nclex-simulator

Clinical Review Log

A running record of question changes driven by clinical accuracy — chiefly guidelines that moved after a question was written.

Questions in this bank are not static facts. Screening ages shift, drug-hold protocols get rewritten, and definitions are revised. An item that was correct when it was authored can quietly become wrong, and a practice bank nobody re-reads will keep teaching the old answer indefinitely. This file is how we keep track.

How to use it: before changing an answer key, check whether the item appears below. When you do change one, add an entry with the guideline, the issuing body and the year.

2026 — Guideline currency review

Reported by Mrs. Sandra Brown, who reviewed the bank against current practice guidelines and identified a set of questions whose answers or rationales had gone out of date. Every item below traces to that review.

Answer keys corrected

These questions had a keyed answer that is now wrong.

q63 — Colorectal cancer screening start age

Was: keyed option read “Begin colonoscopy at age 50 and repeat every 10 years.”
Now: “Begin colonoscopy at age 45 and repeat every 10 years.”
Why: The American Cancer Society lowered the average-risk start age to 45 in 2018; the USPSTF followed in 2021 (Grade B, ages 45–49). Age 50 is the retired recommendation.
Also: the item contradicted q239, q486, q694 and q903 in the same bank, all of which already taught 45. The rationale and tip were rewritten to match.

q151 — Breast cancer screening per USPSTF

Was: keyed option read “Begin mammograms at age 50; annual or biennial screening until age 74 (per USPSTF).”
Now: “Begin biennial mammograms at age 40 and continue through age 74 (per USPSTF 2024).”
Why: The USPSTF 2024 final recommendation is biennial screening mammography for women aged 40–74 (Grade B). The item’s own rationale already said so — the keyed option contradicted it.

q428 — Metformin and iodinated contrast

Was: keyed option read “Hold metformin 48 hours before the procedure and restart only when creatinine is confirmed normal.”
Now: “Check the eGFR — metformin is withheld from the time of contrast and for 48 hours after only if eGFR is below 30 or acute kidney injury is present.”
Why: The ACR–NKF consensus statement (2020), carried into the current ACR Manual on Contrast Media, does not require holding metformin before or after IV contrast in clients with eGFR ≥30 mL/min/1.73 m² and no AKI. The hold applies from the time of contrast through 48 hours after, and only when eGFR is <30, AKI is present or suspected, or contrast is given intra-arterially with likely renal artery exposure. The blanket “hold 48 hours before every scan” rule is obsolete.

Rationales and tips corrected

The keyed answers here were still defensible, but the teaching text stated a superseded guideline as current.

Question	Topic	Correction
q25	STEMI NGN matrix	Metformin hold tied to eGFR / AKI / intra-arterial contrast rather than “any contrast procedure”.
q43	Metformin teaching	Removed “MUST be held 24–48 hours before contrast”. The distractor stating that rule as fact was reworded, since under current guidance it is no longer a correct statement.
q204	Metformin before CT	Rationale now gives the eGFR-based ACR standard; tip no longer says “hold 48 h before”.
q417	Contrast nephropathy prevention	Same correction to the metformin aside.
q789	Pre-angiogram assessment	Same correction to the metformin aside.
q1029	CKD medication review	Hold now described as from the time of contrast through 48 hours after.
q1228	Metformin with raised creatinine	Rationale and tip rewritten around eGFR thresholds.
q1421	Metformin + linagliptin	Tip rewritten; removed “not all institutions hold 48 h before — know your policy”, which taught the obsolete rule as an acceptable variant.
q1576	Metformin before cardiac cath	Removed “this is a STANDARD pre-procedure protocol for all patients on metformin”.
q396	Infective endocarditis prophylaxis	Clindamycin removed from the penicillin-allergy regimens. The AHA dropped it in its 2021 update (Circulation) over C. difficile and other adverse effects. Current alternatives: cephalexin, azithromycin or clarithromycin, or doxycycline.
q423	Postpartum hemorrhage NGN bow-tie	PPH definition updated to ACOG Practice Bulletin 183.
q661	Boggy uterus / bladder distension	Same; the old 500 mL vaginal / 1,000 mL cesarean split is explicitly flagged as outdated.
q1751	Third-stage PPH	Same.
q306	Breast self-examination	Rationale said BSE “should be performed by ALL women”. The ACS no longer recommends routine BSE for average-risk women; it has not been shown to reduce mortality.

Postpartum hemorrhage, current definition (ACOG Practice Bulletin 183, 2017): cumulative blood loss of ≥1,000 mL, or blood loss accompanied by signs or symptoms of hypovolemia, within 24 hours of birth — regardless of delivery route. Blood loss over 500 mL after a vaginal birth remains abnormal and warrants evaluation, but it is no longer the diagnostic threshold.

2026 — Structural faults found by validation

Not guideline changes, but questions that could not be answered correctly. Found by tools/validate-questions.mjs, which was written during this review.

Questions	Fault	Effect
q489, q532, q590	Stem says “select all that apply” and `correct` is an array, but `type` was `"mcq"`	Rendered as radio buttons; the scorer compared a number to an array, so they could never be answered correctly. Retyped as `sata`.
q123, q1267	`rationale` key misspelled as `bationale` and `resale`	The feedback panel displayed the literal text “undefined” instead of the explanation.
8 bow-tie questions	Typed `ngn_bowtie` (the renderer dispatches on `ngn_bowties`), and using a data shape the renderer never supported	The exam screen threw and went blank whenever one appeared. Fixed by correcting the type and adding `bowtieModel()` to normalise both shapes.
6 matrix questions	Row-object shape with no `cols`/`correct`	`renderMatrix()` threw on the missing `cols` array. Fixed by adding `matrixModel()`.
68 questions	`domain: "psychosocial"` while the filter chip used `"psycho"`	Selecting Psychosocial Integrity reached only 18 of 86 psychosocial questions; the other 68 showed on the dashboard under a raw, unlabelled row. Normalised on `psychosocial`.

Reporting a suspected error

Open an issue with:

the question id
what the current answer or rationale says
what it should say
the guideline, the issuing body, and the year

You do not need to supply a patch. A well-sourced report is the hard part. Reports that lead to a fix are credited in ACKNOWLEDGMENTS.md and logged here.

This site is open source. Improve this page.