Early adolescent outcomes of institutionally deprived and non-deprived adoptees. III. quasi-autism

Background: Some young children reared in profoundly depriving institutions have been found to show autistic-like patterns, but the developmental significance of these features is unknown.Methods: A randomly selected, age-stratified, sample of 144 children who had experienced an institutional upbringing in Romania and who were adopted by UK families was studied at 4, 6, and 11 years, and compared with a non-institutionalised sample of 52 domestic adoptees. Twenty-eight children, all from Romanian institutions, for whom the possibility of quasi-autism had been raised, were assessed using the Autism Diagnostic Interview-Revised (ADI-R) and the Autism Diagnostic Observation Schedule (ADOS) at the age of 12 years.Results: Sixteen children were found to have a quasi-autistic pattern; a rate of 9.2% in the Romanian institution-reared adoptees with an IQ of at least 50 as compared with 0% in the domestic adoptees. There were a further 12 children with some autistic-like features, but for whom the quasi-autism designation was not confirmed. The follow-up of the children showed that a quarter of the children lost their autistic-like features by 11. Disinhibited attachment and poor peer relationships were also present in over half of the children with quasi-autism.Conclusions: The findings at age 11/12 years confirmed the reality and clinical significance of the quasi-autistic patterns seen in over 1 in 10 of the children who experienced profound institutional deprivation. Although there were important similarities with ‘ordinary’ autism, the dissimilarities suggest a different meaning.Despite the evidence that autism constitutes a disorder that is strongly influenced by genetic factors, patterns that appear similar to autism have been reported in both congenitally blind children (Brown, Hobson, & Lee, 1997) and in children exposed to profound early institutional deprivation (Hoksbergen, ter Laak, Rijk, vanDijkum, & Stoutjesdijk, 2005; Rutter et al., 1999). At 4 years of age, the pattern of autistic-like behaviour was indistinguishable from that seen in a prospectively studied sample of ‘ordinary’ children with autism investigated by Catherine Lord and her colleagues. By age 6 years, the quasi-autistic features in the sample of children from Romanian institutions had diminished and a number of atypical features were noted. The children showed more flexibility in communication than would ordinarily be expected with autism; several showed substantial social approach (albeit of an abnormal kind), and a few showed indiscriminately friendly behaviour of a kind that is more ordinarily associated with disinhibited attachment. Three of the children with quasi-autistic patterns were also severely mentally retarded, but the remainder were not. The quasi-autistic pattern seen in one girl was quite transient and it was regarded as not falling into the same group. Thus, the overall sample at that time comprised 3 children with quasi-autistic patterns associated with severe mental retardation, 7 with similar patterns that were not accompanied by severe retardation, and 1 child with a transient pattern who was excluded from further consideration. In addition there were another 10 cases on whom less information was available, who seemed to show milder quasi-autistic patterns. A systematic follow-up of this group of children was necessary in order to examine a number of outstanding issues concerning the meaning of the observed pattern.The original report commented on the high frequency of very marked circumscribed interests of a repetitive stereotyped pattern that had developed after the children had left institutions and joined their adoptive families. These interests made the picture rather different from the social deficits seen in severely deprived children within an institutional setting. We hypothesised that the impaired social relationships might have something in common with the so-called indiscriminate friendliness associated with disinhibited attachment seen in children who have experienced institutional rearing (Rutter et al., 2007), and the current follow-up tests the hypothesis that, by the time the children had reached age 11 years, the distinctively autistic features would have faded away and be replaced by disinhibited attachment disorder.Second, the quasi-autistic pattern might be a non-specific consequence of the cognitive impairment seen in some of the children. If so, the persistence of the quasi-autistic pattern should be evident only in those who showed cognitive impairment at age 11.Third, the profound deprivation might have led to an impairment in mentalising skills and in psychological perspective taking (see Frith, 2003), and that it was these features that had underpinned the quasi-autistic pattern. If so, any persistence of the quasi-autistic pattern should be confined to the children who showed such problems in social cognition.A further query was whether or not the quasi-autistic pattern constituted a categorically distinct syndrome or, rather, whether it was simply the end of a continuum of similar features seen in the institution-reared children as a whole. The follow-up of the children with milder features, together with the characteristics of the children with raised scores on the Social Communication Questionnaire (a scale devised to assess autistic-like features – Berument, Rutter, Lord, Pickles, & Bailey, 1999; Rutter, Bailey, & Lord, 2003a) should provide the answer.Because the possible association between institutional deprivation and autism became evident only after the study had started, the sampling for the children to be studied in detail at age 6 using standardised diagnostic instruments for autism was less thorough with the older children at the beginning of the study than it was later on. Therefore, a prior methodological point that needed to be dealt with before proceeding with the more substantive issues was whether or not cases with quasi-autistic features had been missed. Accordingly, for the follow-up at age 11 years, all children who received scores at or above 14 on the Social Communication Questionnaire (SCQ), and/or for whom any of the research assessments suggested the possibility of autism, were routinely included in the special follow-up sample