A system for solution-orientated reporting of errors associated with the extraction of routinely collected clinical data for research and quality impr

Abstract Background: We used clinical data routinely collected over 10 years in epidemiological studies and quality improvement studies. We will extract pseudonyms and link data from a heterogeneous distributed database; we will inevitably encounter errors and problems. Objective: Develop a solution-oriented error reporting system to achieve appropriate corrective actions. Method: Please check 94 errors that occurred in September 2008. Previously, I explained about the failure when data is missing from the response file, but little information on causation was provided.

Please select quality rather than quantity. Today, data scientists have discovered that the vast amount of data they collect is of poor quality. One example is clinical genetics and the "database database" system has been established to understand the data set as the source of the data used to analyze genetic sequence variation is highly contradictory . For example, in a genetic analysis system more than 200 separate databases are often used. Banks often need to extract data from at least 15 external systems. If there is no system infrastructure to select and select data, differences in data may affect the effectiveness of the AI ​​system.

It is necessary to identify data related to drug error and data collection process before and after system introduction. These data must first be based on 5 - R - correct medication, correct dosage, correct path, correct time, accurate patient measurement error and proximity error. This data is used as a basis for quality assurance measurements to show how the system affects drug safety. Data collection after MBCPC may be automated by the system itself. The process of collecting data before MBCPC (if not done yet) needs to be started before proceeding beforehand (eg 2-3 months). Data before MBCPC should be carefully collected through hospital's drug error reporting system and direct observation. Such observations may reveal that it is close to errors or errors that were not detected or reported. 6.6 Process of Supplier Evaluation and Selection

Because the hospital completely enters the world reporting quality data, I understand the difference between collecting financial data and collecting quality data. The number of individuals who generate, summarize, report and receive financial data within the facility is fairly limited, but the quality data is reported throughout the hospital. The quality problem arises from the change, and no one knows the cause and solution of the change as much as the employees at the site. Hence, hospital leaders must expect that unit and department care teams identify key process variables, measure them, report the results extensively, and make improvements as necessary. This may require structural and cultural modifications along the power / authoritarian continuum (Figure 1). In order to manage the current complexity and future uncertainty in modern medicine, quality improvement is not just a department but a method of thinking and behavior.