{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreibamzqi3e7f54remyzklb7aoeyzbopixo443bpjq3cwmqo3ugfs6y",
    "uri": "at://did:plc:wwyqal4cnqhuwyacdj7rqq3n/app.bsky.feed.post/3mm7raaknld72"
  },
  "path": "/t/collider-in-rct-subgroup-analysis/28689#post_20",
  "publishedAt": "2026-05-19T14:34:49.000Z",
  "site": "https://discourse.datamethods.org",
  "textContent": "I don’t follow. Do you mean b to be a subgroup variable acting as an effect modifier or\na subgroup variable being post-randomization or treatment-associated?\n\ns_doi:\n\n> with RECOVERY but the authors went with b) and all I am saying is that b) just represents overfitting because there is no clear indication in my mind that ventilation status represents a variable associated with dexamethasone in a post-randomization sense",
  "title": "Collider in RCT Subgroup Analysis"
}