Exploration of provider information processing in the care of patients with diabetes in primary care

The purpose of this study was to explore the schemata primary care providers use to process information during an encounter with a patient with diabetes, the sources of information, and the completeness of those sources. This study was descriptive and qualitative using techniques from focused ethnography. A purposive sample of primary care providers in an urban area of Utah participated. The sample included 1 nurse practitioner, 2 family practice residents, and 7 primary care physicians.;The data were collected using one-to-one, unstructured interviews, passive participation observation, and semistructured interviews. All interviews were audio recorded and transcribed. Observations took place in each clinic as the providers cared for patients with diabetes. Field notes were written for each observation. The semistructured interviews were based upon analysis of the field notes. Each transcript was coded line by line using descriptive codes to identify the characteristics of patients with diabetes and the sources and content of information gathered by providers. Coding then continued to build the smaller codes into clusters and networks.;The data analysis revealed four main schemata of patients with diabetes: the "autopilot," the "off and on" patient, the "back burner" patient, and the "wonder juice" patient. The classification of the patient was based upon the provider's perception of the patient's degree of engagement in their diabetes management; this perception also reflected the degree of provider work required. Providers did not use these schemata to gather information for the patient encounter. Instead they used a generalized schema that was based upon various clinical practice guidelines. Providers used the information tools available to them in a variety of ways that incompletely integrated the electronic health record and other tools. The cafeteria menu style of guideline usage and piecemeal tool usage may have resulted in incomplete diabetes assessments. Providers expressed frustration over missing and inaccurate information. Additionally, the reminders on the diabetes worksheet were often incorrect.;The main implications of this study are that providers require cues that are integrated into the workflow in order to perform comprehensive diabetes care and that informational continuity is still not present despite mature EHRs.