[
 {
  "trial": "T093",
  "source": "library",
  "arms": {
   "Esketamine": {
    "randomized": "54",
    "analysed": "52"
   },
   "Placebo (20 mL normal saline)": {
    "randomized": "54",
    "analysed": "52"
   }
  },
  "depression": [
   {
    "measure": "MADRS (response = 50% reduction on POD 3; remission <=10)",
    "day": -1,
    "form": "median_iqr",
    "location": "Table 1 (MADRS score, median (IQR))",
    "values": {
     "Esketamine": {
      "median": "20.00",
      "q1": "17.00",
      "q3": "23.25",
      "n": "52"
     },
     "Placebo (20 mL normal saline)": {
      "median": "21.00",
      "q1": "17.00",
      "q3": "24.00",
      "n": "52"
     }
    }
   },
   {
    "measure": "MADRS (response = 50% reduction on POD 3; remission <=10)",
    "day": 1,
    "form": "difference",
    "definition": "odds ratio of MADRS response (≥ 50 % reduction from baseline), esketamine vs placebo; responders, not participants with depression (33 % vs 15 % responders)",
    "location": "Results 3.1, second paragraph (text, postoperative day 1; Fig. 2A, C)",
    "estimate": "2.7",
    "lower": "1.0",
    "upper": "6.9",
    "level": "0.95",
    "scale": "odds_ratio",
    "values": {
     "Esketamine": {
      "n": "52"
     },
     "Placebo (20 mL normal saline)": {
      "n": "52"
     }
    }
   },
   {
    "measure": "MADRS (response = 50% reduction on POD 3; remission <=10)",
    "day": 3,
    "form": "count",
    "definition": "MADRS response, ≥ 50 % reduction from baseline (events are responders, not participants with depression)",
    "location": "Table 2 (MADRS-based response at postoperative day 3); Results 3.1",
    "values": {
     "Esketamine": {
      "events": "14",
      "n": "52"
     },
     "Placebo (20 mL normal saline)": {
      "events": "7",
      "n": "52"
     }
    }
   }
  ],
  "pain": [
   {
    "measure": "NRS pain at 48 h (severe pain = any NRS >=7)",
    "day": 2,
    "form": "count",
    "definition": "severe pain, any NRS ≥ 7 (Methods: pain evaluated at 48 postoperative hours; Table 2 row 'Servere pain within 72 h'; text: no patient had a postoperative NRS ≥ 7)",
    "location": "Table 2 (Servere pain within 72 h); Results 3.1 (no patient in either group had a postoperative NRS ≥ 7)",
    "values": {
     "Esketamine": {
      "events": "0",
      "n": "52"
     },
     "Placebo (20 mL normal saline)": {
      "events": "0",
      "n": "52"
     }
    }
   }
  ],
  "adverse_events": [
   {
    "window": "postoperative",
    "values": {
     "Esketamine": {
      "dizziness": "21",
      "hallucination": "0",
      "visual": "1",
      "dissociation": "0"
     },
     "Placebo (20 mL normal saline)": {
      "dizziness": "16",
      "hallucination": "0",
      "visual": "0",
      "dissociation": "0"
     }
    }
   }
  ],
  "blinding_check": null,
  "enrollment": {
   "start": "2023-09-06",
   "end": "2024-07-08"
  },
  "registration": {
   "number": "ChiCTR2300071062",
   "date": "2023-05-04"
  },
  "note": "108 randomized (54 per arm); surgery was cancelled in 4, leaving 52 per arm in the modified intention-to-treat analysis (last observation carried forward). RESPONSE, NOT DEPRESSION: postoperative MADRS is reported only as response (≥ 50 % reduction from baseline), the opposite direction to a count of depressed participants. Day 3: Table 2 14 (27) vs 7 (13), OR 2.37 (0.87 to 6.47); text 27 % (14/52) vs 13 % (7/52), odds ratio 2.4 [0.9 to 6.6]; Table 2 footnote b says two patients were missing at day 3, but the counts are printed over 52. With missing data imputed: 13 (25) vs 7 (13), OR 2.14 (0.78 to 5.91), not coded. Day 1 is from the text: 'the esketamine group had a significantly higher proportion of responders (33 %) than the placebo group (15 %) (odds ratio 2.7, 95 % CI, 1.0 to 6.9)', esketamine versus placebo; counts are not printed, so it is entered as a difference, and its n is the 52 per arm of the modified intention-to-treat population that Fig. 2's caption says the response analyses used, not a number printed beside the estimate. Not coded because not tied to a day: GLMM group main effects for response OR 5.95 (0.49 to 71.71) and remission OR 1.45 (0.13 to 16.25), and MMRM group main effects for MADRS scores MD -0.08 (−1.83 to 1.67) and change from baseline MD -2.54 (−4.47 to −0.61). MADRS by visit (postoperative days 1 to 5, 4 and 12 weeks) is only in Fig. 3 and the supplement, without printed values. PHQ-9 is not in the design and is reported only at baseline (Table 1, median (IQR): 10.00 (8.00, 12.00) vs 10.00 (7.75, 12.00)); not entered. Severe pain: the Methods evaluate pain at 48 postoperative hours with severe pain as any NRS ≥ 7; Table 2's row is 'Servere pain within 72 h' (0 vs 0) and the text says no patient in either group had a postoperative NRS ≥ 7, which includes 48 h, so it is coded as day 2. NRS scores are only in eTable 6. Adverse events (Table 3, within 3 postoperative days): diplopia coded as visual and Dissociative (CADSS ≠ 0) as dissociation; Delirium (3D-CAM) 2 (4) vs 1 (2) not coded (daily postoperative delirium screening, not emergence agitation); manic (YMRS ≥ 5) and psychotic (BPRS) symptoms 0 vs 0 are not codebook categories.",
  "reason": "Four differences. Day 3 primary outcome: form 'count' (coder A), not 'change_count' (coder B), since it counts responders (≥ 50 % MADRS reduction), which the codebook's count form takes with a definition saying so; there is no change_count form; 14/52 vs 7/52 as printed. Day 1: kept coder A's difference entry (response OR 2.7, 1.0 to 6.9, Results 3.1), since the design lists MADRS at day 1 and the source prints only this estimate and percentages; added a definition marking it as response and the arm n of 52 (the mITT population Fig. 2's caption names). Baseline PHQ-9: the coders differed only in its name, but the design lists no PHQ-9 measure, so it is left out and its values are in the note. Severe pain: kept coder A's 0/52 vs 0/52 at day 2; the design lists it at day 2, the Methods define it at 48 h, and the text says no patient had any postoperative NRS ≥ 7."
 },
 {
  "trial": "T122",
  "source": "fulltext",
  "arms": {
   "Esketamine group": {
    "randomized": "70",
    "analysed": "69"
   },
   "Control group": {
    "randomized": "70",
    "analysed": "70"
   }
  },
  "depression": [
   {
    "measure": "PHQ-9 (>=10)",
    "day": -1,
    "form": "median_iqr",
    "location": "Table 1 (Preoperative PHQ-9 score)",
    "values": {
     "Esketamine group": {
      "median": "2",
      "q1": "0",
      "q3": "5",
      "n": "69"
     },
     "Control group": {
      "median": "3.5",
      "q1": "1",
      "q3": "6",
      "n": "70"
     }
    }
   },
   {
    "measure": "PHQ-9 (>=10)",
    "day": -1,
    "form": "count",
    "definition": "preoperative depressive symptom, PHQ-9 ≥ 10",
    "location": "Table 1 (Preoperative depressive symptom)",
    "values": {
     "Esketamine group": {
      "events": "7",
      "n": "69"
     },
     "Control group": {
      "events": "9",
      "n": "70"
     }
    }
   },
   {
    "measure": "PHQ-9 (>=10)",
    "day": 7,
    "form": "median_iqr",
    "location": "Table 4 (PHQ score, POD 7)",
    "values": {
     "Esketamine group": {
      "median": "3",
      "q1": "1",
      "q3": "6",
      "n": null
     },
     "Control group": {
      "median": "4",
      "q1": "1",
      "q3": "7",
      "n": null
     }
    }
   },
   {
    "measure": "PHQ-9 (>=10)",
    "day": 7,
    "form": "count",
    "definition": "postoperative depressive symptom, PHQ-9 ≥ 10",
    "location": "Table 4 (Postoperative depressive symptom, POD 7)",
    "values": {
     "Esketamine group": {
      "events": "9",
      "n": null
     },
     "Control group": {
      "events": "10",
      "n": null
     }
    }
   }
  ],
  "pain": [
   {
    "measure": "verbal NRS pain",
    "day": 1,
    "form": "median_iqr",
    "location": "Table 4 (NRS, 24 h)",
    "values": {
     "Esketamine group": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "69"
     },
     "Control group": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "70"
     }
    }
   },
   {
    "measure": "verbal NRS pain",
    "day": 2,
    "form": "median_iqr",
    "location": "Table 4 (NRS, 48 h)",
    "values": {
     "Esketamine group": {
      "median": "1",
      "q1": "1",
      "q3": "2",
      "n": "69"
     },
     "Control group": {
      "median": "1",
      "q1": "1",
      "q3": "2",
      "n": "70"
     }
    }
   },
   {
    "measure": "morphine consumption over 48 h",
    "day": 2,
    "form": "median_iqr",
    "location": "Table 4 (Morphine, 48 h (mg))",
    "values": {
     "Esketamine group": {
      "median": "6",
      "q1": "3",
      "q3": "8",
      "n": "69"
     },
     "Control group": {
      "median": "3.5",
      "q1": "1",
      "q3": "8",
      "n": "70"
     }
    }
   }
  ],
  "adverse_events": [
   {
    "window": "emergence",
    "values": {
     "Esketamine group": {
      "agitation": "4",
      "hallucination": "0",
      "nightmare": "0",
      "dissociation": "0",
      "visual": "0"
     },
     "Control group": {
      "agitation": "1",
      "hallucination": "0",
      "nightmare": "0",
      "dissociation": "0",
      "visual": "0"
     }
    }
   }
  ],
  "blinding_check": null,
  "enrollment": {
   "start": "2022-09",
   "end": "2024-09"
  },
  "registration": {
   "number": "ChiCTR 2200062726",
   "date": "2022-08-16"
  },
  "note": "Randomized per arm from Methods ('we enrolled 70 patients per group'); Results: 140 randomized, one esketamine patient lost to follow-up, 139 in the intention-to-treat analysis. PHQ-9 is reported both as scores (median, IQR) and as counts of PHQ-9 ≥ 10 at baseline and on POD 7; both are the design's one measure, 'PHQ-9 (>=10)', so each of those times has two entries under that name, one per form. Table 4's columns say n = 69 and 70, but the POD 7 percentages (9, 13.4%; 10, 14.5%) fit 67 and 69 assessed, so n is left null for the POD 7 entries; the other entries take the column n. Table 4 effects (esketamine vs control): PHQ score POD 7 median difference 0 (−1 to 1); depressive symptom POD 7 RR 0.92 (0.35 to 2.42); NRS 24 h 0 (0 to 0.5); NRS 48 h 0 (0 to 0); morphine 48 h 1 (0 to 3). Morphine over the first 48 h coded as day 2; NRS at 1 h and 2 h and morphine in PACU not coded. The preoperative PHQ-9 score was lower in the esketamine group (P = 0.036). Adverse events are for the PACU stay (Table 3), coded as window emergence: Delirium 4 (5.8) vs 1 (1.4) coded as agitation; the zeros for hallucination, nightmare, dissociation and diplopia come from the text ('No cases of hallucinations, nightmares, dissociation, diplopia, or low oxygen saturation were documented'); dizziness was planned but not reported. Subacute pain (POD 7) and chronic pain (3 months) are outside the coded days. Enrollment dates from Results.",
  "reason": "The coders differed only in the name of the PHQ-9 score entries: coder B named the median (IQR) entries 'PHQ-9 score', which is not a design name, to keep one entry per measure and day. The design names one depression measure, 'PHQ-9 (>=10)', and the source reports it at baseline and POD 7 both as scores and as counts of PHQ-9 ≥ 10; both forms are that measure, so they stay as two entries, one per form, under the design's name (coder A). All values checked against Tables 1 and 4; the POD 7 n stays null because the printed percentages fit 67 and 69 assessed, not the column n of 69 and 70."
 },
 {
  "trial": "T023",
  "source": "abstract",
  "arms": {
   "Group C": {
    "randomized": null,
    "analysed": "49"
   },
   "Group K": {
    "randomized": null,
    "analysed": "48"
   }
  },
  "depression": [],
  "pain": [],
  "adverse_events": [],
  "blinding_check": null,
  "enrollment": null,
  "registration": null,
  "note": "Abstract only. 100 randomized 1:1; per-arm randomized numbers not printed. 'analysed' is the number with the primary outcome (QoR-40 on POD1) available, 49 in group C and 48 in group K; no analysed numbers are given for HADS or NRS. HADS (2 days, 3 months), NRS (4, 24, 48 h) and chronic pain are reported only as no difference between groups (p > .15 for each), without numbers. A post hoc QoR-40 pain subsection score on POD1 ('pain scores were significantly higher in group K than in group C (29.8 ± 3.8 vs. 31.7 ± 2.1, 95% CI -1.81 (-3.00, -0.62), p = .003)'; a recovery subscale, higher is better, its arm order ambiguous as printed) is not a design pain measure and is not entered. No adverse events, dates or registration given.",
  "reason": "Coder B entered the post hoc QoR-40 pain subsection score on POD1 as a pain entry. It is a quality-of-recovery subscale (higher is better), not one of the design's pain measures (NRS at 4, 24 and 48 h; chronic pain at 3 months), and the codebook takes only the measures the design lists, so it is left out (coder A) and described in the note with its ambiguous arm order. The abstract prints no numbers for the design's HADS, NRS or chronic pain measures."
 },
 {
  "trial": "T060",
  "source": "fulltext",
  "arms": {
   "Morning operation with esketamine": {
    "randomized": "108",
    "analysed": "107"
   },
   "Morning operation with saline": {
    "randomized": "108",
    "analysed": "105"
   },
   "Afternoon operation with esketamine": {
    "randomized": "108",
    "analysed": "106"
   },
   "Afternoon operation with saline": {
    "randomized": "108",
    "analysed": "108"
   }
  },
  "depression": [
   {
    "measure": "Hospital Anxiety and Depression Scale (HAD)",
    "day": -1,
    "form": "median_iqr",
    "location": "Table 3 (Preoperation)",
    "values": {
     "Morning operation with esketamine": {
      "median": "4",
      "q1": "3",
      "q3": "6",
      "n": "108"
     },
     "Morning operation with saline": {
      "median": "5",
      "q1": "3",
      "q3": "7",
      "n": "108"
     },
     "Afternoon operation with esketamine": {
      "median": "4",
      "q1": "2",
      "q3": "6",
      "n": "108"
     },
     "Afternoon operation with saline": {
      "median": "4",
      "q1": "3",
      "q3": "6",
      "n": "108"
     }
    }
   },
   {
    "measure": "Hospital Anxiety and Depression Scale (HAD)",
    "day": 1,
    "form": "median_iqr",
    "location": "Table 3 (Postoperative day 1)",
    "values": {
     "Morning operation with esketamine": {
      "median": "8",
      "q1": "7",
      "q3": "10",
      "n": "107"
     },
     "Morning operation with saline": {
      "median": "8",
      "q1": "7",
      "q3": "10",
      "n": "105"
     },
     "Afternoon operation with esketamine": {
      "median": "8",
      "q1": "6",
      "q3": "10",
      "n": "106"
     },
     "Afternoon operation with saline": {
      "median": "8",
      "q1": "6",
      "q3": "10",
      "n": "108"
     }
    }
   },
   {
    "measure": "Hospital Anxiety and Depression Scale (HAD)",
    "day": 3,
    "form": "median_iqr",
    "location": "Table 3 (Postoperative day 3)",
    "values": {
     "Morning operation with esketamine": {
      "median": "8",
      "q1": "6",
      "q3": "10",
      "n": "107"
     },
     "Morning operation with saline": {
      "median": "8",
      "q1": "6",
      "q3": "9",
      "n": "105"
     },
     "Afternoon operation with esketamine": {
      "median": "7",
      "q1": "6",
      "q3": "9",
      "n": "106"
     },
     "Afternoon operation with saline": {
      "median": "8",
      "q1": "6",
      "q3": "10",
      "n": "108"
     }
    }
   },
   {
    "measure": "Hospital Anxiety and Depression Scale (HAD)",
    "day": 7,
    "form": "median_iqr",
    "location": "Table 3 (Postoperative day 7)",
    "values": {
     "Morning operation with esketamine": {
      "median": "7",
      "q1": "6",
      "q3": "9",
      "n": "107"
     },
     "Morning operation with saline": {
      "median": "7",
      "q1": "5",
      "q3": "9",
      "n": "105"
     },
     "Afternoon operation with esketamine": {
      "median": "7",
      "q1": "5",
      "q3": "9",
      "n": "106"
     },
     "Afternoon operation with saline": {
      "median": "7",
      "q1": "5",
      "q3": "9",
      "n": "108"
     }
    }
   }
  ],
  "pain": [
   {
    "measure": "postoperative NRS",
    "day": 1,
    "form": "median_iqr",
    "location": "Table 6 (Postoperative day 1)",
    "values": {
     "Morning operation with esketamine": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "107"
     },
     "Morning operation with saline": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "105"
     },
     "Afternoon operation with esketamine": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "106"
     },
     "Afternoon operation with saline": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "108"
     }
    }
   },
   {
    "measure": "postoperative NRS",
    "day": 2,
    "form": "median_iqr",
    "location": "Table 6 (Postoperative day 2)",
    "values": {
     "Morning operation with esketamine": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "107"
     },
     "Morning operation with saline": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "105"
     },
     "Afternoon operation with esketamine": {
      "median": "2",
      "q1": "1",
      "q3": "3",
      "n": "106"
     },
     "Afternoon operation with saline": {
      "median": "3",
      "q1": "1",
      "q3": "4",
      "n": "108"
     }
    }
   }
  ],
  "adverse_events": [
   {
    "window": "emergence",
    "location": "Table 9 (PACU rows)",
    "values": {
     "Morning operation with esketamine": {
      "agitation": "19",
      "dizziness": "33"
     },
     "Morning operation with saline": {
      "agitation": "31",
      "dizziness": "24"
     },
     "Afternoon operation with esketamine": {
      "agitation": "18",
      "dizziness": "21"
     },
     "Afternoon operation with saline": {
      "agitation": "28",
      "dizziness": "18"
     }
    }
   },
   {
    "window": "postoperative",
    "location": "Table 9 (Postoperation rows)",
    "values": {
     "Morning operation with esketamine": {
      "dizziness": "34"
     },
     "Morning operation with saline": {
      "dizziness": "46"
     },
     "Afternoon operation with esketamine": {
      "dizziness": "42"
     },
     "Afternoon operation with saline": {
      "dizziness": "40"
     }
    }
   }
  ],
  "blinding_check": null,
  "enrollment": {
   "start": "2022-01-04",
   "end": "2023-03-07"
  },
  "registration": null,
  "note": "2x2 factorial (operation time x esketamine 0.2 mg/kg single dose before induction). Tables abbreviate the arms: AES = Afternoon operation with esketamine, ANS = Afternoon operation with saline, MES = Morning operation with esketamine, MNS = Morning operation with saline. 432 randomized 1:1:1:1 (Results); the text prints no other per-arm count of those randomized (Fig. 1's values are not shown), so randomized is the n Table 3 prints for every arm at the preoperative visit, 108, which sums to the 432 randomized. Analysed 426: AES 106, ANS 108, MES 107, MNS 105 (Tables 1, 3, 6, 8, 9); patients whose operation was canceled were excluded. 'HAD' scores are copied as the paper reports them (a single HAD score, no subscales). The first postoperative follow-up came at least 20 h after morning and 14 h after afternoon operations (Methods), so the afternoon arms' 'postoperative day 1' may be under 24 h; coded by the day the source names. GEE estimates (esketamine vs saline: B = -0.054 for HAD, -0.010 for NRS) not entered. PACU rows of Table 9 coded as emergence (EA = emergence agitation coded as agitation); 'Postoperation' rows as postoperative; intraoperative rows (injection pain, cough, hypotension, hypertension, SpO2 < 95%) hold no psychotomimetic category; postoperative delirium (3D-CAM, the primary outcome) is not entered as an adverse event. Enrollment dates are from Methods ('enrolled between 4 January 2022 and 7 March 2023'). No trial registration number appears in the text (ethics approval WDRY2021-K124, 18 August 2021).",
  "reason": "The coders differed only on per-arm randomized numbers. The text gives 432 randomized 1:1:1:1 and no per-arm count except Table 3's preoperative row, which prints n = 108 for each of the four arms; these sum to the 432 randomized, and the later rows' n (106, 108, 107, 105) are the 426 analysed, so 108 is the number randomized to each arm as printed (coder B). Everything else, identical between the coders, was checked against Tables 3, 6 and 9 and kept."
 },
 {
  "trial": "T107",
  "source": "fulltext",
  "arms": {
   "Bolus group": {
    "randomized": "168",
    "analysed": "153"
   },
   "Infusion group": {
    "randomized": "168",
    "analysed": "156"
   },
   "Control group": {
    "randomized": "167",
    "analysed": "150"
   }
  },
  "depression": [
   {
    "measure": "EPDS",
    "day": -1,
    "form": "median_iqr",
    "location": "Table 1 (Baseline EPDS scores)",
    "values": {
     "Bolus group": {
      "median": "5",
      "q1": "2",
      "q3": "7",
      "n": "153"
     },
     "Infusion group": {
      "median": "5",
      "q1": "2",
      "q3": "7",
      "n": "156"
     },
     "Control group": {
      "median": "4.5",
      "q1": "1",
      "q3": "6",
      "n": "150"
     }
    }
   },
   {
    "measure": "EPDS",
    "day": 7,
    "form": "count",
    "definition": "positive PPD screen, EPDS 'more than 10' as printed in Methods (main rows probably EPDS >= 10; see note)",
    "location": "Table 2 (PPD, 1 wk)",
    "values": {
     "Bolus group": {
      "events": "16",
      "n": "147"
     },
     "Infusion group": {
      "events": "17",
      "n": "148"
     },
     "Control group": {
      "events": "22",
      "n": "145"
     }
    }
   },
   {
    "measure": "EPDS",
    "day": 7,
    "form": "median_iqr",
    "location": "Table 2 (EPDS scores, 1 wk)",
    "values": {
     "Bolus group": {
      "median": "3",
      "q1": "0",
      "q3": "6",
      "n": "147"
     },
     "Infusion group": {
      "median": "4",
      "q1": "1",
      "q3": "6",
      "n": "148"
     },
     "Control group": {
      "median": "3",
      "q1": "0",
      "q3": "6",
      "n": "145"
     }
    }
   },
   {
    "measure": "EPDS",
    "day": 28,
    "form": "count",
    "definition": "positive PPD screen, EPDS 'more than 10' as printed in Methods (main rows probably EPDS >= 10; see note)",
    "location": "Table 2 (PPD, 4 wk)",
    "values": {
     "Bolus group": {
      "events": "18",
      "n": "150"
     },
     "Infusion group": {
      "events": "15",
      "n": "147"
     },
     "Control group": {
      "events": "23",
      "n": "145"
     }
    }
   },
   {
    "measure": "EPDS",
    "day": 28,
    "form": "median_iqr",
    "location": "Table 2 (EPDS scores, 4 wk)",
    "values": {
     "Bolus group": {
      "median": "2",
      "q1": "0",
      "q3": "6",
      "n": "150"
     },
     "Infusion group": {
      "median": "2",
      "q1": "0",
      "q3": "7",
      "n": "147"
     },
     "Control group": {
      "median": "2",
      "q1": "0",
      "q3": "7",
      "n": "145"
     }
    }
   },
   {
    "measure": "EPDS",
    "day": 42,
    "form": "count",
    "definition": "positive PPD screen, EPDS 'more than 10' as printed in Methods (main rows probably EPDS >= 10; see note)",
    "location": "Table 2 (PPD, 6 wk)",
    "values": {
     "Bolus group": {
      "events": "14",
      "n": "153"
     },
     "Infusion group": {
      "events": "18",
      "n": "156"
     },
     "Control group": {
      "events": "29",
      "n": "150"
     }
    }
   },
   {
    "measure": "EPDS",
    "day": 42,
    "form": "median_iqr",
    "location": "Table 2 (EPDS scores, 6 wk)",
    "values": {
     "Bolus group": {
      "median": "2",
      "q1": "0",
      "q3": "5",
      "n": "153"
     },
     "Infusion group": {
      "median": "2",
      "q1": "0",
      "q3": "5.75",
      "n": "156"
     },
     "Control group": {
      "median": "3",
      "q1": "0",
      "q3": "7",
      "n": "150"
     }
    }
   }
  ],
  "pain": [
   {
    "measure": "NRS at rest",
    "day": 1,
    "form": "median_iqr",
    "location": "Results, Pain Control paragraph (postoperative day 1)",
    "values": {
     "Bolus group": {
      "median": "3",
      "q1": "2",
      "q3": "3",
      "n": null
     },
     "Infusion group": {
      "median": "3",
      "q1": "2",
      "q3": "3",
      "n": null
     },
     "Control group": {
      "median": "3",
      "q1": "2",
      "q3": "3",
      "n": null
     }
    }
   },
   {
    "measure": "NRS at rest",
    "day": 2,
    "form": "median_iqr",
    "location": "Results, Pain Control paragraph (day 2)",
    "values": {
     "Bolus group": {
      "median": "1",
      "q1": "0",
      "q3": "2",
      "n": null
     },
     "Infusion group": {
      "median": "1",
      "q1": "0",
      "q3": "2",
      "n": null
     },
     "Control group": {
      "median": "1",
      "q1": "0",
      "q3": "2",
      "n": null
     }
    }
   },
   {
    "measure": "NRS with movement (coughing or walking)",
    "day": 1,
    "form": "median_iqr",
    "location": "Results, Pain Control paragraph (postoperative day 1)",
    "values": {
     "Bolus group": {
      "median": "4",
      "q1": "4",
      "q3": "5",
      "n": null
     },
     "Infusion group": {
      "median": "4",
      "q1": "4",
      "q3": "5",
      "n": null
     },
     "Control group": {
      "median": "4",
      "q1": "3",
      "q3": "5",
      "n": null
     }
    }
   },
   {
    "measure": "NRS with movement (coughing or walking)",
    "day": 2,
    "form": "median_iqr",
    "location": "Results, Pain Control paragraph (day 2)",
    "values": {
     "Bolus group": {
      "median": "3",
      "q1": "3",
      "q3": "4",
      "n": null
     },
     "Infusion group": {
      "median": "3",
      "q1": "3",
      "q3": "4",
      "n": null
     },
     "Control group": {
      "median": "3",
      "q1": "3",
      "q3": "4",
      "n": null
     }
    }
   },
   {
    "measure": "PCIA usage (fentanyl)",
    "day": 2,
    "form": "mean_sd",
    "location": "Results, Pain Control (PCIA paragraph, 'total agent'); Methods (PCIA discontinued within 48 hours postoperatively)",
    "values": {
     "Bolus group": {
      "mean": "69.25",
      "sd": "12.04",
      "n": null
     },
     "Infusion group": {
      "mean": "68.73",
      "sd": "12.31",
      "n": null
     },
     "Control group": {
      "mean": "69.92",
      "sd": "12.75",
      "n": null
     }
    }
   }
  ],
  "adverse_events": [
   {
    "window": "intraoperative",
    "location": "Table 3 (During surgery)",
    "values": {
     "Bolus group": {
      "dizziness": "22",
      "nightmare": "6",
      "hallucination": "13",
      "any_ae": "70"
     },
     "Infusion group": {
      "dizziness": "47",
      "nightmare": "9",
      "hallucination": "5",
      "visual": "0",
      "any_ae": "102"
     },
     "Control group": {
      "dizziness": "5",
      "nightmare": "0",
      "hallucination": "0",
      "visual": "0",
      "any_ae": "33"
     }
    }
   },
   {
    "window": "postoperative",
    "location": "Table 3 (Within 6 weeks postpartum, In total)",
    "values": {
     "Bolus group": {
      "any_ae": "17"
     },
     "Infusion group": {
      "any_ae": "23"
     },
     "Control group": {
      "any_ae": "20"
     }
    }
   }
  ],
  "blinding_check": null,
  "enrollment": {
   "start": "2024-04",
   "end": "2025-04"
  },
  "registration": {
   "number": "ChiCTR2400082919",
   "date": null
  },
  "note": "Randomized 168 bolus, 168 infusion, 167 control; analysed (per protocol, completed 6 weeks) 153, 156, 150. PPD definition: Methods say 'EPDS scores more than 10', but the Statistical analysis calls >= 10 the usual cut-off, prenatal screening excluded EPDS >= 10, and the sensitivity rows at >= 11 differ from the main rows, so the main rows probably use >= 10. Sensitivity counts (Table 2; control, bolus, infusion), not entered: EPDS >= 11 at 1 wk 16/145, 10/147, 13/148; 4 wk 19/145, 15/150, 11/147; 6 wk 25/150, 8/153, 12/156; EPDS >= 12 at 1 wk 6/145, 8/147, 12/148; 4 wk 12/145, 12/150, 9/147; 6 wk 18/150, 7/153, 9/156. ITT worst-case results (17.85% vs 17.26%, infusion vs bolus) not entered. Table 2 prints no n for the EPDS median rows; their n is the PPD denominator at the same visit. Pain and PCIA values come from the text, which lists three values per time without naming the groups; read in the order control, bolus, infusion, the order of every table (the bolus-attempt percentages in the same section, 50.66% vs 54.90% vs 55.12%, fit denominators 150, 153, 156 in that order). The infusion value at rest on day 1 is printed '3 3 (2–3)'. Day-3 NRS left out (outside days 1-2). PCIA 'total agent' consumption (unit not stated; PCIA of 1 mg fentanyl in 100 mL) is entered at day 2: the PCIA was discontinued within 48 hours, so its total is the cumulative consumption at 48 h; its n is not printed. PCIA bolus frequencies 2 (0–6) vs 2 (0–4) vs 2 (0–5) (IQR or range not said) and bolus-attempt incidence not entered. Adverse events during surgery (asked at the end of surgery under spinal anaesthesia): 'Dreamscape' coded as nightmare; control and infusion print 0 for both diplopia and nystagmus, so their visual count is 0, while the bolus group prints diplopia 1 and nystagmus 1 as separate rows, so its visual count (1 or 2) is not entered. 'In total' rows coded as any_ae; each equals the sum of its rows, so it may count events rather than participants; the 6-week rows hold no psychotomimetic category. Enrollment dates are the trial period in Methods ('between April 2024 and April 2025').",
  "reason": "Three differences. Coder B added the sensitivity counts at EPDS >= 11 and >= 12 as extra count entries at each visit; the codebook's count is participants with depression by the source's definition, which is the main PPD rows, so the sensitivity rows are left out (coder A) and listed in the note. PCIA total consumption (69.92±12.75 vs 69.25±12.04 vs 68.73±12.31, control, bolus, infusion): the Methods say the PCIA was discontinued within 48 hours, so the total is the cumulative consumption at 48 h, entered at day 2 as mean_sd with n not printed (coder B). Visual events: control and infusion print 0 for both diplopia and nystagmus, so visual is 0 for them (coder A); the bolus group's 1 diplopia and 1 nystagmus may be one patient, so its count isn't printed and is left out."
 }
]