Surgery Question Quality Review
Executive Summary
This review covers a candidate sample of 100 validated non-gold Surgery questions (60 generic + 40 risky), benchmarked against 8 gold-standard benchmark items and 12 recent PYQs. The subject pool contains 11,629 questions total.
The most striking finding in this sample is the severe Bloom's level compression toward the bottom of the taxonomy. The candidate distribution shows 31 questions at Bloom's 1 and 60 at Bloom's 2, against a benchmark/PYQ set that consistently operates at Bloom's 3–4 for its highest-value items. Only 7 candidate questions reach Bloom's 3 and 2 reach Bloom's 4. This is not a minor calibration issue — it represents a structural quality deficit that will make Surgery tests feel like rote-recall exercises rather than clinical reasoning assessments.
Beyond the Bloom's problem, the reviewed set contains a meaningful cluster of wrong-subject placements (dental and obstetric content filed under Surgery topics), at least two factually unsafe or wrong-key items, a broken image-dependent question that cannot function without its image, and a large body of low-yield trivia that adds no discriminatory value for INICET/NEET-PG preparation.
Summary counts across the reviewed sample:
| Issue Category | Approximate Count |
|---|---|
| Wrong Key or Factually Unsafe | 4–5 |
| Wrong Subject or Wrong Topic Placement | 8–10 |
| Broken Delivery (image-dependent, no image) | 2 |
| Low-Value But Correct (trivia, rote recall, low yield) | 35–40 |
| Repetitive or Duplicative Coverage | 4–5 |
| Worthwhile Concept, Weak Execution | 10–12 |
The net recommendation is to disable approximately 40–45 questions outright, fix 10–12, and retain the remainder with minor tagging corrections. The subject needs a deliberate effort to commission Bloom's 3–4 clinical vignette items to replace the disabled volume.
What Good Looks Like
The benchmark and PYQ sets establish a clear quality bar that the candidate sample largely fails to meet. The following features define a strong Surgery question in this context:
Clinical anchoring. Every high-value item in the gold set embeds the question in a patient scenario. The Alvarado score question (4c08e4d1) gives age, sex, symptom migration, fever, nausea, tenderness, and WBC count — the candidate must apply the scoring rubric to a real presentation, not just recall that the score exists. The tension pneumothorax question (f9438ce4) provides BP, tracheal deviation, neck vein distension, and absent breath sounds — the diagnosis is stated and the candidate must select the correct immediate intervention, distinguishing needle thoracostomy from chest tube insertion. The periampullary carcinoma question (45880585) uses waxing-and-waning jaundice, alcohol history, and double-duct sign on CT to force differential reasoning.
Distractor quality. Gold-standard distractors are clinically plausible and represent genuine decision points. In the tension pneumothorax item, "chest tube insertion" is a wrong answer that a poorly prepared candidate will choose — it is the right procedure for the wrong timing. In the head injury/absent breath sounds item (e13711cc), "ICD insertion" is tempting but wrong because airway takes priority over chest decompression when GCS is 8. These distractors test whether the candidate understands the hierarchy of management, not just the name of a procedure.
Appropriate Bloom's level. The best gold items operate at Bloom's 3 (application) or 4 (analysis). The colorectal staging question (e9f2ab61) requires the candidate to map histopathological findings to TNM staging — a multi-step reasoning task. The Graves' disease preoperative question (a7db10b6) requires understanding why potassium iodide reduces vascularity, not just naming a drug.
Factual precision. The ABPI question (e9ce7e11) is a clean Bloom's 1 recall item, but it is acceptable because the threshold values are specific, the distractors are numerically close, and the concept (imminent gangrene at ABPI < 0.3) is directly exam-relevant and frequently tested. Even simple recall items in the gold set are high-yield.
Topic integrity. Every gold and PYQ item belongs unambiguously to Surgery. There is no dental, obstetric, or ENT content misfiled in the gold set.
The candidate sample should be measured against all of these criteria.
Main Issue Categories
1. Wrong Key or Factually Unsafe
Why this pattern is bad. A wrong key is the most serious quality failure in a question bank. It directly harms candidates who reason correctly and are penalized for it. It also erodes trust in the platform. Factually unsafe items — where the stated correct answer is contested, outdated, or contradicted by current standard references — carry the same risk even if the error is not absolute.
How it shows up. In this sample, the pattern appears as questions where the marked correct answer conflicts with current standard surgical textbooks or guidelines, or where the answer is defensible only under a narrow or outdated framing that is not signaled to the candidate.
Example question IDs and explanations:
39635255 (Hepatobiliary Surgery — "procedure of choice for elective removal of CBD stones"): The marked correct answer is "Open choledocholithotomy." This is factually unsafe by current standards. The procedure of choice for elective CBD stone removal in most patients is endoscopic sphincterotomy and stone extraction (ERCP), not open surgery. Open choledocholithotomy is reserved for failed endoscopy or specific anatomical situations. This answer would be marked wrong in any current INICET or NEET-PG key. Disable.
f79e19d6 (Colorectal Surgery — "dietary factors associated with increased risk of colorectal cancer"): The marked correct answer is "Low carbohydrate diet." This is factually wrong. The established dietary risk factor for colorectal cancer is a high-fat, low-fibre diet. A low-carbohydrate diet is not a recognized independent risk factor. The question appears to have been constructed with an inverted or confused key. Disable.
c42d012a (Gastrointestinal Surgery — "persistent vomiting causes which electrolyte disturbance"): The marked correct answer is "Hyponatremic hyperchloremia and paradoxical aciduria." Persistent vomiting causes hypochloremic (not hyperchloremic) hypokalemic metabolic alkalosis with paradoxical aciduria. The correct answer option in this question is actually option C ("Hypokalemic metabolic alkalosis and paradoxical aciduria"), not the marked option A. This is a wrong-key item. Disable (the concept is worth testing but the key is wrong and the option wording is internally inconsistent across all four choices).
951a3a4d (Pancreatic Surgery — "medical treatment of acute pancreatitis"): The marked correct answer is "Calcium." This is misleading at best and wrong at worst. Calcium is given to treat hypocalcemia as a complication of acute pancreatitis, not as a primary medical treatment of the condition itself. The question stem asks for "medical treatment of acute pancreatitis," which should point to supportive care (IV fluids, analgesia, NPO) or somatostatin analogues. Aprotinin (a protease inhibitor) was historically used and is a more defensible answer in an older framing, but the marked key "Calcium" is not a treatment of pancreatitis. Disable.
3e1a1b9f (Vascular Surgery — "all done for DVT except"): The marked correct answer is "Bed rest." This is factually unsafe. Current DVT management guidelines (and standard surgical teaching) recommend early ambulation, not prolonged bed rest. Bed rest is no longer considered standard care and is not an active treatment modality that would be listed alongside heparin and thrombolytics. The question appears to be based on outdated teaching. Disable.
Recommended disposition: All five items above should be disabled. None are salvageable by simple editing because the factual error is in the key itself.
2. Wrong Subject or Wrong Topic Placement
Why this pattern is bad. Subject contamination degrades the diagnostic value of subject-specific tests. When a candidate takes a Surgery mock test and encounters dental extraction technique questions or obstetric procedure questions, it signals poor curation and wastes testing time on content that belongs in a different subject's bank. It also inflates Surgery question counts artificially.
How it shows up. In this sample, the pattern appears in two forms: (a) questions about dental/oral surgical procedures filed under General Surgery or Head and Neck Surgery, and (b) questions about obstetric/gynaecological procedures filed under Urology.
Example question IDs and explanations:
9cd9a2b3 (filed under Urology — "NOT used for diagnosis of a parotid tumor"): Parotid tumor diagnosis belongs to Head and Neck Surgery or ENT, not Urology. The topic placement is wrong. Additionally, the question itself has a factual nuance issue — open surgical biopsy is indeed avoided for parotid tumors due to risk of facial nerve injury and tumour seeding, so the key is correct, but the subject tag is wrong. Fix topic tag (move to Head and Neck Surgery); the question itself is acceptable if correctly placed.
d89360b4 (filed under General Surgery Principles — "rubber band extraction in haemophiliac patient"): This is a dental extraction technique question. It belongs to Dentistry or Oral Surgery, not General Surgery. The content has no relevance to the Surgery PG curriculum for INICET/NEET-PG. Disable (wrong subject entirely).
2e836621 (filed under General Surgery Principles — "principal problem with tuberosity reduction"): Tuberosity reduction is a dental surgical procedure. This is Oral and Maxillofacial Surgery / Dentistry content, not General Surgery. Disable (wrong subject).
ee4e2e9e (filed under Head and Neck Surgery — "bleeding doesn't stop on applying pressure, intermittent pain, treatment"): The answer is "Pulpectomy" and the distractors include "DPC" (direct pulp capping) and "Pulpotomy." This is unambiguously dental/endodontic content. It does not belong in Surgery. The stem is also vague to the point of being uninterpretable without a dental context. Disable (wrong subject).
3b8bb0c2 (filed under Urology — "Tubectomy is performed on which part of the fallopian tube"): Tubectomy is a gynaecological/obstetric procedure. It belongs in Obstetrics & Gynaecology, not Urology. Disable (wrong subject).
f8ccbeaa (filed under Bariatric Surgery — "histologically, what type of bone is produced in distraction osteogenesis"): Distraction osteogenesis is an orthopaedic/maxillofacial surgery concept. It has no relationship to Bariatric Surgery. The topic placement is doubly wrong — wrong subtopic and arguably wrong subject. Fix topic tag if the question is to be retained in Surgery (move to Plastic and Reconstructive Surgery or Orthopaedics); however, given the low yield for INICET/NEET-PG Surgery, consider disabling.
441737dd (filed under Bariatric Surgery — "meconium ileus is associated with which condition"): Meconium ileus belongs to Pediatric Surgery, not Bariatric Surgery. The answer (fibrocystic disease of pancreas = cystic fibrosis) is correct, but the topic tag is wrong. Fix topic tag (move to Pediatric Surgery).
ee8b00fb (filed under Bariatric Surgery — "ulcerative colitis almost always involves which part of the colon"): Ulcerative colitis belongs to Colorectal Surgery or Gastrointestinal Surgery, not Bariatric Surgery. Fix topic tag.
928c3308 (filed under Trauma — "Champy's plates are used for fixation in which manner"): Champy's plates are used for mandibular fracture fixation — this is Oral and Maxillofacial Surgery content. While mandibular fractures can appear in trauma surgery curricula, the specific question about plate mechanics (monocortical vs bicortical screws) is dental/OMFS-level detail not tested in INICET/NEET-PG Surgery. Disable (wrong level of detail for this subject; borderline wrong subject).
Recommended disposition: Disable items that are unambiguously from a different subject (d89360b4, 2e836621, ee4e2e9e, 3b8bb0c2, 928c3308). Fix topic tags for items that are in Surgery but misfiled within it (9cd9a2b3, f8ccbeaa, 441737dd, ee8b00fb).
3. Broken Delivery (Missing Image, Malformed Options, Incomplete Stem)
Why this pattern is bad. A question that depends on an image to be answerable is completely non-functional when the image is absent. The candidate cannot reason toward any answer, and any response is random. This is a delivery failure that makes the question worse than useless — it actively wastes candidate time and generates noise in performance data.
How it shows up. In this sample, two questions reference an image in the stem but no image is present in the data. One additional question has a stem so vague that it is functionally incomplete without additional context.
Example question IDs and explanations:
d8da9c11 (General Surgery Principles — "Which scoring system is used to diagnose the condition suggested by the surgical procedure shown?"): The stem says "the surgical procedure shown" but no image is present. The answer is "Alvarado" (implying appendicectomy), but without the image, the question is unanswerable. A candidate who knows that Alvarado scores appendicitis and Ranson scores pancreatitis cannot determine which is being shown. The Mantrels option adds further confusion (Mantrels is essentially the same as Alvarado). Disable until image is confirmed attached and functional.
37d3fdfd (Preoperative and Postoperative Care — "What does the image show? (AIIMS Nov 2018)"): The stem is entirely image-dependent. The answer is "Pneumatic compression stockings." Without the image, this is a guessing exercise. The AIIMS 2018 tag suggests this is a legitimate PYQ, but it cannot function without its image. Disable until image is confirmed attached and functional. If the image can be restored, this is a keep — it is a legitimate PYQ at an appropriate difficulty level.
ee4e2e9e (Head and Neck Surgery — "When bleeding doesn't stop on applying pressure and the patient has intermittent pain too, what should be the treatment taken into consideration?"): Beyond the wrong-subject problem noted above, the stem is clinically incomplete. There is no anatomical location, no patient context, no indication of what structure is bleeding. The answer "Pulpectomy" only makes sense if the reader already knows this is a dental question about pulp exposure. The stem is functionally broken for any candidate who does not already know the answer. Disable (compound failure: wrong subject + broken stem).
Recommended disposition: Disable all three. The two image-dependent items (d8da9c11, 37d3fdfd) may be reinstated if images are confirmed present and functional, but should not be live in the bank in their current state.
4. Low-Value But Correct (Too Simple, Low-Yield, Trivia-Heavy, Weak Exam Relevance)
Why this pattern is bad. This is the largest single quality problem in the reviewed sample. 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 is unlikely to appear in INICET or NEET-PG, it consumes a slot in the question bank that should be occupied by a higher-quality item. The Bloom's distribution data confirms this is a systemic problem: 31 of 100 candidate questions are at Bloom's 1 and 60 are at Bloom's 2, meaning 91% of the sample requires only recall or basic comprehension.
The specific harm is that when these questions appear in mock tests, they artificially inflate candidate scores, reduce test discrimination, and give candidates false confidence. They also crowd out the harder items that would actually prepare candidates for the real exam.
How it shows up. The pattern appears as bare-fact recall questions with no clinical context, questions where the correct answer is immediately obvious from the stem, and questions testing eponymous trivia that has no management implication.
Example question IDs and explanations:
d1631e36 (Breast Surgery — "Most common type of breast carcinoma is"): Options are Lobular, Sarcoma, Ductal, Granuloma. Three of four options are not even plausible breast carcinoma types. Any candidate who has opened a surgery textbook knows the answer is ductal. Bloom's 1, no clinical reasoning required. Disable.
01ca29a7 (Breast Surgery — "What is the most common type of breast cancer?"): This is a near-duplicate of d1631e36 with slightly different option wording. Both ask the same fact at the same level. Disable (also addressed under Repetitive Coverage below).
992a0998 (Colorectal Surgery — "What is the most common site for colon cancer?"): Answer is rectosigmoid. This is a single-fact recall item with no clinical context. It is also complicated by the fact that ba367932 in the same sample asks "most common site for cancer of the large bowel" with the answer "Rectum" — these two questions give different answers to essentially the same question, which is a factual consistency problem as well as a duplication problem. Disable one; the other needs key review.
f96e43e0 (Colorectal Surgery — "Delorme's Procedure is used for which condition?"): Answer is rectal prolapse. This is a pure eponym-to-condition mapping question at Bloom's 1. The PYQ set already contains a better version of this concept (2ed4ab4d asks which procedure is NOT a perianal approach for rectal prolapse, requiring the candidate to know multiple procedures and distinguish abdominal from perineal approaches). Disable in favour of the richer PYQ version.
6d6b9a9c (Colorectal Surgery — "What is the surgical procedure to create an external opening for the discharge of colonic contents?"): Answer is colostomy. This is a definition question. The word "colostomy" is derived directly from the Greek roots in the question. No medical knowledge is required to answer this. Disable.
2181fb32 (Gastrointestinal Surgery — "What is the length of a standard proctoscope?"): Answer is 4 inches. This is instrument-dimension trivia. It has no clinical reasoning component, no management implication, and is not a concept tested in recent INICET or NEET-PG papers. Disable.
a29bf971 (Gastrointestinal Surgery — "Which surgical approach was first described by Orringer for esophageal carcinoma?"): Answer is transhiatal. This is eponym-to-technique trivia at Bloom's 1. The question tests who described a technique, not when or why to use it. Disable.
c5f661bb (Trauma — "What is the definition of splenosis?"): Answer is rupture of the spleen with distribution of its tissue on the peritoneum. This is a definition question. The concept of splenosis is low-yield for INICET/NEET-PG Surgery. Disable.
f21d7aa4 (Vascular Surgery — "What is the most common vessel affected in acute limb ischemia?"): Answer is superficial femoral artery. Bare-fact recall, no clinical context. Disable.
3b6c834e (Vascular Surgery — "Which vein is preferable for coronary bypass grafting?"): Answer is saphenous vein. This is a Bloom's 1 recall item. The radial artery distractor is actually used in CABG, making this question potentially misleading — the saphenous vein is the most commonly used conduit historically, but the radial artery is increasingly preferred in current practice. The question is both low-yield and potentially outdated. Disable.
d9faab63 (General Surgery Principles — "Which gastrointestinal secretion contains the highest amount of carbohydrate?"): Answer is pancreatic juice. This is basic physiology trivia with no surgical relevance. It belongs in Physiology, not Surgery. Disable.
debee966 (General Surgery Principles — "What type of wound is characterized by the scraping away of the skin?"): Answer is abrasion. This is a definition question that any layperson could answer. Disable.
a18fcb40 (Trauma — "In the context of triage, what color would be assigned to a moribund patient?"): Answer is black. This is a single-fact recall item. While triage is a legitimate Surgery topic, the question requires only memorization of a colour code with no clinical reasoning. Disable (the concept can be tested better in a scenario format).
6708db7e (Trauma — "In which of the following conditions is surgery not useful?"): Answer is cerebral edema. This is a low-complexity recall item. The concept is correct but the question adds no discriminatory value. Disable.
f0ecff65 (Head and Neck Surgery — "Which nerve lies in association with Wharton's duct?"): Answer is lingual nerve. Pure anatomy recall, Bloom's 1. Disable.
6dda874b (Oncological Surgery — "What is the most common histological type of thyroid carcinoma?"): Answer is papillary type. This is one of the most basic facts in surgical oncology. Every candidate knows this. Disable.
0644473c (Urology — "What is the most common cause of severe obstructive uropathy in children?"): Answer is posterior urethral valves. Correct and relevant, but bare-fact recall with no clinical context. Could be improved with a vignette. Disable in current form; concept worth retaining in a better item.
fe2f3346 (Urology — "Normal urine flow rate in healthy adults in urodynamic study"): Answer is 20 ml/sec. Instrument-parameter trivia. Disable.
6396ba7f (General Surgery Principles — "Most common site of a felon"): Answer is thumb. Bloom's 1 trivia. Disable.
c3ce7361 (Trauma — "Most common indication for tracheostomy"): Answer is prolonged mechanical ventilation. This is a legitimate fact but the question is bare recall with no clinical scenario. The concept is worth testing but the format is too simple. Disable in current form.
f0efd184 (Trauma — "Lucid interval is classically seen in"): Answer is extradural hematoma. This is a classic exam fact but the question is pure recall. The concept is high-yield but the execution is Bloom's 1. Disable in current form; replace with a clinical scenario.
Recommended disposition: Disable all items listed above. The concepts underlying some of these (posterior urethral valves, lucid interval, tracheostomy indications) are worth testing but should be replaced with clinical vignette versions operating at Bloom's 3.
5. Repetitive or Duplicative Coverage
Why this pattern is bad. Duplicate or near-duplicate questions waste bank capacity, create inconsistency when two versions give slightly different answers to the same question, and reduce the effective diversity of a test. In a bank of 11,629 questions, duplication is expected, but when duplicates appear within a 100-question sample, the density is high enough to be operationally significant.
How it shows up. In this sample, the pattern appears as multiple questions testing the same narrow fact with the same format, and as questions that are functionally identical to gold-standard PYQs already in the bank.
Example question IDs and explanations:
d1631e36 and 01ca29a7 (both ask "most common type of breast cancer/carcinoma"): These two questions test the same single fact (infiltrating/invasive ductal carcinoma) with nearly identical stems and overlapping options. Having both in the bank adds no value. Disable both (the concept is too simple to warrant even one item in its current form).
66c9f020 and 846a9499 (both ask about peau d'orange): 66c9f020 asks "Peau d'orange develops due to which of the following?" with the answer "Lymphatic obstruction." 846a9499 asks "Peau d'orange appearance is due to:" with the answer "Blockade of subdermal lymphatics." These are the same question with slightly different wording. Disable one (846a9499 is marginally more precise in its wording; disable 66c9f020).
992a0998 and ba367932 (both ask about most common site of colorectal cancer): 992a0998 answers "Rectosigmoid" and ba367932 answers "Rectum." These are not only duplicates but give inconsistent answers. This is a factual consistency problem as well as a duplication problem. The rectum is the single most common site; rectosigmoid is the combined most common region. Both questions are low-yield recall items. Disable both.
bac85cdf and 91b2fa28 (both about Wilms' tumour): bac85cdf asks which statement about Wilms' tumour is false (answer: spreads mostly by lymphatics — it spreads by haematogenous route). 91b2fa28 asks the commonest site of metastasis of Wilms' tumour (answer: lungs, consistent with haematogenous spread). These two questions test the same underlying concept (haematogenous spread of Wilms' tumour) from different angles. This is not a strict duplicate, but the conceptual overlap is high. Keep 91b2fa28 (more direct, correct key); fix bac85cdf if the false-statement format is desired, but verify the key carefully — the statement "spreads mostly by lymphatics" is indeed false, so the key appears correct.
f96e43e0 (Delorme's procedure for rectal prolapse) duplicates the concept already covered by PYQ 2ed4ab4d (which procedure is NOT a perianal approach for rectal prolapse). The PYQ version is superior. Disable f96e43e0.
Recommended disposition: Disable d1631e36, 01ca29a7, 66c9f020, 992a0998, ba367932, f96e43e0. Keep 846a9499 (if not disabled for low value), 91b2fa28 (if not disabled for low value), bac85cdf (after key verification).
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 exam-relevant and appropriately placed within Surgery, but the execution — stem construction, distractor quality, or clinical framing — falls short of the gold standard. Disabling these would waste a valid concept; fixing them is the right call. However, fixing should be conservative: only items where the concept is clearly high-yield and the fix is straightforward should be prioritised.
How it shows up. The pattern appears as correct-key questions where the stem is a bare fact-recall format that could be elevated to a clinical vignette, or where distractors are implausible and give away the answer.
Example question IDs and explanations:
7f06345a (General Surgery Principles — tracheostomy tube blockage): The stem describes a tracheostomised patient with sudden complete tube blockage and asks for the best next step. The answer is "Immediate removal of the tracheostomy tube." This is a legitimate clinical scenario at Bloom's 3 and the key is correct. However, the distractors are weak — "jet ventilation" and "suction with sodium bicarbonate" are not realistic clinical options that a prepared candidate would seriously consider. The question would be stronger with distractors like "deflate the cuff and attempt suction," "call for emergency bronchoscopy," or "attempt bag-mask ventilation." Fix distractors.
ead1980a (Preoperative and Postoperative Care — DVT diagnosis post-gastrectomy): This is a well-constructed clinical vignette (58-year-old woman, post-gastrectomy day 4, leg pain and swelling, DVT suspected, initial diagnostic test). The answer is venous duplex ultrasound, which is correct. The distractors include venography (gold standard but invasive and not initial), impedance plethysmography (outdated), and radio-labeled fibrinogen (research tool). This is a good question. The only weakness is that "venography" as a distractor might confuse candidates who know it is the gold standard — the stem should specify "initial non-invasive test" to remove ambiguity. Fix stem (add "non-invasive" qualifier); otherwise keep.
52cf8d88 (Gastrointestinal Surgery — acute diverticulitis management): The stem describes a 55-year-old man with first-attack acute diverticulitis, left lower abdominal pain, and a palpable tender mass. The question asks what management steps in the first 24 hours should include IV fluids. The answer is "Broad-spectrum antibiotic therapy." This is correct and the clinical scenario is appropriate. However, the stem construction is awkward — "What steps in his management during the first 24 hours after admission should include intravenous fluids?" reads as if IV fluids are a given and the question is asking what else should be added. The stem should be rewritten for clarity: "In addition to IV fluids and bowel rest, which of the following is the most appropriate initial management?" Fix stem wording.
47fb574d (Endocrine Surgery — thyroid carcinoma causing pain and breathing difficulty in multinodular goiter): The clinical scenario is appropriate and the answer (anaplastic carcinoma) is correct. However, the stem says "a patient with multinodular goiter developed pain and difficulty in breathing" — this conflates a pre-existing multinodular goiter with a new carcinoma developing within it, which is a reasonable clinical scenario for anaplastic transformation. The question would be stronger if it specified rapid enlargement and the timeframe, which are the key clinical clues for anaplastic carcinoma. Fix stem to add "rapid enlargement over 4 weeks" to make the clinical reasoning more explicit.
8ae6cd64 (Preoperative and Postoperative Care — complications of enteral nutrition): This is a UPSC-CMS PYQ using the "select the correct combination" format. The answer is "1, 2 and 3" (tube malposition, diarrhoea/constipation, predisposition to systemic sepsis). The key is defensible — electrolyte imbalance (option 4) is more commonly associated with parenteral nutrition than enteral nutrition, though it can occur with enteral feeding too. The question is borderline on the key for option 4. The format is appropriate for the exam level. Keep with a note to verify whether electrolyte imbalance should be excluded from enteral nutrition complications in the standard reference used.
877a594e (Trauma — rupture of the diaphragm): The answer is "Repair of a ruptured diaphragm is best approached via laparotomy." This is correct for acute traumatic diaphragmatic rupture. However, the distractors include "Laparoscopy is the standard approach for repair" — laparoscopy is actually used for delayed/elective repair of diaphragmatic hernias, making this distractor partially true in a different context. The question would benefit from specifying "acute traumatic rupture" in the stem to remove ambiguity. Fix stem.
20dcb01d (Hepatobiliary Surgery — gallbladder polyps): The answer is "Adenomyomatosis is typically less than 1 cm and pedunculated" (marked as false/incorrect). Adenomyomatosis is indeed typically sessile, not pedunculated — cholesterol polyps are pedunculated. The key appears correct. The question tests a clinically relevant distinction (polyp morphology and malignancy risk). The distractors are plausible. This is a reasonable question. Keep.
c21281dd (Pancreatic Surgery — treatment of annular pancreas): Answer is duodeno-duodenostomy. This is correct — the treatment bypasses the obstruction rather than dividing the pancreatic ring. The question is bare recall but the concept is high-yield and the distractors (duodenojejunostomy, gastrojejunostomy, gastro-duodenostomy) are plausible bypass procedures. A clinical vignette (neonate with bilious vomiting and double-bubble sign) would elevate this to Bloom's 3. Fix by adding a clinical scenario.
e2566f42 (Head and Neck Surgery — facial nerve palsy and vertigo after mastoid surgery): The answer is "Vertical segment." This is a legitimate Bloom's 4 question about surgical anatomy. The vertical (mastoid) segment of the facial nerve is the most commonly injured during mastoid surgery. The geniculate ganglion and tympanic segment are also at risk but less commonly injured in mastoidectomy specifically. The question is well-placed and the key appears correct. The distractors are anatomically appropriate. Keep.
7863ded7 (Pancreatic Surgery — "all of the following can be used to predict severe acute pancreatitis except"): The answer is "C-reactive protein < 100." This is correct — CRP > 150 mg/L (not < 100) indicates severe pancreatitis. The question is a legitimate recall item about severity scoring. However, the option text uses inconsistent formatting (line breaks within options) and the CRP threshold stated in the correct answer (< 100) differs from the threshold in the companion question 3d61db94 (which states CRP > 100 mg/dL as a wrong answer for severe pancreatitis). There is internal inconsistency in the CRP threshold across the two pancreatitis questions in this sample. Fix the CRP threshold to be consistent with the standard reference (CRP > 150 mg/L at 48 hours is the standard threshold for severe pancreatitis); also fix option formatting.
Recommended disposition: Fix 7f06345a, ead1980a, 52cf8d88, 47fb574d, 877a594e, c21281dd, 7863ded7. Keep 20dcb01d, e2566f42, 8ae6cd64 (with key verification). The fixes are targeted and achievable without full rewrites.
Prioritization
The following table ranks issue categories by urgency and operational impact for the content team.
| Priority | Issue Category | Action Required | Estimated Items Affected |
|---|---|---|---|
| P1 — Immediate | Wrong Key or Factually Unsafe | Disable before next test cycle | 4–5 |
| P1 — Immediate | Broken Delivery (image-dependent) | Disable until image confirmed | 2–3 |
| P2 — High | Wrong Subject or Wrong Topic Placement | Disable (wrong subject) or retag (wrong subtopic) | 8–10 |
| P2 — High | Low-Value But Correct (trivia/rote recall) | Disable; commission Bloom's 3–4 replacements | 35–40 |
| P3 — Medium | Repetitive or Duplicative Coverage | Disable weaker duplicate; keep stronger version | 4–6 |
| P3 — Medium | Worthwhile Concept, Weak Execution | Fix stem/distractors/vignette | 10–12 |
Structural gap requiring commissioning work. The Bloom's distribution in the candidate sample (91% at levels 1–2) versus the benchmark (majority at levels 3–4 for high-value items) represents a gap that cannot be closed by fixing existing questions alone. The content team should commission approximately 30–40 new Bloom's 3–4 clinical vignette items for Surgery to replace the disabled low-value items. Priority topics for new commissioning, based on gaps observed in this sample, include: hepatobiliary management decisions (ERCP vs surgery), colorectal cancer staging and management, trauma management hierarchies (ATLS principles), endocrine surgery (thyroid/parathyroid workup), and vascular surgery (acute limb ischaemia management).
Example Keep / Fix / Disable Calls
The following table provides a consolidated reference for the content operations team covering representative items from the reviewed sample.
| Question ID | Topic | Disposition | Reason |
|---|---|---|---|
| 39635255 | Hepatobiliary Surgery | DISABLE | Wrong key — open choledocholithotomy is not the current procedure of choice for CBD stones; ERCP is |
| f79e19d6 | Colorectal Surgery | DISABLE | Wrong key — low carbohydrate diet is not a risk factor for colorectal cancer; question appears to have an inverted key |
| c42d012a | Gastrointestinal Surgery | DISABLE | Wrong key — persistent vomiting causes hypochloremic (not hyperchloremic) alkalosis; marked answer is factually incorrect |
| 951a3a4d | Pancreatic Surgery | DISABLE | Wrong key — calcium is not a treatment of acute pancreatitis; it treats a complication (hypocalcaemia) |
| 3e1a1b9f | Vascular Surgery | DISABLE | Factually unsafe — bed rest is no longer standard DVT management; current guidelines recommend early ambulation |
| d89360b4 | General Surgery Principles | DISABLE | Wrong subject — dental extraction technique (rubber band/PDL necrosis); not Surgery curriculum |
| 2e836621 | General Surgery Principles | DISABLE | Wrong subject — tuberosity reduction is a dental procedure; not Surgery curriculum |
| ee4e2e9e | Head and Neck Surgery | DISABLE | Wrong subject + broken stem — pulpectomy/pulpotomy is dental endodontics; stem is uninterpretable without dental context |
| 3b8bb0c2 | Urology | DISABLE | Wrong subject — tubectomy is an obstetric/gynaecological procedure; belongs in O&G |
| d8da9c11 | General Surgery Principles | DISABLE | Broken delivery — stem references "surgical procedure shown" but no image is present |
| 37d3fdfd | Preoperative and Postoperative Care | DISABLE (pending image restoration) | Broken delivery — entirely image-dependent; if image is confirmed present, this is a legitimate PYQ worth keeping |
| d1631e36 | Breast Surgery | DISABLE | Low-value trivia — most common breast carcinoma type; Bloom's 1, no clinical reasoning, duplicate of 01ca29a7 |
| 01ca29a7 | Breast Surgery | DISABLE | Low-value trivia + duplicate of d1631e36 |
| 6d6b9a9c | Colorectal Surgery | DISABLE | Low-value trivia — definition of colostomy derivable from word roots |
| 2181fb32 | Gastrointestinal Surgery | DISABLE | Low-value trivia — proctoscope length; instrument-dimension fact with no clinical reasoning |
| a29bf971 | Gastrointestinal Surgery | DISABLE | Low-value trivia — eponym (Orringer) for transhiatal oesophagectomy; Bloom's 1 |
| f96e43e0 | Colorectal Surgery | DISABLE | Duplicate of PYQ 2ed4ab4d; inferior version of the same concept |
| 992a0998 | Colorectal Surgery | DISABLE | Duplicate + inconsistent key with ba367932; both low-value recall |
| ba367932 | Colorectal Surgery | DISABLE | Duplicate + inconsistent key with 992a0998 |
| debee966 | General Surgery Principles | DISABLE | Low-value trivia — definition of abrasion; layperson-level knowledge |
| 6dda874b | Oncological Surgery | DISABLE | Low-value trivia — most common thyroid carcinoma type; Bloom's 1 |
| d9faab63 | General Surgery Principles | DISABLE | Wrong subject level — basic physiology trivia; belongs in Physiology not Surgery |
| 9cd9a2b3 | Urology | FIX topic tag | Move to Head and Neck Surgery; question content (parotid tumour diagnosis) is correct and relevant |
| 441737dd | Bariatric Surgery | FIX topic tag | Move to Pediatric Surgery; meconium ileus/cystic fibrosis is correct content, wrong subtopic |
| ee8b00fb | Bariatric Surgery | FIX topic tag | Move to Colorectal Surgery; ulcerative colitis content is correct, wrong subtopic |
| 7f06345a | General Surgery Principles | FIX | Good clinical scenario (tracheostomy tube blockage); fix distractors to be more clinically plausible |
| ead1980a | Preoperative and Postoperative Care | FIX | Good vignette (post-op DVT); add "non-invasive" qualifier to stem to remove venography ambiguity |
| 52cf8d88 | Gastrointestinal Surgery | FIX | Good clinical scenario (acute diverticulitis); rewrite stem for grammatical clarity |
| 47fb574d | Endocrine Surgery | FIX | Good concept (anaplastic thyroid carcinoma); add "rapid enlargement" to stem to make clinical reasoning explicit |
| 877a594e | Trauma | FIX | Good concept (diaphragm rupture repair); specify "acute traumatic" in stem to remove laparoscopy ambiguity |
| c21281dd | Pancreatic Surgery | FIX | Good concept (annular pancreas treatment); add neonatal clinical vignette to elevate to Bloom's 3 |
| 7863ded7 | Pancreatic Surgery | FIX | Good concept (pancreatitis severity scoring); fix CRP threshold to standard reference value and fix option formatting |
| e2566f42 | Head and Neck Surgery | KEEP | Well-constructed Bloom's 4 question on facial nerve anatomy in mastoid surgery; correct key, plausible distractors |
| 20dcb01d | Hepatobiliary Surgery | KEEP | Correct key, clinically relevant (gallbladder polyp morphology), plausible distractors |
| 8ae6cd64 | Preoperative and Postoperative Care | KEEP (verify key) | Legitimate UPSC-CMS PYQ on enteral nutrition complications; verify whether electrolyte imbalance should be excluded |
| dba0c0d2 | Breast Surgery | KEEP | Good clinical vignette (22-year-old, mobile breast mass, fibroadenoma); Bloom's 3, appropriate distractors |
| 821de009 | Trauma | KEEP | Good clinical scenario (cannot intubate/ventilate after maxillofacial trauma); Bloom's 3, correct key (cricothyrotomy) |
| bac85cdf | Pediatric Surgery | KEEP (verify key) | Wilms' tumour spread — false statement format; key appears correct (haematogenous not lymphatic spread); verify against standard reference |
| 8acddcca | Hepatobiliary Surgery | KEEP | Calot's triangle anatomy — correct key (portal vein does not pass through), appropriate difficulty, PYQ-tagged |
| be89d35a | Trauma | KEEP | First-degree burns — correct key (blisters do not form in first-degree burns), appropriate difficulty |