Managing the Unwanted: Medical Errors in Primary Care

Medical errors can potentially induce adverse consequences of death, permanent disability, psychological distress, or financial impact for patients, health care professionals, and health care organizations. The seminal report To Err is Human catalyzed the patient safety movement two decades ago, and safety in primary care has now emerged as a priority. This dissertation comprises three empirical studies that investigated how health care professionals and managers in primary care organizations approach patient safety and medical errors. Using interview and focus group data collected from a total of 166 informants across 10 Patient-Centered Medical Homes (PCMH) in four states, the first study is a qualitative investigation of the conceptual meanings of patient safety by front line clinicians, administrators, staff, and patients. It reveals four overarching principles and work functions, which should be considered in patient safety activities in primary care in addition to existing conceptualizations that are more expert-driven. The second study uses the same qualitative interview data to examine how high-performance primary care organizations like PCMHs address errors that may occur in the future. I find that health care professionals and managers anticipate errors through omission-based thinking, which in turn leads to a set of informational search strategies to help them manage errors prospectively. The findings led to a proposed comparative organizational model for prospective error management to help us better understand the organizational factors that can shape patient safety performance. The third study is an experiment that used a pilot sample of 58 primary care physicians to test how the error attributes of severity, frequency, and concentration interact to affect the way errors are identified and addressed (or not). I find that among these error attributes being tested, frequency has an independent effect on errors to be highlighted for organization action and seen as urgent. This also suggests that errors that are severe and/or concentrated may be under-detected, under-reported, and under-addressed. Collectively, these studies demonstrate the utility of integrating perspectives from organization science, especially from the areas of organizational cognition and organizational errors, into patient safety research