphanages, Foster Families, and Culture Wars

A struggle over the right way to care for unparented children has been shaping for a number of years now. The difficulties of doing good outcome research on this issue are such that evidence tends to be overborne by values, opening the door to confirmation biases. The current arguments in Russia about orphanages, and the encouragement by some parties of a movement toward foster-family care for all unparented children, are examples of a confusion between systematic evidence and values as reasons for making practical decisions.

Child-care institutions have existed in some form back to at least the Middle Ages, when monasteries accepted children as oblates, cared for them, educated them, and reared them to be part of the monastic group. Often, but not always, orphanages were run by religious or other charitable groups. During the Great Depression in the United States, institutional care was common, not only for genuine orphans, but for children whose parents could not afford to care for them at home. Although some institutions involved cruel experiences for the children, others did excellent jobs and were remembered with pleasure by their graduates. During the 1994 Congressional upheaval in the U.S., Newt Gingrich (I never thought I’d be quoting him!) referred to those successful orphanages and suggested that they might do a better job than foster families do.

But today we have people in a range of countries speaking loudly against institutional care for children and advocating family care systems. How did this shift come about? The initial push came from the revelations about the horrible Ceausescu-era Romanian orphanages. This information became public in 1996, following the death of Ceausescu, when news sources published extremely disturbing photographs of malnourished young children tied to cribs and groups of older children naked.  The term “orphanage” quickly took on a powerful negative coloring when people identified highly inappropriate practices as equivalent to all institutional care for children.

A few years after the Romanian revelations, several researchers became interested in the outcomes for children who were adopted from Romanian orphanages. The English-Romanian Adoptees project, directed in part by Michael Rutter , followed these children into young adulthood and found that on the whole their development was quite satisfactory. An American research group, the Bucharest Early Intervention Project, headed by Charles Zeanah, began in about 2000 to conduct an investigation that would compare young children randomly assigned to remain in the orphanage to others randomly assigned to foster families. Zeanah and the rest of the BEIP group reported later that the fostered children showed better development than those who stayed in the orphanage, and that therefore family care had a better effect on development than institutional care did. This conclusion, repeated in several professional journals, in the magazine Science, and in the publication Zero to Three, as well as in popular news sources, has had a strong influence and has been used to support the argument that foster care or adoption are demonstrably better methods for unparented children than institutional care.

However, as I have pointed out in this blog and in letters published in Science and Zero to Three, the design of the BEIP studies does not permit the conclusion that has been publicized. The Romanian foster families received special training and funding, as well as access to frequent consultation with child development experts. The orphanage staff received none of these, nor were they helped to provide more consistent staff assignments that would allow infants and toddlers to be with a few familiar caregivers. The comparison was thus between the best possible foster care and poor institutional care, so the outcome was not surprising. Indeed, it would be interesting to know what would have happened if the institutional staff had been specially trained, paid more, and had consultants available to them, while the foster families were left to get along as best as they could--  that would have been a meaningful test of the advantages of foster care per se over institutional care.

As I have also pointed out on this blog, there is important research that contradicts the BEIP conclusions. That work, by Kathryn Whetten, was a nonrandomized study of children in low- and middle-income countries that compared development of fostered children with that of children in institutions and did not find an advantage for foster care. Whetten’s work has received much less attention than the BEIP work, presumably because her design was nonrandomized, whereas the BEIP involved random assignment to groups—but the fact is that randomization is used to attempt to isolate a variable, and the remainder of the BEIP design did not successfully isolate the variable of type of care, instead conflating type of care with availability of training and resources. In my opinion, the two studies are about equal in appropriateness of design and outcome validity.

A new study in which Whetten was involved is also relevant to the comparison of institutional and family care. This study (and my thanks to Yulia Massino for calling my attention to it!) is by Christine Gray et al., “Prevalence and incidence of traumatic experiences among orphans in institutional and family-based settings in 5 low- and middle-income countries: A longitudinal study” (http://www.gshpjournal.org/content/3/3/395.abstract). The researchers looked at physical and sexual abuse experiences among children in institutions and in foster care, and found no more such experiences in institution-reared children than in those cared for by foster families.

Gray et al. concluded that “understanding the specific context, and elements contributing to potential harm and benefits in both family-based and institutional care, are essential to promoting the best interest of the child.” This comment should be related to recent ideas in clinical psychology, where professionals are beginning to recognize that there are adverse events in psychology just as there are in medicine, and that these should be investigated, recognized, and reported along with  demonstrated benefits.

Decisions about best practice need to take unwanted events into account as well as considering the level of evidence supporting a practice. Yet, as we see in the Russian disagreements about orphanages and about adoption, as well as in the American movement from reuniting families at all costs some years ago, to the current concern about putting child safety first, many such decisions are driven by value considerations other than the best interest of the child, and adverse events are not mentioned. An egregious example would be the push for multiple adoptions by fundamentalist Christians in the U.S., as described by Kathryn Joyce in The child catchers.

As Gray and her co-authors imply, one size does not fit all unparented children. When anyone suggests that there is a single best care approach, we should ask what value considerations are guiding this claim. Culture wars are not a way to provide the best outcomes for children, who are likely to experience “collateral damage”.