National EMSC Data Analysis Resource Center
It doesn't matter if you've purchased the greatest software and hardware system on the planet, if the data you get out of your system is of no use.
The importance of quality data goes to the saying, “Do you want it done or do you want it done right?” The primary purpose of collecting data is to analyze a variety of projects and systems.
Software database systems and the people using them are fallible. As a result, turning data into information can sometimes be problematic and cause questionable results. The users of data must recognize that in order create information, they must first identify existing problems and clean the data.
Most managers might conclude poor data comes from deficient data entry. However, what is not usually recognized are the problems caused by software design too. Poor data entry and software design combine to cause a significant amount of bad data going into an EMSC record. Some general field types and its associated problems are listed below.
Programmers can add rules to a software application that prevent certain fields from being filled out when they should not be. They can also add rules preventing an incorrect input for particular fields. (e.g. The writer of a report should not fill out a pulse quality of “strong” when the patient does not have a pulse.) The programmer can also force dates to be entered correctly or a numeric variable to only accept numeric data. Make sure your database designer is aware of some of these common issues and implements edit checks BEFORE you collect the data.
Edit checks serve several functions:
- First, an edit check looks at the completeness of the report. For example, a database could look to see if a patient was in cardiac arrest. If so, the database program can prompt the data entry personnel to complete the section before entering any further data.
- Another function is comparing two fields against each other. For example, if a pediatric patient was reported as “tachycardiac” in the initial assessment, then the initial vital statistics should report a heart rate over 100 BPM - verifying that the patient was indeed "tachycardiac."
- Nominal ranges: A good example of a nominal range is that a patient’s age could range from 0 to 125. How often do providers treat 125-year-old patients? Often personnel make mistakes, and computer systems can be helpful reminders to something that might be within the limits of the range, but should be checked. For instance, a data system could highlight data fields in yellow if the number was acceptable, but still questionable.