Inpatient clinicians’ approach to diagnosis of urinary tract infections in older adults using the COM-B model: a qualitative assessment

How they used Delve

Infectious diseases researchers at Duke used Delve to deductively code 14 interviews with inpatient clinicians against the COM-B behaviour model, identifying opportunity and motivation as the drivers of overdiagnosis and overprescribing for asymptomatic bacteriuria in older adults.

“Transcripts were coded by one trained coder (SDA) using deductive coding (Supplement 2). Any questions about coding were discussed with the senior team member (NB). Delve qualitative data analysis software was used for coding and analysis.”

Field

Antimicrobial stewardship and infectious diseases; why inpatient clinicians overdiagnose and overtreat asymptomatic bacteriuria in hospitalised older adults

Data

Interviews with 14 of 16 selected inpatient clinicians (physicians and physician assistants), 25-30 minutes each over Zoom, audio-recorded and transcribed, following a piloted guide with probes

Approach

Deductive coding against a codebook built from the six components of the Capability, Opportunity, Motivation-Behaviour (COM-B) model and their definitions. The codebook was piloted on the first two interviews and updated with clarifications of how each COM-B component applied to specific questions, then one trained coder coded all transcripts, raising any coding questions with the senior team member. Themes were categorised back onto the COM-B framework.

Data types

Interviews