Routinely recorded patient safety events in primary care: a literature review.

BACKGROUND Existing patient-level data can be used to measure and monitor patient safety. Data from sources including electronic patient records are routinely collected in primary care and may be suitable for adverse event screening, such as patient safety indicators. To inform the feasibility of developing primary care measures of patient harm, information about routinely collected data is needed. OBJECTIVE A literature review was conducted to determine the types of adverse events that are routinely recorded in primary care. METHODS We searched ASSIA, Cochrane Library, Embase, HMIC, ISI Web of Science, Medline and PsycInfo databases, grey literature and websites. We included only original research studies in English where routinely collected patient data were used to identify adverse events occurring in primary or ambulatory care settings. Adverse events were defined as unexpected and undesirable patient outcomes arising from health care contact. RESULTS Of 5029 citations identified, 15 were reviewed. Twelve studies used multiple data sources. Approximately 6.5% of adult emergency admissions were due to drug-related events (n = 1225). Between 0.7% and 2.3% of deaths following adverse events were attributed to treatment in primary care. A large proportion of adverse events resulting in the most severe harm may be preventable. For example, one study estimated that 42% of serious adverse drug events were avoidable. CONCLUSIONS There is limited use of routinely collected data to measure adverse events in primary care despite large volumes of data generated. The potential for using readily available data recorded in primary care for active patient safety surveillance needs further exploration.