HAC Healthcare Analyst Data Analysis & Interpretation 2 — Questions and Answers
Question 1: A healthcare analyst observes that hospital readmission rates spike every January. Which analytical approach best determines if this is a seasonal pattern versus random variation?
- Run a chi-square test on January vs. other months
- Apply time-series decomposition to isolate seasonal components (Correct answer)
- Calculate the mean readmission rate and compare January to it
- Use a scatter plot of readmissions over the 12 months
Correct answer: Apply time-series decomposition to isolate seasonal components
Time-series decomposition separates trend, seasonal, and residual components, directly identifying recurring seasonal patterns.
Question 2: When analyzing a dataset with a heavily right-skewed distribution of patient lengths of stay, which measure of central tendency best represents a 'typical' patient stay?
- Mean
- Mode
- Median (Correct answer)
- Geometric mean
Correct answer: Median
The median is resistant to extreme high values (outliers) that cause right skew, making it the most representative measure for skewed distributions.
Question 3: A hospital wants to predict 30-day readmissions using age, comorbidity count, and discharge disposition. Which model type is most appropriate?
- Linear regression
- Logistic regression (Correct answer)
- K-means clustering
- Principal component analysis
Correct answer: Logistic regression
Logistic regression is designed for binary outcomes (readmitted vs. not readmitted) and produces probability estimates.
Question 4: An analyst finds a Pearson correlation of r = 0.85 between hospital staffing ratios and patient satisfaction scores. What is the most accurate interpretation?
- Staffing ratios cause higher patient satisfaction
- There is a strong positive linear association between the two variables (Correct answer)
- 85% of the variation in satisfaction is explained by staffing
- The relationship is statistically significant at p < 0.05
Correct answer: There is a strong positive linear association between the two variables
Correlation measures the strength and direction of linear association; r = 0.85 indicates a strong positive relationship but does not imply causation.
Question 5: A healthcare analyst is tasked with identifying patient subgroups with similar utilization patterns without predefined categories. Which technique is most suitable?
- Decision tree classification
- K-means clustering (Correct answer)
- Logistic regression
- ANOVA
Correct answer: K-means clustering
K-means clustering is an unsupervised method that groups observations by similarity without requiring predefined outcome labels.
Question 6: In a clinical dataset, an analyst notices that 15% of BMI values are missing. The missing values appear to be concentrated among patients over age 80. This pattern is best described as:
- Missing completely at random (MCAR)
- Missing at random (MAR) (Correct answer)
- Missing not at random (MNAR)
- Structurally missing
Correct answer: Missing at random (MAR)
MAR means the probability of missingness depends on observed variables (age), not on the missing value itself.
Question 7: A healthcare analyst compares average cost per discharge across five hospital departments using one-way ANOVA and finds F = 4.2, p = 0.003. What is the correct next step?
- Conclude all department means are different from each other
- Perform post-hoc pairwise comparisons to identify which departments differ (Correct answer)
- Re-run the analysis with a larger sample
- Accept the null hypothesis because F > 1
Correct answer: Perform post-hoc pairwise comparisons to identify which departments differ
ANOVA only signals that at least one mean differs; post-hoc tests (e.g., Tukey's HSD) are needed to identify which specific pairs differ.
HAC Healthcare Analyst Data Analysis & Interpretation 2
A healthcare analyst observes that hospital readmission rates spike every January.
Which analytical approach best determines if this is a seasonal pattern versus random variation?