*4.4. Limitations*

Several limitations should be considered when interpreting results of this study. First, prevalence rates reported are based on a sample of convenience rather than using a population-based sample, although it is important to note that the participants were not recruited based on ASD status. With this in mind, caution should be used when interpreting the prevalence rates reported. Second, participants' ASD status was determined solely based on the ADOS-2, making it impossible to determine whether the findings reported here will be replicable using other diagnostic measures. Third, based on inclusion criteria for the original study, which required a specific level of language (i.e., a minimum of at least occasional three-word phrase speech), our findings may not be representative of all individuals with DS, such as those who may be minimally verbal or nonverbal. This emphasizes the need for replication of these findings using samples that include individualizing with a range of language abilities, utilizing the full range of ADOS-2 modules. The inclusion of a sample representing the full range of language abilities would allow a better understanding of how ASD symptomatology present differentially among individuals with differing language abilities. Moreover, our findings of higher rates of classification for youth with more limited language abilities, highlights the importance of considering language skills relative to ASD symptomatology; studies focused on these

associations in youth with DS who are in the prelinguistic and first words stages of language development are critical to more fully understand the relation between ASD and language skill. Finally, we did not include other non-DS comparison groups. In particular, it would be useful in future studies of ASD in DS to make comparisons with appropriately matched groups of individuals with non-syndromic ASD and ID of a different origin (e.g., fragile X syndrome). Comparisons between individuals with DS + ASD and matched group of individuals with non-syndromic ASD would allow for a deeper understanding of how the DS phenotypes moderates the expression of ASD and the specific phenotypic factors that play a role in that moderation. Although there is a small body of work that examines the differences in presentation of ASD symptomatology among individuals with DS and those with non-syndromic ASD, e.g., [15,18], further research is needed to truly understand which symptoms are attributable to DS phenotype and which are attributable to ASD among individual with DS + ASD. Similarly, there is a fledgling body of work examining differences between individuals with DS + ASD and matched individuals with ID of other etiologies [45]. Given the dearth of research in this area, future research is needed to are needed help identify which aspects of the expression of ASD that are unique to DS rather than common to those with ID.
