Position statement
Posted: Oct 24, 2019
Jessica A. Brian, Lonnie Zwaigenbaum, Angie Ip; Canadian Paediatric Society, Autism Spectrum Disorder Guidelines Task Force
Paediatr Child Health 2019 24(7):444–451.
The rising prevalence of autism spectrum disorder (ASD) has created a need to expand ASD diagnostic capacity by community-based paediatricians and other primary care providers. Although evidence suggests that some children can be definitively diagnosed by 2 years of age, many are not diagnosed until 4 to 5 years of age. Most clinical guidelines recommend multidisciplinary team involvement in the ASD diagnostic process. Although a maximal wait time of 3 to 6 months has been recommended by three recent ASD guidelines, the time from referral to a team-based ASD diagnostic evaluation commonly takes more than a year in many Canadian communities. More paediatric health care providers should be trained to diagnose less complex cases of ASD. This statement provides community-based paediatric clinicians with recommendations, tools, and resources to perform or assist in the diagnostic evaluation of ASD. It also offers guidance on referral for a comprehensive needs assessment both for treatment and intervention planning, using a flexible, multilevel approach.
Keywords: Autism spectrum disorder; Diagnostic evaluation; Intervention planning
In most provinces and territories, only physicians or psychologists are licensed to diagnose autism spectrum disorder (ASD) [1]. In some communities, appropriately trained nurse practitioners may also make this diagnosis. Emerging evidence suggests that a trained sole practitioner can diagnose less complex cases of ASD [1]–[4], yet most clinical guidance documents recommend a team-based approach, led by a primary care provider, paediatric specialist, or clinical child psychologist who is trained to diagnose ASD [1][3]–[15].
Children with suspected ASD are often first identified by a paediatrician, family physician, parent, or another caregiver, and can present with a wide range and severity of symptoms. A ‘one-size-fits-all’ multidisciplinary team diagnostic approach is inefficient, and contributes to long wait times [1][16]. This statement proposes three ASD diagnostic approaches, the choice of which depends upon the paediatric care provider’s clinical experience and judgment, and the complexity of symptom presentation [8]–[15] and/or psychosocial history (Figure 1). Regardless of the approach taken, open communication, collaboration, and consent to share information among professionals may help to achieve diagnostic accuracy and avoid duplication of effort.
When a child’s symptoms clearly indicate ASD, an experienced or trained sole paediatric care provider can independently diagnose ASD, based on clinical judgement and DSM-5 criteria [17], with or without data obtained using a diagnostic assessment tool (Table 1). However, this approach is not sufficient for accessing specialized services in some Canadian jurisdictions [1].
In the shared care model, a clinician has joint responsibility with another health care provider for patient care, which involves exchanging patient information and clinical knowledge. When a child’s symptom presentation is milder, atypical, or complex, or a child is under 2 years of age, a paediatric care provider may use information from an ASD diagnostic assessment tool, and consult with another health care professional with specialized knowledge (e.g., a psychologist) to inform a diagnosis.
In a team-based approach, diagnostic assessment is performed by health care professionals in an interdisciplinary or a multidisciplinary team. While interdisciplinary teams work collaboratively in an integrated, coordinated fashion, multidisciplinary team members work independently from one another but share information, and may (or may not) reach a diagnostic decision by consensus. In some Canadian jurisdictions, only a team-based diagnostic approach is accepted for accessing specialized services [1].
A community clinician may refer to an ASD diagnostic team when a child’s presentation is subtle, or complicated by co-existing health concerns, or when a child has a complex medical or psychosocial history. Any of these factors can make diagnostic determinations more difficult. Although diagnostic certainty may be improved by drawing on specialists’ expertise [3]–[15], this team approach is not always required and may prolong wait times unnecessarily. Diagnostic delays can be a barrier to early interventions, especially when some communities have limited access to specialized services or teams [1][2][16]. However, multidisciplinary team evaluations can also help to capture information for treatment or intervention planning, and to optimize access to supportive programs.
Figure 1. Algorithm and components of an autism spectrum disorder diagnostic assessment. |
There are no diagnostic biomarkers for ASD. This condition is diagnosed clinically, based on information gathered from a detailed history, physical examination, and the observation of specific characteristic behaviours. Most guideline documents have not offered a maximal acceptable wait time for diagnosis [8][11]–[15], although three reputable guidelines recommend a 3- to 6-month interval between referral and assessment [6][7][10]. An expedited ASD diagnostic assessment, or a referral for one, facilitates earlier and potentially concurrent access to educational interventions and community-based services. The three key objectives of the ASD diagnostic assessment are to:
1. Provide a definitive (categorical) diagnosis of ASD. In ambiguous cases (e.g., milder presentation, under the age of 2), a provisional diagnosis can be made, but the child must be monitored carefully, and referred for further, in-depth evaluation. In many jurisdictions, specialized ASD interventions are not available to children with a provisional diagnosis.
2. Explore conditions or disorders that mimic ASD symptoms and identify co-morbidities.
3. Determine the child’s overall level of adaptive functioning, including specific strengths and challenges, and personal interests, to help with intervention planning.
A family-centred approach is essential in the ASD diagnostic process because a substantial amount of time and care is required to listen to and talk with parents and other family members. Understanding the child’s family and medical history, along with current behaviours and individual and family treatment goals, is central to family-centred care. This approach also requires flexible appointment scheduling and anticipating potential barriers to health care access (e.g., socio-economic adversity), as well as considering cultural and religious values [8]–[10]. When families and clinicians do not speak the same language, involving interpretation services is essential for culturally competent care.
The essential elements of an ASD diagnostic assessment are described below.
Most diagnosing clinicians begin by reviewing a child’s previous records and interviewing parents and caregivers about their concerns. This process is typically followed by taking a detailed developmental history and scheduling an appointment for direct observation and assessment.
Information is obtained by asking semi-structured, open-ended questions, and may be integrated with information from a standardized questionnaire completed before or during the interview. Topics include:
Table 1. Commonly used ASD diagnostic tools | |||||
Diagnostic tool |
Age group |
Time for completion |
Test performance [18] |
Test sample |
Comments |
Behavioural observation (all available for purchase) | |||||
Autism Diagnostic Observation Schedule – 2nd edition (ADOS-2)* |
12 months to adult |
45–60 minutes |
Se: 92–100% [3] Sp: 61–65% PPV: 80–84% NPV: 81–100% |
ASD and non-ASD patients; clinical sampling |
Social interaction, play, communication, behaviours and interests |
Childhood Autism Rating Scale – 2nd edition (CARS-2)* |
2+ years |
20–30 minutes |
Se: 89–94% [3] Sp: 61–100% PPV: 71–100% NPV: 46–56% |
Clinically referred children |
15-item checklist. Incorporates information from parents |
Parent/Caregiver interviews (all available for purchase) | |||||
Diagnostic tool |
Age group |
Time for completion |
Test performance [18] |
Test sample |
Comments |
Autism Diagnostic Interview- Revised (ADI-R)* |
2 years to adult |
1.5–3 hours |
Se: 53–90% [3] Sp: 67–94% PPV: 82–98% NPV: 26–68% |
Clinically referred. High to low risk for ASD |
Social interaction, communication, behaviours and interests. Mostly used for research |
Social Responsiveness Scale – 2nd edition (SRS-2); preschool and school-aged versions |
2.5–4.5 years; 4–18 years |
15–20 minutes |
Se: 75–78% Sp: 67–94% PPV: 93% |
442 children with or without ASD |
65 items: social awareness, cognition, communication and motivation, repetitive behaviours, and interests |
Diagnostic Interview for Social and Communication Disorders (DISCO)* |
Any age |
Up to 3 hours |
Se: 96% [19] Sp: 79% |
Children with different levels of ID |
Detailed, semi-structured interview, with a dimensional approach. Mostly used for research |
Developmental, Dimensional and Diagnostic Interview (3di)* [20] (Computerized) |
3+ years |
Up to 2 hours |
Se: 84–100% [3] Sp: 54–98% |
Children with or without ASD |
740 items: demographics, intensity of ASD symptoms, and potential co-morbidities |
Data drawn from references [8][9][14][17]-[20]. *Training is required to use all tools except for the SRS-2. ASD Autism spectrum disorder; ID Intellectual disability; NPV Negative predictive value; PPV Positive predictive value; Se Sensitivity; Sp Specificity. |
This step involves observing and interacting with the child to assess social interaction and communication abilities. Inquire about or observe for patterns of behaviour or interests, with focus on DSM-5 diagnostic criteria for ASD [17]. Behavioural observation in multiple settings is preferred (e.g., clinic, home, or child care), but most assessments occur in clinical settings. An ASD-specific diagnostic tool may also be used.
There are two broad categories of ASD diagnostic tools: those based on coding observations and direct interactions with the child (e.g., ADOS-2, CARS-2) [3][18][21], and those based on parent or caregiver interviews (e.g., ADI-R) or questionnaires (e.g., SRS-2) [3][21][22]. Findings from an ASD diagnostic assessment tool cannot be used alone to diagnose ASD [3]. Rather, results should complement the diagnostic process and inform clinical judgement. Clinically useful ASD diagnostic assessment tools have a sensitivity (i.e., they correctly identify children with ASD) and specificity (i.e., they correctly identify children without ASD) of at least 80%. Common age-specific ASD diagnostic tools are summarized in Table 1 [8][9][14][17]-[20].
The most recent Cochrane review of ASD diagnostic tools for preschoolers found that the ADOS has the highest sensitivity and comparable specificity to the CARS and ADI-R [22]. In some jurisdictions, the ADOS and ADI-R are required to inform ASD diagnosis [2][5].
The physical exam assesses whether there are medical causes or associations with the child’s behavioural presentation. Findings can help to guide treatment and intervention planning and include:
When making an ASD diagnosis, it is important to consider other disorders that might overlap or mimic ASD symptoms. Table 2 summarizes common differential diagnoses that may also co-occur with ASD.
A trained clinician uses DSM-5 criteria along with clinical judgment to differentiate ASD from other developmental disorders [1]. For a description of DSM-5 criteria, see Table 1, Early Detection for Autism Spectrum Disorder, in the companion statement published in this issue.
Ideally, both parents and the child (when age and otherwise appropriate) should be invited for a face-to-face appointment shortly after the ASD diagnostic evaluation. This visit should be conducted in the family’s primary language, using an interpreter if necessary, with discussion of assessment findings, prognosis, and recommended supportive resources. When an in-person appointment is not possible due to remote location or travel restrictions, a telephone- or video-conference should be arranged instead.
This appointment must be handled in a sensitive, supportive manner. Parents should be given time to process the information being given and to ask questions, without distractions or interruptions. Asking parents what questions they have (rather than if they have questions) is one helpful way to open a conversation. When both parents are involved in the child’s care, they should be present together for the feedback meeting. A lone parent should be encouraged to bring a support person to this appointment.
The information below should be provided verbally and as part of a comprehensive, informative written report, with terms and references suitable for a lay audience.
Table 2. Common differential diagnoses and co-occurring conditions in ASD [10][14] |
Neurodevelopmental disorders Attention-deficit hyperactivity disorder |
Mental/Behavioural disorders Anxiety disorders |
Genetic conditions Fragile X syndrome |
Neurological and other medical conditions Cerebral palsy |
*Social (pragmatic) communication disorder and ASD are mutually exclusive. **There are many other chromosomal abnormalities and genetic syndromes associated with ASD [26] ASD Autism spectrum disorder. |
Share the written report with the child’s family, the referring physician, other health care providers, and educational professionals who are involved with the child’s care plan, with appropriate consents.
What to do when a categorical ASD diagnosis cannot be determined at the time of assessment.
When a diagnosis is unclear, consider:
After a child has been diagnosed with ASD, the paediatric clinician’s role is to ensure that a comprehensive needs assessment for treatment and intervention planning is considered, when this step was not part of initial diagnostic evaluation. This may involve advocating for further assessment if it appears that a lack of clarity about the child’s functioning may undermine effective planning. This assessment may recur or be revisited at different points throughout child or adolescent development, and in many cases will occur within the context of the child's educational or treatment program. Understanding overall levels of functioning in several domains, including strengths, skills, challenges, and needs, helps with developing effective, individualized treatment and management planning. Such plans should also consider both individual and family concerns, priorities, and resources. The needs assessment may evaluate:
The extent of comprehensive assessment for intervention planning is influenced by the approach used for initial ASD diagnostic evaluation.
Diagnostic evaluations conducted by a sole paediatric care provider tend to focus on the core domains of ASD. Therefore, referral to an interdisciplinary team or to multiple professionals for a comprehensive needs assessment may be required, particularly when planning will not be conducted in the context of a child’s educational or treatment program.
Diagnostic evaluations carried out with a shared care partner can provide additional information for intervention planning (e.g., insight into cognitive functioning). Further referrals may only be needed to understand additional areas of functioning (e.g., sensory or motor), if indicated.
When a child is evaluated by a multidisciplinary team, diagnostic evaluation and assessment for intervention planning are more likely to occur concurrently. Additional assessments may or may not be necessary.
Although a definitive diagnosis for ASD is possible in children under 2 years of age, this can be challenging. Children this young often exhibit symptoms that are subtle or less distinguishable from other developmental delays—or even from typical development [8][10]. Very young children who receive a provisional ASD diagnosis will need a timely follow-up evaluation or referral for further assessment because their symptoms can change substantially during development [27]. The importance of confirming or ruling out an ASD diagnosis as early as possible cannot be overstated.
ASD is diagnosed four times more frequently in boys than girls [27]. In younger siblings of children with ASD, the boy:girl ratio is approximately 3:1 [28]. When symptoms are equally severe, boys are more likely to be diagnosed. Gendered differences in symptom presentation may contribute to later diagnosis in girls [8]–[10]. Also, the evidence suggests that girls are better able to ‘camouflage’ symptoms and use compensatory strategies to overcome communication and social difficulties [13]. Clinicians should be aware that girls are probably underdiagnosed for ASD [29][30].
Current evidence further suggests that children from racial or ethnic minorities are diagnosed later than their peers in the general population [6][8][22]. It is unclear the extent to which this disparity is attributable to limited access to diagnostic services, interpretation of symptoms by families, health care system-related factors [14], or an interplay thereof. Cultural factors can impact many aspects of health care, including how ASD is understood, interpreted, and accepted in different communities. Cultural factors may, in turn, affect help-seeking behaviours [31].
Children living in rural or remote communities receive an ASD diagnosis later than peers living in urban areas [6][8][32]. Shared care between a local primary care provider and ASD specialist(s) further away can be facilitated through telehealth (especially video-conferencing services) and by itinerant developmental clinics which travel to rural or remote areas on a seasonal schedule [6][8][32]. Capacity building in underserved communities should be a priority.
When an ASD diagnosis has been confirmed, a community-based paediatrician or primary care provider plays a vital role in supporting both child and family. Physicians provide routine health care, manage co-occurring medical conditions, and coordinate care among medical, educational, community, and social care professionals. For additional information on condition management, read the companion statement published in this issue, including links to online resources.
Funding: Production of these guidelines has been made possible through funding from the Public Health Agency of Canada. The views expressed herein do not necessarily represent the view of the Public Health Agency of Canada.
Potential Conflicts of Interest: Dr. Zwaigenbaum reports personal fees from Roche - Independent Data Monitoring Committee (iDMC), outside the submitted work. There are no other disclosures. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
Members: Mark Awuku MD (CPS Community Paediatrics Section), Jessica Brian PhD (co-Chair), Susan Cosgrove, Pam Green NP, Elizabeth Grier MD (College of Family Physicians of Canada), Sophia Hrycko MD (Canadian Academy of Child and Adolescent Psychiatry), Angie Ip MD, James Irvine MD, Anne Kawamura MD (CPS Developmental Paediatrics Section), Sheila Laredo MD PhD (Canadian Autism Spectrum Disorders Alliance), William Mahoney MD (CPS Mental Health Section), Patricia Parkin MD, Melanie Penner MD, Mandy Schwartz MD, Isabel Smith PhD, Lonnie Zwaigenbaum MD (co-Chair)
Principal authors: Jessica A. Brian PhD, Lonnie Zwaigenbaum MD, Angie Ip MD
Disclaimer: The recommendations in this position statement do not indicate an exclusive course of treatment or procedure to be followed. Variations, taking into account individual circumstances, may be appropriate. Internet addresses are current at time of publication.
Last updated: Feb 7, 2024