Anesthesiology Question Quality Review
Executive Summary
This review covers a candidate sample of 100 validated non-gold questions drawn from a pool of 3,585 Anesthesiology items. The sample was reviewed against four benchmark questions and twelve recent PYQs as the quality bar.
The most striking finding in this sample is the severe Bloom's level compression toward the bottom of the taxonomy: 33 of 100 candidate questions sit at Bloom's Level 1 (pure recall), and 58 at Level 2 (basic comprehension), leaving only 9 questions at Levels 3–4 combined. The gold-standard PYQ set, by contrast, includes items at Levels 3, 4, and 5 that require clinical reasoning, data interpretation, and synthesis. This mismatch is the single largest structural problem in the reviewed set.
Beyond the Bloom's distribution, the sample contains a meaningful cluster of factually unsafe or wrong-key items, several broken image-dependent questions that are undeliverable without their images, a large volume of low-yield trivia (veterinary use of ketamine, chemical formula of laughing gas, pin index codes), and a recurring pattern of malformed option sets including duplicate options, "All of the above" catch-alls, and five-option items in a four-option exam format.
The subject also shows topic contamination: at least two questions filed under Anesthesiology topics are clearly Dental Anesthesia items (Gowgates block, bilateral mandibular block), and one question filed under Local Anesthetics tests a neuromuscular blocker property.
Approximately 30–35% of the reviewed candidate questions are recommended for immediate disable; a further 20–25% require substantive fixes before they are exam-ready. Only about 40–45% are serviceable as-is, and even those are predominantly low-Bloom's items that add limited discriminatory value.
What Good Looks Like
The benchmark and PYQ sets establish a clear quality bar. The best items in the gold set share several properties that the candidate sample largely fails to replicate.
Clinical vignette framing with decision-making demand. Question b47460b0 (ARDS patient on ventilator develops hypotension, tachycardia, absent breath sounds, tracheal deviation) forces the candidate to synthesize a constellation of findings into a diagnosis and then identify the mechanism. Question 6828292a (laparoscopic cholecystectomy, sudden EtCO2 drop to 8 mmHg, SpO2 90%, BP 80/50, low airway pressure) requires the candidate to distinguish gas embolism from pneumothorax and anaphylaxis using quantitative monitoring data. These are Bloom's Level 4 items that mirror real exam difficulty.
Meaningful distractors that represent plausible clinical errors. In aabaa85d (failed intubation after rocuronium 85 mg), the distractors Edrophonium and Neostigmine are genuinely tempting wrong answers because they are legitimate reversal agents — just not for this scenario. The distractor set teaches as well as tests.
Quantitative reasoning and data interpretation. Question 77d06928 (ventilated pneumonia patient, SpO2 100% on FiO2 90%, PEEP 12, tidal volume 420 mL) requires the candidate to recognize that the FiO2 is dangerously high relative to the oxygenation achieved and to act on that. This is Bloom's Level 5 and represents the ceiling the candidate pool should be reaching toward.
Appropriate difficulty calibration. Even the simpler PYQ items (e.g., ecb677e0 on highest FiO2 delivery device, c490b26b on second-generation LMA) are clean, unambiguous, and test a concept that genuinely appears in Indian PG exams. They are easy but not trivial.
Stem completeness. Every gold-standard item is self-contained. No item depends on an image that may or may not render.
The candidate sample should be measured against all of these properties.
Main Issue Categories
1. Wrong Key or Factually Unsafe
Why this pattern is bad
A wrong key is the most serious quality defect possible. It actively harms candidates who have studied correctly, rewards those who have not, and — if the item reaches a live test — creates scoring disputes and erodes platform credibility. Factually unsafe items that are not outright wrong but are ambiguous enough to support multiple defensible answers are nearly as damaging.
How it shows up
In this sample, wrong-key and factually unsafe problems appear in at least five questions. The errors range from a clearly incorrect answer key to ambiguous stems where the "correct" option is only one of several defensible choices.
Affected question IDs and explanations
21fb3af2 — "Which neuromuscular blocking agent is most commonly associated with bradycardia during anesthesia?" The marked correct answer is Atracurium. This is factually wrong. Atracurium causes histamine release and can produce tachycardia and hypotension; it does not characteristically cause bradycardia. The agent most associated with bradycardia among non-depolarizing NMBAs is Vecuronium (vagotonic at higher doses) or, more classically, Succinylcholine (which causes bradycardia especially in children and on repeat dosing). Pancuronium, listed as a distractor, is actually known for causing tachycardia via vagolysis. The key is wrong and the distractor set is internally inconsistent.
77aa304d — "Which anesthetic agent is known for providing smooth induction?" The marked correct answer is Isoflurane. This is factually unsafe. Isoflurane is well-documented to cause airway irritation, breath-holding, laryngospasm, and coughing on induction — it is explicitly not the agent of choice for smooth inhalational induction. Sevoflurane is the standard answer for smooth inhalational induction in modern practice, and Halothane was the historical answer. This item is tagged as a NEET-PG 2013 PYQ, but the key as presented does not match the accepted answer for that question.
f222f1f8 — "A throat pack is primarily used for which of the following purposes?" The marked correct answer is "To facilitate easier extubation." This is wrong. The primary purpose of a throat pack is to prevent aspiration of blood, secretions, and debris into the trachea during oropharyngeal or nasal surgery. The correct answer is the first distractor option. The key and the correct distractor have been transposed.
75e62c0b — "Overdosage of a local anesthetic agent would most likely cause which of the following?" The marked correct answer is Hypotension. While hypotension is a late cardiovascular consequence of LA toxicity, the question asks what is "most likely" caused by overdosage. The well-established sequence is CNS excitation (seizures) followed by CNS depression, and then cardiovascular collapse. Cardiac arrhythmias (listed as a distractor) is also a recognized and arguably more prominent early cardiovascular manifestation than simple hypotension. The stem is ambiguous and the key is contestable. This item is also conceptually redundant with a86f7417 in the same sample.
5766242b — "Which of the following characterizes Mallampati class 3?" The marked correct answer is "Both hard palate and soft palate visible." This is factually incorrect. In the standard Mallampati classification, Class 3 is defined as visibility of only the soft palate and base of the uvula (or in some versions, only the soft palate). "Both hard palate and soft palate visible" corresponds to Class 2. Class 4 is hard palate only. The key is wrong by one class.
Recommended disposition: All five items — 21fb3af2, 77aa304d, f222f1f8, 75e62c0b, 5766242b — Disable immediately. The wrong-key items cannot be fixed by stem editing alone; the underlying factual error or key assignment must be corrected and the item re-reviewed before any live use. Given that strong gold-standard coverage exists for all five underlying concepts (NMB side effects, inhalational induction agents, throat pack purpose, LA toxicity sequence, Mallampati classification), disabling and replacing with verified items is safer than attempting in-place repair.
2. Wrong Subject or Wrong Topic Placement
Why this pattern is bad
Topic misclassification corrupts topic-level analytics, causes items to appear in the wrong test plans, and confuses candidates who are studying by topic. It also inflates apparent coverage of one topic while leaving another under-represented. Subject contamination — where a question from a different specialty is filed under Anesthesiology — is a harder problem because the item may be correct within its home specialty but wrong or misleading in the Anesthesiology context.
How it shows up
In this sample, the contamination is primarily from Dental Anesthesia / Oral Surgery, which has its own distinct question bank. Several items in the Regional Anesthesia topic are clearly dental nerve block questions. There is also a topic-level misclassification within Anesthesiology itself.
Affected question IDs and explanations
ed8d0f17 — "Which of the following techniques of local anesthesia requires extraoral landmarks? (Gowgates)" This is a dental inferior alveolar nerve block question. Gowgates, Fischer 1-2-3, Vazirani-Akinosi, and high tuberosity approach are all dental anesthesia techniques. This item belongs in a Dental Surgery or Oral Medicine question bank, not in Anesthesiology Regional Anesthesia.
1d9ebb2c — "What is true regarding a bilateral mandibular block?" Again, this is a dental anesthesia question. The clinical context (bilateral mandibular block, risk of swallowing tongue, space infection) is entirely within the domain of dental practice. It has no place in an Anesthesiology PG exam question bank.
8b7bbc8d — "Which neuromuscular blocking agent releases the maximum amount of histamine?" This item is filed under Local Anesthetics but tests a property of neuromuscular blocking agents. The correct answer (Succinylcholine, though D-tubocurarine is the classical answer for maximum histamine release among non-depolarizing agents — making this key also potentially contestable) belongs under General Anesthesia or Complications in Anesthesia, not Local Anesthetics. This is an intra-subject topic misclassification.
c41a1a03 — "Succinylcholine is used with caution in all conditions except which of the following?" This item is filed under Anesthetic Equipment and Monitoring. Succinylcholine is a pharmacological agent; its contraindications belong under General Anesthesia or Complications in Anesthesia. This is a topic-level misclassification within the subject.
9d7c7d52 — "Veril's sign is seen in which condition?" Filed under Anesthetic Equipment and Monitoring. Veril's sign (venous engorgement of the dorsum of the hand on lowering the arm, seen with diazepam administration) is a pharmacological observation, not an equipment or monitoring concept. Misclassified topic.
Recommended disposition: ed8d0f17 and 1d9ebb2c — Disable from Anesthesiology and route to the Dental Surgery subject bank for separate review. 8b7bbc8d, c41a1a03, 9d7c7d52 — Fix by reassigning to the correct topic within Anesthesiology; no stem changes required.
3. Broken Delivery (Missing Image, Malformed Options, Incomplete Stem)
Why this pattern is bad
A question that depends on an image that does not render is completely unanswerable. The candidate sees a stem that says "identify the type shown" or "the machine shown below" with no image, and every option becomes a random guess. This is not a quality problem — it is a delivery failure that makes the item actively harmful in a live test. Malformed option sets (duplicate options, five options in a four-option format, garbled text) are a separate but equally serious delivery problem.
How it shows up
In this sample, broken delivery appears in three distinct forms: image-dependent stems with no image reference that can be verified, malformed option sets with duplicate text, and a five-option item in a four-option exam format.
Affected question IDs and explanations
b8b8dd08 — "Identify the type of endotracheal tube shown." The stem contains no image. Without the image, the four options (Flexometallic, Laser, RAE, Double lumen) are indistinguishable. The item is undeliverable.
f4cfbaf8 — "The interpretation for the following capnography is:" The stem references a capnography tracing that is not present in the text. Without the waveform image, the options (Endobronchial intubation, Bronchospasm, Esophageal intubation, Curare cleft) cannot be evaluated. This item is in multiple daily plans, meaning it is already being served to candidates in a broken state.
3fc438d2 — "The machine shown below is used for ______:" The stem contains a LaTeX placeholder (\qquad) and no image. The item is undeliverable.
c3fa9ede — "What does the marking X in the endotracheal tube indicate?" The stem references a marking on a tube that requires an image to identify. Without the image, the question cannot be answered. This item is also in multiple daily plans.
c3a4e740 — "What is the primary use of the depicted device?" No image is present. Options reference neuromuscular blockade monitoring, depth of anesthesia monitoring, temperature monitoring, and EtCO2 monitoring — all plausible for different devices. Undeliverable without the image.
d77f5c2f — Duplicate option text. Both Option A and Option B read identically: "In areas with lower pH, the effectiveness of anesthetic agents is reduced." One option is marked correct and the other incorrect, but they are textually identical. This is a broken option set that makes the item unanswerable and will confuse any candidate who reads carefully.
bbd838a9 — Five-option item. This item has five answer choices (Pethidine + Pentobarbital + Scopolamine; Scopolamine alone; Pethidine alone; Pentobarbital alone; Mepiridine). Standard Indian PG exam format is four options. The fifth option also introduces a misspelling ("Mepiridine" for Meperidine/Pethidine) and is conceptually redundant with Option C. The item needs to be reduced to four options with the redundancy resolved.
e1b816f4 — Five-option item with a garbled option. Options include "25%", "50%", "5" (no percent sign), "75%", and "5%" — five options, one of which ("5") is clearly a data entry error for "5%", making it a duplicate of the correct answer. The item is malformed.
cab5cd9d — Garbled option text. Option C reads "0.736111111" — this is clearly a data conversion artifact (likely a fraction that was not properly formatted) rather than a meaningful concentration. The item has a broken option.
Recommended disposition: b8b8dd08, f4cfbaf8, 3fc438d2, c3fa9ede, c3a4e740 — Disable until images are confirmed present and rendering correctly; if images cannot be recovered, disable permanently. d77f5c2f — Disable; the duplicate option error makes the item unrescuable without a full rewrite. bbd838a9, e1b816f4 — Fix by reducing to four clean options and correcting the spelling/formatting errors. cab5cd9d — Fix by replacing the garbled option with the correct concentration label (1:10,000 or 0.1 mg/mL, per standard ACLS dosing).
4. Low-Value But Correct (Too Simple, Low-Yield, Trivia-Heavy, Weak Exam Relevance)
Why this pattern is bad
A question can be factually correct and still be a poor exam item. When a question tests a fact that any first-year MBBS student knows, requires no clinical reasoning, and has no realistic chance of appearing in NEET-PG or INI-CET, it wastes a slot in the question bank and in a candidate's study session. The Bloom's Level 1 items in this sample are disproportionately of this type. More specifically, trivia-heavy items — those testing historical eponyms, veterinary applications, or chemical formulas — have essentially zero exam relevance for Indian PG candidates.
How it shows up
This is the largest single category in the reviewed set. Approximately 30–35 items in the combined generic and risky samples fall into this bucket. The pattern clusters around: (a) pure recall of a single drug property with no clinical context, (b) historical or obsolete agents (ether, chloroform, trilene), (c) veterinary or non-human applications, (d) chemical formulas and pin index codes, and (e) definitional questions where the answer is embedded in the question stem.
Affected question IDs and explanations
ba23ee93 — "What is the chemical formula for laughing gas?" The answer (N2O) is contained in the common name "laughing gas." This is a general chemistry question with no clinical relevance to PG anesthesiology. Bloom's Level 1, zero exam yield.
dab129b2 — "What is the pin index code for nitrous oxide?" Pin index codes (3,5 for N2O) are a memorization fact with no clinical reasoning component. This is the kind of trivia that appears in older question banks but has not appeared in recent NEET-PG or INI-CET papers. Low yield.
bf670bba — "Which of the following agents can be used for anaesthesia in dogs?" This is a veterinary anesthesia question. It has no relevance to Indian human medical PG examinations. The correct answer ("All of the above") is also a weak distractor format.
5633a862 — "Which stage of anesthesia is known as the stage of analgesia?" Stage 1 = analgesia is a first-year pharmacology fact. The answer is definitionally embedded in the question. Bloom's Level 1, no discriminatory value.
748c298e — "Which of the following barbiturates is used for induction of anesthesia?" The distractors are Naloxone, Naltrexone, and Phenobarbitone — none of which are induction agents. Any candidate who has opened a pharmacology textbook will answer this correctly. The item does not discriminate between prepared and unprepared PG candidates.
bad699f3 — "Which intravenous anesthetic agent has a high incidence of psychomimetic reactions during the early recovery period?" Ketamine's emergence reactions are one of the most commonly tested facts in anesthesiology. This item is correct and relevant, but at Bloom's Level 1 with no clinical context, it adds nothing beyond what 51e8c94d (hallucination as a side effect of ketamine) already covers in the same sample. Both are low-yield duplicates of each other.
bd954a0f — "Transient acute adrenal insufficiency is seen with which of the following drugs?" Etomidate's adrenal suppression is a high-yield fact, but the item is pure recall with no clinical framing. A vignette version (e.g., a septic patient given etomidate for RSI who subsequently has refractory hypotension) would be far more valuable.
9e1cf573 — "What is the maximum recommended dose of lignocaine with adrenaline in mg/kg?" Pure numerical recall (7 mg/kg). No clinical context. Bloom's Level 1.
92268e38 — "What is the MAC of halothane?" Pure numerical recall (0.75%). No clinical context. Bloom's Level 1.
713efeef — "100 grams of soda lime absorbs what percentage of CO2?" Obscure numerical trivia (25%) with no clinical application in PG exam context.
9d7c7d52 — "Veril's sign is seen in which condition?" An eponym-based recall question about a sign associated with diazepam administration. This is the kind of historical trivia that has not appeared in recent exam cycles and has no clinical decision-making value.
a3816fe0 — "Which of the following statements is true about ether as an anesthetic agent?" Ether is no longer used clinically in India. Questions about ether's properties (slow induction, flammability, etc.) are historical trivia with minimal current exam relevance.
d90f02f2 — "Levo-bupivacaine is administered by which of the following routes?" The correct answer is "Epidural," but levobupivacaine is also used for spinal, peripheral nerve blocks, and infiltration. The question is both low-yield and potentially misleading because the "All" option was not included and the correct answer is incomplete.
0bfbe853 — "What is the definition of conscious sedation?" A definitional question where the answer (ability to respond to verbal commands) is the definition itself. Bloom's Level 1 with no clinical application.
Recommended disposition: ba23ee93, dab129b2, bf670bba, 5633a862, 748c298e, 713efeef, 9d7c7d52 — Disable. These are trivia or veterinary items with no realistic exam relevance; strong gold-standard coverage exists for the underlying concepts where relevant. bad699f3, 51e8c94d — Disable one (keep whichever has better distractor quality; both test the same ketamine fact). bd954a0f, 9e1cf573, 92268e38, a3816fe0, 0bfbe853 — Fix by adding a clinical vignette frame that elevates the Bloom's level to at least 3; if the concept is already well-covered by a gold-standard item, prefer disable.
5. Repetitive or Duplicative Coverage
Why this pattern is bad
Duplicate or near-duplicate items within the same question bank inflate apparent coverage without adding discriminatory value. When two items test the same fact at the same Bloom's level with the same distractor structure, one of them is wasting a bank slot. In a 3,585-item bank, this is a significant efficiency problem. Repetition also means that candidates who encounter both items in a test plan get an unfair advantage on the second item.
How it shows up
In this sample, repetition appears in two forms: (a) near-identical stems testing the same single fact, and (b) conceptually overlapping items where the same clinical principle is tested twice at the same low Bloom's level.
Affected question IDs and explanations
bad699f3 and 51e8c94d — Both test ketamine's psychomimetic/hallucinatory side effects. bad699f3 asks which IV agent has a high incidence of psychomimetic reactions; 51e8c94d asks which agent causes hallucination as a major side effect. The stem wording differs but the concept, correct answer, and distractor set are functionally identical. One should be disabled.
5a20117a and db5af29b — Both test soda lime composition. 5a20117a is an "all except" format asking which component is NOT in soda lime (answer: Mg(OH)2). db5af29b asks for the full composition with specific percentages. These are testing the same factual domain at Bloom's Level 1. The percentage-based item (db5af29b) is marginally more demanding but both are low-yield. In a bank of 3,585 items, two soda lime composition questions at Bloom's 1 is excessive.
bd954a0f (etomidate/adrenal suppression) and 9840cf98 (etomidate safe in heart failure) — These are not duplicates but they both test etomidate properties at Bloom's Level 1–2 with no clinical context. Together with the benchmark item 2bdbe2e6 (vecuronium reversal), the bank appears to have multiple low-Bloom's etomidate items. The concept deserves one well-constructed clinical vignette, not multiple recall items.
dc6b97a1 (Allen test before ABG) and 6128d9ec (EtCO2 to differentiate endotracheal vs esophageal intubation) — These are not duplicates of each other, but both are simple recall items in the Respiratory and Airway Management topic that test single-fact knowledge. They are noted here because the topic appears to have a high density of Bloom's 1–2 items, and the content team should audit for further repetition at the topic level.
36462612 (Remifentanyl characteristics) — The marked correct answer is "More potent than Alfentanil." However, the other options — "Metabolized by plasma esterase," "Short half-life," and "Dose reduced in hepatic and renal disease" — are all actually TRUE statements about Remifentanyl (it IS metabolized by plasma esterases, it DOES have a short half-life, and it does NOT require dose reduction in hepatic/renal disease — making that option false). The question is asking which is a characteristic, but three of the four options are also characteristics. This is a poorly constructed item that overlaps conceptually with other opioid pharmacology items in the bank.
Recommended disposition: bad699f3 or 51e8c94d — Disable one (prefer to keep the one with better distractor quality and add a clinical frame to the survivor). 5a20117a — Disable (the percentage-based db5af029b is marginally better; keep that one but note it is still low-yield). 36462612 — Fix by restructuring the stem to ask which is NOT a characteristic, or rewrite entirely to avoid the multiple-true-options problem.
6. Worthwhile Concept, Weak Execution (Keep the Concept, Fix the Stem/Options/Vignette)
Why this pattern is bad
Some items in this sample test genuinely high-yield concepts that appear in Indian PG exams but are written in a way that undermines their discriminatory value. The most common execution failures are: (a) a clinically important concept stripped of all context and reduced to a one-line recall question, (b) distractor sets where the wrong answers are obviously wrong rather than plausibly tempting, (c) "NOT true" or "all except" formats applied to a concept that would be better served by a positive stem, and (d) stems that are ambiguous about what is being asked.
How it shows up
This is the second-largest category in the reviewed set. These items are worth fixing because the underlying concept has exam relevance; the problem is purely in the execution.
Affected question IDs and explanations
e48bdb80 — Maximum allowable blood loss calculation for a 65-year-old post-MI patient. This is a genuinely high-yield calculation question (MABL = EBV × [starting Hct − target Hct] / starting Hct). The stem provides weight (73 kg), height (6'2"), and PCV (35%). The concept is excellent and the Bloom's level (3) is appropriate. However, the answer options (600, 680, 700, 780 mL) are very close together, and the correct answer (780 mL) requires the candidate to know the correct EBV formula for a male patient of this height/weight. The stem should specify whether to use ideal body weight or actual body weight, and the clinical context (post-MI, elective surgery) should be used to ask a follow-up decision question rather than just a calculation. Fix: Add clarity on which body weight formula to use and consider adding a decision-making layer (e.g., "at what point would you transfuse?").
9083b6c5 — Perioperative management of losartan in a hypertensive patient. This is a high-yield preoperative pharmacology question. The correct answer (continue till day of operation) is the current consensus for ARBs, though some guidelines recommend withholding on the day of surgery to avoid refractory intraoperative hypotension. The stem is well-constructed but the correct answer is contested in current literature. Fix: Revise the stem to specify the specific guideline being referenced, or change the question to ask about ACE inhibitors (where the evidence is clearer) to avoid ambiguity.
e516e447 — Emergence delirium with ketamine: increased incidence in all except? The concept (risk factors for ketamine emergence reactions) is high-yield. The correct answer (dose less than 1 mg/kg) is appropriate — higher doses are associated with more emergence reactions. However, the "all except" format is weak here because it forces the candidate to evaluate four negative statements simultaneously. Fix: Rewrite as a positive stem: "Which of the following is a risk factor for emergence delirium with ketamine?" and restructure the options accordingly.
4c5eb7fe — CRPS Type 1 occurs following a nerve injury (false). The concept is correct — CRPS Type 1 (Reflex Sympathetic Dystrophy) occurs WITHOUT a defined nerve injury; Type 2 (Causalgia) occurs WITH nerve injury. This is a high-yield distinction. However, the "false" stem format combined with the option "Type 1 CRPS occurs following a nerve injury" is confusing because the candidate must hold a double negative in mind. Fix: Rewrite as: "Which of the following correctly distinguishes CRPS Type 1 from Type 2?" with options that present the distinction directly.
47bee7ad — Balanced resuscitation definition. The correct answer (permitting a degree of hypotension to minimize bleeding) is actually the definition of permissive hypotension or hypotensive resuscitation, not "balanced resuscitation." Balanced resuscitation typically refers to the use of blood products in a balanced ratio (e.g., 1:1:1 PRBC:FFP:platelets). The stem uses the wrong terminology for the concept being tested. Fix: Either change the stem to ask about "permissive hypotension" or change the correct answer to reflect the actual definition of balanced resuscitation.
04766b31 — Complications of stellate ganglion block: all except Mueller's syndrome. The concept (stellate ganglion block complications) is high-yield. However, "Mueller's syndrome" is not a recognized medical eponym in this context — the intended answer is likely Horner's syndrome (which is actually an expected effect of stellate ganglion block, not a complication). The question appears to be testing whether candidates know that Horner's syndrome is an expected outcome rather than a complication, but the use of "Mueller's syndrome" (a non-standard or misspelled eponym) makes the item factually unreliable. Fix: Replace "Mueller's syndrome" with "Horner's syndrome" and reframe the stem to ask which is an expected outcome rather than a complication, or restructure entirely.
dafd90db — Trilene + Sodalime respiratory complication. The concept is correct — Trichloroethylene (Trilene) reacts with soda lime to produce dichloroacetylene, which causes cranial nerve palsies and encephalopathy, not "airway irritation and inflammation." The marked correct answer is factually wrong. The correct answer should reference neurotoxicity (cranial nerve palsies, encephalopathy). This item belongs in Category 1 (Wrong Key) as well as here. Fix/Disable: Correct the key to neurotoxicity/cranial nerve palsies, or disable if the concept is already covered by a gold-standard item.
c2c12cf8 — Best initial approach for difficult intubation history. The correct answer (video laryngoscope) is reasonable but not universally agreed upon — awake fiberoptic intubation is the gold standard for a known difficult airway in many guidelines. The stem says "best initial approach" which is ambiguous. Fix: Add clinical context (e.g., patient is cooperative, Mallampati IV, modified RSI planned) to make the video laryngoscope answer unambiguous, or change the scenario to one where video laryngoscopy is clearly the first-line choice.
9ba5105f — Spinal anesthesia drug deposited between which layers? The correct answer (pia mater and arachnoid mater) is the subarachnoid space. This is correct. However, the item is Bloom's Level 1 anatomy recall. Fix: Elevate to Bloom's Level 3 by adding a clinical scenario (e.g., "During spinal anesthesia, the anesthesiologist notes free flow of CSF — which space has been entered?") or disable if the concept is covered by a better item.
13bdc053 — "Which of the following is NOT commonly used as an induction agent?" (Midazolam = correct). Midazolam IS used as an induction agent in some contexts (e.g., in combination, in ICU, in pediatrics). The "NOT commonly used" qualifier makes this ambiguous. Fix: Change to "Which of the following is NOT a primary induction agent?" or specify the clinical context more precisely.
Recommended disposition: e48bdb80, 9083b6c5, e516e447, 4c5eb7fe — Fix (high-yield concepts, execution problems are correctable). 47bee7ad — Fix (terminology error, concept is salvageable). 04766b31 — Fix (replace Mueller's syndrome with correct eponym and restructure). dafd90db — Fix key (change correct answer to neurotoxicity/cranial nerve palsies) — this also has a wrong-key component. c2c12cf8 — Fix (add clinical context to disambiguate). 9ba5105f, 13bdc053 — Fix or disable depending on whether a better-constructed item for the same concept exists in the gold set.
Prioritization
The table below ranks action items by urgency and impact.
| Priority | Action | Rationale | Affected IDs (representative) |
|---|---|---|---|
| P0 — Immediate | Disable wrong-key items | Active harm to candidates in live tests | 21fb3af2, 77aa304d, f222f1f8, 5766242b, dafd90db |
| P0 — Immediate | Disable broken image items currently in live plans | Undeliverable items already in daily plans | f4cfbaf8, c3fa9ede, b8b8dd08, 3fc438d2, c3a4e740 |
| P1 — High | Fix malformed option sets | Broken delivery, some items in live plans | d77f5c2f, bbd838a9, e1b816f4, cab5cd9d |
| P1 — High | Disable dental/wrong-subject items | Subject contamination, misleads candidates | ed8d0f17, 1d9ebb2c |
| P1 — High | Reassign topic-misclassified items | Corrupts topic analytics | 8b7bbc8d, c41a1a03, 9d7c7d52 |
| P2 — Medium | Disable trivia/zero-yield items | Wastes bank slots, no exam relevance | ba23ee93, dab129b2, bf670bba, 5633a862, 748c298e, 713efeef |
| P2 — Medium | Fix worthwhile-concept weak-execution items | High-yield concepts salvageable with rewrite | e48bdb80, e516e447, 4c5eb7fe, 47bee7ad, 04766b31 |
| P3 — Lower | Disable duplicates | Bank efficiency | bad699f3 or 51e8c94d, 5a20117a |
| P3 — Lower | Elevate Bloom's level on surviving low-Bloom's items | Improve discriminatory value | bd954a0f, 9e1cf573, 92268e38, 0bfbe853 |
Structural note for the content team: The Bloom's distribution in this sample (33% at Level 1, 58% at Level 2, 9% at Levels 3–4) is severely bottom-heavy compared to the gold standard. Even after all the above fixes and disables, the surviving items will still be predominantly Bloom's 1–2. The content team should treat this as a net-new authoring priority: the subject needs a significant injection of Bloom's Level 3–5 clinical vignette items, particularly in the topics of Critical Care Medicine, Complications in Anesthesia, and Neuroanesthesia, where the gold-standard PYQs demonstrate that exam setters consistently operate at Levels 3–5.
Example Keep / Fix / Disable Calls
The following are concrete, actionable disposition calls for individual items reviewed in this sample.
KEEP
a5fac942 — Most suitable agent for intravenous regional anesthesia (Lidocaine). Clean stem, correct key, plausible distractors (Bupivacaine is a genuine wrong-answer trap because it is contraindicated for IVRA due to cardiotoxicity), tagged as FMGE PYQ. Bloom's Level 3. Keep as-is.
1fc84f73 — Post-spinal headache duration (up to 2 weeks). Correct, clinically relevant, appropriate difficulty. The distractor "up to 4 days" is a plausible wrong answer. Keep as-is.
80784521 — Volatile anesthetic causing most renal toxicity (Methoxyflurane). Correct, high-yield, good distractor set (Sevoflurane is a plausible wrong answer due to Compound A concerns). Keep as-is.
f6e71a35 — Nerve fibers blocked first during spinal anesthesia (Autonomic). Correct, clinically relevant, good distractor set. Keep as-is.
FIX
e48bdb80 — MABL calculation. High-yield concept, correct answer, but stem needs to specify body weight formula and could benefit from a clinical decision-making layer. Fix: add clarity on EBV calculation method and consider adding a transfusion trigger decision point.
04766b31 — Stellate ganglion block complications. High-yield topic, but "Mueller's syndrome" is a non-standard or misspelled eponym. Fix: replace with "Horner's syndrome" and reframe as an expected outcome vs. complication question.
e516e447 — Ketamine emergence delirium risk factors. High-yield concept, correct key, but "all except" format is weak. Fix: rewrite as a positive stem asking for a risk factor.
47bee7ad — Balanced resuscitation definition. The concept being tested (permissive hypotension) is high-yield but the terminology in the stem is wrong. Fix: change stem to "permissive hypotension" or correct the answer to match the actual definition of balanced resuscitation.
bbd838a9 — Jorgenson technique drugs. Correct concept, but five options and a misspelled distractor. Fix: reduce to four options, remove the redundant "Mepiridine" option, correct spelling.
DISABLE
21fb3af2 — NMB agent most associated with bradycardia (marked: Atracurium). Wrong key. Atracurium causes histamine release and tachycardia, not bradycardia. Disable immediately.
77aa304d — Agent known for smooth induction (marked: Isoflurane). Wrong key. Isoflurane is pungent and causes airway irritation; Sevoflurane is the correct answer. Disable immediately.
f222f1f8 — Primary purpose of throat pack (marked: facilitate easier extubation). Wrong key. The correct answer (prevent aspiration) is listed as a distractor. Disable immediately.
5766242b — Mallampati Class 3 characteristics (marked: hard and soft palate visible). Wrong key. This describes Class 2. Disable immediately.
b8b8dd08 — Identify type of endotracheal tube shown. No image present. Undeliverable. Disable until image is confirmed.
f4cfbaf8 — Capnography interpretation. No waveform image present. Currently in multiple live daily plans. Disable immediately.
ed8d0f17 — Gowgates block requires extraoral landmarks. Dental anesthesia question with no relevance to Anesthesiology PG exam. Disable from this subject; route to Dental Surgery bank.
bf670bba — Agents used for anaesthesia in dogs. Veterinary question. No relevance to Indian human medical PG. Disable.
ba23ee93 — Chemical formula of laughing gas. General chemistry trivia. No exam relevance. Disable.
dab129b2 — Pin index code for nitrous oxide. Memorization trivia with no clinical application. Disable.
d77f5c2f — Local anesthetic action and pH. Both Option A and Option B are identical text. Broken option set. Disable.