Key Steps for Designing and Conducting Video Elicitation Interviews
| Step | Example From Authors’ Study |
|---|---|
| Conceptualizing a video elicitation study | |
| 1. Choose a research question for which the added value of video elicitation interviews justifies the time and resources required | How do physicians’ and patients’ prioritize discussions of preventive services in primary care? |
| 2. Decide on the scope of data collection | Interviewed participants about the entire interaction; asked participants to make unsolicited comments and respond to questions about preventive services delivery |
| Participants and sampling | |
| 3. Determine the unit of analysis and sampling frame | Interviewed both physicians and patients; used convenience and snowball sampling; recruited physicians before patients; recruited patients (2 per physician) from waiting rooms; paid patients |
| 4. Establish sufficient trust to record physicians | Recruited physicians from investigator’s local department and community; presented study aims to clinic staff using an academic detailing approach; offered individualized feedback sessions to physicians and clinics |
| Data collection and management | |
| 5. Choose an appropriate video camera | Used a professional-grade, portable video camera with a fixed lens; used an adapted medical equipment stand (tripods would not fit in examination rooms) |
| 6. Establish a protocol for recording interactions | Used checklists for equipment setup; frequently reviewed recordings to evaluate quality; obtained backup audio recordings; research assistant set up equipment on site and was not in room during the interaction; physicians covered lens during physical examination |
| 7. Minimize the elapsed time between video recording and elicitation interviews | Interviewed patients immediately after the interaction in available clinic space; used a laptop computer to play videos; interviewed physicians 1 to 2 weeks later as their schedules allowed |
| 8. Determine the elicitation interview structure | Conducted semistructured interviews; investigated both spontaneous comments and discussions about preventive services; interviewers and participants were encouraged to pause recordings and comment; variation among interactions made structured interviews difficult |
| 9. Train interviewers | Investigator trained interviewers |
| 10. Determine a protocol for storing video recordings | Converted digital cassettes to computer files and stored them on secure servers; archived video recordings for future use |
| 11. Choose a transcription protocol for interviews | Used a standard protocol designed for studying medical discourse47 |
| Data analysis | |
| 12. Review interview and data quality regularly | Discussed data and findings in regular debriefing meetings using transcripts; investigators participated in some interviews |
| 13. Decide how to integrate data from video recordings and elicitation interviews | Tried integrating transcripts from video recordings and interviews into a single document (which added considerable time and little additional insight); did not record exact times that participants’ paused videos |
| 14. Choose a method for analyzing elicitation interview data | Developed a theory-driven coding system for assessing statements likely to represent tacit clues |
| Mitigating limitations of video elicitation interviews | |
| 15. Consider what kinds of interview data (recall, reliving, and/or reflection) are needed | Did not address |
| 16. Consider how social pressures related to interview setting may influence interview data | Noted that patients often emphasized features they liked about their physicians; were alert to participant responses during interviews that reflected social desirability |