# Design codebook: what each trial did, read from its Methods alone

Each item is a candidate trial's Methods: the Methods section of its open full text or preprint, or, when neither is open, its abstract up to where the results begin. There is no title, results, discussion, or table, since titles and tables often carry results. Code what the trial did. You never see its results, so nothing you code can depend on them; don't look them up. An abstract that has no labelled sections is given whole (`source` says `abstract`); code its design and ignore any results it states.

## Eligibility

A trial is **eligible** when its Methods show all of these:

1. **Randomized.** Participants were randomly allocated to arms.
2. **A ketamine arm.** At least one arm received ketamine or esketamine (racemic ketamine, S-ketamine, or arketamine), by any route.
3. **An inert control arm.** At least one arm received saline, a placebo, or no study drug, with otherwise the same anesthesia and care as a ketamine arm.
4. **Around a procedure or childbirth.** The drug was given before, during, or after surgery, another procedure under general anesthesia, regional anesthesia, or sedation, or labor or delivery. Electroconvulsive therapy doesn't count.
5. **Depression measured after it.** A depression scale or the presence of depression was assessed at least once after the procedure.

Otherwise give the first reason that applies: `not_randomized`, `no_ketamine_arm`, `no_inert_control`, `not_perioperative`, `no_depression_measure`, `ect`, `other` (explain in `note`). A trial whose ketamine arms differ only in timing, with no inert arm, is `no_inert_control`, and you also set `timing_comparison` to true and code its arms as below.

## Arms

List every arm as the Methods name it, each with:

- `role`: `ketamine` (ketamine or esketamine, alone or with what the control arm also gets), `control` (the inert control), or `other` (any arm that adds or substitutes another active drug, or isn't a randomized arm). Only `ketamine` and `control` arms are analysed.
- For `ketamine` arms:
  - `drug`: `ketamine`, `esketamine`, or `arketamine`.
  - `route`: `iv`, `epidural`, `intrathecal`, `intranasal`, `oral`, `im`, `nerve_block`, `other`.
  - `bolus_mg_per_kg`: the loading or single dose in mg/kg (convert a fixed dose in mg to mg/kg only if the Methods give body weight; otherwise null, and give the dose in `dose_note`).
  - `infusion_mg_per_kg_h` and `infusion_hours`: a continuous infusion's rate and how long it ran, if any (`infusion_hours` null if it ran "until the end of surgery" and the duration isn't given).
  - `pump`: true if the drug was also or only given by a postoperative patient-controlled or continuous analgesia pump, with `pump_total_mg` if stated.
  - `dose_note`: anything else about the dose, in a few words.
  - `start`: when the first dose was given: `before_induction`, `at_induction` (with or as part of the induction of general anesthesia), `after_induction`, `after_delivery` (cesarean, after the baby is delivered), `during_labor`, `end_of_surgery`, `after_emergence`, `postoperative`, `unclear`.
  - `end`: when the last dose ended: `before_end_of_surgery`, `end_of_surgery` (an infusion run until, or a dose given at, the end of surgery or skin closure), `after_emergence` (any dose after the participant woke, including pumps), `unclear`.

## Anesthesia and consciousness

- `anesthesia`: what the participants received for the procedure: `general` (loss of consciousness with airway management), `neuraxial` (spinal, epidural, or combined), `regional` (nerve block), `sedation` (sedation with spontaneous breathing, as for endoscopy), `general_and_regional`, `none`, or `unclear`.
- `conscious_dosing`, the key code, for the ketamine arms together:
  - `none`: every dose was given while the participant was unconscious under general anesthesia, after induction began and ending at or before the end of surgery.
  - `some`: any dose was given while the participant was conscious: under neuraxial or regional anesthesia without general anesthesia, under sedation, before induction, after waking, or by a postoperative pump.
  - `unclear`: the Methods don't say enough to tell.
- `late_dosing`: true if any dose under general anesthesia was given in the last 30 minutes of surgery, at its end, or by an infusion run until its end, so the drug may still have been acting when the participant woke; false if the last dose ended earlier; null if unclear. Only `conscious_dosing: none` trials need it.

## Population

- `setting`: `cesarean`, `vaginal_delivery`, or `surgery` (any other procedure); `procedure` names it in a few words ("breast cancer surgery", "elderly hip fracture surgery", "gastroscopy").
- `depression_selection`: `selected` if only participants with depression or depressive symptoms were enrolled (a diagnosis, or a score at or above a cutoff on a depression scale), `unselected` otherwise (including trials that excluded people with depression). `selection_criterion`: the criterion, in a few words.
- `age_group`: `adult`, `older` (enrollment restricted to ages 60 or 65 and over), `adolescent` (under 18), or `mixed`.

## Measures

- `depression_measures`: each depression measure the Methods name, as `{"measure": "EPDS", "code": "EPDS", "days": [3, 7, 42], "primary": true}`: `measure` as the Methods name it; `code` one of `MADRS`, `HAMD` (any version of the Hamilton Depression Rating Scale), `EPDS`, `PHQ9` (or PHQ-8, PHQ-2), `BDI` (any version), `HADS_D` (the depression subscale of the Hospital Anxiety and Depression Scale), `HADS_TOTAL` (its total score only), `CESD`, `GDS`, `SDS` (Zung), `POMS_D` (the depression subscale of the Profile of Mood States), or `OTHER`; `days` after the procedure, as planned (day 0 the day of surgery or delivery; "postoperative day 1" is 1; "6 weeks" is 42; "1 month" is 30); `primary` true if the Methods name it as the trial's primary outcome. Anxiety scales (HADS-A, GAD-7, SAS, STAI) aren't depression measures.
- `pain_measures`: each pain or opioid-consumption measure the same way, with `code` one of `rest` (pain at rest), `movement` (pain on movement or coughing), `opioid` (opioid consumption), or `other`.
- `psychotomimetic_monitoring`: true if the Methods say adverse events such as hallucinations, nightmares, dissociation, dizziness, or psychiatric symptoms were recorded.

## Conduct

- `blinding`: `participants` true if participants were blinded, `assessors` true if outcome assessors were; null when not said.
- `registration`: `{"registry": "ChiCTR", "id": "ChiCTR2200061234", "date": "2022-06-21"}` as the Methods state them (`date` null if not stated), or null if no registration is mentioned.
- `enrollment`: `{"start": "2022-07", "end": "2023-03"}` as stated (any precision), or nulls.
- `ethics_approval`: true if an ethics committee approval is stated.
- `planned_n`: the sample size the Methods planned, if stated.
- `country` and `centers` (`single` or `multi`).

## Output

One JSON object per trial, in the order given, as a JSON array written to the file named in your task:

```json
{
  "id": "pmid:12345678",
  "eligible": true,
  "exclusion_reason": null,
  "timing_comparison": false,
  "arms": [
    {"name": "Esketamine group", "role": "ketamine", "drug": "esketamine", "route": "iv", "bolus_mg_per_kg": 0.2, "infusion_mg_per_kg_h": null, "infusion_hours": null, "pump": false, "dose_note": "", "start": "after_induction", "end": "before_end_of_surgery"},
    {"name": "Control group", "role": "control"}
  ],
  "anesthesia": "general",
  "conscious_dosing": "none",
  "late_dosing": false,
  "setting": "surgery",
  "procedure": "breast cancer surgery",
  "depression_selection": "unselected",
  "selection_criterion": "",
  "age_group": "adult",
  "depression_measures": [{"measure": "HAMD-17", "code": "HAMD", "days": [1, 3, 7], "primary": true}],
  "pain_measures": [{"measure": "NRS at rest", "code": "rest", "days": [1, 2], "primary": false}],
  "psychotomimetic_monitoring": true,
  "blinding": {"participants": true, "assessors": true},
  "registration": {"registry": "ChiCTR", "id": "ChiCTR2200061234", "date": null},
  "enrollment": {"start": "2022-07", "end": "2023-03"},
  "ethics_approval": true,
  "planned_n": 100,
  "country": "China",
  "centers": "single",
  "note": ""
}
```

`note`: anything a second reader would need to follow your codes, above all for `conscious_dosing` and `late_dosing`. Keep it short. Code from the text given alone. When it doesn't say something, code `unclear` or null rather than guessing.