How to Validate Electronic Medical Record Reports
EMR’s are more than just a place to complete charting. These robust software systems allow healthcare providers and administrators to access a variety of different information on individuals and patient populations.
Unfortunately, EMR’s are no different than any other software system – they do exactly what they’re told. While this is great for generating a report without the fear of a boycott from your EMR, the accuracy of that report is inherently dependent upon two important things:
Software programming
Where patient information has been entered and saved
If EMR programming is out of date, then your reports may not align with current national guidelines. Likewise, if patient information hasn’t been entered in the correct location within your EMR, your reports may not be capturing a patient’s complete medical history. In this post, I’ll provide a step-by-step guide on how to validate an EMR report and discuss its importance.
Why is this important?
Two words - Patient. Care. Let’s be honest, most medical professionals don’t go into medical world to be IT gurus. They care about providing medical services to their patients. Accurate EMR reports are a useful tool in ensuring quality patient outcomes. However, trusting the accuracy of your EMR reports is critical to their value. The best route to gaining this trust is to validate your reports.
From an administration perspective, accurate reports and data can be motivating and inform decision making. Not to mention they are critical in clinic quality improvement efforts. Whether it’s gaining buy-in from clinical staff, tailoring workflows, or tracking improvements in care, accurate reports are critical.
Steps to validate EMR reports
Identify a report you’d like to validate
Identify a measurement period [generally this is a yearlong timeframe]
Using the timeframe identified in Step 2, run the EMR report
Research eligibility criteria for the EMR report you’re validating. This generally includes some of the following information:
Age Range, diagnosis, medication(s), visits, lab values, etc…
Using the report eligibility criteria, and your applicable measurement period, identify an eligible patient list.
In the EMR reporting world, this is generally referred to as: “denominator eligible”
Identify an appropriate sample of the eligible patients to complete a chart review.
I could speak, at length, about representative sample sizes, confidence intervals, and whether to use finite population correction. Maybe another day in a different post.
For ease, see the table below for a general guide on an appropriate sample size.
After identifying an appropriate sample size, randomly select the indicated number of patient charts identified in step 5 for review. This is your Sample n.
Complete chart reviews for the randomly sampled patients and identify the number who met the report criteria. This is your numerator.
Calculate the chart review rate (Divide the numerator by the sample n)
Compare the chart review rate to the EMR report
*~80% Confidence Level and 10% Margin of Error
**~95% Confidence Level and 10% Margin of Error
Is your EMR report accurate? And, what if it’s not?
Generally speaking, your EMR report is accurate if your chart review rates are within ≈10% of your EMR reports.
If your chart review rate and EMR report are >10% different, here are some things to consider:
Consideration and Next Steps
Are you using the same report criteria as your EMR?
Review the report criteria. If your criteria differ from the reporting criteria your EMR uses, your results will be different.
Is your chart review rate noticeably higher than your EMR reported rate?
Generally, though not always, this is a product of your EMR not pulling data from the same locations that you reviewed in your EMR. Determine what field(s) the EMR report is pulling from and ensure staff receive training about where to document patient information.
Could information be put into a “discrete” data field?
Patient history is crucial to a patient’s diagnosis and/or treatment plan. However, if critical components of this history are ONLY documented in a “notes” field, your EMR will likely not capture that information.
Should any patients have been excluded?
Make sure you review the quality metrics to determine if any patients should be excluded.
Can your EMR pull a patient-level list of all patients that met the criteria for a given report?
Some EMR’s have this capacity and others don’t. If your EMR does, you can cross reference the EMR report with your chart review results at the patient level. This will help troubleshoot why your EMR reports and chart review data don’t align.
Accurate EMR reporting is the bedrock to quality improvement. Before implementing changes in your clinic, make sure you have EMR reporting in place and that those reports are accurate. Sharing these data with clinic staff, providers and organizational decision makers then allows you to understand your strengths and opportunities in patient care.