Clinical Informatics Data Analytics Flashcards
7 cards from real Clinical Informatics Certification practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Clinical Informatics Data Analytics flashcards as text
Which SQL clause is used to filter aggregated results in a query that groups patients by diagnosis code?
Answer: HAVING
HAVING filters the results of GROUP BY aggregations, whereas WHERE filters individual rows before grouping.
A clinical informatics team calculates sensitivity and specificity for a sepsis screening alert. If sensitivity is 90% but specificity is 50%, what is the likely operational consequence?
Answer: The alert will fire frequently for patients who do not have sepsis
Low specificity (50%) means half of non-sepsis patients will trigger the alert, leading to alert fatigue and workflow burden.
Which data governance role is typically responsible for ensuring data quality and appropriate use within a specific clinical domain?
Answer: Data steward
A data steward is the subject-matter expert accountable for data quality, definitions, and appropriate use within an assigned domain.
In clinical analytics, what does 'phenotyping' refer to?
Answer: Algorithmically identifying patients with a clinical condition using EHR data
Computational phenotyping uses EHR data elements—diagnoses, labs, medications—to identify patient cohorts with specific clinical characteristics.
Which statistical test is most appropriate for comparing the means of three or more independent clinical groups?
Answer: ANOVA
Analysis of Variance (ANOVA) tests whether the means of three or more independent groups differ significantly.
What is the primary risk of using a predictive model trained on data from one health system when deploying it at a different institution?
Answer: External validity may be poor due to different patient populations or coding practices
Models trained at one site may not generalize because case-mix, documentation patterns, and EHR configurations differ across institutions.
Which technique reduces a high-dimensional clinical dataset to fewer dimensions while preserving the most variance?
Answer: Principal component analysis (PCA)
PCA projects data onto orthogonal principal components ordered by the amount of variance they capture, reducing dimensionality.