*3.1. Article Search Results*

A total of 1938 articles were identified from the initial search. Due to the search term "regression" without further specification, our search was deliberately over-inclusive. It was felt that due to the inconsistencies of terminology and labelling used in referring to this condition this was a necessary step in order to capture as many related articles as possible and to achieve our aim of collating the various terminologies used. Title, keyword and abstract review eliminated the vast majority of articles, leaving 57 articles for full-text review. A further four articles were sourced from references and full-text screening with

the same inclusion criteria applied to these articles was completed. Ultimately 14 articles were retained. A full-text version could not be sourced for three articles, and two others were sourced at a later date, leaving a final total of 13 articles for inclusion in this review. The search pathway is shown in Figure 1 and the final included articles are summarised in Table 1. the same inclusion criteria applied to these articles was completed. Ultimately 14 articles were retained. A full-text version could not be sourced for three articles, and two others were sourced at a later date, leaving a final total of 13 articles for inclusion in this review. The search pathway is shown in Figure 1 and the final included articles are summarised in Table 1.

review. A further four articles were sourced from references and full-text screening with

**Figure 1.** Conceptual framework showing article retrieval and inclusion process. Truncation was used in the search term Down\* syndrome so as to capture alternate references such as Down's and Downs syndrome. **Figure 1.** Conceptual framework showing article retrieval and inclusion process. Truncation was used in the search term Down\* syndrome so as to capture alternate references such as Down's and Downs syndrome.



### Cardinale et al. (2018) [19] 4 4 3:1 Range 17–25 years *3.2. Additional Comments and Exclusions*

Santoro et al. (2019) [4] 0 35 53% female 9–34 years Miles et al. (2020) [20] 7 0 6:1 18–33 years Articles where the age range of the participants extended outside of our 35-year upper age limit underwent an additional level of scrutiny. In research articles where cases could

Mircher et al. (2017) [18] 0 30 20:10 Range 12–30 years

not be distinguished from each other, these were excluded, however in case studies or where this information was available, only the individual cases that did not meet our inclusion criteria were excluded from that particular review. Further exclusions included specified autistic regression and progression to AD. Two individual cases were excluded from this review based on the above criteria respectively. A ten-year-old female [12] and a 44-year-old male [10]. Table 1 reflects the number of participants after these additional exclusions. Case-control studies identified that did not distinguish between young children and adolescents could not be included in the review [4,21,22].
