Diagnostic Performance Comparison of Primary Care Clinicians in Spirometry Interpretation With/Without AI Software (SPIRO-AID)

Diagnostic Performance Comparison of Primary Care Clinicians in Spirometry Interpretation With/Without AI Software

A Randomized Controlled Trial Comparing Diagnostic Performance of Primary Care Clinicians in the Interpretation of SPIROmetry With or Without Artificial Intelligence Decision Support Software

To evaluate whether an artificial intelligence decision support software (ArtiQ.Spiro) improves the diagnostic accuracy of spirometry interpreted by primary care clinicians, as measured by Clinician Diagnostic Accuracy (vs Reference Standard).

This is a randomised controlled study to evaluate an AI support software in aiding interpretation of spirometry in primary care clinicians. Clinicians will be provided with a clinical dataset of 50 entirely anonymous, previously recorded real-world spirometry records to interpret and will be asked to complete specific questions about diagnosis and quality assessment. The records will be randomly selected from a database comprising spirometry records from 1122 patients undergoing spirometry in primary care and community -based respiratory clinics in Hillingdon borough between 2016-2019. The records have been previously assessed by experts for a reference diagnosis and for quality control. Participating clinicians will be allocated at random to receive either spirometry records alone or spirometry records with the addition of the ArtiQ.Spiro report. The clinical spirometry records will be de-identified (name, date of birth, address, postcode, occupation, GP, medications data removed), by a member of the clinical care team. Study participants (participating clinicians) will independently examine the same 50 spirometry records through a bespoke designed trial platform (Qualtrics). For each spirometry record, the primary care clinician participant will answer the following questions on the Qualtrics platform: Provide a technical/pattern interpretation based on spirometry report received. Options: Normal/Obstructive/Restrictive/Mixed. How confident are you in your technical/pattern interpretation using a visual analogue scale (0: Not confident at all to 10: Very Confident)? Most Likely Diagnosis-options: Asthma/COPD/ interstitial lung disease (ILD)/normal lung function/other obstructive disease/other unidentifiable disease. Second most likely diagnosis- options: Asthma/COPD/ILD/normal lung function/other obstructive disease/other unidentifiable disease How confident are you in your diagnosis using a visual analogue scale (0: Not confident at all to 10: Very Confident)? What is the Quality of the spirometry? Options: Acceptable/usable/unacceptable How confident are you in your quality assessment using a visual analogue scale (0: Not confident at all to 10: Very Confident)? The study statistician will be blinded to treatment allocation up to the point of analysis and interpretation. The reference standards for diagnosis will be made by a panel of three respiratory specialists from the clinical care team with access to medical notes to determine the diagnosis but without an ArtiQ.Spiro report. The reference standards for spirometry quality will be made by a senior experienced respiratory physiologist but without an ArtiQ.Spiro report.

Participants report the same 50 spirometry records provided in the control arm with an ArtiQ.Spiro diagnostic and quality assessment report for each spirometry result

Diagnostic performance, defined to be the number of correct cases, expressed as a percentage out of the 50 presented.

Pattern/technical interpretation of the trace, defined to be the percentage of the 50 presented spirometry traces where the pattern/technical interpretation corresponds with the reference interpretation.

Differential diagnostic performance, defined to be the percentage of the 50 presented spirometry traces where the preferred or differential diagnosis corresponds with the reference final diagnosis.

Quality assessment performance, defined to be the percentage of the 50 spirometry traces where the participant's quality assessment category corresponds with the reference quality assessment category.

Time taken to complete diagnostic interpretation and quality assessment of the 50 spirometry records.

Time taken to complete diagnostic interpretation and quality assessment of the 50 spirometry records will be collected.

Inclusion Criteria: Clinicians working in primary care who refer for spirometry or receive spirometry reports (typically GP, practice nurse) Able to access spirometry traces on study platform Provide written informed consent via study platform Exclusion Criteria: Clinicians who have completed specialist training in respiratory medicine and recognised by the General Medical Council with a right to practise as a consultant in respiratory medicine