Certified Health Data Analyst (CHDA) Practice Review

Comprehensive vocabulary flashcards covering analytics types, SQL query syntax, statistical measures, data visualization selections, and wRVU physician productivity calculations for the CHDA exam.

Descriptive Analytics

A category of analytics that summarizes past performance and describes what happened (e.g., quarterly denial rates, an average LOS of $4.8\,\text{days}$, or a monthly report detailing coder productivity, denial rate, and CMI).

Diagnostic Analytics

A category of analytics that investigates causes and explains why an event or trend occurred (e.g., determining why denial rates spiked following a workflow change, or identifying that an LOS increase was concentrated in patients awaiting skilled nursing placement).

Predictive Analytics

A category of analytics that forecasts future outcomes or estimates what is likely to occur (e.g., estimating 30-day readmissions, forecasting emergency department volume, or scoring which new claims are likely to deny based on historical denial characteristics).

Prescriptive Analytics

A category of analytics that recommends specific actions or selects optimized interventions (e.g., advising which clinical intervention should be assigned to a high-risk patient, scheduling a specific number of nurses based on predicted volume, or choosing the lowest-cost risk-reduction strategy).

Data Extraction (SQL Step)

The retrieval of specific input variables from a database table (e.g., selecting age, diagnosis, prior admissions, and LOS from PatientHistory) as a precursor to predictive modeling, distinct from the model execution itself.

SELECT

The SQL keyword that specifies the exact data elements or columns to be returned by a query.

Asterisk ($*$) in SQL

The SQL wildcard character used within a SELECT statement to return all columns from the referenced table.

FROM

The SQL keyword that specifies the source table or data source from which records are retrieved.

WHERE

The SQL keyword that defines row-level filtering criteria that records must satisfy to be included in the output.

AND Operator (SQL)

A logical operator in SQL that requires every specified condition to be true for a record to be returned.

OR Operator (SQL)

A logical operator in SQL that returns a record if at least one of the specified conditions evaluates to true.

Inclusive Boundary Operators ($\ge$, $\le$)

Comparison operators in SQL filtering that include the boundary value itself alongside values greater than or less than the threshold (e.g., $\text{age} \ge 18\,\text{AND}\,\text{age} \le 64$).

AVG()

An aggregate SQL function used to calculate the arithmetic mean of a specified numeric field (e.g., $AVG(\text{los})$).

COUNT(*)

An aggregate SQL function used to compute the total number of rows meeting the query criteria.

INNER JOIN

A join operation that returns only the rows that have matching key values across both joined tables.

LEFT JOIN

A join operation that retains every row from the left table, populating matched columns from the right table or returning NULL when no match exists.

RIGHT JOIN

A join operation that retains every row from the right table, regardless of whether a matching record exists in the left table.

FULL OUTER JOIN

A join operation that preserves all records from both tables, retaining matched rows, left-only rows, and right-only rows.

Unmatched-Record Pattern (LEFT JOIN + IS NULL)

A SQL technique used to identify records in the primary table with no counterpart in a secondary table by applying a LEFT JOIN and filtering with WHERE right_table.key IS NULL.

Range

A measure of statistical dispersion calculated as the difference between the maximum and minimum values in a dataset ($\text{Range} = \text{maximum} - \text{minimum}$).

Mean

The arithmetic average of a dataset, calculated by dividing the sum of all observed values by the total number of observations ($n$).

Median

A measure of central tendency representing the middle value in an ordered series (or the average of the two middle values when $n$ is even) that is resistant to extreme outliers.

Line Graph

The preferred data visualization format for displaying trends and continuous patterns over ordered time intervals (e.g., monthly denial rates from January through December).

Bar Chart

A visualization method best suited for comparing discrete numeric values across distinct categories or groups (e.g., comparing average LOS across four different hospitals).

Histogram

A graphical representation that displays the frequency distribution of a continuous numeric variable grouped into adjacent, non-overlapping intervals or bins (e.g., patient ages grouped into 10-year intervals).

Scatterplot

A visualization used to plot paired data points to investigate the relationship or correlation between two continuous numeric variables (e.g., length of stay versus total charges).

Positive Correlation

A statistical relationship in which two variables tend to increase together, demonstrated graphically by a scatterplot pattern rising from left to right.

Special-Cause Variation

Process instability indicated on a control chart when an observed value plots beyond the upper control limit ($\text{UCL}$) or lower control limit ($\text{LCL}$), signaling an atypical cause that requires investigation.

Two-Sample t-Test

A parametric hypothesis test used to compare the means of two independent groups (e.g., testing for a difference in mean LOS between two independent hospitals).

ANOVA (Analysis of Variance)

A statistical hypothesis test used to evaluate differences among the means of three or more independent groups (e.g., comparing mean LOS across four hospitals).

Chi-Square Test of Independence

A non-parametric statistical test used to evaluate whether an association exists between two categorical variables (e.g., testing the relationship between payer category and discharge disposition).

wRVU (Work Relative Value Unit)

A standardized metric reflecting the relative intensity, time, skill, and complexity of physician work associated with healthcare services and CPT codes.

Total wRVUs

A composite measure of physician productivity calculated by summing the products of visit volumes and assigned wRVU values for each CPT code ($\sum \text{Visits} \times \text{wRVU}$).

Encounter Volume

A productivity metric based strictly on total visit counts, representing patient volume rather than service intensity or physician work effort.

Average wRVU per Visit

An intensity measure calculated by dividing total wRVUs by total visit volume ($\frac{\text{Total wRVUs}}{\text{Total Visits}}$), indicating the average complexity of care provided per encounter.