Nursing informatics practice questions for the NI-BC

A patient-facing app needs to retrieve only the patient's current allergy list from the EHR on demand. Which standard is designed for this type of exchange?

1. HL7 v2 ADT message
   ADT messages push admission and transfer events between systems; they are not a request-response method for patient apps.
2. C-CDA continuity of care document
   A C-CDA is a whole document that bundles many sections; it is not a way to query one current data element on demand.
3. DICOM query and retrieve
   DICOM query/retrieve finds and pulls images from PACS, not allergy data.
4. FHIR RESTful API
   FHIR exposes discrete resources, such as AllergyIntolerance, through RESTful APIs, so an app can request just the current allergy list on demand.

A receiving hospital's EHR recognizes an incoming LOINC-coded potassium result and automatically applies it to its own decision support rules. This demonstrates which level of interoperability?

1. Semantic
   Semantic interoperability means the receiver understands shared coded meaning well enough to act on it, here applying a LOINC-coded potassium to its own rules.
2. Structural
   Structural interoperability gets the message format right so fields can be parsed; it does not guarantee the receiver understands what a code means.
3. Foundational
   Foundational interoperability is simply moving data from one system to another without interpreting it.
4. Organizational
   Organizational interoperability concerns governance, policy, and legal agreements between organizations, not code-level understanding.

A deterioration prediction model performed well at go-live, but its positive predictive value has declined steadily over 18 months as the patient mix changed. This is best explained by:

1. hallucination.
   Hallucination describes generative AI producing fluent but false content; a risk score does not generate text.
2. model overfitting.
   Overfitting shows up right away on new data; a model that worked well at go-live and declined over 18 months points to drift.
3. model drift.
   Model drift is gradual loss of performance after deployment as the patient mix or practice changes away from the data the model learned from.
4. algorithmic bias.
   Algorithmic bias is systematically worse performance for certain groups; the stem describes decline over time for everyone.

A readmission analytics project combines claims, nursing notes, device feeds, and images, each in a different format. This characteristic of big data is called:

1. veracity.
   Veracity concerns the trustworthiness of data, not the number of different formats.
2. variety.
   Variety is the range of data types and formats, here claims, notes, device feeds, and images, that must be combined.
3. volume.
   Volume is the amount of data; mixing four kinds of sources is about kinds, not amount.
4. velocity.
   Velocity is the speed of data generation and processing, such as streaming device data, not the mix of formats.

Infusion pumps, monitors, and smart beds are being connected to the hospital network. Which measure most effectively limits the cybersecurity risk these devices add?

1. Keep each manufacturer's default network settings
   Default settings and passwords are a well-known attack route and are among the first things to change.
2. Install antivirus software on every device
   Many medical devices run locked or embedded operating systems that cannot host antivirus or that the manufacturer will not support.
3. Rely on the perimeter firewall alone
   A perimeter firewall does nothing once a threat is inside; devices on a flat network remain exposed to each other.
4. Isolate devices on a segmented network
   Network segmentation puts connected devices in their own zone so a compromised pump or bed cannot reach the EHR or other systems.

The informatics nurse is building a care plan and needs a nursing-specific terminology to code the action 'fall prevention'. Which terminology is designed for this?

1. Nursing Interventions Classification
   NIC is the standardized terminology for nursing interventions, the actions nurses take, such as fall prevention.
2. Nursing Minimum Data Set
   The NMDS is a minimum set of data elements to collect, not a terminology for coding a specific intervention.
3. NANDA International nursing diagnoses
   NANDA-I codes nursing diagnoses, such as risk for falls, rather than the intervention.
4. Nursing Outcomes Classification
   NOC codes patient outcomes that interventions aim to change, not the action itself.

A device gateway sends only hourly averages of each patient's heart rate to the data warehouse. Analysts studying brief tachycardia episodes find the data inadequate because of limited:

1. timeliness.
   The averages arrive on schedule; timeliness would be at issue only if they were delayed.
2. consistency.
   Consistency is about the same fact being recorded the same way across systems, which is not at issue.
3. granularity.
   Granularity is the level of detail in data; hourly averages smooth over brief tachycardia episodes that minute-level data would show.
4. completeness.
   Completeness is met: every hour has a value, so nothing is missing; each value is simply too coarse.

A staffing tool uses forecasted census and acuity to recommend how many nurses each unit should schedule for the next shift. This type of analytics is:

1. predictive modeling.
   Predictive modeling forecasts what will happen, such as the census; stopping at the forecast confuses predictive with prescriptive analytics.
2. prescriptive.
   Prescriptive analytics recommends an action, here how many nurses to schedule, often built on a forecast.
3. diagnostic.
   Diagnostic analytics explains why something happened, such as why overtime rose.
4. descriptive.
   Descriptive analytics summarizes what already happened, such as last month's census.

Before building a new flowsheet row, the informatics nurse records its name, definition, data type, allowed values, and owner. This information belongs in the:

1. data dictionary.
   A data dictionary is the metadata catalog of each element's name, definition, data type, allowed values, and owner.
2. audit trail.
   An audit trail logs who accessed or changed data and when; it does not define the element.
3. value set.
   A value set lists the permitted codes for one element, only one of the attributes described.
4. interface specification.
   An interface specification defines how data move between two systems, not the definition of every flowsheet row.

A physician orders a tenfold overdose, the pharmacist verifies it, and the nurse scans the patient and medication before giving it. Barcode medication administration did not prevent the error because BCMA:

1. cannot read unit-dose barcodes that pharmacy applies.
   The nurse scanned the medication successfully, so barcode readability was not the issue.
2. stops checking once a pharmacist has verified an order.
   BCMA checks at the bedside after pharmacy verification; the check ran, it simply had a wrong order to compare against.
3. verifies the patient's identity but not the dose.
   BCMA does check the scanned product against the ordered drug and dose, so it does more than confirm identity.
4. matches the dose to the order without judging the order.
   BCMA confirms that the right patient gets the drug and dose that was ordered; if the order itself is a tenfold overdose that pharmacy verified, BCMA has nothing to catch.

Nurses ignore a new deterioration score because it displays only a number, with no indication of why the patient scored high. Which change best supports appropriate use?

1. Show the main factors behind each score
   Showing the main contributing factors makes the score explainable, so nurses can judge whether it fits the patient and act on it.
2. Retrain nurses on the model's accuracy
   Quoting overall accuracy does not tell a nurse why this patient scored high, which is what drives use at the bedside.
3. Convert the score to an interruptive alert
   Interrupting nurses with an unexplained number adds alert burden without building trust in the score.
4. Raise the threshold to reduce false alerts
   A higher threshold changes how often the score flags patients but still shows no reason, so trust does not improve and some deterioration is missed.

Which automated dispensing cabinet report best supports early detection of controlled-substance diversion?

1. The expired medication removal report
   Expired-drug removal is inventory housekeeping and does not track individual users' controlled-substance activity.
2. The cabinet temperature monitoring log
   Temperature logs protect medication storage conditions and have nothing to do with who removes drugs.
3. The daily par-level restocking report for the unit
   Par-level reports guide restocking quantities; diversion hides inside normal restocking volumes.
4. A user's withdrawals compared with peers'
   Diversion surveillance flags users whose controlled-substance withdrawals, wastes, or cancellations are out of line with peers in the same role and unit.

Maintaining one authoritative, shared list of providers, locations, and service codes that all enterprise systems use is called:

1. master data management.
   Master data management maintains one authoritative list of shared reference data, such as providers, locations, and service codes, used by all systems.
2. enterprise data archiving.
   Archiving moves inactive data to long-term storage, not the upkeep of current reference lists.
3. data masking.
   Masking swaps sensitive values for realistic stand-ins, often so test systems never hold real patient data; it builds no shared reference list.
4. data mining.
   Data mining searches large data sets for patterns; it uses data rather than governing reference lists.

An NLP tool counts the radiology phrase 'no evidence of pneumonia' as a case of pneumonia. Which capability needs improvement?

1. Optical character recognition
   Optical character recognition converts scanned images to text; the report is already digital text.
2. Named entity recognition
   Named entity recognition found 'pneumonia' correctly; the failure is not recognizing that it was ruled out.
3. Negation detection
   Negation detection recognizes phrases like 'no evidence of' so a mention of pneumonia is not counted as a positive finding.
4. Speech recognition accuracy
   Speech recognition turns dictation into text; the text here is correct, it is the interpretation that fails.

BCMA reports show that many patient wristbands are scanned at the nurses' station rather than at the bedside. What is the best initial response?

1. Investigate why bedside scanning is failing
   Workarounds signal a system or workflow problem, such as unreadable wristbands, scanner range, or connectivity, so the first step is finding the cause.
2. Turn off wristband scanning to reduce delays
   Removing patient identification scanning eliminates a key safety check instead of fixing why it fails.
3. Discipline nurses who scan away from the bedside
   Disciplining nurses for a workaround before understanding it ignores system causes and drives the problem underground.
4. Send a unit-wide reminder of the BCMA policy
   A reminder assumes the problem is knowledge; nurses who scan at the station usually know the policy and are working around a barrier.

During conversion to a new EHR, the legacy system stores allergies as free text but the new system requires coded allergens. Which conversion deliverable addresses this?

1. a data dictionary describing each allergy field in the new system
   A description of the new allergy fields says nothing about how each legacy value is translated into them.
2. a cutover checklist sequencing the final conversion activities
   A cutover checklist orders go-live tasks; it does not handle translation of the content.
3. a validation report comparing record counts in both systems
   Record-count validation checks that data arrived after conversion; it does not define how free text becomes codes.
4. a data map from each source element to its target code
   A conversion data map defines how each legacy element translates to the target system's structure and codes, such as mapping free-text allergies to coded allergens.

Before a vendor's sepsis prediction model goes live, the informatics nurse is asked to confirm that it suits the hospital's patients. The most important step is to:

1. validate the model's performance on local patient data.
   A vendor model must be validated on the hospital's own patients, including subgroups, because performance often drops when population, documentation, or workflow differ from the development site.
2. review the accuracy the vendor published from its development site.
   Development-site accuracy is the starting claim; relying on it is the error that local validation exists to prevent.
3. verify that its alerts display on every nursing workstation.
   Display testing is a technical step; it says nothing about whether the predictions are right for these patients.
4. confirm the model uses the newest machine learning technique.
   A newer machine learning technique is no guarantee of better performance on a particular hospital's patients.

A primary care network wants to track every patient with uncontrolled hypertension and see who is overdue for follow-up so nurses can do outreach. Which tool should the informatics nurse build?

1. A disease registry
   A disease registry lists every patient with a condition and tracks key measures and due dates, so care teams can find patients with gaps such as overdue follow-up.
2. An enterprise data lake
   A data lake stores large amounts of raw data of every kind; it is not a curated list built for outreach.
3. A master patient index
   A master patient index links each person's identifiers across systems; it does not track clinical status or care gaps.
4. A data dictionary
   A data dictionary is reference documentation, not a list of patients, so it cannot show who is overdue.

Physicians, nurses, and pharmacists disagree on the content of a new sepsis order set. The best way to resolve the dispute for good is to:

1. build a separate version of the order set for each of the disciplines.
   Separate versions break standardization and create conflicting orders for the same patient.
2. let the most senior attending physician make the final content choice.
   Leaving the decision to one senior physician shuts out nursing and pharmacy expertise and weakens buy-in.
3. use an interprofessional content committee with set decision rights.
   An interprofessional governance committee with agreed decision rights gives every discipline a voice and a defined way to settle content disputes.
4. accept the vendor's default order set to avoid the internal conflict.
   Vendor defaults avoid the argument but may not fit local evidence or practice, and the disagreement returns later.

A risk algorithm that uses past health care spending to predict need gives Black patients lower risk scores than White patients who are equally sick. The main cause of this bias is that:

1. the model was overfit to a single hospital's data.
   Overfitting causes poor performance on new data in general, not a systematic gap between equally sick groups.
2. the algorithm did not include race as an input.
   Leaving race out does not remove bias when another input, here spending, carries the inequity; 'race-blind' is not the same as fair.
3. cost is a proxy shaped by unequal access to care.
   Past spending reflects who could get care, not how sick people are; groups with less access spend less at the same level of illness, so a cost-trained model underrates their need.
4. the training data set contained too few patients.
   A larger data set with the same biased label would reproduce the same bias; sample size is not the cause.

Nurses frequently use the copy-forward function in progress notes. The main safety and legal risk of this practice is:

1. violation of the minimum necessary rule
   The minimum necessary standard caps the PHI a use or disclosure involves; copying text forward inside one patient's chart shows it to no one new.
2. loss of the note's audit trail
   The EHR still logs who created and signed the note; the audit trail is kept.
3. outdated findings recorded as current
   Copy-forward can carry earlier assessments into a new note as if they were reassessed, creating inaccurate records and legal exposure.
4. slower signing of progress notes
   Slower signing is not the concern; copy-forward usually speeds documentation, which is its appeal.

TEFCA (the Trusted Exchange Framework and Common Agreement) primarily establishes:

1. the minimum data elements required for exchange.
   The required data elements come from USCDI, a separate ONC standard.
2. the API standard used to exchange health data.
   TEFCA is a governance and trust framework; technical standards such as FHIR and IHE profiles sit underneath it.
3. governance for exchange among national networks.
   TEFCA sets a common agreement and framework so Qualified Health Information Networks can exchange data with each other nationwide.
4. a mandatory federal HIE that replaces regional HIEs.
   TEFCA participation is voluntary, and regional HIEs can take part through QHINs rather than being replaced.

A statistician analyzes a research data set that still contains some admission dates and certifies that the risk of re-identifying any patient is very small. The data set is de-identified under the:

1. minimum necessary standard.
   The minimum necessary standard sets how much PHI a task may involve. It is a scope rule, and following it leaves the data identifiable.
2. expert determination.
   Under expert determination, a qualified statistician certifies that re-identification risk is very small, which allows elements such as some dates to remain.
3. limited data set provision.
   A limited data set is still PHI and requires a data use agreement; it is not a de-identification method.
4. Safe Harbor method.
   Safe Harbor requires removing all 18 identifier types, including dates more specific than the year; a set that keeps admission dates cannot qualify.

An informatics nurse builds an EHR prompt that reminds nurses to offer pneumococcal vaccination to eligible inpatients before discharge. This design choice most directly reflects which ethical principle?

1. nonmaleficence.
   Nonmaleficence is avoiding harm; a vaccination reminder actively promotes good rather than preventing a harm caused by care.
2. autonomy.
   Autonomy is respecting a patient's own choices; offering the vaccine supports a choice, but the design aims at the patient's good.
3. justice.
   Justice concerns fair distribution of resources and care; the prompt's main purpose is the patient's benefit, not fair allocation.
4. beneficence.
   Beneficence is the duty to act for the patient's good; a prompt that makes sure eligible patients are offered a protective vaccine promotes their benefit.

An informatics nurse's main duties are verifying that the organization's data practices meet HIPAA and accreditation requirements and investigating audit-log anomalies. In ANA's Nursing Informatics: Scope and Standards of Practice (3rd ed.), this work belongs to the practice area of:

1. clinical informatics.
   Clinical informatics is optimizing how clinicians use systems in care delivery, such as documentation and order workflows, not regulatory compliance review.
2. compliance and integrity management.
   Compliance and integrity management is the NI practice area for making sure data practices meet laws, regulations, and accreditation requirements, including review of audit logs for inappropriate access.
3. research and evaluation.
   Research and evaluation is generating and testing knowledge about informatics interventions, not checking that current practice meets HIPAA and accreditation rules.
4. management, administration, and leadership.
   Management, administration, and leadership covers directing informatics teams, budgets, and strategy, not auditing compliance with privacy rules.

A survey asks nurses only whether a new mobile documentation app will improve their work and whether it will be easy to learn. The survey measures the two core constructs of the:

1. technology acceptance model.
   The technology acceptance model rests on two constructs, perceived usefulness and perceived ease of use, exactly what the survey asks.
2. diffusion of innovations theory.
   Diffusion of innovations explains how an innovation spreads through adopter groups and attributes such as relative advantage, not these two survey items.
3. theory of planned behavior.
   The theory of planned behavior uses attitude, subjective norms, and perceived behavioral control, not usefulness and ease of use.
4. unified theory of acceptance and use.
   The unified theory of acceptance and use (UTAUT) adds social influence and facilitating conditions to performance and effort expectancy, so two questions do not cover its core.

A hospital's overall rate of timely antibiotic administration for sepsis meets its target. To advance health equity, the informatics nurse should next:

1. add a sepsis best-practice alert for all adult inpatients.
   A blanket alert adds burden without showing whether any group is being left behind.
2. report the overall rate to the board, citing it as evidence of equity.
   An overall rate that meets target says nothing about equity; disparities can sit inside a good average.
3. raise the overall target for timely antibiotic administration.
   A higher overall target still ignores whether some groups get worse care.
4. stratify the measure by race, ethnicity, and language.
   Stratifying an overall rate by race, ethnicity, and language can reveal disparities that a good average hides, which is the first step toward equitable care.

A doctorally prepared nurse conducts research that develops new models for representing nursing knowledge in EHRs. Using Staggers, Gassert, and Curran's levels, this nurse is functioning at which competency level?

1. Informatics specialist
   The informatics specialist has advanced preparation and applies informatics in practice and systems work, but developing new models through research is the innovator level.
2. Informatics innovator
   In Staggers, Gassert, and Curran's levels, the informatics innovator is educationally prepared to do informatics research and develop new theory and models.
3. Experienced nurse
   The experienced nurse is proficient in a specialty and uses information systems well, without informatics specialty preparation.
4. Beginning nurse
   The beginning nurse has basic computer and information management skills.

After a new admission template goes live, the informatics nurse compares documentation completion and time per admission against the targets set during planning. This activity reflects which ANA NI standard of practice?

1. Implementation
   Implementation is carrying out the plan; comparing results with targets comes after.
2. Outcomes identification
   Outcomes identification is setting the expected outcomes and targets during planning, not measuring against them after go-live.
3. Assessment
   Assessment is the standard that opens the process; checking results against targets after go-live closes it.
4. Evaluation
   The Evaluation standard covers measuring progress toward the planned outcomes, here completion and time against the targets set during planning.

A nurse knows that a rising early warning score predicts deterioration. For a patient whose goals of care are comfort only, the nurse sees a rising score and discusses comfort-focused options with the team instead of calling a rapid response. In the DIKW framework, this best illustrates:

1. wisdom.
   Wisdom is applying knowledge with judgment to fit the situation, here recognizing that rapid response would conflict with this patient's comfort-only goals.
2. knowledge.
   Knowledge is the nurse's understanding that a rising score predicts deterioration; choosing to act differently for this patient goes beyond it.
3. data.
   Data are raw, uninterpreted values, such as a single vital sign.
4. information.
   Information is data given context, such as the score trend itself.

An informatics nurse participates in a project that uses documentation audit data to reduce charting errors and then monitors the results. This activity reflects which ANA professional performance standard?

1. Education
   Education is maintaining current knowledge and competence, not leading an improvement project.
2. Quality of practice
   The Quality of practice standard covers taking part in improvement work, using data to find problems, making changes, and monitoring the results.
3. Professional practice evaluation
   Professional practice evaluation is appraising one's own practice against standards and feedback, not a team project to reduce errors.
4. Environmental health
   Environmental health concerns safe and healthy practice environments, such as hazardous exposures, not documentation error rates.

Graves and Corcoran's classic definition describes nursing informatics as a combination of which sciences?

1. Nursing, cognitive, and computer science
   Cognitive science entered later discussions of informatics, not the 1989 definition, which names information science instead.
2. Nursing, computer, and management science
   Management science is not part of the definition; information science is missing.
3. Nursing, information, and decision sciences
   Decision science is not in the classic definition; this swaps computer science for decision science.
4. Nursing, computer, and information science
   Graves and Corcoran defined nursing informatics as a combination of computer science, information science, and nursing science.

Which implementation specification of the HIPAA Security Rule's contingency plan standard is required rather than addressable?

1. Applications and data criticality analysis
   Criticality analysis is also addressable under the contingency plan standard.
2. Automatic logoff
   Automatic logoff is an addressable specification under the access control standard, not part of the contingency plan.
3. Testing and revision procedures
   Testing and revision of contingency plans is an addressable specification, not a required one.
4. Emergency mode operation plan
   The contingency plan standard lists the data backup plan, disaster recovery plan, and emergency mode operation plan as required specifications.

After an EHR upgrade, medication delays rose because pharmacy verification queues changed and pharmacy staffing was not adjusted. Analyzing technology, people, workflow, and organizational factors together reflects which model?

1. Sittig and Singh's sociotechnical model
   Sittig and Singh's sociotechnical model looks at health IT through interacting dimensions, including hardware and software, people, workflow, and organizational factors.
2. Lewin's force-field analysis
   Force-field analysis weighs driving and restraining forces for a planned change, not the causes of a post-upgrade problem.
3. The technology acceptance model
   The technology acceptance model predicts whether individual users will adopt a system; it does not analyze staffing and workflow.
4. Donabedian's structure-process-outcome model
   Donabedian classifies quality measures by structure, process, and outcome; it is not a model of how technology interacts with people and workflow.

Under DEA rules for electronic prescriptions for controlled substances, a prescriber signing an EPCS order must use:

1. two-factor authentication.
   DEA rules for electronic prescribing of controlled substances require two-factor authentication at signing, using two of something you know, have, or are.
2. a hospital-issued smart badge alone.
   A badge alone is one factor, something you have; it needs a second factor to meet the rule.
3. a biometric fingerprint alone.
   A biometric alone is one factor; it must be paired with a second, such as a password or hard token.
4. a unique username and password.
   A username and password is a single factor, something you know, which DEA rules do not accept for signing.

A hospital security committee is updating authentication for remote EHR access. Which approach best aligns with current NIST digital identity guidance?

1. Single sign-on authentication with a password that expires every 90 days
   Single sign-on is convenient, but one expiring password is still a single factor for remote access.
2. Multifactor authentication with long passphrases and no forced resets
   Current NIST SP 800-63B guidance favors length over complexity rules, drops routine forced password changes, and calls for multifactor authentication for remote access to sensitive data.
3. A password plus a security question, with password changes every month
   A security question is another thing you know, so this is not true multifactor authentication, and monthly changes add no protection.
4. Complex eight-character passwords that must be changed every 60 days
   Short complex passwords with forced periodic changes are the older practice NIST now advises against, because users respond with predictable patterns.

An informatics nurse describes the EHR as continuously exchanging data with labs, payers, and patient apps and adapting to new regulations. In general systems theory, this exchange makes the EHR:

1. a feedback loop.
   A feedback loop is one mechanism within a system that returns output to adjust input; it does not label the system as a whole.
2. an open system.
   An open system continuously exchanges matter, energy, or information with its environment and adapts, as the EHR does with labs, payers, apps, and regulations.
3. a suprasystem.
   A suprasystem is the larger system that contains others, such as the health system around the EHR, not a description of the EHR's exchanges.
4. a closed system.
   A closed system has no exchange with its environment, the opposite of what is described.

Each month, the informatics director compares actual help desk response times and spending with the targets in the annual plan and corrects deviations. This management function is:

1. organizing.
   Organizing arranges structure, roles, and resources, not monitoring against targets.
2. coordinating.
   Coordinating aligns the work of different groups; it does not compare results with plan.
3. planning.
   Planning set the targets in the annual plan; checking results against them and correcting course is controlling.
4. controlling.
   Controlling is measuring actual performance against standards, such as response-time and spending targets, and correcting deviations.

A patient is harmed after a nurse overrides a valid allergy alert without reviewing it. In a malpractice claim, showing that this action fell below the expected standard of care establishes which element?

1. duty of care.
   Duty arises from the nurse-patient relationship; it is shown by the relationship, not by the substandard act.
2. breach of duty.
   Breach of duty is shown by proving the nurse's action fell below the standard of care, here overriding a valid allergy alert without review.
3. proximate causation.
   Causation links the breach to the harm; proving the act was substandard does not by itself prove it caused the injury.
4. compensable damages.
   Damages are the actual injury or loss; they are a separate element from whether care fell below the standard.

Under the 21st Century Cures Act, clinical decision support software is excluded from FDA device regulation only if, among other criteria, it:

1. processes medical images or physiologic signals for diagnosis.
   Software that analyzes images or physiologic signals is expressly outside the exclusion and stays a device.
2. lets clinicians independently review the basis for its advice.
   Under the Cures Act, CDS software is excluded from the device definition only if, among other criteria, it enables the clinician to independently review the basis of its recommendation rather than rely on it.
3. is approved by the hospital's pharmacy and therapeutics committee.
   Internal committee approval is local governance and has no effect on FDA jurisdiction.
4. is purchased from a vendor certified under the ONC program.
   ONC certification is a separate program for EHR technology and has no bearing on FDA device status.

Before BCMA go-live, which test is most likely to reveal medication barcodes that the scanners cannot read?

1. scanning the vendor's sample barcode test sheet
   A vendor test sheet has ideal barcodes and cannot reveal problems with the hospital's actual stock.
2. load testing the scanners at peak medication times
   Load testing checks performance under heavy use, not barcode readability.
3. scanning actual unit-dose packages from stock
   Real packages show real-world barcode problems, such as print quality, curved vials, and repackaged or overwrapped items, that a clean test sheet hides.
4. unit testing the eMAR's scan-verification rule
   Testing the verification logic checks software, not whether physical barcodes can be read.

A clinical system takes a full backup on Sunday night and, each night after that, saves only the changes made since the Sunday full backup. After a failure on Thursday morning, restoration requires:

1. Sunday's full backup and every nightly backup since
   Restoring every nightly backup in sequence is how incremental backups work; confusing differential with incremental is the common error.
2. only Sunday's full backup
   The full backup alone would lose every change made Monday through Wednesday.
3. Sunday's full backup and Wednesday night's backup
   With differential backups, each nightly copy holds all changes since the last full backup, so restoring needs only the full backup plus the most recent differential.
4. only Wednesday night's backup
   A differential holds changes only, not the base data, so it cannot be restored without the full backup.

During a parallel go-live, nurses chart vital signs in both the old and the new system. The greatest risk of this strategy is:

1. no chance to compare the two systems' outputs
   Comparing outputs from both systems is exactly what a parallel approach allows.
2. fatigue and mismatched entries from duplicate work
   Parallel conversion doubles documentation work, which tires staff and leads to discrepancies between the two systems about which record is right.
3. no fallback system if the new system fails
   Parallel running keeps the old system available, so a fallback is its main advantage.
4. issues affecting every unit at the same moment
   Issues hitting every unit at the same moment describe the risk of a big-bang cutover, not of parallel running.

A hospital runs the new documentation system on one unit for six weeks and will use the results to decide how, and whether, to roll it out further. This is a:

1. phased rollout
   A phased rollout implements in planned stages toward full adoption; the decision to proceed is already made.
2. parallel conversion
   A parallel conversion runs old and new systems at the same time; nothing in the stem describes double documentation.
3. pilot implementation
   A pilot implements the system in one limited area to evaluate results and decide whether and how to roll it out further.
4. big-bang implementation
   Big bang moves everyone at once with no trial area.

The best way to confirm that nursing units can actually carry out the new EHR downtime procedures is to:

1. run a scheduled drill on each unit and debrief it
   A scheduled drill shows whether staff can actually perform the procedures, and the debrief exposes gaps to correct before a real outage.
2. post the procedures in each unit's break room
   Posting makes the procedures available but tells you nothing about whether staff can use them.
3. include the procedures in annual online training
   Online training builds knowledge but does not show units can perform the procedures in real conditions.
4. have staff sign that they read the policy
   A signature shows the policy was read, not that staff can carry out the procedures.

Which BCMA test case is most important for confirming that a safety function works?

1. scanning a drug the patient is allergic to
   A safety test case must show the system catches an unsafe condition, such as an allergy conflict, not just that the normal path works.
2. scanning the right drug at the scheduled time
   Scanning the right drug at the scheduled time tests only the normal path; it does not show that the safety checks fire.
3. logging in to the eMAR with a valid user ID
   Logging in with a valid user ID checks access, not a medication safety function.
4. scanning a wristband with a fully charged scanner
   Scanning a wristband with a charged scanner tests basic hardware operation, not a safety check.

Two analysts overwrote each other's changes to the same order set in the build environment. Which process would have prevented this?

1. regression testing after each build change
   Regression testing could detect broken behavior after the fact but does not prevent conflicting edits.
2. change control board approval of each request
   A change board decides what changes are allowed, but approval does not stop two approved edits from colliding in the build.
3. version-controlled configuration management
   Version control tracks and locks changes to build objects, so two analysts cannot silently overwrite each other's work and earlier versions can be restored.
4. role-based security for build access
   Both analysts legitimately need build access, so access rights would not prevent the overwrite.

After the laboratory changes the code for a troponin test, a CDS rule that depends on that result silently stops firing. Which practice would most quickly reveal the problem?

1. Tracking override rates for each active alert
   Override rates show how users respond to alerts that fire; a rule that never fires has no overrides to track.
2. Trending each rule's trigger count over time
   Watching how often each rule triggers shows a sudden drop to zero when an upstream code change breaks the rule's trigger, often well before anyone notices missing alerts.
3. Reviewing rule content at the annual CDS review
   An annual review is far too slow to catch a rule that broke after a laboratory code change.
4. Waiting for clinicians to report missing alerts
   Clinicians rarely notice an alert that never appears, so this relies on luck and lets harm happen first.

During testing, triage nurses confirm that the new sepsis screening workflow actually fits how they assess patients on arrival. This activity is best described as:

1. unit testing
   Unit testing checks single components, not whether a whole workflow fits clinical practice.
2. validation
   Validation asks whether the right system was built, meaning it meets users' real needs and workflow, which triage nurses confirm here.
3. verification
   Verification asks whether the system was built right, to specification; mixing up the two is the classic error.
4. regression testing
   Regression testing confirms existing functions still work after a change.

Business continuity reports were last printed at 0400, and an unplanned outage begins at 1000. Before giving an 1100 dose from the printed MAR, the nurse should verify:

1. that the dose also appears on yesterday's MAR
   Yesterday's MAR is older than the 0400 report and cannot show recent changes.
2. that pharmacy has restored its own system
   Pharmacy system status does not tell the nurse whether this patient's order changed.
3. whether the order changed after 0400
   Business continuity reports are a snapshot; orders changed or stopped after they printed are not shown, so active orders must be confirmed before giving a dose.
4. the patient's allergies on the 0400 report
   Checking allergies on the same old report does not address what may have changed since it printed.

A unit dashboard shows 40 metrics in identical font and color, and managers missed a rising central line infection rate. Which redesign best addresses this?

1. Add drill-down tables for all 40 metrics
   More detail behind every metric adds to the clutter without making the important signal visible.
2. Flag out-of-target metrics visually and place them first
   Good dashboard design makes exceptions stand out: color and position flag measures that miss their targets so a rising infection rate is seen at once.
3. Refresh the data every minute instead of daily
   Faster refresh does not help when the display hides what matters; infection rates do not change minute to minute.
4. Sort the metrics alphabetically for scanning
   Alphabetical order has nothing to do with importance or performance, so a problem metric stays buried.

A VTE prophylaxis reminder appears to the unit secretary at discharge instead of to the prescriber during admission ordering. Which CDS Five Rights are violated?

1. Right information and right channel
   The channel, the EHR, is appropriate, and the information itself is not described as wrong.
2. Right information and right format
   Right information and right format are not the failures: the reminder's content and its form as an alert may be fine; who sees it and when is what fails.
3. Right channel and right format
   Right channel and right format are met: the EHR and an alert are reasonable choices; the recipient and the timing are wrong.
4. Right person and right time
   The prompt reaches the unit secretary instead of the prescriber (wrong person) at discharge instead of admission ordering (wrong time).

Obtaining the sponsor's signature on the project charter for a new documentation system occurs in which project management process group?

1. initiating
   Developing and approving the project charter, which authorizes the project and names the sponsor and project manager, happens in the initiating process group.
2. monitoring and controlling
   Monitoring and controlling runs alongside the project work; it does not formally authorize the project.
3. executing
   Executing carries out the plan's work; the charter must already be signed.
4. planning
   Planning builds the detailed scope, schedule, and budget after the charter has authorized the project.

Which requirement for a pressure injury module is written clearly enough to be tested?

1. The module will make pressure injury documentation faster than before
   'Faster' has no baseline or target, so there is no clear pass or fail.
2. Braden documentation will be intuitive and user-friendly for all nurses
   'Intuitive' and 'user-friendly' are subjective; without a measure, such as task time or error rate, they cannot be tested.
3. The module will support best practice in pressure injury prevention
   'Support best practice' is a goal statement with no observable behavior to test.
4. The Braden total displays automatically once all six items are entered
   This requirement names a specific, observable behavior, the Braden total appearing once all six items are entered, that a tester can pass or fail.

Early in planning, the team wants a broad overview of available telehealth products before defining detailed requirements. The appropriate document is a:

1. request for quotation
   A request for quotation (RFQ) asks for prices on clearly specified products, the last step in the sequence, not an overview.
2. request for proposal
   A request for proposal (RFP) asks vendors for detailed solutions against defined requirements, which the team does not yet have.
3. request for information
   An RFI gathers general information about available products and vendors early, before requirements are detailed enough for a proposal request.
4. vendor statement of work
   A statement of work defines tasks and deliverables in a contract after a vendor is chosen.

Which document should be completed before a request for proposal is issued to EHR vendors?

1. the vendor contract
   The contract is signed after RFP responses are evaluated and a vendor is chosen.
2. the detailed design specification
   Detailed design follows vendor selection and turns requirements into the build.
3. the requirements specification
   Vendors respond to an RFP against defined requirements, so the requirements specification must be finished first.
4. the system test plan
   Test plans are written once the system is designed and built, long after the RFP.

During workflow analysis for BCMA, nurses report that patient wristbands often become unreadable after bathing. If this is not addressed, the MOST likely result is:

1. longer pharmacy verification times for new orders
   Pharmacy verification happens before administration and does not depend on wristband readability.
2. more transcription errors when orders are entered
   Orders are entered electronically; wristband condition has no effect on order entry.
3. more wrong-dose errors that the barcode scanner will detect
   An unreadable wristband blocks the patient scan; it does not create dose errors for the scanner to catch.
4. workarounds like scanning spare labels at the station
   When the wristband cannot be scanned, nurses tend to scan spare labels or barcodes away from the patient, defeating the identity check.

To define requirements for a new discharge-education module, the informatics nurse reviews 12 months of readmission data and incomplete-documentation reports. This step belongs to the:

1. post-implementation evaluation.
   A post-implementation evaluation happens after go-live; this data review comes before any requirements exist.
2. needs assessment.
   Reviewing outcome data and documentation reports to establish what the module must address is part of the needs assessment early in planning.
3. system design.
   Design turns already-defined requirements into screens and logic; it does not establish the need.
4. feasibility study.
   A feasibility study judges whether a proposed solution is viable; it follows the needs assessment rather than defining the need.

After reviewing RFP responses, the committee wants to see how a shortlisted system works in daily nursing practice at a similar hospital. The best next step is:

1. a second scripted demonstration.
   A scripted demo is staged by the vendor in a controlled setting and does not show real daily use.
2. a request for quotation.
   A request for quotation (RFQ) asks for prices on specified products; it shows nothing about how the system works in daily practice.
3. a vendor financial stability review.
   Financial review is a necessary step, but it says nothing about how the system works in nursing practice.
4. a site visit to a current customer.
   A site visit shows the system working in daily practice at a comparable hospital and lets nurses talk to real users.

A patient portal shows a simple summary of lab results, and patients can tap a result to see reference ranges and details. This design principle is:

1. error prevention.
   Error prevention designs out slips and mistakes before they happen; tapping for more detail does not prevent an error.
2. consistency and standards.
   Consistency and standards means the same words and controls behave the same way across screens; the layered summary-then-detail design is a different principle.
3. progressive disclosure.
   Progressive disclosure shows the essential information first and reveals detail, such as reference ranges, only when the user asks for it, keeping the first screen simple.
4. user control and freedom.
   User control and freedom is about undo and easy exits from unwanted states, not about hiding detail until it is requested.