treatment_effect
One published RANDOMISED effect per answer: being assigned this arm changed that outcome's rate in a trial. Carries its own interval, the trial population, its PMID and the authors' own sentence. SEPARATE FROM `prognostic_association` ON PURPOSE. A hazard ratio for a pressure predicting an outcome is not the effect of lowering that pressure: the two are measured against different comparators, adjustment is not randomisation, and no arithmetic over one of each is a quantity anybody measured. They are two tools rather than two arguments so that no single call can ask for one of each. `arm` RATHER THAN `intervention`, and the reason is in the corpus: half the records here are told apart by the POPULATION they were measured in, people with and without previous cardiovascular disease, rather than by a different intervention. EVERY FIGURE HERE IS CITED, NEVER MEASURED, and `crosses_no_effect` is recomputed from the bounds rather than copied: a record whose interval includes no effect ships precisely because it does, since a corpus carrying only a paper's significant outcomes would report the paper's best result as the paper's result.
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": {
"arm": {
"description": "Which trial arm, from this corpus's closed vocabulary. `no_prior_cvd` and `prior_cvd` are the two strata of the blood-pressure lowering trials, told apart by the POPULATION they were measured in rather than by a different intervention, which is why this argument names an arm and not a drug. `cardiac_rehabilitation` is the rehabilitation programme. A name outside this list is refused naming the list.",
"enum": [
"no_prior_cvd",
"prior_cvd",
"cardiac_rehabilitation"
],
"type": "string"
},
"outcome": {
"description": "The outcome the trial measured, from this corpus's closed vocabulary. A name outside this list is refused naming the list, and a name inside it that this corpus holds no record for is a separate refusal, because an absence somebody looked for is not an absence nobody did.",
"enum": [
"incident_heart_failure",
"major_cardiovascular_event",
"all_cause_hospitalisation",
"all_cause_mortality"
],
"type": "string"
}
},
"required": [
"arm",
"outcome"
],
"type": "object"
}