ANQA Clinician Survey

AI-Powered ADHD Diagnostics: Healthcare Provider Perspectives

Estimated time: 10–15 minutes

Incentive: Summary of findings and early demo access

Introduction

We are developing an AI-powered platform to support ADHD diagnosis and would value your professional insights. This research will inform the development of clinician-assistive software designed to enhance, not replace, clinical decision-making.

Auto-restore answersEnable to automatically restore half-finished surveys on this device.

Progress (0/0)0%

Section A

Professional Background

A1. Primary professional role

PsychiatristPsychologistPediatricianGeneral Practitioner/Family PhysicianPsychiatric Nurse PractitionerClinical Social WorkerNeurologist

Other:

A2. Years of clinical experience

Less than 2 years2-5 years6-10 years11-20 yearsMore than 20 years

A3. Practice setting

Private practiceHospital/Medical centerCommunity mental health centerUniversity/Academic medical centerSchool-based clinicTelehealth platform

Other:

A4. Geographic location

AustriaGermanySwitzerland

Other:

A5. Patient population (select all that apply)

Children (under 12)Adolescents (12-17)Adults (18-64)Older adults (65+)

A6. Approximate number of ADHD patients you see per month

0-56-1516-3031-50More than 50

Section B

Current ADHD Diagnostic Workflow

B1. Diagnostic tools used (select all that apply)

Clinical interview (DSM-5/ICD-11 criteria)Conners Rating ScalesAdult ADHD Self-Report Scale (ASRS)Wender Utah Rating Scale (WURS)QbTest/QbCheckComputerized continuous performance testsPsychological testing batteriesCollateral information from family/teachers

Other:

B2. Average time spent on complete ADHD diagnostic evaluation

30-60 minutes1-2 hours2-3 hours3-4 hoursMore than 4 hours

B3. Appointments typically required for ADHD diagnosis

1 appointment2 appointments3 appointments4+ appointments

B4. Current wait time for new ADHD evaluations

Same week1-2 weeks3-4 weeks1-2 months3-6 monthsMore than 6 months

B5. Most significant bottlenecks (select up to 3)

Limited appointment availabilityTime-intensive assessment protocolsDifficulty obtaining collateral informationComplex documentation requirementsInsurance pre-authorizationDifferential diagnosis complexityPatient no-shows/cancellationsWaiting for psychological testing

Other:

Section C

Documentation & Reporting

C1. Time spent on documentation per ADHD evaluation

15-30 minutes31-60 minutes1-2 hours2-3 hoursMore than 3 hours

C2. Burden of current diagnostic documentation

1 = No burden, 5 = Excessive burden

12345

C3. Components of an ideal ADHD diagnostic report (select all that apply)

Symptom severity scoresDSM-5/ICD-11 criteria checklistFunctional impairment assessmentDifferential diagnosis considerationsTreatment recommendationsComorbidity screening resultsObjective test resultsPatient strengths and capabilitiesFamily/collateral perspectivesRisk factors and protective factors

C4. Preferred format for AI-generated diagnostic reports

Structured template with scored sectionsNarrative summary with bullet pointsVisual dashboard with charts/graphsIntegration into existing EHR templatesCombination of structured and narrative elements

Section D

Technology Integration

D1. Current Electronic Health Record (EHR) system

EpicCernerAllscriptseClinicalWorksathenahealth

Other:

Paper-based records

Other:

D2. Importance of EHR integration for new diagnostic tools

1 = Not important, 5 = Essential

12345

D3. Experience with AI/machine learning tools in clinical practice

Extensive experience and very comfortableSome experience and generally comfortableLimited experience but open to learningNo experience but interestedNo experience and skeptical

D4. Preferred method for accessing AI diagnostic tools

Integrated within EHRStandalone web applicationMobile appDesktop softwareNo preference

Section E

AI-Assisted ADHD Diagnosis

E1. Interest level in AI-powered ADHD diagnostic support tools

Very interestedSomewhat interestedNeutralSomewhat hesitantNot interested

E2. Acceptable false positive rate for AI ADHD screening tool

Less than 5%5-10%11-15%16-20%More than 20%Unsure

E3. Acceptable false negative rate for AI ADHD screening tool

Less than 5%5-10%11-15%16-20%More than 20%Unsure

E4. Most valuable AI capabilities for ADHD diagnosis (select top 3)

Automated symptom severity scoringPattern recognition in patient responsesObjective behavioral analysis (video/audio)Risk stratification and triageDifferential diagnosis supportTreatment recommendation algorithmsOutcome prediction modelingDocumentation automation

E5. Primary concerns about AI-assisted ADHD diagnosis (select all that apply)

Diagnostic accuracy and reliabilityLegal liability and malpractice concernsPatient acceptance and trustLoss of clinical intuition/skillsData privacy and securityImplementation complexityCost and reimbursementRegulatory complianceOver-reliance on technologyNone - generally supportive

Section F

Multimodal Assessment

F1. Value of the following AI assessment modalities (1-5)

1 = Not important, 5 = Essential

Voice/speech pattern analysis

12345

Eye-tracking and gaze patterns

12345

Facial expression analysis

12345

Motor movement patterns

12345

Cognitive performance testing

12345

Conversational pattern analysis

12345

F2. Comfort level with patients providing video/audio data for AI analysis

Very comfortableSomewhat comfortableNeutralSomewhat uncomfortableVery uncomfortable

F3. Patient populations where AI screening might be most beneficial (select all that apply)

Adult women (often underdiagnosed)Adolescents with complex presentationsAdults with suspected inattentive typePatients with comorbid conditionsNon-native speakersPatients in rural/underserved areasAll populations equallyNone - prefer traditional methods

Section G

Clinical Decision Support

G1. Preferred presentation of AI diagnostic suggestions

Probability scores with confidence intervalsRisk categories (low/medium/high)Binary recommendations (positive/negative)Detailed symptom profiles with explanationsComparison to normative dataMultiple diagnostic hypotheses ranked

G2. Importance of AI transparency/explainability

1 = Not important, 5 = Essential

12345

G3. Preferred level of AI involvement in diagnosis

Screening tool only - I make all diagnostic decisionsDiagnostic support with detailed rationaleCollaborative decision-making with AIAI recommendation with override capabilityMinimal AI involvement

G4. Would you use an AI tool that identifies patients who may benefit from ADHD evaluation?

Definitely yesProbably yesMaybeProbably notDefinitely not

Section H

Implementation & Training

H1. Training requirements for adopting AI diagnostic tools

Self-directed online learning (1-2 hours)Half-day hands-on workshopFull-day comprehensive trainingMultiple sessions over weeksOngoing support and consultation

H2. Factors that would facilitate adoption (select all that apply)

Evidence of improved diagnostic accuracyReduction in assessment timeBetter documentation efficiencyPositive patient feedbackPeer recommendations and testimonialsIntegration with existing workflowRegulatory approval and guidelinesInsurance/reimbursement supportComprehensive training program

H3. Barriers to implementing AI diagnostic tools (select all that apply)

Lack of evidence for clinical utilityConcerns about diagnostic accuracyPatient privacy and data securityHigh implementation costsWorkflow disruptionLack of technical supportRegulatory uncertaintyStaff resistance to changeLimited time for training

Section I

Market & Economics

I1. Current reimbursement rate for ADHD evaluation (EUR)

€100-€200€201-€400€401-€600€601-€800More than €800Not sure/Variable

I2. Willingness to pay for an AI ADHD diagnostic tool

Yes, as subscription feeYes, as per-use feeYes, as one-time purchaseOnly if reimbursed by insuranceNo, would not pay

I3. Reasonable monthly cost for AI diagnostic platform (EUR)

€0-€100€101-€300€301-€500€501-€1000More than €1000Should be free/included in EHR

Section J

Quality & Outcomes

J1. Most important outcome measures for AI diagnostic tools (select top 3)

Diagnostic accuracy vs. gold standardTime savings in assessmentImproved patient satisfactionBetter documentation qualityReduced missed diagnosesEnhanced treatment planningDecreased clinician burdenCost-effectiveness

J2. How would you evaluate the success of an AI ADHD diagnostic tool?

J3. Specific features you would want in an AI ADHD platform?

Section K

Professional Development

K1. Interest in research collaboration with AI diagnostic platform developers

Very interestedSomewhat interestedMaybe, depending on detailsNot interested

K2. Willingness to participate in clinical validation studies

Yes, definitelyYes, with appropriate compensationMaybe, depending on requirementsProbably notNo

K3. Interest in early access/beta testing of AI platform

Very interestedSomewhat interestedNot interested

Section L

Final Comments

L1. Additional thoughts on AI in ADHD diagnosis?

L2. What would make you most likely to adopt an AI ADHD diagnostic tool?

L3. Contact information (optional)

Preferred communication:EmailPhoneNone

This survey is conducted by ANQA Digital Health for research and development purposes. All responses will be kept confidential and are used only for product development.

Thank you for your professional insights!