Primary outcome measures were the percentage of med- ication orders with one or more ME and the percentage of patients with one or more pADE (both analyzed with inter- rupted time series analysis). Secondary measures of effect were types of medication errors and length of hospital stay (analyzed with Mann–Whitney U test). A ME was defined as any error in the process of prescribing and transcribing. An ADE was defined as any untoward medical occurrence that may present during treatment with a pharmaceutical product (WHO: www.who-umc.org). A pADE was defined as an ADE that occurred due to a medication error with a possible or probable causal relationship with the medication error.
In studies using an interrupted time series design the outcome measures are evaluated as a function of time. The CPOE/CDSS implementation is the interruption. The medica- tion error rates that were calculated at weekly intervals and pADEs rates that were calculated at monthly (4 week) intervals are the time series data. We then used segmented regres- sion models to analyze the impact the implementation of CPOE/CDSS had on level and trend of medication errors and pADEs.