diff --git "a/assets/Books Chunks/Encyclopedia of autism spectrum disorders/book0_cleaned_chunk_12.txt" "b/assets/Books Chunks/Encyclopedia of autism spectrum disorders/book0_cleaned_chunk_12.txt"
new file mode 100644--- /dev/null
+++ "b/assets/Books Chunks/Encyclopedia of autism spectrum disorders/book0_cleaned_chunk_12.txt"
@@ -0,0 +1,356 @@
+Autism Cymru
+**Major Areas or Mission Statement**
+Autism Cymru is Wales’s pioneering national charity for Wales. It is a practitioner-led charity set up in 2001 to improve the lives of people in Wales with an autistic spectrum disorder and their families. It has a dedicated national brief in Wales and in the projection of Welsh practice within and outside Wales. Autism Cymru takes the view that everyone with an autistic spectrum disorder in Wales should receive a service appropriate to their assessed needs, whatever their age and wherever they live. In order to achieve this, Autism Cymru actively promotes at both national and local levels the practice of strategic, collaborative, and multidisciplinary partnerships and highly focused coordination of services to people with autistic spectrum disorders and their families.
+Autism Cymru’s primary task was successfully to influence the Welsh Assembly Government to establish the world’s first government-led strategy for autism, which was launched at Autism Cymru’s third International Autism Conference in Cardiff in April 2008. Autism Cymru’s Chief Executive, Hugh Morgan OBE, heads up the implementation of the Assembly Government’s Action Plan for Autism.
+
+**Landmark Contributions**
+Wales is the only country in the world with a national strategy for autism.
+
+**Major Activities**
+Autism Cymru runs the pioneering bilingual websites, Awares (www.awares.org). Every November, Autism Cymru’s editor, Adam Feinstein, runs the Awares international online autism conference (www.awares.org/conferences), the largest event of its kind in the world, with more than 60 world autism experts taking part, along with thousands of delegates. Professor Simon Baron-Cohen has called this event “the finest online conference on the planet.” Autism Cymru together with Autism Northern Ireland, Scottish Society for Autism, and the Irish Society for Autism has launched the Celtic Nations Autism Partnership. This will lead to shared opportunities for joint working in Northern Ireland, Scotland, and Republic of Ireland, in addition to Autism Cymru’s existing work in Wales.
+
+Autism Cymru carries out research in partnerships with universities in Wales. In 2010, Professor Sue Leekam became the first chair of autism at Cardiff University and head of the new Welsh Autism Research Centre, based at the university’s school of psychology and supported by Autism Cymru. Autism Cymru works in partnership with Mudiad Ysgolian Meithrin with funding from Children in Need to train Welsh medium playgroup leaders across Wales. Autism Cymru works in partnership with Autism Northern Ireland (PAPA) on UK and European campaigns to improve the lives of those with autism and to project best practice in each country. Autism Cymru works with local authorities and local health boards to develop local strategies for autism. Autism Cymru works with local education authorities in Wales to deliver its Inclusive Schools and ASDs: Whole School Training and Research Project. Autism Cymru works in partnership with the North Wales Police and Dyfed Powys Police to operate the Emergency Services ASD Attention Card Scheme which raises awareness of autism among members of the emer-gency services in Wales.
+Autism Cymru has worked with Bro Morgannwg NHS Trust on a research project connected to the criminal justice system. Autism Cymru operates the AWARES EDUNET website and School Fora for education professionals. Autism Cymru publishes books and bilingual infor-mation booklets for professionals, parents, and peo-ple with autism. Publications include All About Autistic Spectrum Disorders and My Brother Gwern, a book for siblings of children with autism which won an award at Welsh Language in Healthcare Awards 2006. Autism Cymru’s work also takes place on an international stage and with European partners, including Autism-Europe. For example, its 2009–2012 European-funded Deis Cyfle project (Opportunities for people with autism in education and employment) reached out to over 5,700 people across Wales and Ireland. The charity is also the sole national autism charity governed by those living in Wales. Autism Cymru’s Chair is Professor Bill Fraser CBE. Its Patron is Lord Dafydd Wigley and its President is Dame Stephanie Shirley.
+
+Autism Diagnostic Interview-Revised
+**Abbreviations**
+ADOS: Autism diagnostic observation schedule
+ASD: Autism spectrum disorders
+
+**Synonyms**
+ADI-R
+
+**Description**
+The Autism Diagnostic Interview-Revised (ADI-R; Le Couteur et al. 2003; Lord et al. 1994) is a standardized, semistructured, investigator-based interview administered by trained clinicians to parents or caregivers of individuals referred for a possible autism spectrum disorder (ASD). The ADI-R includes 93 items in three domains of functioning: communication; reciprocal social interactions; and restricted, repetitive, and stereotyped patterns of behavior, as well as other aspects of behaviors. All items in the ADI-R are coded for current and past behavior. Current refers to whether the behavior has occurred in the past 3 months. For some items, “past” refers to whether the behavior “ever” occurred, whereas others ask whether the behavior was present during a specifically defined period between 4 and 5 years of age (referred to as “most abnormal 4 to 5”).
+
+Up to 42 of the interview items are systemati-cally combined to produce a formal, diagnostic algorithm for autism based on the ICD-10 (World Health Organization [WHO] 1990) and DSM-IV (American Psychiatric Association [APA] 1994) criteria as specified by the authors. In addition to the three domains of behavior, there is a fourth domain, abnormality of development evident at or before 36 months, to indicate whether the child meets criteria for age of onset. Each domain has a cutoff; a child must meet or exceed cutoffs in all four areas to receive an ADI-R classification of “autism.” Separate cutoffs are available for the communication domain, depending on whether or not the child is verbal (defined as showing “functional use of spontaneous, echoed, or stereo-typed language that, on a daily basis involves phrases of three words or more that at least some-times include a verb and are comprehensible to other people,” a score of 0 on item 30 overall level of language). Other criteria including using lower cutoffs with the same set of items have been used to create an algorithm for broader classification of autism spectrum disorders (ASD) as in several collaborative studies (Dawson et al. 2004; Lainhart et al. 2006; Risi et al. 2006). The diag-nostic algorithm for children 4 years old and above is based on the “ever/most abnormal” codes, but current behavior algorithm forms are available to facilitate a clinical diagnosis for children from 2 years old and above.
+
+A toddler version of the ADI-R was also devel-oped several years ago to provide descriptive data for research with children under 4 years of age. The Toddler ADI-R has a total of 125 items, including 32 new questions and codes about the onset of autism symptoms and general develop-ment. Other items are identical to the ADI-R, with the exception that the Toddler ADI-R does not have codes for behaviors between 4 and 5 years of age.
+
+Previous analyses suggested that the diagnos-tic algorithm was useful for children with a non-verbal mental age above 2 years (Le Couteur et al. 1989; Lord et al. 1994; Rutter et al. 2003). Thus, the interview had been appropriate for the diag-nostic assessment of any person within the age range extending from early childhood to adult life, provided that they have a nonverbal mental age above 2 years. Recently, however, newly devel-oped algorithms for toddlers and young pre-schoolers have shown improved predictive validity compared to the preexisting algorithms for young children from 12 to 47 months of age (Kim and Lord 2011). These algorithms extend the use of the ADI-R to children as young as 12 months and a nonverbal developmental level of at least 10 months. In addition, these new algorithms include items present in both the tod-dler and standard versions of the ADI-R, allowing for use of the algorithms with either version.
+
+Most items in the ADI-R relate to behaviors that are rare in individuals who do not have ASD and/or who do not have profound intellectual dis-abilities. Thus, numerical estimates of the scores of typically developing children based on general population have not been obtained. However, there have been several comparisons to children and adolescents with other disorders, which have been used in the development of the diagnostic algorithms (Le Couteur et al. 1989; Lord et al. 1994; Kim and Lord 2011). Researchers have used individual domain scores or an overall total of the three domains as estimates of autistic symp-tom severity, though the validity of this approach has not been directly tested. Scores have been published for many research populations but not yet systematically dimensionalized.
+
+**Historical Background**
+The ADI was first developed in 1989 (Le Couteur et al. 1989), which was modified in 1994 (Lord et al. 1994). The 1994 version was somewhat shorter than the original in order to make the interview more feasible in both clinical and research settings. The current version of the ADI-R was published in 2003 by Western Psy-chological Services.
+
+The development of the toddler version of the ADI-R was completed in 2006 for research pur-poses. Following the development of the toddler version of the ADI-R, there was an increase in demand for diagnostic instruments for very young children, which prompted the development of the new diagnostic algorithms for toddlers and young preschoolers (Kim and Lord 2011). The final algorithms for toddlers and young preschoolers contain fewer items than the original algorithms and are appropriate for use with children 12–47 months of age.
+
+**Psychometric Data**
+Psychometric properties for the original ADI were reported for a sample of 16 children and adults with autism and 16 children and adults with intel-lectual disabilities; each group included individ-uals that spanned wide ranges of age and performance IQ (with a mean age of 12.28 years and a standard deviation of 3.43 from a perfor-mance IQ of 43 to 71). Participants were carefully selected and blindly interviewed and coded. Interrater reliability was assessed, with multirater kappas ranging from 0.25 to 1 for each item. Intraclass correlations were above 0.94 for all subdomain and domain scores. The majority of individual items showed good discriminative validity between the autism group and the group of individuals with nonautism intellectual disabil-ities (Le Couteur et al. 1989).
+
+Psychometric properties for the development of the algorithms for the current ADI-R were based on a sample of 25 children with autism and 25 children with intellectual disabilities who were carefully selected and blindly interviewed and coded (Lord et al. 1994; Rutter et al. 2003). These children ranged in chronological age from 36 to 59 months, with nonverbal mental ages ranging from 21 to 74 months. Using a sample of 10 children, interrater reliability was assessed; multirater kappas ranged from 0.63 to 0.89 for each item. Using the same sample, intraclass cor-relations were above 0.92 for all subdomain and domain scores. In addition, after the initial stan-dardization of the ADI-R in 1989, a separate sam-ple of 53 children with autism and 41 nonautistic children with intellectual disabilities or language impairments was used to further assess the valid-ity of the ADI-R (Lord et al. 1993). The results of the study showed that the interrater reliability was as high as the initial study, with multirater kappas ranging from 0.62 to 0.96 for individual items. Test-retest reliability was also very high, with all coefficients in the 0.93–0.97 range.
+
+The majority of individual items in the current ADI-R showed good discriminative validity between children with autism and children with intellectual disabilities (see Lord et al. 1994). The existing algorithms differentiated children with autism over 36 months of age from children with nonspectrum disorders, showing high sensitivity and specificity (both over 0.90). Further analyses of data from preschool children revealed that the ADI-R algorithms differentiated children over 2 years with ASD from those with other develop-mental disorders. However, for children under 2 years, discrimination between nonverbal chil-dren with ASD and nonverbal children without ASD was poor, resulting in low specificity, espe-cially for children with mental ages under 18 months, (Lord et al. 1993).
+
+In a more recent study including a larger sam-ple (Risi et al. 2006), the ADI-R showed high sensitivity (above 80%) for children with ASD under 3 years of age, but lower specificity for the comparison of nonautism ASD versus non-spectrum disorders (around 70%). Ventola et al. (2006) reported that, for children between 16 and 37 months of age, the diagnostic classifications made based upon the ADI-R algorithm resulted in lower sensitivity than those made using the Autism Diagnostic Observation Schedule (ADOS; Lord et al. 1999), Childhood Autism Rating Scale (CARS; Schopler et al. 1980), or clinical judgment using the DSM-IV criteria. Wiggins and Robins (2008) also found that ADI-R algorithms resulted in poor sensitivity for children in the same age range when the standard cutoff for the RRB domain was included in the diagnostic criteria. Given the low sensitivities and specificities being reported for young children, new ADI-R algorithms were developed for tod-dlers and preschoolers between 12 and 47 months of age using a sample of 491 children with ASD, 136 with nonspectrum disorders (NS), and 67 with typical development (Kim and Lord 2011). The new ADI-R algorithms consist of two different cutoff scores: one for research (more restrictive, higher specificity with lower sensitivity) and one for clinical purposes (more inclusive, higher sensitivity with lower specific-ity). They also include “ranges of concern” for clinical use (discussed below). In this sample, sensitivity using the clinical cutoff ranged from 80% to 94% and specificity ranged from 70% to 81% for the comparison of nonautism ASD vs. NS. Using the research cutoffs, the comparison of nonautism ASD vs. NS resulted in sensitivity ranging from 80% to 84% and specificity ranging from 85% to 90%. Another multi-site study (Kim et al. 2013) using two independent datasets pro-vided by National Institute of Health funded con-sortia, the Collaborative Programs for Excellence in Autism, and Studies to Advance Autism Research and Treatment (n = 641) and the National Institute of Mental Health (n = 167) rep-licated the results from the original psychometric study, including the diagnostic validity and factor structure of the new algorithms for toddlers and young preschoolers (Kim and Lord 2011). Results suggested that the new ADI-R algorithms can be appropriately applied to existing research data-bases with children from 12 to 47 months and down to nonverbal mental ages of 10 months for diagnostic grouping. With a non-US sample, sen-sitivities, especially for those with phrase speech, were lower, using the new algorithms for toddlers and young preschoolers, suggesting that the algo-rithms need to be replicated more with other inde-pendent, non-US samples (de Bildt et al. 2015).
+
+**Clinical Uses**
+The ADI-R offers a profile of a child, adolescent, or adult which includes information regarding reciprocal social interactions, language and com-munication, and restricted, repetitive, and stereo-typed behaviors and interests. Items are scored based on caregivers’ detailed descriptions of the history and behaviors of their child, thus allowing the clinician to gather both quantitative and qual-itative information. One important caveat for clinical users to recognize is that diagnostic clas-sifications based on the algorithms and true clin-ical diagnoses are not the same. Clinical diagnosis is based on multiple sources of information, including direct observations (Le Couteur et al. 2007; Risi et al. 2006; Kim and Lord 2012). Risi et al. (2006) found a better balance of sensitivity and specificity when the ADI-R and the ADOS were used in combination compared to when each instrument was used alone. The combined use of these instruments resulted in sensitivity and spec-ificity of 82% and 86%, respectively, for children with autism compared to children with non-spectrum disorders over age 3 years. For younger children, sensitivity and specificity for the same diagnostic comparison using both instruments were 81% and 87%, respectively. In contrast, when each instrument was used alone, specific-ities ranged from 59% to 72%. Le Couteur and her colleagues (2007) also examined the combined use of the ADOS and ADI-R for preschoolers with ASD using revised ADOS algorithms (Gotham et al. 2007). Consistent with Risi’s 2006 study, the authors found that combining information from both ADOS and ADI-R pro-vided improved diagnostic accuracy compared to either instrument in isolation. Thus, even though the ADI-R provides information about the individual’s history and description of his or her current functioning from a broad range of contexts, the ADI-R alone cannot be used to make a clinical diagnosis.
+
+The diagnostic algorithm cutoffs allow classi-fication of ASD based on patterns of behavior, meeting the current DSM-IVor ICD-10 diagnostic criteria for autistic disorder. In addition to single cutoff scores, the new algorithms for toddlers and young preschoolers provide clinicians and researchers with several different options for the diagnostic classification of young children. For clinical purposes, ranges of concern (little-to-no concern, mild-to-moderate concern, and moderate-to-severe concern) that represent the severity of autism symptoms in young children are also provided. A clinician or a researcher can use these ranges of concern to inform decisions about whether or not a child should be followed up with further assessments or should be quickly referred for treatment services irrespective of diag-nostic cutoffs. Scores that fall in the little-to-no range of concern indicate that the child is reported to have no more behaviors associated with ASD than children in the same age range who do not have ASD. On the contrary, a child who scores in the mild-to-moderate range has a number of behav-iors consistent with, but perhaps not unique to, ASD. For clinical purposes, children in the mild-to-moderate or moderate-to-severe ranges of con-cern should receive further evaluation and follow-up, including other cognitive and language assess-ments, and recommendations for treatment. In addition to ranges of concern, single cutoff score can be used when more strictly stratified groupings are necessary, such as for intervention, neuroimag-ing, or genetic research. These different alterna-tives allow clinicians and researchers to be transparent about the choices they make, recogniz-ing that diagnostic decisions about ASD in very young children are less stable and precise than for older children and adolescents.
+
+In addition to the diagnostic algorithms, the ADI-R includes a current behavior algorithm form that can be used in clinical settings to assess changes that occur during or after interventions or that may reflect increasing developmental matu-rity or changing life circumstances. Because the current behavior algorithm form has not been empirically validated, it is not intended to be used as a diagnostic algorithm. The development of a new algorithm is underway by the authors in anticipation of an updated protocol and algorithm with new criteria. A shorter version of the ADI-R that can be used over the phone is also in the process of being developed and validated.
+
+The ADI-R provides a useful structure to obtain history and understand a caregiver’s per-spective on his or her child’s symptoms associated with ASD. However, it requires substantial prac-tice to administer reliably, and it takes approxi-mately 2–3 h to administer. The ADI-R should only be used by appropriately experienced clini-cians who are familiar with ASD and relevant behaviors. Training workshops and videotapes are available to help clinicians and researchers understand the scoring and administration of the ADI-R. For research use, interviewers must meet standards for reliability.
+
+In a recent effort to identify children with ASD more efficiently, a brief parent interview, Autism Symptom Interview (ASI; 15–20 min), has been designed primarily as a case confirmation tool for ASD (Bishop et al. 2017). The ASI has been based on questions from the ADI-R. Based on school-age children ranging from 5–12 years of age, the verbal algorithm yielded a sensitivity of 0.87 (95% CI = 0.81–0.92) and a specificity of 0.62 (95% CI = 0.53–0.70). When used in conjunction with the ADOS, sensitivity and specificity were 0.82 (95% CI = 0.74–0.88) and 0.92 (95% CI = 0.86–0.96), respectively. Internal consis-tency and test-retest reliability were both excel-lent. Based on these results, the authors have con-cluded that particularly for verbal school age children, the ASI may serve as a useful tool to more quickly ascertain or classify children with ASD for research or clinical triaging purposes. Additional data collection is underway to deter-mine the utility of the ASI in children who are younger and/or nonverbal.
+
+**See Also**
+*   Autism Diagnostic Observation Schedule
+*   Autism Diagnostic Observation Schedule (ADOS): Toddler Module
+
+Autism Diagnostic Observation Schedule
+**Synonyms**
+ADOS
+
+**Description**
+The Autism Diagnostic Observation Schedule (ADOS) is a semi-structured observation scale designed to observe social behavior and commu-nication in children and adults referred for possi-ble diagnosis of autism spectrum disorder (ASD). Originally developed as a research instrument, it became commercially available through Western Psychological Services in 2001 (Lord et al. 1999) and is used widely in clinical, school, community, and research settings. The goal of the ADOS is twofold: to help clinicians and researchers discriminate autism from other disorders and typically developing individuals and to character-ize social and communicative behaviors associ-ated with autism (Lord et al. 1989). It is often used in conjunction with the Autism Diagnostic Interview-Revised (ADI-R; Rutter et al. 2003), a parent interview. When used by a skilled clinician, together, these two instruments form the “gold standard” for the diagnosis of ASD.
+
+The format of the ADOS is unique. It is a structured interaction between an examiner and individual in which the examiner’s behavior is standardized using a hierarchy of structured and unstructured social behaviors. The examiner cre-ates a “social world” in which occasions for spe-cific behaviors are purposefully orchestrated in order to observe the presence – or absence – of an expected response. For example, with an older child or adult with fluent language, the examiner might initiate a conversation and observe whether the individual participates in a reciprocal exchange or asks about the examiner’s experi-ences. With a child or adolescent with limited language, the examiner might observe whether the individual conveys shared enjoyment in an activity, such as bubble play, by smiling, laughing, or requesting for the activity to con-tinue. The ADOS goes beyond measuring the frequency of behaviors and also focuses on the quality of social behavior, allowing the examiner to make informed decisions regarding the pres-ence of features associated with a diagnosis of ASD. Because of the movement between struc-tured and unstructured tasks, and the need for keen observation within such tasks, it is impera-tive the ADOS is administered by a skilled exam-iner familiar with ASD.
+
+The original version of the ADOS (Lord et al. 2000) consists of four modules based on age and language level, with “higher” modules generally requiring more language and social demands. Each module takes approximately 35–60 min to administer. Module 1 is for individuals with a minimum of no speech or the emergence of simple phrases. Module 2 is designed for individuals who use flexible three-word phrases, but are not yet speaking fluently. Modules 3 and 4 are for indi-viduals with fluent speech. For the purposes of the ADOS, three-word phrases are defined as “regular spontaneous meaningful use of three-word utter-ances including a verb,” while fluent speech is defined as “producing a range of flexible sentence types, providing language behavior the immediate context and describing logical connections within a sentence” (Lord et al. 1999).
+
+Though each module of the ADOS has differ-ent language requirements, the overall format and structure is the same. In fact, there is considerable overlap of tasks across modules. In each module, the examiner interacts with the individual, admin-istering a series of tasks, or “presses” for particular social behaviors. Modules 1 and 2 are conducted while moving around a room and include play-based tasks appropriate for young children or individuals with very limited language. Modules 3 and 4 generally take place while sitting at a table and include tasks involving more conversation.
+
+Immediately after the administration of all tasks, the examiner rates the individual’s behavior on items across domains including communica-tion, social interaction, play or imagination, and stereotyped behaviors and restricted interests. Ratings, or codes, are made on an ordinal scale from 0 to 3, with 0 indicating no evidence of abnormality related to autism and 3 indicating definite evidence, such that behavior interferes with interaction. Selected items from each domain are used to generate a diagnostic algorithm. These items were selected for their ability to discrimi-nate between ASD and nonspectrum disorders and also for their relevance to DSM-IV and ICD-10 criteria. A classification of autism or non-autism ASD is made when thresholds on the social affect and restricted and repetitive behavior domains, and a combined social affect and restricted and repetitive behavior total, are exceeded. When combined with information from other sources, including but not limited to a parent interview and clinical judgment, an ADOS classification of autism or ASD may lead to a diagnosis on the spectrum.
+
+Since its publication by WPS in 1999, the ADOS has expanded considerably. Revised algo-rithms for modules 1–3 were developed to improve the instrument’s sensitivity and specific-ity (Gotham et al. 2007), and a toddler module appropriate for children under 30 months old has been available for research purposes (Luyster et al. 2009). The revised ADOS algorithms and the new toddler module were released commer-cially by WPS in 2012 in the second edition of the ADOS (ADOS-2; Lord et al. 2012a, b) (see Table 1 for a summary of ADOS algorithms). Adapted versions of modules 1 and 2 with modi-fied tasks and materials are in development for adolescents and adults with limited language (Hus et al. 2011).The ADOS-Change (ADOS-C; Colombi et al. 2011), a measure using ADOS item descriptions with expanded codes ranging from 0 to 5, has also been created. This measure is scored by watching an unstructured interaction between an adult and child and will be used to measure response to intervention in young children.
+
+**Historical Background**
+The first version of the ADOS (Lord et al. 1989) was intended for individuals between five and 12 years old, with an expressive language level of at least three years. It included only eight tasks, with two sets of materials based on developmental level and chronological age. The validation sam-ple included 20 children and adolescents with autism and 20 children with intellectual disability matched for chronological age, verbal IQ, and gender. The measure showed promise in distinguishing children with autism from those with intellectual disability.
+
+As public awareness of autism increased and more younger and nonverbal children were referred to clinics for diagnostic evaluations, there became a need to develop a “downward extension” of the ADOS that would be appropri-ate for younger children with no-phrase speech. The Pre-Linguistic Autism Diagnostic Observa-tion Schedule (PL-ADOS; DiLavore and Lord 1995) was intended for children less than 6 years old with limited language. It included 12 tasks with 31 overall ratings. All tasks were adminis-tered in the context of play and were informed by the increasing amount of research on early indi-cators of autism, particularly those studies focus-ing on joint attention, functional and symbolic play, imitation, and early patterns of language development. The PL-ADOS was validated on a sample of 63 children with autism or developmen-tal delay and matched for chronological age or language level. Overall, the algorithm was suc-cessful at differentiating autism from develop-mental delay, but its performance was not as good when discriminating verbal children with autism from nonverbal children with develop-mental delay, and children with autism who had some expressive language tended to be underclassified by the instrument.
+
+The ADOS-Generic (ADOS-G; Lord et al. 2000) was developed directly from its original version (Lord et al. 1989) and the PL-ADOS (DiLavore and Lord 1995). It aimed to improve the tendencies to overdiagnose autism in children with insufficient language ability and underdiag-nose children with higher language abilities. Fur-thermore, it sought to extend the current tasks to be appropriate for adolescents and adults. The ADOS-G differed from its predecessors in that it spanned a broader developmental and age range and was the first to introduce the use of modules across different developmental and language levels. It was also the first version to provide continuous scores from ASD to autism, thus mak-ing it applicable for children with broader ranges of social and communication impairments.
+
+The ADOS-G was normed on a sample of 381 children, adolescents, and adults spanning a broader diversity of spectrum and nonspectrum disorders. The sample included a group of individuals diagnosed with autism, PDD-NOS, and a group designated as “nonspectrum,” which included individuals with diagnoses of mental retardation, language disorder, attention-deficit/ hyperactivity disorder, oppositional defiant disorder, anxiety, depression, and obsessive-compulsive disorder and children who were typi-cally developing. The ADOS-G algorithms were successful at discriminating ASD from non-spectrum, but were not as good at making distinc-tions between children with milder forms of ASD. Upon WPS publication of the ADOS-G in 1999, the “G” was dropped and the instrument became solely known as the ADOS. Gotham et al. (2007) and colleagues sought to improve the diagnostic validity of the ADOS by validating revised algo-rithms for modules 1–3 on a significantly larger sample of children with ASD and nonspectrum diagnoses. The new algorithms were grouped into developmental cells to reduce the effects of age and IQ and included more similar items across modules with the same number of items per algorithm to increase comparability. Factor analyses yielded two domains representing features of social affect and restricted and repetitive behaviors (RRBs); thus, the new algorithms required thresholds to be met in social affect, RRB, and a combined total, in order to meet classification criteria for autism or ASD. This was a significant departure from earlier versions of the ADOS in which RRBs were not included on the algorithm and social interaction and communication were considered separately. Specificity in children with nonverbal mental ages of 15 months and younger continued to pose prob-lems in distinguishing children with ASD from those with other language-based disorders or intel-lectual disability. Since the publication of the revised algorithms, however, several replications with larger and more diverse samples have been conducted with consistent results supporting the improved diagnostic validity of the new algorithms.
+
+Though higher scores on the ADOS do indi-cate a greater number of behaviors consistent with core deficits of ASD and, to some degree, greater severity of impairment, ADOS scores were not standardized for this purpose. The creation of revised algorithms paved the way for the devel-opment of calibrated severity scores (Gotham et al. 2009). Severity scores that reduced the effects of IQ and chronological age were devel-oped to promote the comparison of ADOS assess-ments over time, age, and module and to identify trajectories of autism severity. Raw scores have been mapped onto a 10-point severity metric with lower scores indicating less autism impairment.
+
+As calibrated severity scores were being devel-oped, a new module of the ADOS, the ADOS-Toddler, was also underway. Advancements in the understanding of autism in very young children, particularly infants and toddlers, increased the need for diagnostic tools appropriate for use in that developmental level. Because the ADOS, even with revised algorithms, had limited appli-cability for children with nonverbal mental ages below 15 months, the toddler module was created. The toddler module consists of a combination of ADOS and some new tasks and is intended for use in children 12–30 months chronological age, with nonverbal mental ages of at least 12 months, and who are walking independently. It includes two algorithms, nonverbal 12–20 months/ 12–30 months and verbal 21–30 months. Because of the relative instability of diagnostic classifica-tions in very young children, the toddler algo-rithms differ from those of the ADOS-G in two ways. First, they yield research classifications of ASD or nonspectrum and do not make distinc-tions between autism and ASD, and second, they provide clinical “ranges of concern,” (little-to-no, mild-to-moderate, and moderate-to-severe con-cern for ASD) indicating the degree of need for continued clinical monitoring.
+
+The ADOS has developed considerably since the first 1989 version, and research on expanded applications of the instrument continues today. Continued testing of the ADOS is occurring in clinical and community-based settings, in addi-tion to the application of translated versions for use in languages other than English.
+
+Autism Diagnostic Observation Schedule, Table 1 ADOS algorithms
+| Module | T | Module T | Module 1 | Module 1 | Module 2 | Module 3 | Module 4 | Adapted module 1 | Adapted module 2 |
+|---|---|---|---|---|---|---|---|---|---|
+| Age | No words | Some words | <5 words | Single words | Phrases | Fluent | Fluent | No words | Some words |
+| 12–30 m | X | | | | | | | | |
+| 21–30 m | | X | X | | | | | | |
+| 30–35 m | | | X | X | | | | | |
+| 3–4 years | | | X | X | X | X | | | |
+| 5–9 years | | | | | X | X | X | | |
+| 10+ years | | | | | X | X | X | X | X |
+
+Autism Diagnostic Observation Schedule, Table 2 History of the ADOS in JADD publications
+| Publication | Contribution |
+|---|---|
+| Autism Diagnostic Observation Schedule: A Standardized Observation of Communicative and Social Behavior (Lord et al. 1989) | First published version of the ADOS |
+| The Pre-Linguistic Autism Diagnostic Observation Schedule (DiLavore and Lord 1995) | Introduction of alternate version of ADOS more appropriate for individuals with very limited language |
+| The Autism Diagnostic Observation Schedule-Generic: A Standard Measure of Social and Communication Deficits Associated with the Spectrum of Autism (Lord et al. 2000) | Consolidation of ADOS and PL-ADOS<br>Introduction of four module structure<br>Appropriate for broader range of social communication deficits and age<br>Accompanied by commercial release of ADOS by Western Psychological Services (Lord et al. 1999) |
+| The Autism Diagnostic Observation Schedule: Revised Algorithms for Improved Diagnostic Validity (Gotham et al. 2007) | Revised algorithms for improved diagnostic validity<br>Algorithms grouped by developmental and language ability<br>Inclusion of restricted and repetitive behaviors in algorithm totals |
+| The Autism Diagnostic Observation Schedule-Toddler Module: A New Module of a Standardized Diagnostic Measure for Autism Spectrum Disorders (Luyster et al. 2009) | Introduction of ADOS-Toddler<br>Appropriate for use in children under 30 months with mental age of at least 12 months |
+| Standardizing ADOS Scores for Measure of Severity in Autism Spectrum Disorders (Gotham et al. 2009) | Created standardized severity metric to measure change in ADOS assessments over time, age, and module |
+
+**Psychometric Data**
+Reliability. Across all ADOS modules, intraclass correlations for the social, communication, social communication, and restricted and repetitive domains were 0.93, 0.84, 0.92, and 0.82, respec-tively, and mean weighted kappas across items ranged from 0.65 to 0.78. Test-retest reliability ranged from 0.59 to 0.82. For the toddler module, intraclass correlation was 0.96 for the entire pro-tocol and mean weighted kappa was 0.67. Test-retest reliability was 0.86 for the 12–20/21–30 nonverbal algorithm and 0.95 for verbal 21–30. Interrater reliability across all modules is reported in Table 3.
+
+Diagnostic validity. Algorithm cutoffs for the ADOS were excellent for autism and ASD rela-tive to nonspectrum disorders, with even greater performance with the introduction of revised algorithms. Algorithm cutoffs for the toddler module yielded high sensitivity and specificity. Sensitivities and specificities for current and revised algorithms of the ADOS are reported in Tables 4 and 5 and in Table 6 for the toddler module algorithms.
+
+Autism Diagnostic Observation Schedule, Table 3 Interrater reliability: percent agreement
+| | Toddler | Module 1b | Module 2b | Module 3b | Module 4b |
+|---|---|---|---|---|---|
+| Interrater (items) | 84 | 91.5 | 89 | 88.2 | 88.3 |
+| Interrater (algorithm) | 87 | 93 | 87 | 81 | 84 |
+aLuyster et al. 2009
+bLord et al. 2000
+
+Autism Diagnostic Observation Schedule, Table 4 Sensitivities and specificities for current and revised ADOS algorithms: autism versus nonspectrum (Gotham et al. 2007)
+N = 1157
+| | Current ADOS classification Se | Current ADOS classification Sp | Revised ADOS classification Se | Revised ADOS classification Sp |
+|---|---|---|---|---|
+| Mod 1, no words, nvma <=15 | 100 | 19 | 97 | 50 |
+| Mod 1, no words, nvma >15 | 97 | 91 | 95 | 94 |
+| Mod 1, some words | 88 | 96 | 97 | 91 |
+| Mod 2, younger | 97 | 93 | 98 | 93 |
+| Mod 2, age 5+ | 96 | 97 | 98 | 90 |
+| Mod 3 | 86 | 89 | 91 | 84 |
+| Mod 4 | 93 | 93 | N/A | N/A |
+
+Autism Diagnostic Observation Schedule, Table 5 Sensitivities and specificities for current and revised ADOS algorithms: non-autism ASD versus nonspectrum (Gotham et al. 2007)
+N = 685
+| | Current ADOS classification Se | Current ADOS classification Sp | Revised ADOS classification Se | Revised ADOS classification Sp |
+|---|---|---|---|---|
+| Mod 1, no words, nvma <=15 | 95 | 6 | 95 | 19 |
+| Mod 1, no words, nvma >15 | 88 | 67 | 82 | 79 |
+| Mod 1, some words | 67 | 84 | 77 | 82 |
+| Mod 2, younger | 76 | 70 | 84 | 77 |
+| Mod 2, age 5+ | 86 | 77 | 83 | 83 |
+| Mod 3 | 68 | 77 | 72 | 76 |
+| Mod 4 | 86 | 93 | N/A | N/A |
+
+**Clinical Uses**
+The ADOS is intended for use by clinicians famil-iar with autism. Valid administration and interpre-tation of results is dependent on the clinical skill of the examiner and requires substantial training. The ADOS can be used clinically upon comple-tion of a two-day WPS-certified clinical course or from WPS training DVDs. Even with training, however, administration of the ADOS should not be attempted without significant practice in administering the tasks, in observing features of autism as specified by the ADOS items, and in scoring. For those using the ADOS in research settings, more rigorous requirements for use exist. Individuals must attend a standardized training workshop and then obtain reliability with work-shop leaders and within the research site. As specified in Lord et al. (2000), research reliability is defined as agreement of 80% or above on ADOS protocols and algorithms on three consec-utive scorings for modules 1 and 2 and modules 3 and 4, separately.
+
+Selecting the correct module for use in the ADOS is also crucial for obtaining an accurate classification. Clinicians and researchers can use the results of standardized tests or parent report to inform module choice, but as an individual’s lan-guage often varies in unstructured versus struc-tured environments, the collection of a spontaneous language sample at the beginning of an ADOS administration is highly recommended. Administration of an “easier” module (e.g., selecting module 2 for a child with fluent speech because the tasks are “more fun” when module 3 would be more appropriate) can result in under classification. When in doubt, however, a clini-cian should adopt a conservative approach and chose a lower module as language difficulties may confound the social demands of a higher one.
+
+Perhaps the most important practice in using the ADOS is to recognize its limitations. The ADOS is only one of multiple sources of information that should be considered when determining whether criteria for ASD are met. It is possible to meet classification thresholds on the ADOS algorithm and not meet formal criteria for an autism diagno-sis. Conversely, a clinician with information from parent report and observations in different settings may assign a diagnosis of ASD even without an accompanying ADOS classification. The ADOS was developed as a companion instrument to the ADI-R, and indeed, both the ADOS and ADI yield higher sensitivities and specificities together than when used separately (Risi et al. 2006). In the hands of a skilled clinician with ample training and mul-tiple sources of information, the ADOS provides a unique contribution to the observation of social and communicative features of autism and greatly aids in the diagnosis of ASD.
+
+**See Also**
+*   Autism Diagnostic Interview-Revised
+*   Autism Diagnostic Observation Schedule (ADOS): Toddler Module
+*   Prelinguistic Autism Diagnostic Observation Schedule
+
+Autism Diagnostic Observation Schedule (ADOS): Toddler Module
+**Synonyms**
+ADOS-T
+
+**Description**
+The Autism Diagnostic Observation Schedule – Toddler Module (or ADOS-T; Luyster et al. 2009; Lord et al. 2012) – is a semi-structured assessment of social engagement, communication, and play using a set of planned “presses” within a natural-istic social interaction. It is intended for children under 30 months of age who have a nonverbal mental age of at least 12 months. Other guidelines for use include independent walking and minimal language; once the child masters three-word phrases, the Toddler Module is no longer consid-ered appropriate.
+
+Eleven activities are included in the Toddler Module, along with 41 overall codes. Two algo-rithms are associated with the module, including one for all children between 12 and 20 months of age and nonverbal children between 21 and 30 months of age and a second algorithm for verbal children between 21 and 30 months of age. These algorithms include formal cutoffs, which are primarily intended for research use and provide a binary classification of ASD or nonspectrum. Each algorithm also has three “ranges of concern,” which are intended for clin-ical use and provide three classifications of con-cern: little to no, mild to moderate, and moderate to severe. The Toddler Module can be adminis-tered in a professional’s office or playroom, although a familiar caregiver must be present. Codes are completed immediately after Toddler Module completion and are based on all behaviors during the administration. Each code can be scored between 0 and 3, with higher scores indic-ative of greater abnormality.
+
+**Historical Background**
+The Toddler Module was developed in response to a research and clinical need for a standardized instrument for use in very young children at high risk for, or suspected of having, an autism spec-trum disorder (ASD). Research had indicated that the ADOS Module 1 was over-inclusive (meaning it exhibited relatively poor specificity) for chil-dren with nonverbal mental ages under 16 months (Gotham et al. 2007). The Toddler Module was developed for use in this very young population and was intended to aid in both clinical and research efforts targeted at children who fell below the floor of the ADOS.
+
+The creation of the Toddler Module was based primarily on the Module 1 of the ADOS (Lord et al. 2000), which provides a series of semi-structured, play-based tasks and activities to probe for a range of behaviors. Module 1 items that were appropriate for infants and toddlers were included, and additional tasks were created based on a review of the literature on early social and communicative development. Some other impor-tant changes were made based on current knowl-edge of early development in children with ASD, including a shift from three classifications on the algorithm (autism, ASD, nonspectrum) to two (ASD, nonspectrum), based on extensive evi-dence of the instability of specific diagnoses within the autism spectrum. For similar reasons, an emphasis was placed on using algorithm ranges of concern in order to encourage a focus on clinical monitoring and follow-up rather than assigning a formal diagnosis to a very young child.
+
+**Psychometric Data**
+Instrument development involved both validity and reliability studies (Lord et al. 2012). The validity study was completed using data from 182 children. Analyses were repeated using two overlapping samples, one of which included each child only once and a second that included multi-ple visits from some children. The final set of 41 codes was selected in order to yield markedly different distributions across diagnostic groups or to have high clinical or theoretical importance. In addition, codes were chosen in a manner that minimized collinearity with other codes or sample characteristics. Two algorithms were generated by selecting items that met theoretical and empirical thresholds for optimal group classification. Each algorithm includes items in two domains – social affect (SA) and restricted, repetitive behaviors (RRB) – and cutoff scores were selected based on maximal sensitivity and specificity. Using for-mal cutoffs, sensitivity and specificity exceeded 86% on the younger/nonverbal algorithm, and they exceeded 83% on the verbal algorithm.
+
+The reliability study included ratings from 7 independent, “blind” raters on 14 Toddler Mod-ule administrations (8 from children with ASD, 3 from typically developing children, and 2 from children with non-ASD developmental disabil-ities, one child contributed two administrations). Inter-rater reliability was evaluated using weighted kappas for nonunique pairs of raters, with kappas between 0.4 and 0.74 considered good and kappas at or above 0.75 considered excellent. Three codes were not included in the reliability analyses because of limited variability; 30 codes had kappas equal to or above 0.60, and the remaining eight codes exceeded 0.45. Inter-rater item reliability was measured using percent agreement and the full range of 0–3 scores: the mean percent agreement was 84%. All items exceeded 71%, and 30 of 41 items had exact agreement of at least 80%. Inter-rater agreement on the algorithms’ (younger/nonverbal and ver-bal) diagnostic cutoffs was 97% and 87%, respec-tively; inter-rater agreement for ranges of concern was 70% and 87%, respectively. Test-retest reliability was also satisfactory across both algorithms.
+
+Note that although standardized calibrated severity scores are not available as a formal com-ponent of the instrument (Lord et al. 2012), research suggests that they may be helpful in reducing the effects of language level on algo-rithm totals (Esler et al. 2015).
+
+Autism Diagnostic Observation Schedule, Table 6 Sensitivities and specificities for toddler module algorithms: ASD versus nonspectrum (Luyster et al. 2009)
+N = 234
+| | Se | Sp |
+|---|---|---|
+| 12–20/nonverbal 21–30 | 87 | 86 |
+| Verbal 21–30 | 81 | 83 |
+
+**Clinical Uses**
+Clinical usage of the Toddler Module should be accompanied by other sources of information. The ranges of concern may be useful in providing an indication of the degree to which a child is exhibiting symptoms consistent with an ASD, but in some cases, these behaviors may be attrib-utable to other, non-ASD etiologies. Therefore, informed clinical judgment is critical in inter-preting results within a broader developmental framework. Examining the profile of scores across the 41 codes may be useful in identifying areas of difficulty for the child and can help in education and intervention planning.
+
+**See Also**
+*   Prelinguistic Autism Diagnostic Observation Schedule
+
+Autism Family Experience Questionnaire (AFEQ)
+**Synonyms**
+Autism spectrum disorder (ASD); General health questionnaire-12 (GHQ-12); Pre-school autism communication trial (PACT trial); Pediatric autism communication therapy (PACT therapy); United kingdom (UK); Vineland adaptive behavior scales (VABS); Warwick-edinburgh mental wellbeing scale (WEMWBS)
+
+**Description**
+The Autism Family Experience Questionnaire (AFEQ; Leadbitter et al. 2018) is an ecologically valid questionnaire that measures the intervention priorities of parents/caregivers of children with autism spectrum disorder (ASD) and assesses the impact of interventions on family experience and quality of life. The AFEQ was developed in con-sultation with parents/caregivers of children with ASD and was designed to address the paucity of outcome measures that assess parent/caregiver-nominated intervention outcomes for autistic chil-dren and their families (McConachie et al. 2015; Morris et al. 2014, 2015). It can be used for both research and clinical purposes.
+
+The AFEQ has 48 items. Items are organized into four domains: (1) experience of being a parent of a child with autism (13 items); (2) family life (9 items); (3) child development (development, understanding, and social relation-ships; 14 items); and (4) child symptoms (feelings and behavior; 12 items). The questionnaire is designed to be self-rated by parents. Instructions are: “Please read each statement carefully and tick the box which you think best fits your feelings about you, your child with autism and your family life.” Items include both positively and negatively worded statements and are scored on an order scale: 1 = always to 5 = never, with an option for “not applicable”. Items that are negatively worded are reverse scored; individual missing data points can be prorated with the mean score of all items for that participant. The AFEQ pro-duces a total score (range = 48–240) and domain scores which can be used to assess group differ-ences or within-participant/within-group change over time. On the AFEQ a lower score indicates a positive outcome, and a higher score is a poor outcome. There are no clinical or case thresholds. The AFEQ is being used as an outcome measure in current United Kingdom (UK) and international clinical trials. It is also being used in observational studies in several different countries. We invite other researchers and clinicians to utilize the AFEQ. A copy of the questionnaire can be accessed in open-access at: https://link.springer.com/article/10. 1007/s10803-017-3350-7#SupplementaryMaterial. Scoring guidelines can be accessed directly from the author (Kathy.Leadbitter@manchester.ac.uk). The questionnaire has been translated from English into several other languages. Please con-tact the author for further information on available languages and/or the translation procedures. We ask that you appropriately cite our work and keep us informed about your research, its findings, and any planned publications or outputs relating to the AFEQ measure.
+
+**Historical Background**
+The AFEQ was developed prior to and in prepa-ration for the UK Medical Research Council Pre-school Autism Communication Trial (PACT Trial) as part of a broader strategy adopted by the trial team and funders to promote the involve-ment of service users in pretrial research design and out of the recognition that there was a mea-surement gap in relation to parent-generated out-come measures and assessments of family experience and child and family well-being. The PACT Trial was a two-arm parallel-group ran-domized controlled trial of Pediatric Autism Com-munication Therapy (PACT Therapy) – a parent-mediated video-aided communication-focused intervention for preschool children with ASD and their parents. The trial ran between 2006 and 2009 across three UK centers and evaluated the effectiveness of PACT therapy plus treatment-as-usual against treatment-as-usual alone in 152 pre-school children with core autism (Green et al. 2010). A subsequent follow-up study assessed outcomes at 6 years after the end of the treatment phase (Pickles et al. 2016). The PACT Trial commissioning and protocol included a pre-specified strategy to develop a new parent-generated change measure of child and family well-being that could be used within the trial and suitable for future intervention research.
+
+The AFEQ questionnaire was developed through three phases. Firstly, a series of pretrial focus groups was conducted with 31 parents of children with ASD to generate a core set of param-eters that parents identified as the most important outcomes of a preschool intervention for ASD. Focus group transcripts were analyzed with the-matic analysis. A set of 78 individual statements were abstracted from the thematic analysis themes to serve as response items within the draft ques-tionnaire. Secondly, a large pretrial web-based con-sultation in collaboration with the UK National Autistic Society (www.autism.org.uk) subjected the 78 statements to wider review to evaluate the clarity and usefulness of the items. Using the data from the online consultation, low-performing items were discarded, resulting in a questionnaire with 56 well-performing items. This questionnaire was rated by 152 parents as part of the baseline and 12-month follow-up assessments of the PACT Trial. Parents reported that they found the question-naire easy to complete and that they valued the opportunity to report real-life experiences for their children and family on metrics nominated by other parents of autistic children. Following further data cleaning, an additional eight items were excluded as they generated too much missing data. The resulting 48-item questionnaire was named the Autism Family Experience Questionnaire. The 48-item AFEQ was then used in the PACT trial 6-year follow-up study.
+
+A planned analysis within the PACT trial pro-tocol was to use data from the trial to evaluate the psychometric properties of the AFEQ (see below; also Leadbitter et al. 2018). It was also planned to use the questionnaire as a trial outcome measure, to assess the estimation of treatment effect of PACT therapy over treatment-as-usual on parent-prioritized outcomes, family experience, and quality of life. On the 48-item AFEQ total score, there was a statistically significant improvement in the PACT group over the treatment-as-usual group at both trial endpoint (Cohen’s d = -0.29) and at 6-year follow-up (Cohen’s d = -0.49). There were also treatment effects at trial endpoint on two domain scores: the “expe-rience of being a parent” domain and the “child development” domain. These findings provided evidence that the AFEQ total score, and to some extent its domain scores, were sensitive to change in response to a parent-mediated intervention for young children with ASD.
+
+**Psychometric Data**
+Descriptive data: In the validation cohort (PACT Trial sample – see above), the range of AFEQ total scores across three measurement timepoints was 64–196, with means of 132–141, medians of 133–141, and standard deviations of 21.3–24.6.
+
+Internal consistency: The AFEQ has good internal consistency. To assess the internal consis-tency of the AFEQ, we examined the scale reli-ability based on Cronbach’s alpha for the total score and domain scores, calculated from PACT Trial baseline data. The total score and all domains demonstrated excellent levels of reliabil-ity: AFEQ total (alpha = 0.92), parent (0.85), family (0.83), child development (0.81), and child symptoms (0.79).
+
+Criterion validity: We assessed the external cri-terion validity of the AFEQ against (a) the parental Vineland Adaptive Behavior Scales (VABS; Second Edition; Sparrow et al. 2006), a well-validated parent-rated scale of child adaptive functioning; (b) the General Health Questionnaire-12 (GHQ-12; Goldberg 1992), a well-established measure of adult mental health; and (c) the Warwick-Edinburgh Men-tal Wellbeing Scale (WEMWBS; Tennant et al. 2007), a widely used measure of adult well-being. The correlations between the VABS total score and AFEQ child development domain score (items 23–36; 14 items) were moderate to strong across the three timepoints (r = -0.48 to -0.71; positive outcome indicated by a low score on the AFEQ and a high score on the VABS). The correlation between the parent domain score (items 1–13) and GHQ-12 total score was Spearman’s Rho = 0.408 (p < 0.001, n = 101; a Spearman’s rank correlation was conducted as the GHQ-12 distribution was highly positively skewed). The association between the parent domain score and the WEMWBS total score at trial follow-up was r = -0.528 (p < 0.001, n = 103). The AFEQ therefore showed good con-vergent validity with well-established measures of child adaptive functioning, parental mental health, and parental well-being.
+
+The psychometric evidence is based on a sam-ple of children with “core autism” aged 2–12 years. It is not yet known how the question-naire would work outside of these parameters.
+
+**Clinical Uses**
+The AFEQ can be applied in a range of healthcare, child development, educational, or social care settings and developmental settings that support families of children with ASD and similar neurodevelopmental conditions. Its main use would be to make pre- and post-within-participant or within-group comparisons to evaluate the effect of an intervention on parent-nominated intervention priorities, family experience, and quality of life. At the time of publication, it was in use in this way by several clinicians/clinical teams internation-ally. The AFEQ could also be used to quantify the experience of families and to make between-family comparisons, in order to identify families who are having a particularly difficult experience and who could benefit from further support.
+
+**See Also**
+*   Assessing Quality of Life in Autism
+*   Developmental Intervention Model
+
+Autism Family Experience Questionnaire (AFEQ), The
+**Synonyms**
+Autism Spectrum Disorder (ASD); General Health Questionnaire (GHQ-12); Pre-school Autism Communication Therapy (PACT); Vineland Adaptive Behavior Scales (VABS); Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS)
+
+**Description**
+The Autism Family Experience Questionnaire (AFEQ) is a measure developed to reflect the intervention priorities of parents of children with autism spectrum disorder (ASD), and to assess the impact of interventions on family experience and quality for life (The PACT Consortium et al. 2018).
+
+The AFEQ consists of 48 items in the domains: (1) experience of being a parent of a child with autism; (2) family life, (3) child development (development, understanding, and social relationships); and (4) child symptoms (feelings and behavior) (The PACT Consortium et al. 2018). All items are scored on an order scale: 1 = always to 5 = never, with an option for “Not Applicable.”
+
+**Historical Background**
+Quality of life has been widely studied; however, the research on quality of life for people with autism is scarce (Burgess and Gutstein 2007). The AFEQ was developed and tested within the large PACT intervention, Trial and Follow-Up (The PACT Consortium et al. 2018). The AFEQ is developed based on focus and online consultations with parents (The PACT Consortium et al. 2018).
+
+In the focus group phase, 31 parents of pre-school or school-aged children with an autism diagnosis were recruited. The participants were recruited from both local clinical services and parent-support groups (The PACT Consortium et al. 2018). The participants attended one of five focus groups, convened and led by members of the PACT Principal Investigator team, and an independent qualitative researcher. The focus groups explored the specific parameters identified as the most important outcomes in a pre-school communication intervention for parents. A qualitative analysis of the focus groups resulted in 78 questionnaire items. Thirty-five parents rated the 78 questionnaire for clarity and usefulness online.
+
+Based on the online rating, revised question-naire with 56 items were applied with parents within the PACT trial. A further cleaning resulted in the 48-item questionnaire: the Autism Family Experience Questionnaire (AFEQ).
+
+**Psychometric Data**
+To date there is only one study exploring the psychometric date of the AFEQ (The PACT Consortium et al. 2018). The scale reliability of the AFEQ was initially examined based on 140 participants based on Cronbach’s alpha for the domain scores and the total AFEQ score. All domains and the total score demonstrated excellent reliability: parent (alpha = 0.85), family (0.83), child development (0.81), child symptoms (0.79), and total AFEQ (0.92) (The PACT Consortium et al. 2018).
+
+External criterion validity was explored for the child development domain against parental Vineland Adaptive Behavior Scales (VABS). The correlation were significant (r = -0.478 to -0.710, p <0.001) and indicating a moderate to strong association between AFEQ and VABS at three different time-points (The PACT Consortium et al. 2018). The external criterion validity of the parent domain of the AFEQ was assessed against GHQ-12 and WEMWBS at one time-point. The correlation between the parent domain score and GHQ-12 was Spearman’s Rho = 0.408 (p <0.001) and with the WEMWBS was r = -0.528 (p < 0.001).
+
+**Clinical Uses**
+The AFEQ are used in the PACT 6-year follow-up study (Sigafoos and Waddington 2016) and may be used as described in the objective – to reflect the intervention priorities of parents of children with autism spectrum disorder (ASD) and to assess the impact of interventions on family expe-rience and quality for life.
+
+**See Also**
+*   Assessing Quality of Life in Autism
+*   Parental Response to Diagnosis
+*   Quality of Life for Transition-Age Youth with ASD
+*   Social Validity
+*   Vineland III
+
+Autism in Ecuador
+**Historical Background**
+As in other countries from the region, the field of mental health in Ecuador has been documented during pre-Columbian times, the colonial period, and the modern republics. Children and adults with autism spectrum disorder (ASD), especially those situated in the most severe end of the spec-trum, may have received treatments available for individuals with atypical and challenging behav-iors long before ASD was recognized and named. Different treatments used for mental health con-ditions have been documented since pre-Columbian times, such as the use of hallucinogens (ayahuasca, bejuco), animals (chickens, guinea pigs), songs, and dances with mythical and reli-gious allusions (Naranjo 1983). In different parts of the country, it is still possible to observe prac-tices of shamanism or crusaderism, offered to individuals with mental health difficulties (Zuniga Carrasco and Riera Recalde 2018), or read billboards advertising the services particu-larly for children’s difficulties, mentioning autism among them. The first hospice and psychiatric asylum, Hospicio Jesus, María y José, was founded under the initiative of the Catholic Church in 1785 in the city of Quito, then a colo-nial center of the Spanish Crown. Operating on a Western prison model, this center had a population of individuals suffering from mental health problems but also orphans and beggars (Landazuri 2008).
+
+The birth of psychology in Ecuadorian aca-demic circles is situated toward the end of the nineteenth century. The first chair of psychology was given in 1897 on subjects relating to hypno-tism and suggestion by professors of general med-icine. The lessons were addressed to teachers trained in philosophy and pedagogy, although the biological paradigm appeared to dominate. The first experimental studies took place shortly after, affirming psychology as a scientific area. The creation of various chairs of psychiatry in different cities succeeded until 1926, influenced by the conference “Psychology and Pedagogy,” dictated at the first Ecuadorian Congress of Med-icine in 1919 (Zuniga Carrasco and Riera Recalde 2018). By 2007, a total of 17 faculties of psychol-ogy had been created in different Ecuadorian uni-versities. One of the most important, the Faculty of Psychological Sciences of the Central Univer-sity of Ecuador, in Quito, was created in 1972 as a single school, covering four specializations: clin-ical psychology, special education and psycho-rehabilitation, industrial psychology, and legal psychology.
+
+The first identifications of autism cases likely took place in the 1980s, in the context of psychi-atric public and private professional practices, following official recognition of autism by the Diagnostic and Statistical Manual of Mental Disorders, third edition (DSM-III), in 1980 and under the influence of European or North Ameri-can textbooks that university libraries possessed at that time. From the treatment point of view, psychotherapies and psychopharmacology, such as insulin shock therapy, were used for schizo-phrenia and related disorders from the 1950s (Aguilar 2013). In the 1960s, nongovernmental institutions were created in order to provide care for individuals with mental health conditions. The first organizations created with this purpose were religious, such as the Order of the Sisters of the Hospital, whose work was based on an agreement with the Ministry of Social Protection and Labor. Other nonprofit and for-profit laic foundations, set up to meet the needs of adults and children with different types of health problems, appear in the following years. The first private centers intending to provide specific assistance to chil-dren with autism, by separating them from chil-dren with intellectual disabilities, appeared in the late 1980s (Aguirre et al. 2017 cited in Zuniga Carrasco and Riera Recalde 2018).
+
+Currently, it is not possible to identify any specific programs for autism in the registers of public health agencies in Ecuador. The Ministry of Health published the first Guide for Clinical Practice: Diagnosis, Treatment, Rehabilitation, and Case Management in 2017. The Ecuadorian health system is based on public and private prac-tices. The public system comprises two subsys-tems, hospitals, and state social security system institutions, on the one hand, and institutions dependent on the Ministry of Health, on the other. Two major pediatric hospitals, Hospital Baca Ortiz in Quito and Hospital Dr. Francisco de Icaza Bustamante in Guayaquil, depend on the Ministry of Health and provide general and spe-cialized services. There are also private entities which operate in the public sector, such as the Welfare Board of Guayaquil (Hospital Dr. Robert Gilbert E. Guayaquil), the Child Pro-tection Association of Guayaquil, and the Ecua-dorian Red Cross. Public services are funded from the general state budget, extra-budgetary funds, and funds from national and international projects and agreements. Private services are funded by selling health-care services to the public sector; by private health insurers, mainly for the middle- and high-income population (Pan American Health Organization 2017); and by the families themselves. Services for people with autism are offered by pediatricians, neuro-pediatricians, psy-chiatrists, clinical psychologists, and educators, as well as speech and language therapists and occu-pational therapists, working in public or private centers and private practices but also in nonprofit foundations which are mainly located in the cities.
+
+As it has been the case in other countries, parents’ associations have contributed signifi-cantly to obtain recognition of ASD through pub-lic conferences and free-of-charge training to parents and teachers. In Quito, a group of parents met in 2012 to ask for official recognition of autism as a handicap. Parents obtained access to a Disability Card (Carnet de Discapacidad) from the National Parliament, which allows individuals with autism to certain rights. In March 2013, those parents united into an association, APADA, aiming to contribute to the development of awareness programs and to work together with the Ministry of Education and the Ministry of Labor on special education plans. This association has also contributed to the development of the National Agenda for Equality in Disabilities 2017–2021, aiming to support the autonomy and productivity of people with disabilities. This agenda also examines how “to define a national instrument for the diagnosis of the Autism Spec-trum” and “to implement the screening and diag-nosis of the Autism Spectrum in the national territory” (National Council for Equality in Disabilities 2017, p. 54–55). Together with nine other parent associations, Guayaquil (3), Quito (1), Cuenca (1), Machala (1), Santo Domingo (1), Ibarra (1), Los Ríos (1), and Loja (1), it has formed the “Ecuadorian Federation of Autism Spectrum” (Organizaciones de Autismo buscan 2017). Their objective is to protect the well-being of individuals with ASD and their families, promote public policies, and support the work of other organizations. Other parents’ associations are currently being organized in different cities across the country.
+
+**Legal Issues, Mandates for Services**
+Ecuador is a member of the International Conven-tion on the Rights of the Child (The United Nations 1989), which recognizes and protects access to health and education services, and was the 20th state to ratify the Convention on the Rights of Persons with Disabilities (The United Nations 2006) and its optional protocol that entered into force in 2008. The Ecuadorian State has transposed and clarified those rights in its national legislation. The Law on Disabilities (LOD) of the National Health System (National Assembly of the Republic of Ecuador 2012) guar-antees the rights of people with ASD to free access to medicines and equipment, technical and technological aids, adaptations of study plans, and permanent accompaniment of guides (LOD, art. 33). Other benefits, such as financing con-struction or remodeling housing and a reduction of 50% in the services of water and electricity, are considered by this law. In the field of education, an agreement intends to guarantee the access of individuals with special needs to special educa-tion (Ministerio de Educación 2013). In order to be eligible, individuals with ASD must request the Disability Card. Potential beneficiaries need to justify that they are suffering from a “non-evident and non-visible disability.” They also need to present a report from a medical practitioner or a specialist and the results of additional examina-tions, which may only be issued by the units of the Complementary and Integral Public Health Net (Ministry of Health n.d.). Autism is defined in this context as a “catastrophic disease,” namely, a pathology or chronic disease which poses a grave risk to the life of the person. Their treatment has a high economic cost and social impact, and, being of a prolonged or permanent nature, they must be part of long-term health plan and gener-ally have little or no insurance coverage (Ministry of Health 2012a). In the particular case of individ-uals in this situation who are living in critical socioeconomic circumstances, this ministerial agreement contemplates the allocation of a monthly voucher of 240 USD under certain con-ditions. As described, an important legal frame-work intending to support the development of individuals with ASD exists in Ecuador. How-ever, some critical gaps in terms of training and access to information on good practices exist, hindering the implementation of rules and policies (Educación Inclusiva en Ecuador hay ley 2019).
+
+**Overview of Current Treatments and Centers**
+Many different treatments are offered in public and private centers, as well as in private practices. They include a considerable variety of methods, such as Floortime, Tomatis, hippotherapy, and others, sometimes not specified. Speech therapies, sensory therapies, and cognitive-behavioral therapies are also commonly offered. Specific evidence-based treatment services are still scarce. To date, two professionals are registered within the Behavior Analyst Certification Board as BCBAs, one in Quito and one in Guayaquil. Only one professional, in Quito, is currently reg-istered on the official list of certified therapists of the ESDM model. TEACCH strategies and alter-native/complementary systems of communication are also used within a variety of settings. Drug treatments, generally intended for comorbidities, are prescribed by pediatricians, neuro-pediatricians, and psychiatrists. The annex num-ber ten of the Guide for Clinical Practice: Diag-nosis, Treatment, Rehabilitation, and Case Management (Health Minister 2017) provides a list of medication endorsed by this document.
+
+**Overview of Research Directions**
+The document “Priority research areas 2013–2017” from the Ministry of Health defines a certain number of fields that had been chosen according to a list of health problems identified in official registers. Mental health issues are the 11th among 19 categories, in which autism and Asperger’s are considered (Ministry of Healths 2012b). However, current research literature indi-cates that areas related to ASD have been under-explored and studies on prevalence at a national level have not yet been conducted. According to the Guide for Clinical Practice (Ministry of Health 2017), ASD prevalence in a child population of 5 years old or less was estimated to be 0.28% (0.18–0.41%) in 2015. According to data pro-vided by the National Directorate of Disabilities of the Ministry of Health, in 2016, 1,266 people diagnosed with ASD were reported. Of these, 254 cases have been registered with a diagnosis of atypical autism, 792 with a diagnosis of child-hood autism, 205 with Asperger syndrome, and 15 with Rett syndrome (as cited in Ministry of Health 2017, p. 11). The reasons why estimates of ASD prevalence in Ecuador are remarkably lower than those reported in Western countries remain unclear. As preliminary evidence, a study aiming to estimate school attendance of children with an ASD diagnosis in the city of Quito found a pro-portion of 0.11% among 453 pupils in 161 in regular schools, assessed through interviews with school directors (Dekkers et al. 2015).
+
+Preliminary research on the field of assistive technology has also been conducted in different universities. The research group in Artificial Intel-ligence and Assistive Technology of Salesian Polytechnic University of Ecuador (https://www. ups.edu.ec/giiata) has carried out a pilot project aiming to explore the functionality of a mobile tool and a robotic assistant for the diagnosis and intervention of children with ASDs (Galán-Mena et al. 2016, June). A project intending to develop an application to support verbal communication and personal autonomy in children and young people has been carried out at the Universidad de las Fuerzas Armadas ESPE (Cárdenas et al. 2015, October).
+
+A research project aiming to identify potential barriers to diagnosis in pediatric environments has been conducted in cooperation with the School of Pediatrics of the Faculty of Medicine at the Cath-olic University of Quito, the Faculty of Psychol-ogy at the University of Geneva, and the AJ Drexel Autism Institute in Philadelphia, with the endorsement of the Ecuadorian Society of Pediat-rics (ESP). The results suggest that, as in many other countries, the pediatric community in Ecua-dor may be facing obstacles in terms of time for screening, training, and resources adapted to their clinical practices. The results also point to a low number of autism cases identified during the pro-fessional life of the participants (Buffle et al. 2019). An additional study, aiming to examine the pediatric community’s perception on screening procedures and tools, suggests a preference for observational procedures, over paper parent-administered questionnaires, as well as a clear interest among professionals in acquiring knowl-edge and expertise on the identification of early signs (Buffle and Gentaz 2019). A pilot project aiming to study the visual social attention with an eye-tracking is currently being conducted in neurotypical preschool age children in Quito. Eye-tracking measures are increasingly proposed as sensitive biomarkers for ASD, particularly con-venient to assess the core social attention deficits contributing to ASD. Remote eye gaze tracking is a noninvasive technique not requiring participants’ overt responses and has not significant technical or ethical limitations (Frazier et al. 2018). Further-more, measures can be rapidly collected across a wide range of ages and probably in different cul-tural settings. The present study aims to explore the adaptability of this technique and procedures to an Ecuadorian context (www.unige.ch/fapse/babylab/ le-babylab/equipe/projet-en-equateur).
+
+An overview of digital repositories from dif-ferent universities in Ecuador shows an increasing interest among young bachelors in psychology and pedagogy in fields related to intervention in children with ASD, suggesting a potential for the development of new lines of research. An impor-tant research priority is the understanding of the cultural fit and adaptations required to implement evidence-based practices originating from the West (Vivanti 2019).
+
+**Overview of Training**
+Ecuadorian universities currently offer postgrad-uate studies in pediatrics and neurology. The spe-cialization in neuro-pediatrics is not presently available, and this field relies on professionals trained in other countries who return to work in Ecuador. Currently, a training module for pedia-tricians and pediatric interns on ASD evidence-based practices is being developed in cooperation with the Department of Pediatrics of the Faculty of Medicine of the Catholic University of Quito (www.unige.ch/fapse/babylab/le-babylab/equipe/ projet-en-equateur).
+
+Aiming to facilitate the dissemination of scien-tific research in the field of ASD, the ESP included a session on “Validated screening tools in Spanish and their importance for a diagnostic process” during the 19th Ecuadorian Pediatric Congress in 2017. The newly created Ecuadorian Society of Neuropediatrics, supporting continuous educa-tion on early detection, included a session on “The challenges to diagnostic faced in pediatric settings and validated methods of intervention,” during the First Ecuadorian Neuropediatric Congress in 2019. The ESP has also supported in February 2018 the organization of workshops intended for service providers on an evidence-based early intervention model in Quito, in collaboration with trainers certified from the UC Davis MIND Institute.
+
+Education intended to service providers, such as speech therapists, psychologists, and occupa-tional therapists, is primarily offered as a bache-lor’s degree at many official universities across the country. Training on specific topics has been provided within universities, such as an introduc-tory module on ASD evidence-based practices addressed to students on special education of the Faculty of Education within the Universidad de las Américas in 2018. Several universities in major cities hold ad hoc conferences on ASD from a variety of theoretical backgrounds. Free-of-cost conferences organized by different par-ents’ associations try to raise awareness among the general public and particularly among teachers.
+
+**Social Policy and Current Controversies**
+Current controversies include the terminology used for the diagnostic of ASD. Ecuador’s public system mainly relies on the World Health Organi-zation’s International Classification of Diseases (ICD-10) (WHO 1993), one of the two official diagnostic systems. As ICD-10, which has con-served the traditional three categories dating back to Rutter’s (1978) criteria, coexists with the autism spectrum disorder’s description of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) (American Psychiatric Asso-ciation 2013), confusion about the diagnostic is common among parents and individuals with ASD. Furthermore, some families strongly iden-tify to their child’s diagnosis of Asperger syn-drome, perceived as less stigmatizing and more descriptive of milder conditions. For this reason, some professionals and parents oppose the disappearing of Asperger’s diagnosis (APADA, personal communication, December 2, 2019). Adapted and validated treatments are another important source of controversy. Media has become an authoritative source for families, and it is quite common to observe parents requesting advice and recommendations through this chan-nel. Furthermore, a miscellaneous offer of ser-vices, particularly for children, with sound, little, or no evidence, is advertised through the Internet. Some parents’ associations, mainly based in Spain, provide information on good practices and are becoming well known among family cir-cles in Ecuador (e.g., Autismo Diario). However, information about treatments that have little or no evidence (i.e., “Beware of non-evidence-based treatments,” 2019) are still scarce in Spanish. This situation highlights the importance of mak-ing knowledge easily accessible for all families, independently of their socioeconomic status or level of education, in order to facilitate informed decisions.
+
+Many other areas of controversy can be iden-tified, such as the profile of professionals qualified to give a diagnosis and to carry out interventions. For the time being, multidisciplinary teams do not seem to be constituted at the public level. In private sectors, a common source of concern for families is related to situations where the profes-sional giving a diagnosis of ASD also carries out an intervention with a method that is familiar to the professional but does not correspond to an individual’s need, thus excluding other models or strategies of intervention that could be more suitable. Also, in the field of services, the impact of very short-term trainings open to the public and certifications on methods of evaluation and inter-vention that have significant variability in terms of duration, theoretical background, evidence, and professional supervision need to be studied.
+
+Finally, an important concern for families relates to the lack of visibility given to the needs of adults on the spectrum, especially in the cases when the level of functioning compromises autonomy. Those themes illustrate that an absolute priority must be to provide families an access to knowl-edge on good practices that enable informed deci-sions on the assessment, intervention, and support processes at each stage of their children’s devel-opment. Attitudes toward individuals with ASD are changing as society becomes more aware of needs. Initiatives are no longer taken only by parents’ associations but also by civil society, as suggested by the organization of a discussion forum on the social integration of individuals with ASD, by an Ecuadorian NGO traditionally interested in sustainable development (Territorios Sostenibles, October 2019). The needs and con-straints of individuals with ASD may be more visible nowadays. However, awareness among the general population specifically about symp-toms’ manifestation may still be frail. Indeed, early identification of young children requires families’ participation in the decision process that leads to professional assistance. This process may not take place if autistic behaviors do not raise a certain level of concern among parents and professionals or if those behaviors are not understood as signs of a potential developmental disorder. In Ecuador, a study conducted on 183 adults concluded that most participants did not endorse many socio-communicative core symptoms as concerning enough to require pro-fessional assistance. Only language impairment and self-injurious behaviors attracted attention as concerning behaviors in young children by more than half of the respondents. On the other hand, most of the participants attributed the causes of autistic behaviors to factors unrelated to ASD or neurodevelopmental difficulties, such as child personality (Buffle et al. 2020). Those results suggest that “red flags” may not be recognized by families and non-trained professionals, which may lead to missing critical developmental oppor-tunities. It also suggests that a substantial number of cases may remain invisible, preventing the estimation of individuals needing services in Ecuador.
+
+Autism in Higher Education: Access, Challenges, and Support Strategies
+**Background**
+The Interagency Autism Coordinating Committee identified the development of services to support transition to adulthood as its first objective within the domain of lifespan research (IACC 2019). Around 50,000 adolescents with autism spectrum disorder (ASD) enter adulthood each year (Shattuck et al. 2012), many of whom have both an interest in higher education and the intellectual capacity to succeed academically. The develop-mental period bridging adolescence to adulthood is when multiple essential milestones are gener-ally achieved, including increased independence, autonomy, and responsibility (Arnett 2000). Unfortunately, many youths with ASD either fail to meet typical adult developmental milestones or decline in functioning during late adolescence (Picci and Scherf 2014).
+
+Young adults with ASD often experience lower quality of life than do age- and cognitive ability-matched healthy adults (Bishop-Fitzpatrick et al. 2017). They also face sustained challenges with living independently (Flynn and Healy 2012; Steinhausen et al. 2016) and finding and keeping gainful employment (Engstrom et al. 2003). One pathway to independence in adulthood and finan-cial mobility is through higher education. Based on data from the Bureau of Labor Statistics, edu-cation predicts income and people who obtain bachelor’s degrees, on average, earn more than those with just a high school diploma (BLS; Torpey 2018). As such, understanding how soci-ety, and especially those in mental health and higher education, can best support the needs of college-enrolled and college-bound people with ASD is important.
+
+**Transition from High School to College**
+Federal policies such as the No Child Left Behind Act of 2001 and the Individuals with Disabilities Education Improvement Act have greatly improved educational outcomes for students with ASD in elementary and secondary education (Smith 2005). But while these policy changes have provided more opportunity for students with ASD to pursue a college education, this is not necessarily a smooth or successful transition. Transitioning into college can be both thrilling and difficult for anyone (Shea 2019), but a grow-ing body of research indicates that emerging adults with ASD face unique challenges that can impede access to higher education. Adolescents and emerging adults with ASD often exhibit underdeveloped independence, lagging interper-sonal skills, and impaired ability to manage stress (Elias et al. 2019; Elias and White 2018). Partly because of these factors, it is estimated that only 50% of young adults with ASD pursue a college degree (Taylor and Seltzer 2011), a rate that is lower than the U.S. average college enrollment rate of 70% (U.S. Dept. of Health and Human Services 2017). Prior research also indicates that college students feel more supported academi-cally than socially (Cai and Richdale 2016), a trend which can lead colleges to overlook the social deficits inherent to ASD. However, despite these factors, colleges are still seeing a nationwide increase in applications from students with ASD, and more and more high school students with ASD decide to pursue postsecondary education every year (CITE).
+
+**Different College Pathways**
+Once an adult with ASD decides to go to college, several choices await them. The student must decide on a major, a living arrangement, and indeed the type of college or university they will apply to. These choices are important for all stu-dents to consider, especially so for students with ASD. There are multiple different pathways an adolescent with ASD could take, including attending vocational high school, studying at a community college, or pursuing a 4-year degree.
+
+Firstly, where do students with ASD attend? Many students with ASD enroll in community colleges and local institutions. One study reported that around 80% of postsecondary students with ASD were in a community college for at least part of their education (Wei et al. 2014). Some pro-fessionals have recommended this pattern as ben-eficial for students with ASD. For example, according to Adreon and Durocher (2007), Lars Perner (2002) suggested that the increased per-sonal attention found in community colleges can help a student with ASD adjust to the routines and responsibilities of college. Conversely, large uni-versities provide far less opportunity to receive individualized attention, especially in large clas-ses (Freedman 2010). It may also be worth con-sidering what sort of programs colleges have to offer students with disabilities; for example, Shmulsky et al. (2015) demonstrated that partici-pation in a holistic transition program resulted in students with ASD having a higher first-year com-pletion rate than unsupported typically develop-ing students.
+
+Secondly, what majors do students with ASD choose? Research on this topic is minimal, but studies on the majors chosen by students with ASD suggest that STEM majors are the most commonly chosen areas of study (Baron-Cohen et al. 2007; Fessenden 2013; Wei et al. 2013). In particular, computer science seems to be a popular area of study for these students (Wei et al. 2013). Further research is needed to better understand what students with ASD are interested in study-ing, so as to better-inform program administrators.
+
+Finally, where do students with ASD choose to live? Some students commute while others live in campus housing. Students with ASD may choose to commute for several reasons. It may be due to family finances and commuting can prevent a family from being charged a housing fee by the college or university (Buescher et al. 2014; Shimabukuro et al. 2008; Wei et al. 2014). It may also be due to personal factors such as defi-cits in independent living skills (Elias and White 2018; Steinhausen et al. 2016; Van Hees et al. 2015). Conversely, if the student decides to live in campus housing, they may have a roommate. Studies have shown that certain factors in room-mates, such as aloofness (preference of solitary activity and decreased social involvement) can play a key role in fostering good relationships with students with ASD, with phenotypes closer to the autism phenotype resulting in higher rela-tionship satisfaction from both roommates (Faso et al. 2016).
+
+**Strengths of College Students with ASD**
+Students with ASD typically bring a variety of strengths with them to college. The prevalence of special abilities and talents among those with ASD has long been recognized (Asperger 1944; Kanner 1943). Around two-thirds of individuals with ASD are thought to possess special isolated skills (Meilleur et al. 2015) which can be harnessed for success in higher education. Atten-tion to detail, strong memory, adherence to rules and guidelines, passionate interests, and intense knowledge of a particular subject area are com-monly noted strengths of these students (Anderson et al. 2017; Gobbo and Shmulsky 2014), as well as openness to feedback and sug-gestions (Elias et al. 2019). Other strengths which may prove useful in higher education include enhanced perceptual functioning (Mottron et al. 2006), superior pitch discrimination (Heaton et al. 2008), hyperlexia (Ostrolenk et al. 2017), mathe-matical/calculating skills (Howlin et al. 2009), as well as musical, artistic, and other abilities (Meilleur et al. 2015).
+
+Previous research suggests individuals with ASD tend to do particularly well in STEM areas of study. These students often have a cognitive style that lends itself particularly well to STEM fields: an ability to observe, identify, construct, and apply logical rule-based systems of reasoning to explain the world around them (Cox et al. 2016), i.e., “systemizing” (Baron-Cohen 2009). It has even been suggested that those with ASD may have an innate predisposition for STEM with higher autism rates among children whose parents work in STEM fields (Baron-Cohen 1998; Baron-Cohen and Hammer 1997). Students with ASD may approach problems in science and engineer-ing in unique ways, develop novel and divergent solutions, and show strength, resilience, and determination (Baron-Cohen 2009). Interestingly, students with ASD who pursue STEM fields tend to progress further in their education and are more likely to finish all 4 years or transfer from a com-munity college into a 4-year college or university (Wei et al. 2014). As previously mentioned, col-lege students with ASD are more likely than stu-dents in other disability categories, and students in general, to gravitate towards STEM fields, and the efforts of these students are a credit to their fields.
+
+**Challenges Experienced by College Students with ASD**
+Emerging adults with ASD may experience sig-nificant difficulties during their college career. Social interactions, organization and time man-agement, managing anxiety and depression, maintaining motivation, and sensory overload have all been noted as areas of both frequent and severe difficulty for many students with ASD (Alverson et al. 2015; Trembath et al. 2012; White et al. 2016). Loneliness and isolation are common problems for college students with ASD (Madriaga and Goodley 2010). Knowing where to meet other students with similar interests, initiat-ing and maintaining conversations with class-mates, and findings ways to connect with other students can all be challenging. Other common situations such as group projects, maintaining appropriate classroom behavior, and in-class debates all demand complex social skills which students with ASD may not have (Cullen 2015). By the students’ own admissions, these social needs often go under-supported (Cai and Richdale 2016).
+
+Furthermore, the college environment is dra-matically different from high school, particularly concerning changes in schedule and routine, self-autonomy, and the need for self-advocacy skills (Van Hees et al. 2015). Living on campus also may require negotiating with roommates and independently managing a range of personal responsibilities such as doing laundry, cleaning, self-care, and handling mealtimes. High schools often have a somewhat invisible support system where teachers, staff, and classmates know and understand the student with ASD. Particularly on a large campus, these invisible supports are diffi-cult to replicate in a college environment.
+
+**Disclosure of ASD Diagnosis**
+An additional challenge for students with ASD in college is the issue of disclosure. Many students with a diagnosis of ASD choose not to disclose (Van Hees et al. 2015) for a variety of reasons, which may include relation to self-identity, expected benefits from disclosing, and previous experiences with disclosure. Many college stu-dents with disabilities may want to eliminate the label of being disabled to reset their social identity (Marshak et al. 2010). Data from the National Longitudinal Transition Study-2 indicated that around 33% of students with ASD did identify as “disabled” (Shattuck et al. 2014). The colleges themselves may not do an adequate job of provid-ing information to incoming students about the services and supports which may be available to them and how to navigate the disability services system.
+
+Further complicating the situation, disclosure does not guarantee assistance. One study of dis-closing students attending 2-year colleges found that less than half reported receiving any ser-vices or accommodations (Roux et al. 2015). Others have reported reluctance to disclose their ASD diagnosis until they encounter a sig-nificant problem or are unable to cope (e.g., Gobbo and Shmulsky 2014; Van Hees et al. 2015). Finally, unlike in high school, in college it is the student’s responsibility to take initiative and seek help. Students with ASD, who may have depended upon parents and teachers to set goals (Elias et al. 2019), may fail to access these resources simply because they are not used to doing so.
+
+**Attitudes Towards College Students with ASD**
+Attitudes among students, faculty, staff towards students with ASD are another area of concern. Research has shown a significant lack of knowl-edge and understanding of ASD among faculty and staff (Glennon 2016; Tipton and Blacher 2014), leading to frustration among faculty and inaccurate interpretations of inappropriate class-room behavior. These and other misconceptions about autism have led to stigmatization and exclu-sion of these students (Gillespie-Lynch et al. 2015; Gobbo and Shmulsky 2014; Schindler et al. 2015; Wenzel and Brown 2014).
+
+What about students? Student responses vary and seem influenced by major and previous expe-rience. Students often distance themselves from students with ASD (Gardiner and Iarocci 2014) but those more familiar with autism seem to be more accepting (Nevill and White 2011). Students studying engineering and physical sciences have been shown to be more willing to interact with a student with ASD compared to those majoring in arts and social sciences (Nevill and White 2011). Inclusion in training programs presenting infor-mation about ASD has been shown to increase knowledge and decrease stigma among college students (Gillespie-Lynch et al. 2015), since understanding motivation for behavior can help reduce stigma (Butler and Gillis 2011). However, studies by both Gillespie and colleagues and by Matthews et al. (2015) found that training has a greater impact on behavior and cognitive attitudes towards individuals with ASD and less impact on the affect experienced by these trained students.
+
+Related to this work, White et al. (2019) exam-ined student knowledge and attitudes towards other college students with ASD, the underlying factors contributing to such attitudes, and whether attitudes changed over a 5-year period. While the later cohort had greater knowledge and more pos-itive attitudes towards students with ASD, there was no significant relationship between knowl-edge and attitudes. Even after being presented with an accurate list of traits that might be seen in a student with ASD, students who had previ-ously identified a higher number of aggressive or misleading traits still demonstrated less positive attitudes – their own beliefs still trumped new factual knowledge. These findings imply that despite increasing knowledge and understanding of ASDs in society, negative attitudes remain resistant to change.
+
+Students who personally knew someone with ASD had more positive attitudes toward their peers with ASD, consistent with other research in this area (Gillespie-Lynch et al. 2015; Nevill and White 2011). Students who did not know someone with ASD were more likely to endorse inaccurate traits related to cognitive deficits, per-haps reflective of stereotypes about disability more broadly, which are often perpetuated by a lack of contact. The conclusion that knowledge about ASD does not necessarily mediate attitudes toward peers with ASD is consistent with much of the research on attitudes toward members of minority populations, including those with dis-abilities and mental health issues (Allport 1954; May 2012; McManus et al. 2011).
+
+**Strategies to Support College Students with ASD**
+With all of this in mind, how can we best support students with ASD? Given the potential for suc-cess among college students with ASD, formulat-ing effective support strategies is a priority for an increasing number of higher education institu-tions. Existing supports available through univer-sity counseling centers and learning and academic support services including tutoring and advising are often helpful. Disability services offices also play an important role in setting up academic accommodations which, depending on a student’s disabilities and eligibilities, might include extended time for exams, having a note-taker dur-ing class, or taking verbal exams (Egan and Giuliano 2009). However, barriers to accessing accommodations are multilayered, beginning with concerns of disclosure, as outlined above. There is a clear need to identify cost-effective programming that can be implemented with effi-cacy and which improve retention and success for students with ASD (Barnhill 2016).
+
+**Supports for Social Skills**
+As stated previously, college students with ASD often state that they are not receiving adequate social or educational support in this new setting (Cai and Richdale 2016). Students themselves seem willing to attend support groups where they can meet other students with an ASD diagnosis (Van Hees et al. 2015), although colleges offering such groups have reported concerns that students were not receptive to the group, did not show up at the expected time, or that breaking down social skills was not helpful (Barnhill 2016). While an increasing number of institutions are developing tailored support groups, mentoring programs, and special tutoring, empirical support for many of these activities is minimal (Gelbar et al. 2014), and it is not yet clearly understood what supports are most helpful for students with ASD (Cox et al. 2016). It is clear, however, that support for social difficulties is critical (White et al. 2016) and likely has major implications for retention and success.
+
+In part to address these problems, Hillier et al. (2018a) provided a support group program, “Con-nections,” for college students with ASD which had a broad curriculum addressing not only social skills but a range of other potential challenges including academic skills, time and stress man-agement, managing group work, and future plans. Group members indicated significant reductions in loneliness and anxiety and increase in self-esteem at the end of the program. Focus groups were conducted to examine functional changes in academic and social skills, and to hear directly from students themselves, a notable gap in the literature focused on students with ASD (Cox et al. 2016; Gelbar et al. 2014). Five prominent themes were identified in the focus group analysis which reflected how the program had positively impacted participants’ skills and coping: execu-tive functioning; goal setting; academics and resources; stress and anxiety; and social. Given that college students typically make their own decisions regarding interventions and services they are willing to receive, the program’s social validity was also assessed and participants indi-cated that the group was acceptable, socially rel-evant, and useful to them.
+
+**Mentoring Programs**
+Mentoring is a well-established strategy for improving outcomes for a broad range of populations: at-risk youth (Britner et al. 2006), fos-ter children (Rhodes et al. 1999), workplace men-torship (Janssen et al. 2015), and individuals with disabilities (Daughtry et al. 2009). Increasing atten-tion has been paid to the possibility of implementing mentoring programs in higher education institutions for students with ASD. While many factors affect a program’s success (Rhodes and DuBois 2008), a meta-analysis of 73 youth mentoring studies identi-fied “best practices” including ongoing training for mentors, recruiting mentors with a background in helping roles or professions, structured activities for mentors and youth, expectations for frequency of contact, mechanisms for support and involvement of parents, and monitoring of overall program implementation (DuBois et al. 2011).
+
+Hillier et al. (2019a) reported on a mentoring program for college students with disabilities, the majority of whom had an ASD diagnosis. The program supported freshmen for one semester and consisted of one-on-one hour-
\ No newline at end of file