Study design and data collection
A retrospective cross-sectional analysis of the prevalence of DDIs and was conducted in the cardiovascular intensive care unit (CVICU) at MM, Ann Arbor, MI, USA. A cohort of 1,320 patients admitted to the CVICU between Jan 2018 – Jan 2019, was obtained from DataDirect, a tool providing access to EMR data from patients at MM. The sample size for this representative analysis was calculated using the formula for known populations 16, (the anticipated prevalence was considered as 50% (P = 0.5); 5% margin of error was considered (d = 0.05); and 95% confidence level (Z = 1.96) was used)11, resulting in a sample of 298 patients, which was rounded to 300. Three hundred patients who were admitted to the CVICU for at least 24 hours and prescribed 2 or more drugs were randomly selected from the cohort for inclusion in the study and their full medication profile during their CVICU stay were obtained from MiChart, the EMR of MM, while only concomitant medications were included in the analysis for DDIs. Topical medications, sterile bulk preparations and medications administered on an as needed basis were excluded from the analyses.