Diagnostic evaluation

Evaluation answers a defined diagnostic question

The current CDC clinical diagnosis page says autism diagnosis usually draws on descriptions of a child's development from parents or caregivers and professional observation. It also states that no single diagnostic tool should serve as the basis for diagnosis.

For an adult or a person able to report directly, their own developmental history, sensory experience, communication, relationships, daily demands, masking or compensation, strengths, and support needs are important sources. Family or old records may add context when available and authorized. Their absence should be documented rather than filled with assumptions.

A diagnostic conclusion answers whether the available evidence meets current criteria. The evaluation can also identify questions needing medical, genetic, hearing, vision, communication, cognitive, adaptive, learning, mental health, motor, or other assessment. Each question belongs to a professional with the relevant competence and authority.

Screening, diagnosis, and eligibility differ

The CDC screening page states that screening does not diagnose autism. A high- or low-likelihood result remains one input. Permission to administer a tool does not confer authority to diagnose.

For school services, IDEA regulation 34 CFR 300.8 defines a child with a disability using both an identified category and a need for special education and related services. Its educational autism definition and decision purpose differ from a health-care diagnosis. State rules and the complete evaluation requirements still apply.

Components should fit the person and question

The 2025 NIMH autism publication describes possible components for children, including medical and neurological examination, cognitive and communication assessment, observation, detailed developmental history, adaptive functioning, family history, and sometimes hearing or blood testing. It also notes that adult evaluation pathways continue to evolve.

The AAP autism clinical report supports a clinical history and examination, assessment of development and adaptive function, etiologic and co-occurring-condition evaluation as indicated, and shared planning. The exact team and tests depend on age, presentation, existing evidence, and the questions being answered. A long battery is not automatically more valid.

Before the appointment, ask:

• Which diagnosis or referral question will this evaluation address?
• Who makes the diagnostic decision, and under which license or authority?
• Which interviews, observations, measures, records, and medical checks are planned?
• How are language, culture, hearing, vision, motor, sensory, fatigue, and literacy needs handled?
• How can the person use AAC, take breaks, decline an activity, or correct the record?
• What will the written report include, and when will results be explained?
• Which costs, prior authorization, or network rules apply?

Communication and assent affect evidence quality

An evaluation should not require speech, eye contact, still hands, pretend play, or one motor response when an accessible alternative can answer the question. The ASHA AAC practice portal says AAC users should always have access to their tools or devices.

Record the system, vocabulary, access method, positioning, partner support, and wait time used. If a tool was standardized without those supports or an item was changed, document the deviation and interpret cautiously. An evaluator should not score an access failure as absence of an underlying ability.

Consent from an authorized decision-maker does not make assent irrelevant when assent applies. Explain the activities in accessible language, recognize stop or pause signals, and separate necessary health or safety action from optional test participation.

Results need explanation and a next-step map

A useful report distinguishes direct observation, self-report, caregiver or teacher report, records, test findings, clinical interpretation, diagnosis, and recommendations. It states limitations, contradictory evidence, tool versions, conditions of administration, and unanswered questions.

Possible outcomes include a supported diagnosis, criteria not met, an uncertain or deferred conclusion, or another referral. A diagnosis does not prescribe one service, number of hours, school placement, or life outcome. A decision not to diagnose should still address documented needs and explain when reevaluation may be appropriate.

The current CDC autism FAQ confirms that adults can be diagnosed and notes that incomplete early history and overlapping prior diagnoses can make evaluation harder. Difficulty does not justify substituting a child tool outside its age range or ignoring the adult's own account.

A fictional evaluation example

Rowan is a fictional nine-year-old who uses speech and AAC. A screening result, school concerns, and Rowan's own questions lead to an autism diagnostic evaluation. The clinician reviews developmental and health history, Rowan's communication and sensory account, school records, direct observation, adaptive functioning, and hearing information.

Across six planned observation blocks, Rowan's AAC is available and positioned correctly in four. The evaluator records the other 2 of 6 as administration and access failures rather than evidence of missing communication. In the four accessible blocks, Rowan chooses AAC, speech, or both and is offered breaks before each activity.

The evaluator integrates all sources, explains which criteria are and are not supported, documents limits, and identifies follow-up questions. The family receives a written report and explanation. No percentage from these six blocks becomes Rowan's probability of autism, and the diagnosis does not automatically establish school eligibility, ABA need, payer coverage, or service intensity.