prognostic_association
One published observational association per answer: this exposure predicted that outcome, in people nobody randomised. Carries its own interval, the population it was measured in, its PMID and the authors' own sentence. NAME THE `exposure`. This world holds more than one observational corpus and ships no default, because a lookup naming no exposure would answer about nothing in particular. `masld_nafld` is the fatty-liver review and takes a `stratum`; `clinic_bp` and `nighttime_ambulatory_bp` are the blood-pressure records. EVERY FIGURE HERE IS CITED, NEVER MEASURED. This world read each of these out of somebody else's meta-analysis or cohort. It observed none of the populations and computed none of the numbers, and every answer says so in its `basis`. ONE RECORD PER ANSWER, and that is the refusal rather than a limitation. Where a corpus holds a severity-stratified estimate and an unstratified one, they share a comparator, so the stratified figure REPLACES the unstratified one for a staged population instead of multiplying it. A payload carrying both would hand you the multiplication, and their product is a hazard nobody measured. AND IT NEVER CROSSES CORPORA. Two exposures out of two reviews are two facts, not one measurement, and no arithmetic over them is a quantity anybody observed. A randomised effect is NOT here. Lowering a pressure is a different question from a pressure predicting an outcome, measured against a different comparator; that is `treatment_effect`, and the two tools are separate so that no single call can ask for one of each.
Use this form for new arguments. Follow concrete links for known values. Fetch-only readers may be unable to submit new URLs.
Exact input contract
{
"additionalProperties": false,
"properties": {
"exposure": {
"description": "Which observational exposure. `masld_nafld` is the fatty-liver review and requires `stratum`. `clinic_bp` and `nighttime_ambulatory_bp` are the blood-pressure prognostic records. A name not in this list is refused against the ones held, because an unheld name is a fact about this index and never about the literature.",
"enum": [
"masld_nafld",
"clinic_bp",
"nighttime_ambulatory_bp"
],
"type": "string"
},
"outcome": {
"description": "The outcome the association runs to. Which outcomes are held depends on the exposure, and an outcome this corpus does not hold for that exposure is refused by name against the ones it does.",
"enum": [
"cardiovascular_event",
"incident_type_2_diabetes",
"liver_specific_mortality",
"all_cause_mortality",
"incident_heart_failure"
],
"type": "string"
},
"stratum": {
"description": "Required for `masld_nafld` and meaningless for the others. `unstratified` is the whole cohort of the review at any severity; `higher_fibrosis` is the severity-stratified subset, in practice the biopsied one. This is this world's index into the corpus and not a phrase any paper printed.",
"enum": [
"unstratified",
"higher_fibrosis"
],
"type": "string"
}
},
"required": [
"exposure",
"outcome"
],
"type": "object"
}