CHDA study plan

Define and distinguish descriptive, diagnostic, predictive, and prescriptive analytics with a healthcare example of each

Map the major healthcare data sources (EHR/EMR, claims, registries, HIE, public datasets) to the question types each can answer

Identify common healthcare data standards and terminologies (ICD-10, CPT, LOINC, SNOMED, HL7 v2, FHIR) and state what each encodes

Trace the path of a single clinical event from capture at the point of care to its appearance in an analytic dataset

Classify data elements by type (nominal, ordinal, interval, ratio) and select the appropriate summary statistic for each

Compute and interpret measures of utilization, cost, quality, and readmission rate from a small synthetic dataset

Explain the difference between source-system data models and analytic data models (normalized vs. dimensional)

Evaluate data acquisition methods (extract, interface, feed, API) and name the tradeoffs in latency, completeness, and cost

Write a data requirements specification that names the measure, population, denominator, numerator, time window, and exclusions

Distinguish data validation, data cleaning, and data transformation activities and sequence them correctly

Select appropriate statistical tests for comparing groups on continuous and categorical outcomes

Detect and describe common data quality defects (duplicates, missing values, outliers, unit mismatch) and propose a remediation step for each

Build a small analytic dataset from raw extracts by applying inclusion/exclusion logic and documenting each step

Differentiate between association and causation and identify confounding in a healthcare comparison

Translate a stakeholder's business question into an analytic question with defined success criteria

Conduct a stakeholder needs assessment that separates stated wants from underlying decisions to be supported

Construct a requirements traceability matrix linking business need, data element, measure, and report output

Select visualization types appropriate to comparison, trend, distribution, and part-to-whole questions

Design a dashboard layout that supports a specific decision rather than general awareness

Identify the ethical and equity considerations that arise when segmenting performance by patient group

Write narrative interpretations that state the finding, the magnitude, the comparison, and the limitation

Apply the correct disclosure and aggregation rules when reporting on small cell sizes

Distinguish data governance roles and artifacts (data steward, data owner, data dictionary, data lineage, stewardship council)

Classify data elements by sensitivity and map each class to a handling requirement under HIPAA and organizational policy

Document data lineage for one report from source field to displayed value

Evaluate a data-access request against minimum-necessary and role-based access principles

Describe the components of a data quality program (stewardship, profiling, issue logging, SLA) and how each prevents downstream analytic error

Re-derive the foundational analytics concepts cold, without notes, and flag the weakest area for targeted review

Complete two full-length timed practice exams and record the score by domain

Re-teach the three lowest-scoring domains from the weak-area list using self-generated summaries

Resolve every missed item by writing the underlying rule or definition, not just the correct answer

Finalize a one-page formula and terminology reference sheet for last-day review