Does increasing the size of bi-weekly samples of records influence results when using the Global Trigger Tool? An observational study of retrospective record reviews of two different sample sizes

OBJECTIVES: To investigate the impact of increasing sample of records reviewed bi-weekly with the Global Trigger Tool method to identify adverse events in hospitalised patients. DESIGN: Retrospective observational study. SETTING: A Norwegian 524-bed general hospital trust. PARTICIPANTS: 1920 medical records selected from 1 January to 31 December 2010. PRIMARY OUTCOMES: Rate, type and severity of adverse events identified in two different samples sizes of records selected as 10 and 70 records, bi-weekly. RESULTS: In the large sample, 1.45 (95% CI 1.07 to 1.97) times more adverse events per 1000 patient days (39.3 adverse events/1000 patient days) were identified than in the small sample (27.2 adverse events/1000 patient days). Hospital-acquired infections were the most common category of adverse events in both the samples, and the distributions of the other categories of adverse events did not differ significantly between the samples. The distribution of severity level of adverse events did not differ between the samples. CONCLUSIONS: The findings suggest that while the distribution of categories and severity are not dependent on the sample size, the rate of adverse events is. Further studies are needed to conclude if the optimal sample size may need to be adjusted based on the hospital size in order to detect a more accurate rate of adverse events.