EHR Data Analytics & Population Health 3 — Questions and Answers
Question 1: Which data visualization type is best suited for showing trends in patient wait times over 12 months?
- Pie chart
- Scatter plot
- Line chart (Correct answer)
- Heat map
Correct answer: Line chart
Line charts are optimal for displaying continuous data trends over time, making month-over-month patterns immediately visible.
Question 2: A care coordinator reviews a panel report showing patients with HbA1c > 9%. This is an example of which population health activity?
- Registry-based outreach (Correct answer)
- Utilization management
- Claims adjudication
- Formulary review
Correct answer: Registry-based outreach
Pulling a registry of patients meeting a clinical threshold—like uncontrolled diabetes—supports targeted outreach and proactive management.
Question 3: What does the term 'data normalization' mean in the context of EHR analytics?
- Encrypting data at rest
- Structuring data to reduce redundancy and improve integrity (Correct answer)
- Removing all PHI before analysis
- Converting data from CSV to HL7 format
Correct answer: Structuring data to reduce redundancy and improve integrity
Normalization organizes database tables to minimize duplication and ensure data dependencies make logical sense, improving query accuracy.
Question 4: Which type of study design is most appropriate for evaluating whether an EHR intervention reduced hospital-acquired infections?
- Case series
- Cross-sectional survey
- Before-and-after quasi-experimental study (Correct answer)
- Ecological correlation study
Correct answer: Before-and-after quasi-experimental study
A before-and-after (pre-post) design compares outcomes before and after implementation of an intervention in the same population.
Question 5: In predictive analytics, 'overfitting' occurs when a model:
- Performs equally well on training and test data
- Performs well on training data but poorly on new data (Correct answer)
- Uses too few predictor variables
- Fails to converge during training
Correct answer: Performs well on training data but poorly on new data
An overfitted model memorizes training noise rather than learning generalizable patterns, causing poor performance when applied to unseen data.
Question 6: Which FHIR resource would most commonly be used to retrieve a patient's list of active diagnoses for population analytics?
- Appointment
- Condition (Correct answer)
- Encounter
- Medication
Correct answer: Condition
The FHIR Condition resource stores clinical diagnoses and problems, making it the primary resource for pulling diagnosis-based population cohorts.
Question 7: A health system wants to measure preventable hospitalizations related to poor chronic disease management. Which measure set should they use?
- HEDIS Effectiveness of Care measures
- AHRQ Prevention Quality Indicators (PQIs) (Correct answer)
- CMS Hospital Readmissions Reduction Program rates
- Joint Commission National Patient Safety Goals
Correct answer: AHRQ Prevention Quality Indicators (PQIs)
AHRQ's PQIs are designed specifically to flag hospitalizations that could have been prevented with good ambulatory/chronic disease care.
EHR Data Analytics & Population Health 3
Which data visualization type is best suited for showing trends in patient wait times over 12 months?