ENT Question Quality Review
Executive Summary
This review covers a candidate sample of 100 validated non-gold ENT questions (60 generic + 40 risky), evaluated against 4 benchmark items and 12 recent PYQs as the quality bar. The subject pool contains 4,280 questions total.
The most serious finding in this sample is a severe Bloom's-level compression toward the bottom of the taxonomy. In the candidate set, 44 of 100 questions sit at Bloom's Level 1 and 47 at Level 2, leaving only 9 questions at Levels 3–4 and none at Level 5. The benchmark and PYQ sets, by contrast, operate predominantly at Levels 3–5, with clinical vignettes, applied reasoning, and management decisions as the norm. This gap is not marginal — it represents a structural mismatch between what the candidate pool is delivering and what Indian PG exams (INICET, NEET-PG) actually test.
Beyond the Bloom's problem, the reviewed set shows four additional actionable patterns: a factual error and a wrong-key item that must be corrected before any further use; a broken image-dependent question that cannot function without its image; a large cluster of low-value, trivia-level recall items that add no discriminatory power; and a meaningful duplication problem where the same concept is tested by near-identical questions within the same sample.
The proportion of questions that are genuinely exam-ready without any intervention is estimated at roughly 15–20% of the candidate sample. The remainder require either disabling or substantive revision.
What Good Looks Like
The benchmark and PYQ items establish a clear quality bar. The defining features are:
Clinical vignette framing. Questions like the BPPV case (acafcb97) and the perilymphatic fistula case (8d6609dd) embed the concept in a realistic patient scenario with age, sex, symptom timeline, and relevant negatives. The candidate must reason from presentation to diagnosis, not simply recall a label.
Applied management decisions. The cochlear implant candidacy question (4463b4ad) requires the candidate to distinguish sensorineural from conductive pathology and apply selection criteria. The CSOM-with-cholesteatoma management question (d04b59dd) in the generic sample is a good local example of this pattern.
Meaningful distractors. In the piriform fossa foreign body question (4d94eca8), all four options are plausible nerve-function consequences; the correct answer requires understanding the anatomical relationship between the piriform fossa and the internal branch of the superior laryngeal nerve. Distractors are not obviously wrong.
Appropriate difficulty calibration. Benchmark items cluster at difficulty 2–3 with Bloom's 3–5. Even the one Bloom's-1 PYQ (cbba76d5, olfaction tests) is a genuine recall of a specific test name that appears in exam papers — it is not a definition of a body part.
Negative-stem discipline. When "NOT/EXCEPT" format is used (b66df1e8, da7e715d), the correct answer is a genuinely plausible-sounding distractor that requires knowledge to reject, not an obviously absurd option.
Main Issue Categories
1. Wrong Key or Factually Unsafe
Why this pattern is bad. A wrong key is the most dangerous defect in a question bank. It actively penalises candidates who know the correct answer and rewards those who do not. Even a single wrong-key item in a high-stakes test damages validity and generates student complaints that erode platform trust.
How it shows up. In this sample, two items carry factual errors serious enough to constitute wrong keys or clinically unsafe statements, and one additional item has a key that is defensible but contested enough to require editorial review.
Specific items:
1df06001 — "Which of the following is an ototoxic drug?" The marked correct answer is Ampicillin. This is factually incorrect. Ampicillin is a penicillin-class antibiotic with no established ototoxic profile. The well-recognised ototoxic drugs in the aminoglycoside class (gentamicin, streptomycin, amikacin), loop diuretics (furosemide), and platinum-based chemotherapy agents are the standard answers. None of the other options (tetracycline, vincristine, penicillin) are ototoxic either, which means the question has no correct answer among the options as written. This is a double defect: wrong key plus no valid correct option. Disable immediately.
4e4b1fb5 — "What is the function of the larynx?" The marked correct answer is "Immunity enhancement." The primary functions of the larynx are phonation, airway protection (prevention of aspiration), and participation in the Valsalva manoeuvre. "Immunity enhancement" is not a recognised laryngeal function in any standard ENT text. The three distractors (phonation, protection of lower respiratory tract, prevention of food regurgitation) are all correct functions, making this question structurally inverted — the distractors are true and the key is false. Disable immediately.
d9532013 — "What is true about quinsy?" The marked correct answer is "Commonly occurs bilaterally." Peritonsillar abscess (quinsy) is characteristically unilateral; bilateral quinsy is rare and considered atypical. The standard teaching is that quinsy is a unilateral condition. This key is wrong. The correct true statement among the options would be that penicillin (or amoxicillin-clavulanate) is used in treatment, or that immediate tonsillectomy is not indicated. Disable or fix with corrected key.
c40cbc20 — "Which of the following is not true about Bell's palsy?" The marked correct answer is "Spontaneous remission." Bell's palsy is well-known for spontaneous recovery in approximately 70–85% of cases; spontaneous remission is one of its defining features and is therefore true, not false. The correct "not true" statement in this set would be "Recurrent" — Bell's palsy is typically a single episode, though recurrence is possible, making this option debatable. The key as marked is factually wrong. Fix: correct the key to "Recurrent" or rewrite the stem and options.
Recommended disposition: 1df06001 and 4e4b1fb5 — disable immediately. d9532013 and c40cbc20 — disable pending editorial review; do not serve until corrected.
2. Wrong Subject or Wrong Topic Placement
Why this pattern is bad. Misplaced questions distort topic-level analytics, cause incorrect tagging in study plans, and confuse candidates who are drilling a specific ENT subtopic. When a question about the maxillary sinus appears under "Otology," or a question about neck dissection appears under a topic that does not cover it, the item pollutes both the source and destination topic.
How it shows up. In this sample, the problem is moderate in scale and involves two distinct sub-types: questions placed in a topic that does not match their clinical content, and questions placed in a generic "Otolaryngology Basics" bucket that should be assigned to a specific disease topic.
Specific items:
777988bc — "Antrum of Highmore is: Maxillary" is tagged under "Otology." The antrum of Highmore is the maxillary sinus — a rhinology/paranasal sinus topic. This has no business in Otology. Fix: reclassify to Rhinology or Diseases of the Nose and Paranasal Sinuses.
dbe17e3b — "An abscess related to the sternocleidomastoid muscle is known as: Bezold abscess" is tagged under "Neurotology." Bezold's abscess is a complication of mastoiditis and is appropriately an otology/diseases-of-the-ear topic, not neurotology. Fix: reclassify to Diseases of the Ear or Otology.
82050b59 — "Areas of spontaneous healed perforation of the tympanic membrane" is tagged under "Neurotology." This is a basic otology/tympanic membrane anatomy question. Fix: reclassify to Diseases of the Ear.
be851edb — "Mild hoarseness with stridor is seen in: Bilateral abductor palsy" is tagged under "Otolaryngology Basics." This is a laryngology question and belongs under Diseases of the Larynx. Fix: reclassify.
8ed3f04f — "A Blom-Singer prosthesis is used" is tagged under "Otolaryngology Basics." This is a post-laryngectomy voice rehabilitation question and belongs under Diseases of the Larynx or ENT Oncology. Fix: reclassify.
6b493b18 — "What is true regarding functional aphonia?" is tagged under "Otolaryngology Basics." Functional aphonia is a laryngology topic. Fix: reclassify to Diseases of the Larynx.
Recommended disposition: All six items above — fix by reclassification. The factual content of most is acceptable; the placement is the primary defect.
3. Broken Delivery (Missing Image, Malformed Options, Incomplete Stem)
Why this pattern is bad. A question that depends on an image to be answerable becomes completely non-functional when the image is absent. The candidate sees a stem with no clinical information and four options with no basis for selection. This is worse than a bad question — it is an unanswerable question that wastes exam time and generates justified complaints.
How it shows up. Two image-dependent questions in the reviewed set have stems that are entirely uninformative without the referenced image.
Specific items:
b34be1c4 — "Identify the condition of the given image" with options including acquired cholesteatoma, congenital cholesteatoma, rupture of tympanic membrane, and keratosis obturans. The stem provides zero clinical information. Without the image, this is a random four-option guess. The question is tagged as a NEET-PG 2019 PYQ, which means the original image existed in the source exam. If the image cannot be reliably attached and verified, this item cannot function. Disable unless image can be confirmed and attached.
cdac9a47 — "Identify the instruments shown in the image" with options pairing Politzer bag/Hartmann catheter vs. Siegel speculum/Higginson catheter. Same problem — the stem is entirely image-dependent. Without the image, the question is unanswerable. Disable unless image can be confirmed and attached.
A secondary delivery issue appears in several questions using "All of the above" or "Both of the above" as the correct answer. While not broken in the strict sense, this construction is a known item-writing flaw because it rewards partial knowledge (a candidate who identifies any one true option can select "all of the above" without knowing the others). Examples include 998c65f0, 9e28df96, a1fa6822, 5fd795ea, and c8f51559. These are not broken but are weak delivery choices that should be flagged for revision.
Recommended disposition: b34be1c4 and cdac9a47 — disable unless images are confirmed. "All of the above" items — flag for revision in the next editorial cycle.
4. Low-Value But Correct (Too Simple, Low-Yield, Trivia-Heavy, Weak Exam Relevance)
Why this pattern is bad. Questions at Bloom's Level 1 that test pure definitional recall of a single fact — with no clinical context, no reasoning requirement, and no discriminatory power — do not prepare candidates for INICET or NEET-PG. They inflate the question count without improving learning outcomes. When a candidate drills these items, they are practising a skill (rote recall of isolated facts) that is not what the exam rewards. The benchmark data shows that the gold standard operates at Bloom's 3–5; the candidate sample has 44 questions at Bloom's 1 and 47 at Bloom's 2, with only 9 at Bloom's 3–4.
How it shows up. This is the single largest problem category in the reviewed set. The pattern appears as: one-line stems asking for a definition, a single anatomical fact, or a named association; options that include obviously wrong choices (e.g., "Touch" as a function of the semicircular canals); and correct answers that any first-year MBBS student would know.
Specific items (representative, not exhaustive):
69113e63 — "The semicircular canals primarily respond to which type of movement?" Options include "Touch" and "Gravity." This is a first-year physiology recall item. No PG exam tests this in isolation. Disable.
5baedf6d — "The organ of Corti is situated in which of the following structures?" with "None of the above" as a distractor. Basic anatomy, Bloom's 1. Disable.
3e0c7bbc — "Endolymph resembles which of the following fluid?" Bloom's 1 physiology recall. Disable.
762650fe — "The Arnold nerve is a branch of which of the following nerves?" Single-fact anatomy recall. Disable.
661b6442 — "The glycerol test is used for the diagnosis of which condition?" This fact is worth knowing, but the question is a pure one-liner with no clinical context. A vignette version (patient with fluctuating hearing loss, tinnitus, and vertigo — what diagnostic test involves oral glycerol administration?) would be exam-relevant. As written, disable or rewrite as vignette.
09b0f597 — "Mikulicz and Russell bodies are characteristic of which condition?" Pure histopathology recall, Bloom's 1. Disable.
b4ae788b — "Palatine tonsils are lined by which type of epithelium?" Basic histology, no clinical relevance at PG level. Disable.
a69855b6 — "A Tornwaldt cyst is: A nasopharyngeal cyst" Three of four options are anatomically absurd (laryngeal cyst, ear cyst). No discriminatory power. Disable.
c7952997 — "Mulberry mucosa is seen in which condition?" One-liner sign-to-diagnosis recall. Disable.
3399a692 — "Rhinitis medicamentosa is due to which of the following?" Bloom's 1, single-fact. Disable.
feb60e79 — "What is the presenting symptom of nasal myiasis?" (also duplicated as 3edb8bfe — see Category 5). Disable.
e0d106c3 — "Granular cell myoblastoma of the tongue is: A benign tumor" The distractors include "lymphatic enlargement" and "developmental anomaly" — not plausible alternatives. Disable.
8946c5bb — "Cochlear implants convert what form of energy to which form of energy?" Bloom's 1 device mechanics. Disable.
6629fbea — "Where is the scutum located?" Single anatomical fact. Disable.
9e2ae494 — "Allergic salute is seen in which of the following conditions?" Bloom's 1 sign recall. Disable.
4e4b1fb5 — already flagged as wrong key (Category 1), but also exemplifies this pattern.
Recommended disposition: The majority of Bloom's-1 items in this sample should be disabled. Where the underlying concept is high-yield (e.g., glycerol test for Meniere's, ototoxic drugs), the concept should be retained but rewritten as a clinical vignette at Bloom's 3 or higher. Where the concept is genuinely low-yield at PG level (epithelium of tonsil, location of scutum as a standalone question), disable without replacement.
5. Repetitive or Duplicative Coverage
Why this pattern is bad. Duplicate or near-duplicate questions within the same pool waste slots, skew topic weighting in test generation, and give candidates a false sense of coverage breadth. When two questions test the same fact with the same stem structure and the same correct answer, serving both in the same test or study plan is a quality failure.
How it shows up. Several exact or near-exact duplicates are observed within the reviewed candidate sample itself — meaning the duplication rate in the full pool of 4,280 is likely substantially higher.
Specific items:
998c65f0 and 9e28df96 — Rhinolith symptoms. Both questions ask about clinical manifestations of a rhinolith. 998c65f0 asks "What are the common clinical manifestations?" and 9e28df96 asks "What can a rhinolith cause?" Both have "All of the above" as the correct answer, with overlapping option sets (nasal obstruction, foul-smelling discharge, pain in one; nasal obstruction, epistaxis, epiphora in the other). These are functionally duplicate items on the same low-yield topic. Disable both — the topic is low-yield at PG level and neither question is well-constructed.
9a0e5d84 and 6f1407bd — VEMP and inferior vestibular nerve. Both questions ask which structure VEMP assesses, with the same correct answer (inferior vestibular nerve). 9a0e5d84 asks "VEMP has been used in the assessment of which structure?" and 6f1407bd asks "VEMP detects lesions of which structure?" The options are nearly identical. Disable 9a0e5d84 (the easier, less precisely worded version); keep 6f1407bd if the concept is to be retained, but consider upgrading to a clinical vignette.
b0c45485 and ed821285 — Chevallet fracture mechanism. b0c45485 asks about the mechanism of Chevallet fracture of the nasal bone (correct answer: blow from the side) and ed821285 asks about Chevallet fracture of the nasal septum (correct answer: blow from below). These are actually testing different facts — nasal bone vs. nasal septum — but the stems are so similar that they will appear duplicative to candidates and may reflect a factual inconsistency between the two items. This requires editorial clarification: if both are correct for their respective structures, they should be clearly differentiated in the stem; if one is wrong, it should be corrected. Fix both stems to make the distinction explicit, or disable the less accurate one.
bf157acb and f9d7317e — Diabetic patient with blackish nasal discharge / mucormycosis. bf157acb is a one-liner ("A young diabetic patient presents with blackish nasal discharge. What is the most likely diagnosis?") and f9d7317e is a richer vignette with CT findings and bone erosion. Both arrive at mucormycosis. The vignette version (f9d7317e) is clearly superior and exam-relevant. Disable bf157acb; keep f9d7317e.
feb60e79 and 3edb8bfe — Nasal myiasis presenting symptom. These are exact duplicates with identical stems, identical options, and identical correct answers. Disable 3edb8bfe; keep feb60e79 (though even the retained item is low-yield and should be reviewed under Category 4).
128f05d8 and 4fa2a4b1 — Referred otalgia. 128f05d8 asks "Referred pain in the ear is commonly associated with which of the following conditions?" with carcinoma of the tongue as the correct answer. 4fa2a4b1 asks "Referred otalgia can be due to which of the following?" with "All of the above" (carcinoma of larynx, oral cavity, tongue) as the correct answer. These test the same concept with contradictory keys — one says tongue specifically, the other says all three. This is both a duplication and a potential factual conflict. Disable 4fa2a4b1 (the "all of the above" version is weaker and the conflict needs resolution); review 128f05d8 for accuracy.
9c1a088e and 9d1b3048 — JNA clinical presentation. 9c1a088e presents a 10-year-old with unilateral nasal obstruction, epistaxis, and cheek swelling (correct: angiofibroma). 9d1b3048 presents a 16-year-old with recurrent epistaxis and a firm purplish mass on the posterior choanae (correct: surgical excision). These are not exact duplicates — one tests diagnosis, one tests management — but they cover the same condition at a similar level and both are already well-covered by the gold-standard PYQ (6e819832). Keep 9d1b3048 (management question, higher Bloom's); disable 9c1a088e (diagnosis question, lower Bloom's, weaker vignette than the PYQ).
Recommended disposition: Disable the weaker item in each duplicate pair as described above. Flag the Chevallet fracture pair for editorial fact-check before any disposition decision.
6. Worthwhile Concept, Weak Execution (Keep the Concept, Fix the Stem/Options/Vignette)
Why this pattern is bad. These questions test concepts that are genuinely high-yield and exam-relevant, but the execution — stem construction, distractor quality, or clinical framing — is weak enough that the item underperforms its potential. Disabling these would waste a good concept; rewriting them would produce exam-ready items.
How it shows up. The pattern appears as: correct-answer items where the stem is a one-liner that could be converted to a vignette; items where one or more distractors are implausible or obviously wrong; and items where the "NOT/EXCEPT" format is used but the correct answer is not genuinely surprising or requires no reasoning.
Specific items:
ff3e578e — Hypopharyngeal carcinoma TNM staging. This is a genuinely high-yield, Bloom's-3 question with a clinical vignette, a specific staging decision, and plausible distractors. The concept is excellent. However, the stem says "At surgery, you find a 3 cm tumour affecting the right pyriform fossa and extending into the post-cricoid region" — the key clinical feature for T2 staging of hypopharyngeal carcinoma is involvement of more than one subsite or extension to an adjacent site without fixation. The stem should explicitly state whether there is laryngeal fixation (which would push to T3) and whether the tumour is limited to one subsite. As written, the staging rationale is ambiguous. Fix: clarify the staging criteria in the stem to remove ambiguity about T2 vs T3.
83c9e735 — CSF rhinorrhea EXCEPT. The correct answer is "Glucose content less than 10 mg/dL." This is a good concept — CSF glucose is typically greater than 30 mg/dL (not less than 10). However, the stem is a bare "All are true EXCEPT" format with no clinical context. A vignette (patient after head trauma with clear watery nasal discharge — which of the following is NOT a feature of CSF rhinorrhea?) would be more exam-aligned. Fix: add clinical context.
716d26c2 — Unilateral secretory otitis media in an adult. Correct answer is nasopharyngeal carcinoma. This is a high-yield clinical pearl that appears in PG exams. The question is a one-liner but the concept is important enough to keep. The distractors are reasonable. Fix: convert to a brief vignette (e.g., a 45-year-old male with unilateral conductive hearing loss and a Type B tympanogram — what must be excluded?) to raise the Bloom's level.
d04b59dd — CSOM with cholesteatoma and vertigo. This is one of the better questions in the generic sample. The clinical scenario is clear, the correct answer (immediate mastoid exploration) is appropriate, and the distractors are plausible. The only weakness is that the stem does not specify what type of vertigo or whether there are signs of labyrinthine fistula, which would strengthen the clinical reasoning chain. Keep as is or minor fix to add fistula sign detail.
d0fd9d0a — Acute mastoiditis in a child. Good clinical vignette with appropriate management question. The correct answer (mastoidectomy with incision and drainage and antibiotics) is appropriate for coalescent mastoiditis with subperiosteal abscess. Keep.
df34dde0 — Tuning fork test interpretation. The clinical scenario (Weber louder than normal, Schwabach bone conduction better than normal, Rinne negative) pointing to conductive deafness is a good applied reasoning question. However, the stem says "Weber test — sound from a vibrating tuning fork was louder than normal" which is ambiguous — louder than normal in which ear? Weber lateralisation should specify the affected side. Fix: clarify Weber lateralisation in the stem.
4313d25f — Parapharyngeal abscess NOT true. The correct answer is Horner's syndrome. This is a reasonable question — Horner's syndrome is actually a recognised feature of parapharyngeal abscess due to involvement of the sympathetic chain in the posterior compartment. This makes the marked key potentially wrong. Horner's syndrome CAN occur in parapharyngeal abscess. The question needs fact-checking. If Horner's is indeed a recognised feature, the key is wrong and this moves to Category 1. Flag for urgent fact-check; provisionally treat as Category 1 risk.
97c0c8e3 — Hot potato voice EXCEPT. Correct answer is glottic cancer. This is a good concept — hot potato voice is a supraglottic/oropharyngeal phenomenon, not a glottic one. The distractors are appropriate. The stem is a bare one-liner. Fix: add brief clinical context (e.g., "A patient presents with a muffled, thick voice. This voice quality is seen in all of the following EXCEPT:") to improve clinical relevance.
0eb76cca — Rhinoscleroma NOT a characteristic. Correct answer is "Caused by a fungus" (it is caused by Klebsiella rhinoscleromatis, a bacterium). Good concept, reasonable distractors. The stem is a bare "NOT a characteristic" format. Keep or minor fix to add clinical context.
aef76d38 — Inverted papilloma. Correct answer is "Can be premalignant." However, option B ("Arises from the lateral wall") is actually true — inverted papilloma does arise from the lateral nasal wall. This means option B is also a correct statement, making the question ambiguous if the intent is to identify the one true statement. Fix: review all options for factual accuracy and rewrite to ensure only one option is correct.
Recommended disposition: ff3e578e, 83c9e735, 716d26c2, df34dde0, 97c0c8e3 — fix stems. d04b59dd, d0fd9d0a — keep. 4313d25f — urgent fact-check before any disposition. aef76d38 — fix options. 0eb76cca — keep or minor fix.
Prioritization
The following table summarises the recommended action sequence for the content operations team.
Tier 1 — Immediate action required (do not serve)
These items carry wrong keys, factual errors, or broken delivery that makes them actively harmful or unanswerable.
| Priority | Question ID | Issue | Action |
|---|---|---|---|
| 1 | 1df06001 | Wrong key: Ampicillin is not ototoxic; no correct option exists | Disable immediately |
| 2 | 4e4b1fb5 | Wrong key: "Immunity enhancement" is not a laryngeal function; distractors are all correct | Disable immediately |
| 3 | d9532013 | Wrong key: Quinsy is characteristically unilateral, not bilateral | Disable immediately |
| 4 | c40cbc20 | Wrong key: Spontaneous remission IS a feature of Bell's palsy | Disable immediately |
| 5 | b34be1c4 | Broken delivery: image-dependent stem with no image | Disable unless image confirmed |
| 6 | cdac9a47 | Broken delivery: image-dependent stem with no image | Disable unless image confirmed |
| 7 | 4313d25f | Potential wrong key: Horner's syndrome may be a feature of parapharyngeal abscess | Urgent fact-check; suspend pending review |
Tier 2 — Fix before next deployment cycle
These items have correct keys but require stem, option, or classification fixes to be exam-ready.
| Priority | Question ID | Issue | Action |
|---|---|---|---|
| 8 | ff3e578e | Ambiguous T-staging criteria in stem | Fix stem |
| 9 | df34dde0 | Weber lateralisation ambiguous | Fix stem |
| 10 | aef76d38 | Option B is also factually true; ambiguous question | Fix options |
| 11 | b0c45485 / ed821285 | Chevallet fracture pair: nasal bone vs. septum distinction unclear | Fact-check and fix stems |
| 12 | 777988bc | Wrong topic: Antrum of Highmore in Otology | Reclassify |
| 13 | dbe17e3b | Wrong topic: Bezold abscess in Neurotology | Reclassify |
| 14 | 82050b59 | Wrong topic: TM perforation in Neurotology | Reclassify |
| 15 | be851edb | Wrong topic: Abductor palsy in Otolaryngology Basics | Reclassify |
| 16 | 8ed3f04f | Wrong topic: Blom-Singer in Otolaryngology Basics | Reclassify |
| 17 | 6b493b18 | Wrong topic: Functional aphonia in Otolaryngology Basics | Reclassify |
| 18 | 83c9e735 | Good concept, bare stem — add clinical context | Fix stem |
| 19 | 716d26c2 | Good concept, one-liner — convert to vignette | Fix stem |
| 20 | 97c0c8e3 | Good concept, bare stem — add clinical context | Fix stem |
Tier 3 — Disable (low-value, duplicative, or both)
These items are either too simple for PG-level use, exact/near-exact duplicates of better items, or both.
Disable: 69113e63, 5baedf6d, 3e0c7bbc, 762650fe, 09b0f597, b4ae788b, a69855b6, c7952997, 3399a692, 8946c5bb, 6629fbea, 9e2ae494, e0d106c3, 661b6442 (or rewrite as vignette), 3edb8bfe (duplicate of feb60e79), 9a0e5d84 (duplicate of 6f1407bd), bf157acb (duplicate of f9d7317e), 4fa2a4b1 (duplicate/conflict with 128f05d8), 9c1a088e (weaker duplicate of PYQ 6e819832), feb60e79 (low-yield even if unique), 998c65f0 and 9e28df96 (both low-yield duplicates).
Example Keep / Fix / Disable Calls
The following examples illustrate the reasoning applied across the three dispositions.
KEEP — d04b59dd "A patient with CSOM has cholesteatoma and presents with vertigo. What is the treatment of choice?" This is a clean management question at Bloom's 3. The clinical scenario is specific (CSOM + cholesteatoma + vertigo = labyrinthine fistula risk), the correct answer (immediate mastoid exploration) is unambiguous, and the distractors (antibiotics alone, myringoplasty, labyrinthectomy) are all plausible wrong choices that require reasoning to reject. No fix needed.
KEEP — d0fd9d0a "A 12-year-old child with fever, postauricular pain, mastoid bulging, and loss of bony trabeculae..." Good vignette, appropriate management question for coalescent mastoiditis with subperiosteal abscess. Correct answer (mastoidectomy + I&D + antibiotics) is well-supported. Distractors represent common management errors. Keep.
FIX — ff3e578e "At surgery, you find a 3 cm tumour affecting the right pyriform fossa and extending into the post-cricoid region. According to TNM staging, what T-stage is this malignancy?" The concept (hypopharyngeal carcinoma staging) is high-yield and the vignette format is appropriate. The defect is that the stem does not state whether there is laryngeal fixation or hemilarynx fixation, which is the critical discriminator between T2 and T3 in hypopharyngeal carcinoma. Add a line explicitly stating "without fixation of the hemilarynx" to make T2 unambiguous. Also confirm that involvement of two subsites (pyriform fossa + post-cricoid) is the basis for T2 in the current AJCC/UICC edition being used.
FIX — df34dde0 "Weber test — sound from a vibrating tuning fork was louder than normal; Schwabach — bone conduction better than normal; Rinne — air conduction did not outlast bone conduction. What is the most likely diagnosis?" The concept (interpreting a battery of tuning fork tests to diagnose conductive hearing loss) is exactly the kind of applied reasoning that PG exams test. The defect is that "Weber test — sound louder than normal" is ambiguous — it should specify lateralisation to the affected ear. Rewrite the Weber description as "Weber lateralises to the affected ear" or specify a side to make the interpretation unambiguous.
DISABLE — 1df06001 "Which of the following is an ototoxic drug? Correct answer: Ampicillin" Ampicillin has no established ototoxic profile. The question has no correct answer among the options as written. This is a double defect (wrong key + no valid correct option) and must be disabled immediately. If the concept of ototoxic drugs is to be tested, a new question should be written with aminoglycosides, loop diuretics, or cisplatin as the correct answer and appropriate distractors.
DISABLE — 69113e63 "The semicircular canals primarily respond to which type of movement? Options: Linear acceleration / Angular acceleration / Gravity / Touch" "Touch" as a distractor for a vestibular physiology question is not a plausible alternative. Any candidate who has attended a single physiology lecture will answer this correctly. The question adds no discriminatory value and does not reflect the reasoning style of any PG exam in the reviewed set. Disable. If angular vs. linear acceleration is to be tested, it should be embedded in a clinical scenario (e.g., a patient with BPPV — which structure detects the angular acceleration that triggers the nystagmus?) to reach Bloom's 3.
DISABLE — 3edb8bfe "What is the presenting symptom of nasal myiasis? Correct answer: Severe irritation with sneezing" This is an exact duplicate of 998c65f0 (same stem, same options, same key). One copy must be disabled. Additionally, nasal myiasis is a low-yield topic at INICET/NEET-PG level and the question is Bloom's 1. Disable 3edb8bfe; the retained copy (998c65f0) should also be reviewed for low-yield status under Category 4.