I85PREP
⌘ K
CHDA study materials
CHDA study resources

CHDA practice questions

One question at a time. Build your CHDA understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Foundational Knowledge of Analytics in Healthcare

Which type of analytics primarily summarizes historical data to explain what happened?

  1. Prescriptive analytics
  2. Predictive analytics
  3. Descriptive analytics Correct answer.
  4. Diagnostic analytics

Explanation

Descriptive analytics summarizes past performance using metrics, reports, and dashboards. Diagnostic analytics examines causes, while predictive and prescriptive analytics focus on future outcomes and recommended actions.

Question 2Foundational Knowledge of Analytics in Healthcare

Which data quality dimension is concerned with whether a value correctly reflects the real-world fact it represents?

  1. Granularity
  2. Completeness
  3. Accuracy Correct answer.
  4. Timeliness

Explanation

Accuracy measures whether data values are correct and agree with the source or real-world event. Timeliness concerns availability when needed, completeness concerns missing values, and granularity concerns level of detail.

Question 3Foundational Knowledge of Analytics in Healthcare

Which code set is specifically designed for identifying laboratory tests and clinical observations?

  1. CPT
  2. LOINC Correct answer.
  3. ICD-10-CM
  4. NDC

Explanation

LOINC provides standardized codes for laboratory tests and clinical observations. ICD-10-CM codes diagnoses, CPT codes procedures, and NDC codes drugs.

Question 4Foundational Knowledge of Analytics in Healthcare

Which standard is most commonly associated with transmitting admission, discharge, and transfer messages between clinical systems?

  1. FHIR
  2. HL7 version 2 Correct answer.
  3. DICOM
  4. X12 837

Explanation

HL7 version 2 messaging is widely used for ADT event notifications such as admissions and discharges. DICOM handles imaging, X12 837 handles claims, and FHIR is a newer REST-based standard.

Question 5Foundational Knowledge of Analytics in Healthcare

Under the HIPAA Safe Harbor method, how many categories of identifiers must be removed to consider data de-identified?

  1. 21
  2. 12
  3. 15
  4. 18 Correct answer.

Explanation

HIPAA Safe Harbor lists 18 specific identifier categories that must be removed, such as names, dates, and geographic subdivisions. Removing fewer identifiers does not meet Safe Harbor.

Question 6Foundational Knowledge of Analytics in Healthcare

Which organization develops and maintains HEDIS quality measures?

  1. AHRQ
  2. The Joint Commission
  3. CMS
  4. NCQA Correct answer.

Explanation

The National Committee for Quality Assurance maintains HEDIS, a widely used set of healthcare quality measures. CMS, The Joint Commission, and AHRQ have other quality and accreditation roles.

Question 7Foundational Knowledge of Analytics in Healthcare

Which chart type is most appropriate for displaying a continuous trend over time?

  1. Scatter plot
  2. Pie chart
  3. Line chart Correct answer.
  4. Bar chart

Explanation

A line chart plots values across ordered time periods, making trends and shifts easy to see. Pie charts show parts of a whole, scatter plots show relationships between two variables, and bar charts compare categories.

Question 8Foundational Knowledge of Analytics in Healthcare

In healthcare claims data, what does the place of service code identify?

  1. Service setting Correct answer.
  2. Diagnosis type
  3. Provider's specialty
  4. Patient's home address

Explanation

Place of service indicates the setting where a service was delivered, such as an outpatient hospital or physician office. It does not describe the provider specialty, patient address, or diagnosis.

Question 9Foundational Knowledge of Analytics in Healthcare

A screening test has 90% sensitivity and 80% specificity. If disease prevalence is 10%, what is the approximate positive predictive value?

  1. 90%
  2. 80%
  3. 33% Correct answer.
  4. 50%

Explanation

With 100 people, 9 have disease and 9 test positive; among 90 without disease, 18 falsely test positive. PPV is 9 divided by 27, which is about 33%.

Question 10Foundational Knowledge of Analytics in Healthcare

Which practice is designed to create and maintain consistent reference data, such as patient identity, across multiple systems?

  1. Long-term archival storage
  2. Scheduled data purging
  3. Master data management Correct answer.
  4. Automated data mining

Explanation

Master data management establishes governed, consistent core data such as patient, provider, and facility identities across systems. Archiving, mining, and purging do not resolve cross-system identity consistency.

Question 11Foundational Knowledge of Analytics in Healthcare

A clinic's readmission rate is lower than the national rate, but its patients are on average more severely ill. Which analytic adjustment is needed for a fair comparison?

  1. Sampling error
  2. Seasonality
  3. Risk adjustment Correct answer.
  4. Data latency

Explanation

Risk adjustment accounts for differences in patient severity and case mix when comparing outcomes. Without it, a favorable readmission rate may simply reflect a different patient population.

Question 12Foundational Knowledge of Analytics in Healthcare

Which best describes the primary purpose of the OMOP common data model?

  1. Clinic appointment scheduling
  2. Claims billing and payment only
  3. Research data standardization Correct answer.
  4. Medical imaging archive storage

Explanation

OMOP provides a common structure and vocabulary for standardizing heterogeneous health data for observational research. It is not a billing-only, imaging-storage, or scheduling system.

Question 13Foundational Knowledge of Analytics in Healthcare

A predictive model for 30-day readmission was trained on one hospital's historical data. What is the main concern when applying it to a different patient population?

  1. Generalizability Correct answer.
  2. Data granularity
  3. Network latency
  4. Encryption strength

Explanation

A model trained on one population may not perform well in another because patterns, coding, and care processes differ. This is a generalizability, or external validity, concern.

Question 14Foundational Knowledge of Analytics in Healthcare

A dashboard metric does not match the source system total. Which area should an analyst examine first to trace the discrepancy?

  1. User interface settings
  2. Transformation logic Correct answer.
  3. Printer configuration
  4. Network bandwidth

Explanation

ETL transformation logic, such as filters, joins, or code mappings, commonly changes totals between source and report. Tracing lineage through those transformations is the appropriate first step.

Question 15Foundational Knowledge of Analytics in Healthcare

A hospital's case mix index increases from 1.0 to 1.2. What does this most likely indicate?

  1. Higher acuity Correct answer.
  2. Fewer comorbidities
  3. Lower average costs
  4. Shorter average stays

Explanation

Case mix index reflects the relative resource intensity and acuity of a patient population. A higher CMI generally indicates more complex or severely ill patients.

Question 16Foundational Knowledge of Analytics in Healthcare

When estimating a mean length of stay, what happens to the 95% confidence interval as sample size increases, all else equal?

  1. Narrows the interval Correct answer.
  2. Changes the point estimate systematically
  3. Has no effect on the interval
  4. Widens the interval

Explanation

A larger sample reduces the standard error, which narrows the confidence interval around the estimate. The point estimate may change slightly, but it is not systematically shifted by sample size alone.

Question 17Foundational Knowledge of Analytics in Healthcare

Which of the following is a technical safeguard under the HIPAA Security Rule?

  1. Disaster recovery plan
  2. Employee training
  3. Access controls Correct answer.
  4. Workstation use policies

Explanation

Technical safeguards include controls such as access management, audit logs, and encryption. Workstation policies and training are administrative safeguards, and disaster recovery planning is also administrative.

Question 18Foundational Knowledge of Analytics in Healthcare

A population health team needs to identify diabetic patients who are overdue for an A1c test. Which data source is most appropriate?

  1. EHR registry Correct answer.
  2. Social media
  3. Patient surveys
  4. Claims only

Explanation

An EHR-based registry contains clinical results, diagnoses, and due dates needed to identify care gaps. Claims may lack lab values, surveys are not timely clinical sources, and social media is inappropriate.

Question 19Foundational Knowledge of Analytics in Healthcare

In a linear regression model, an R-squared value of 0.65 means that:

  1. 65% chance of model error
  2. 65% of variance explained Correct answer.
  3. 65% of predictions are accurate
  4. Correlation is 0.65

Explanation

R-squared measures the proportion of variance in the dependent variable explained by the model. It is not a prediction accuracy percentage, a correlation coefficient, or an error probability.

Question 20Foundational Knowledge of Analytics in Healthcare

What is the primary role of a data steward in a healthcare analytics program?

  1. Software vendor
  2. Domain data custodian Correct answer.
  3. External auditor
  4. Privacy officer only

Explanation

A data steward is responsible for the quality, definitions, and appropriate use of data within a specific domain. This operational custodian role differs from vendor, auditor, or solely privacy responsibilities.

Question 21Foundational Knowledge of Analytics in Healthcare

A hospital wants to compare its mortality rate with another hospital that treats more severely ill patients. Which approach best supports a fair comparison?

  1. Unadjusted control chart
  2. Simple moving-average rate
  3. Unadjusted crude mortality rate
  4. Standardized mortality ratio Correct answer.

Explanation

The standardized mortality ratio compares observed deaths with expected deaths after accounting for patient risk factors. Crude rates, moving averages, and unadjusted charts do not adjust for case mix.

Question 22Foundational Knowledge of Analytics in Healthcare

On a control chart, eight consecutive points fall above the centerline. What does this pattern most likely indicate?

  1. Special-cause variation Correct answer.
  2. No variation at all
  3. Only common-cause variation
  4. Normal seasonal variation

Explanation

A run of consecutive points on one side of the centerline is a recognized signal of special-cause variation. Common-cause variation appears as random fluctuation around the centerline.

Question 23Foundational Knowledge of Analytics in Healthcare

Which is a key benefit of HL7 FHIR compared with older messaging approaches?

  1. Reduces storage requirements
  2. Standardized granular access Correct answer.
  3. Replaces all clinical coding
  4. Eliminates privacy concerns

Explanation

FHIR uses modular resources and RESTful APIs to provide standardized, granular access to clinical data. It does not replace coding systems, guarantee storage savings, or remove privacy obligations.

Question 24Foundational Knowledge of Analytics in Healthcare

Which is a major limitation of using claims data alone for clinical quality measurement?

  1. Excessive timeliness
  2. Too much laboratory detail
  3. Overly granular clinical notes
  4. Limited clinical detail Correct answer.

Explanation

Claims data are generated for payment and often lack laboratory values, severity, and detailed clinical findings. They may be timely for billing but are not a rich clinical source.

Question 25Foundational Knowledge of Analytics in Healthcare

Two reports show different visit counts because one includes cancelled appointments and the other does not. Which data quality dimension is most directly affected?

  1. Precision
  2. Completeness
  3. Timeliness
  4. Consistency Correct answer.

Explanation

Consistency means the same concept is defined and calculated the same way across reports. Differing inclusion rules create inconsistent counts even if each report is timely and complete for its own definition.

Question 26Foundational Knowledge of Analytics in Healthcare

Which standard is designed to identify specific laboratory tests and clinical observations, such as a serum potassium result?

  1. ICD-10-CM diagnosis codes
  2. NDC package codes
  3. LOINC codes Correct answer.
  4. CPT Category I codes

Explanation

LOINC supplies universal identifiers and units for laboratory tests and clinical observations so the same result can be recognized across disparate systems. Diagnosis classifications, procedure code sets, and drug product identifiers serve different purposes in the record.

Question 27Foundational Knowledge of Analytics in Healthcare

A disease registry contains an A1c result for only 62% of its enrolled diabetic patients. Which data quality dimension does this most directly describe?

  1. Precision
  2. Timeliness
  3. Completeness Correct answer.
  4. Uniqueness

Explanation

Completeness measures the degree to which required data elements are present for the population of interest. Missing values on 38% of records is a gap in presence of data rather than in currency, duplication, or repeatability of measurement.

Question 28Foundational Knowledge of Analytics in Healthcare

Which task is an example of inferential statistics?

  1. Ranking the ten most frequent primary diagnoses
  2. Calculating the mean age of patients discharged last quarter
  3. Projecting statewide rates from a hospital sample Correct answer.
  4. Counting admissions by service line for a monthly report

Explanation

Inferential methods generalize from a sample to a larger population, usually with an estimate of uncertainty. Summarizing or tabulating the full set of observed records is descriptive.

Question 29Foundational Knowledge of Analytics in Healthcare

A hospital's length of stay is strongly right-skewed because a small number of stays last many weeks. Which statistic best represents the typical stay?

  1. Standard deviation
  2. Range
  3. Median Correct answer.
  4. Mean

Explanation

For skewed distributions the median is resistant to extreme values and better reflects the center of the bulk of observations. The mean is pulled upward by a few very long stays, while measures of spread do not summarize the center at all.

Question 30Foundational Knowledge of Analytics in Healthcare

An analytics team builds a repository that contains only oncology data to support the cancer service line, drawing its data from the organization's enterprise repository. Which term describes this repository?

  1. Data mart Correct answer.
  2. Master patient index
  3. Data warehouse
  4. Operational data store

Explanation

A data mart is a subject-oriented subset of a warehouse scoped to a department or use case, which matches a single-service-line analytic repository. A warehouse integrates data across the whole enterprise rather than one subject area.

Question 31Foundational Knowledge of Analytics in Healthcare

In HL7 version 2 messaging, which message type carries patient admission, discharge, and transfer events?

  1. ORM
  2. SIU
  3. ORU
  4. ADT Correct answer.

Explanation

ADT messages communicate admissions, discharges, and transfers and are the backbone of census and encounter tracking. The other message types handle orders, observation results, and scheduling respectively.

Question 32Foundational Knowledge of Analytics in Healthcare

A comparison of two care-management programs reports p = 0.03 for the difference in readmission rates. Which interpretation is correct?

  1. Under no true difference, a result this extreme arises 3% of the time Correct answer.
  2. There is a 3% probability that the null hypothesis of no difference is true
  3. The finding would replicate in 97 of 100 future samples
  4. The two programs differ by 3 percentage points in readmission rate

Explanation

A p-value is the probability of observing a result at least as extreme as the one obtained, assuming the null hypothesis is true. It is not the probability that the null is true, not a measure of effect size, and not a replication guarantee.

Question 33Foundational Knowledge of Analytics in Healthcare

An analyst finds that counties with higher average income have higher diabetes diagnosis rates and concludes that wealthier individuals are more likely to be diabetic. Which error is most likely?

  1. Selection bias
  2. Ecological fallacy Correct answer.
  3. Immortal time bias
  4. Confounding by indication

Explanation

The ecological fallacy is inferring an individual-level relationship from an association observed at the group level. County-level averages can reflect composition, screening access, or diagnosis patterns that do not hold for individuals.

Question 34Foundational Knowledge of Analytics in Healthcare

A hospital wants cost estimates for inpatient stays but its billing data contain only charges. Which approach is most defensible?

  1. Apply a cost-to-charge ratio by department Correct answer.
  2. Substitute the DRG payment amount for cost
  3. Divide total charges by length of stay
  4. Use the charge amount without adjustment

Explanation

Cost-to-charge ratios, ideally applied at the departmental level, convert billed charges into estimated resource cost and are the standard approach when only charges are available. Charges reflect list prices rather than resource consumption.

Question 35Business Needs Assessment

What is the primary purpose of a business needs assessment in a healthcare analytics project?

  1. To assign accountability for prior project failures
  2. To identify gaps between current and desired performance Correct answer.
  3. To document the final report formatting standards
  4. To determine which software vendor submitted the lowest bid

Explanation

A needs assessment establishes the difference between the organization's present capability and the capability it requires to meet its objectives. Vendor selection, formatting standards, and assignment of blame are later or unrelated activities that do not define need.

Question 36Business Needs Assessment

In a needs assessment, which term describes the individuals and groups who affect or are affected by a proposed analytics solution?

  1. The data steward council
  2. The internal audit committee
  3. External accreditation surveyors
  4. Stakeholders Correct answer.

Explanation

Stakeholders are anyone with an interest in or influence over the initiative, including sponsors, clinicians, analysts, and patients. The other named bodies may be stakeholders in a specific project, but the general term for the population is stakeholders.

Question 37Business Needs Assessment

A gap analysis compares which two states?

  1. Hardware capacity versus network bandwidth
  2. Staff headcount versus patient volume
  3. Vendor pricing versus budget allocation
  4. Current state versus desired future state Correct answer.

Explanation

Gap analysis quantifies the distance between where the organization is today and where it needs to be. The other pairs are specific measures that might be examined, but they do not define the analytical structure of a gap analysis.

Question 38Business Needs Assessment

Which type of requirement describes a mandatory constraint imposed by HIPAA privacy rules?

  1. A user interface preference
  2. A regulatory requirement Correct answer.
  3. A performance benchmark
  4. An optional enhancement

Explanation

Requirements traceable to statutes, regulations, or accreditation standards are classified as regulatory requirements and are non-negotiable. Preferences and enhancements are discretionary and do not carry legal obligation.

Question 39Business Needs Assessment

Which data quality dimension describes the degree to which all required data elements are present?

  1. Timeliness
  2. Validity
  3. Consistency
  4. Completeness Correct answer.

Explanation

Completeness measures whether expected fields are populated for every required record. The other dimensions address when data are available, whether values agree across sources, and whether values conform to defined formats and ranges.

Question 40Business Needs Assessment

Which technique visually documents the sequence of steps in a clinical or administrative process?

  1. Regression modeling
  2. Process mapping Correct answer.
  3. Data profiling
  4. Cost-benefit analysis

Explanation

Process mapping produces a diagram of the steps, decisions, and handoffs that make up a workflow. The other techniques analyze financial value, inspect data characteristics, or model statistical relationships rather than depict process flow.

Question 41Business Needs Assessment

A key performance indicator is best described as which of the following?

  1. A metric tied to a strategic objective Correct answer.
  2. Any clinical data element stored in the electronic health record
  3. A vendor contractual service level agreement
  4. A single finding produced by an internal audit

Explanation

A key performance indicator is a quantifiable measure deliberately chosen because it reflects progress toward a strategic goal. Stored data elements, contractual service levels, and one-time audit findings are not by themselves strategic performance measures.

Question 42Business Needs Assessment

Which elicitation technique is most appropriate for gathering detailed, open-ended information from a single subject matter expert?

  1. A one-on-one interview Correct answer.
  2. A closed-ended mailed survey
  3. Observation of a public waiting area
  4. Review of an existing data dictionary

Explanation

A one-on-one interview allows follow-up probing and captures nuanced detail from an individual expert. Closed-ended surveys limit response depth, observation reveals behavior rather than rationale, and document review captures only what was previously recorded.

Question 43Business Needs Assessment

Using MoSCoW prioritization, how should a regulatory reporting requirement that must be delivered in the first release be classified?

  1. Could have
  2. Must have Correct answer.
  3. Won't have
  4. Should have

Explanation

MoSCoW assigns 'must have' to items that are mandatory for the release to be acceptable, which includes legally required reporting. 'Should have' and 'could have' denote important and desirable items that can be deferred under pressure.

Question 44Business Needs Assessment

A clinic reports that 30% of claims are denied for missing authorization. Which tool best helps the analyst move beyond the symptom to underlying causes?

  1. A Gantt chart
  2. A RACI matrix
  3. A burn-down chart
  4. A fishbone diagram Correct answer.

Explanation

A fishbone, or cause-and-effect, diagram organizes potential contributing causes across categories such as people, process, and technology. The other tools address scheduling, responsibility assignment, and progress tracking rather than causal diagnosis.

Question 45Business Needs Assessment

In a cost-benefit analysis for a new analytics platform, which item should be treated as a tangible benefit?

  1. Better brand reputation in the community
  2. Stronger executive confidence in the data
  3. Fewer hours spent on manual chart abstraction Correct answer.
  4. Improved clinician satisfaction with the workspace

Explanation

Tangible benefits can be measured in money or countable units such as labor hours. Satisfaction, reputation, and confidence are intangible and are typically described qualitatively or with proxy assumptions.

Question 46Business Needs Assessment

A project costs $200,000 and is expected to return $300,000 in benefits over one year. What is the return on investment?

  1. 150%
  2. 50% Correct answer.
  3. 33%
  4. 67%

Explanation

Return on investment equals net benefit divided by cost, so ($300,000 minus $200,000) divided by $200,000 yields 0.50, or 50%. The distractors result from dividing the wrong quantities in the wrong direction.

Question 47Business Needs Assessment

What is the primary purpose of defining project scope during a needs assessment?

  1. To finalize the software license agreement
  2. To guarantee the project will be completed under budget
  3. To set boundaries on what the solution will address Correct answer.
  4. To eliminate the need for later stakeholder review

Explanation

Scope definition states which problems, users, and functions the initiative will and will not cover, which controls expectations and change requests. It does not guarantee cost outcomes, replace review, or settle contractual terms.

Question 48Business Needs Assessment

Who is typically accountable for approving data definitions and resolving data quality disputes in a healthcare organization?

  1. The help desk supervisor
  2. The data governance council Correct answer.
  3. The scheduling coordinator
  4. The network administrator

Explanation

Data governance bodies own decisions about standard definitions, ownership, and dispute resolution across the enterprise. Operational and technical roles handle tickets, infrastructure, and scheduling rather than enterprise data policy.

Question 49Business Needs Assessment

An analyst compares the hospital's 30-day readmission rate to published rates from similar hospitals. What is this activity called?

  1. Benchmarking Correct answer.
  2. Root cause analysis
  3. Cost-benefit analysis
  4. Data profiling

Explanation

Benchmarking compares an organization's performance against external reference points such as peer groups or national standards. Profiling describes data content, root cause analysis explains a specific problem, and cost-benefit analysis weighs financial alternatives.

Question 50Business Needs Assessment

In agile requirements documentation, which structure is commonly used to express a user story?

  1. The report shall run nightly at midnight
  2. Given the system shall process data when triggered
  3. As a role, I want a goal so that a benefit is realized Correct answer.
  4. The database must contain fully normalized tables

Explanation

User stories follow a role-goal-benefit template that keeps the focus on who needs the capability and why. The other statements are system-level or technical specifications without an explicit user and rationale.