Qualitative data collection and analysis methods: the INSTINCT trial.

Patient care practices often lag behind current scientific evidence and professional guidelines. The failure of such knowledge translation (KT) efforts may reflect inadequate assessment and management of specific barriers confronting both physicians and patients at the point of treatment level. Effective KT in this setting may benefit from the use of qualitative methods to identify and overcome these barriers. Qualitative methodology allows in-depth exploration of the barriers involved in adopting practice change and has been infrequently used in emergency medicine research. The authors describe the methodology for qualitative analysis within the INcreasing Stroke Treatment through INteractive behavioral Change Tactics (INSTINCT) trial. This includes processes for valid data collection and reliable analysis of the textual data from focus group and interview transcripts. INSTINCT is a 24-hospital, randomized, controlled study that is designed to evaluate a system-based barrier assessment and interactive educational intervention to increase appropriate tissue plasminogen activator (tPA) use in ischemic stroke. Intervention hospitals undergo baseline barrier assessment using both qualitative as well as quantitative (survey) techniques. Investigators obtain data on local barriers to tPA use, as well as information on local attitudes, knowledge, and beliefs regarding acute stroke treatment. Targeted groups at each site include emergency physicians, emergency nurses, neurologists, radiologists, and hospital administrators. Transcript analysis using NVivo7 with a predefined barrier taxonomy is described. This will provide both qualitative insight on thrombolytic use and importance of specific barrier types for each site. The qualitative findings subsequently direct the form of professional education efforts and system interventions at treatment sites.