When I work with clinics on quality improvement projects, it is sometimes difficult to engage teams in the work. They agree that improving the quality of care they provide to their patients is important, but they lack confidence in the data being used to monitor their improvements. Once they believe the data is accurate and reliable, they are more willing to fully participate in QI projects. These are two of the tactics I recommend for improving data validation:
Conduct a Mini Chart Audit
Mini chart audits are an effective and efficient method for verifying the veracity of data. It’s important that the audit team includes both a clinical staff member and an electronic health record analyst who understands where in the EHR the report data is being pulled from. In most cases, the numerator is the problem. Compare the report with recent patient charts to determine whether patients were correctly included or excluded from the numerator. If there are no significant inclusion or exclusion errors, examine where staff are inputting data; another common issue is data being entered in the wrong place in the EHR.
I recently worked with a clinic that was monitoring the smoking clinical quality measure for patients 13 and older. For several months the report showed a success rate in the 90th percentile. However, when we ran the quarterly report, the performance plummeted to 13%.
We were shocked. What had gone wrong? Had the workflow changed? Were the reports ever validated in the first place?
A mini chart audit revealed there had been a software upgrade we weren’t aware of. Software upgrades are frequent, and they can change where data is stored and contribute to flawed reports. Paying attention to these upgrades and keeping track of the changes is key to improving data validation.
Appoint a Data Validation Lead
Leveraging data to measure quality is becoming increasingly important. Whether working on quality improvement or other aspects of the business, you can set yourself up for success by making a single team member responsible for all data validation throughout the entire organization.
Often, when something is everyone’s job it ends up being no one’s job. Assigning this crucial role to just one person creates the focus and accountability necessary for high-quality data and reporting. In the clinic in my previous example, a data validation lead would have been responsible for keeping track of software updates and their impact on reporting.
For more detailed information about developing reports for quality improvement, I recommend our white paper, “Producing Accurate Clinical Quality Reports for Population Health: A Delivery System-Oriented Approach to Report Validation.”