In this study, adolescents and adults dually diagnosed with fragile X syndrome and autism manifested a behavioral phenotype that overlapped with but was distinct from their age peers with either fragile X syndrome-only or autism-only. To our knowledge, this is the first study to describe how the behavioral phenotype of comorbid fragile X syndrome and autism in adolescence and adulthood differs from these two comparison groups, and the findings show the unique vulnerabilities of the dually diagnosed group. Based on findings from prior research on children with fragile X syndrome (Kaufmann et al., 2004
; Rogers et al., 2001
), we hypothesized that adolescents and adults dually diagnosed with fragile X syndrome and autism would be more similar to individuals with autism-only than to individuals with fragile X syndrome-only in terms of their autism symptoms. Although this was the case for overall level of autism symptoms, important differences emerged when examining the three core domains of autism separately. In terms of the scores in the social reciprocity and communication domains, the autism-only group was the most impaired, followed by the dually diagnosed fragile X syndrome and autism group, and then the fragile X syndrome-only group. One notable exception is that both fragile X syndrome groups were much more likely than the autism-only group to have a friend.
In contrast to the findings of similarity between the dually diagnosed fragile X syndrome and autism group and the autism-only group in overall autism symptoms, the dually diagnosed fragile X syndrome and autism group had higher levels of problems in repetitive behaviors and restricted interests than either the autism-only or the fragile X syndrome-only groups. The repetitive behaviors and restricted interests domain has been shown to have the least amount of improvement from childhood to adolescence for individuals with fragile X syndrome (McDuffie et al., 2010
). This unique presentation of autism symptoms in individuals dually diagnosed with fragile X syndrome and autism highlights the value of examining specific types of behaviors at multiple points in the life course in addition to global and lifetime ratings for understanding the nuances of the behavioral phenotype of individuals dually diagnosed fragile X syndrome and autism (Demark et al., 2003
). In addition, the differences in symptoms between those with fragile X syndrome and autism and those with autism-only underscore the need to study in more depth the pathways leading to autistic traits in individuals with autism of known and unknown etiologies.
As hypothesized, all groups had elevated levels of behavior problems and psychological symptoms relative to national norms. However, counter to our expectation, adolescents and adults dually diagnosed with fragile X syndrome and autism had the highest level of behavior problems and psychological symptoms relative to the other two groups. This pattern held for specific items derived from two well-validated measures, including socially offensive behavior, withdrawn–inattentive behavior, and uncooperative behavior (from the SIB-R) and intrusive behavior and inattention (from the CBCL and ABCL). These findings suggest that for individuals with fragile X syndrome, a profile of especially high externalizing behavior problems, not necessarily autism-specific symptoms, is associated with an additional diagnosis of autistic disorder. However, individuals in the dually diagnosed fragile X syndrome and autism group also displayed greater impairments in social reciprocity and communication than the fragile X syndrome-only group, suggesting that those with fragile X syndrome and autism in this study exhibited a global profile of social communication symptoms consistent with an autism diagnosis. Together, these findings suggest that for individuals with fragile X syndrome, symptoms may be significant enough to meet diagnostic criteria for autism, but the specific behavioral presentation of impairments is different for individuals dually diagnosed with fragile X syndrome and autism than for individuals with autism alone.
We also hypothesized that adolescents and adults with fragile X syndrome-only would have higher levels of adaptive functioning than those in either the fragile X syndrome and autism group or autism-only group. This hypothesis was only partially supported. We found that both fragile X syndrome groups had fewer difficulties in adaptive behavior than the autism-only group. Unlike past studies comparing children with fragile X syndrome-only to children dually diagnosed with fragile X syndrome and autism (Fisch et al., 2002
; Hernandez et al., 2009
; Kaufmann et al., 2004
), there was not a difference between the two groups of adolescents and adults with fragile X syndrome in the present study on overall adaptive behavior scores. It may be that daily living skills are a relative strength for individuals with fragile X syndrome, at least at this later point in development. In addition, both groups with fragile X syndrome had higher scores on the strengths subscale of the CBCL and ABCL than the autism-only group. These findings, as well as those pertaining to friendships, underscore how although individuals dually diagnosed with fragile X syndrome and autism share many behavioral features in common with or more severe than those with autism alone, the overall profile of strengths and weaknesses for the fragile X syndrome and autism group is unique, at least during adolescence and adulthood. Furthermore, it is important that the elevated levels of behavior problems and autism symptoms in the fragile X syndrome and autism group did not interfere with adaptive functioning in the same way as in the autism-only group.
The pattern of significant covariates yielded additional insights about the factors that are associated with the behavioral profile of adolescents and adults with these developmental disabilities. Consistent with much past research on the impact of intellectual disability on behavior (Totiska & Hastings, 2009
), having an intellectual disability was associated with poorer functioning during adolescence and adulthood in terms of behavior problems, adaptive behavior, and autism symptoms. In addition, child age was significantly associated with multiple aspects of behavioral functioning in the present study. Specifically, being older was related to fewer restricted–repetitive behaviors, better adaptive functioning, and fewer total behavior problems. The findings regarding the association between age and restricted–repetitive behaviors are similar to what has been found in cross-sectional and longitudinal studies of individuals with autism, in that these types of behaviors often partially abate with age (Esbensen, Seltzer, Lam, & Bodfish, 2009
; Shattuck et al., 2007
). Similarly, adaptive behavior skills have been shown to improve with age for children with fragile X syndrome (Hatton et al., 2003
). These findings suggest that there may be continued improvements in several behavioral domains for individuals with fragile X syndrome as they mature into adulthood.
In contrast, gender and residential status were not significant predictors of total scores of behavior in the present analyses, highlighting the possible relative importance of other covariates (e.g., intellectual disability, age) in contributing to behavioral outcomes. Particularly for individuals with fragile X syndrome, factors related to gender (e.g., intellectual disability status) rather than gender itself may be better predictors of symptoms. In a study of children with fragile X syndrome, Hatton et al. (2006)
found that girls had fewer autism symptoms than boys, although this relationship was nonsignificant when the level of FMRP was included in the model. Large, longitudinal studies of individuals with varying lengths of CGG repeat and levels of FMRP are needed to further disentangle the relationships among these covariates during adolescence and adulthood.
The present study was not without weaknesses. First, there was a wide range of ages of participants, limiting our ability to understand the phenotype at specific points in the life course. In addition, the autism-only group was considerably older than the fragile X syndrome groups. However, age was statistically controlled in the analyses, and the variability in age made it possible to interpret the effect of age on the behavioral phenotype. Second, the sample had little racial and ethnic diversity; however, the fragile X syndrome sample had excellent geographic representation, with individuals residing in 38 different U.S. states and 1 Canadian province. Third, IQ scores were not available on all sample members, which is why we used intellectual disability status as a control variable instead of a full-scale IQ score. This is a limitation of the current analysis, given that past research has demonstrated that individuals dually diagnosed with fragile X syndrome and autism have lower IQs compared with individuals with fragile X syndrome-only, even in samples including only individuals with intellectual disability (Lewis et al., 2006
). Last, it is important to note that autism symptoms were measured using the ADI-R in the autism-only group and using the SCQ in the fragile X syndrome groups. Although the items are identical between the two measures, the ADI-R items were re-coded to reflect absence–presence to match the coding used in the SCQ.
The present study also had several valuable strengths. Studies of individuals with fragile X syndrome often only include males and have small sample sizes, whereas the current study explored the behavioral phenotype in a large sample of males and females with fragile X syndrome. In addition, few prior studies have focused specifically on adolescents, with very few studies examining the behavioral profiles of adults with fragile X syndrome (for a notable exception, see Hartley et al., 2011
). In the present study, we addressed this gap by examining the behavioral phenotype of fragile X syndrome in both adolescents and adults. We elaborated on the phenotype by dividing those dually diagnosed with fragile X syndrome and autism from those with fragile X syndrome-only. In addition, as noted above, in the present study we incorporated an analysis of a wide range of positive and negative behaviors (adaptive behaviors and behavior problems) and symptoms (autism symptoms and general psychopathology) and found that not all domains operated the same way, highlighting the value of exploring a variety of behaviors in describing the phenotype.
In conclusion, the present study informs the debate about whether autism in individuals with fragile X syndrome is qualitatively similar to idiopathic autism. The pattern of findings reveals that there is no simple answer to the question posed by Bailey et al. (2004)
. On the one hand, we found that adolescents and adults dually diagnosed with fragile X syndrome and autism tended to have a presentation of communication and social reciprocity impairments and internalizing symptoms more similar to individuals with autism-only than to those with fragile X syndrome-only. On the other hand, the dually diagnosed individuals exhibited higher levels of repetitive behaviors, restricted interests, and challenging behavior problems than those with either fragile X syndrome-only or autism-only. Therefore, we conclude that autism in individuals with fragile X syndrome is at least as impairing as idiopathic autism and, in some respects (repetitive and challenging behaviors), even more impairing. Counterbalancing this profile of vulnerability is the pattern of greater strengths, more friendships, and better adaptive functioning in individuals dually diagnosed with fragile X syndrome and autism than those with autism-only, which only further complicates the picture.
Nevertheless, the elevated level of behavior problems in the dually diagnosed fragile X syndrome and autism group highlights that individuals with fragile X syndrome and comorbid autism may require additional behavioral and/or pharmacological interventions during adolescence and adulthood. Furthermore, parents of individuals dually diagnosed with fragile X syndrome and autism may need extra supports and services to help address their older children’s behavioral needs. It is notable that mothers of adolescents and adults dually diagnosed with fragile X syndrome and autism may be especially vulnerable to the stresses of caregiving due to a heightened level of challenging behaviors and autism symptoms in their children as well as to their own genetic risk as premutation carriers (Bailey et al., 2008
; Coffey et al., 2008
). This double vulnerability makes the need for adequate family-centered planning and services for families of individuals with fragile X syndrome all the more imperative.