{
  "task_id": "F4-conf-02",
  "verdicts": {
    "A1": "supported",
    "A2": "not_supported",
    "A3": "not_supported",
    "A4": "not_supported",
    "A5": "not_assessable",
    "A6": "supported"
  },
  "locators": {
    "A1": "arxiv-2609.27732v1 | Table 6 | row \"ATT\" | column \"5 km\"",
    "A2": "arxiv-2609.27732v1 | Table 7 | row \"HIV\" | column \"1 km\"",
    "A3": "arxiv-2609.27732v1 | Table 6 | row \"% change\" | column \"3 km\"",
    "A4": "arxiv-2609.27732v1 | Table 7 | row \"Hepatitis\" | column \"5 km\"",
    "A5": "not reported in arxiv-2609.27732v1",
    "A6": "arxiv-2606.14621v1 | Table 2 | row \"Weeks on UI\" | column \"Average treatment effect estimate\""
  },
  "reasons": {
    "A1": "Table 6 prints the 5 km ATT as -0.431** with SE (0.210), ** denoting p<0.05, and a % change of -35.0% (100(exp(-0.431)-1) = -35.0%), matching every element of the assertion.",
    "A2": "Table 7 prints the 1 km HIV ATT as +0.063 (SE 0.801), positive and insignificant, so the claimed negative sign (-0.063) and the claimed fall in HIV hospitalisations are contradicted.",
    "A3": "Table 6 converts the 3 km ATT of -0.283 to a -24.7% change (100(exp(-0.283)-1) = -24.7%), not 28.3%, so the stated magnitude is contradicted (28.3% is the log-point coefficient read directly as a percentage).",
    "A4": "With ATT -0.406 and SE (0.250), the 95% CI is -0.406 +/- 1.96*0.250 = [-0.896, 0.084], which includes zero (the asserted [-0.66, -0.16] is only +/- 1 SE), and the estimate carries no significance star.",
    "A5": "The extract reports ATTs only by buffer (Table 6), outcome category (Table 7) and estimator (Table A19) and never by sex, so a men-versus-women comparison can be neither confirmed nor refuted.",
    "A6": "Table 2 gives the ATE of RESEA selection on weeks on UI as -1.99 weeks (95% CI [-2.24, -1.75], p<0.001) with a control mean of 17.3 weeks, and 1.99/17.3 = 11.5% as also stated in the results text."
  },
  "n_supported": 2,
  "n_not_supported": 3,
  "n_not_assessable": 1,
  "feasibility": "complete",
  "feasibility_reason": null,
  "data_issues": [
    {
      "issue": "arxiv-2609.27732v1: Table A19 CS-DR column prints SE (0.218) for the same ATT -0.431 for which Table 6 (5 km) prints SE (0.210).",
      "handling": "A1 judged against Table 6 (the table the assertion describes); significance at 5% holds under either SE (|b|/SE = 2.05 or 1.98 > 1.96), noted in report."
    },
    {
      "issue": "arxiv-2609.27732v1: Table A19 note says 640 never-treated comparison regions, whereas Tables 6, 7, A19 body and Figure 6 note say 608.",
      "handling": "Recorded; not decisive for any assertion."
    },
    {
      "issue": "arxiv-2609.27732v1: abstract says fatal hospitalisations fall by 35%, but Table 7 5 km fatal ATT -0.354 implies -29.8%, matching Section 6.1 ('approximately 30%').",
      "handling": "Recorded; no assertion concerns fatal outcomes; tables preferred over abstract text."
    },
    {
      "issue": "arxiv-2609.27732v1: text announces five diagnostic sub-categories, Figure 7 note says four, Table 7 note gives ICD codes for four.",
      "handling": "Recorded; not decisive for any assertion."
    },
    {
      "issue": "arxiv-2606.14621v1: rendering defect 'p ¡ 0.001' in the short-term weeks-on-UI sentence (should read p < 0.001).",
      "handling": "Read as p < 0.001 (consistent with Table 2); not decisive."
    },
    {
      "issue": "arxiv-2606.14621v1: results text prints the weeks-on-UI CI as reduction magnitudes [1.75, 2.24] while Tables 2 and 3 print signed [-2.24, -1.75].",
      "handling": "Treated as equivalent sign conventions; A6 judged against Table 2."
    }
  ],
  "n_input_rows": {
    "assertions.json": 6,
    "README.md": 14,
    "sources/arxiv-2609.27732v1.txt": 134,
    "sources/arxiv-2606.14621v1.txt": 58
  },
  "n_analysed": 6,
  "random_seed": null,
  "baseline_verdicts_number_match": {
    "A1": "all_numbers_found",
    "A2": "all_numbers_found",
    "A3": "some_number_missing",
    "A4": "some_number_missing",
    "A5": "all_numbers_found",
    "A6": "all_numbers_found"
  },
  "computation_details": {
    "A1": {
      "b": -0.431,
      "se": 0.21,
      "z": 2.0524,
      "pct": -35.01
    },
    "A2": {
      "b": 0.063,
      "se": 0.801
    },
    "A3": {
      "b": -0.283,
      "se": 0.279,
      "pct": -24.65,
      "printed_pct": "-24.7%"
    },
    "A4": {
      "b": -0.406,
      "se": 0.25,
      "ci_196": [
        -0.896,
        0.084
      ],
      "ci_200": [
        -0.906,
        0.094
      ],
      "crit_to_exclude_zero": 1.624
    },
    "A5": {
      "n_sex_mentions": 0
    },
    "A6": {
      "ate": -1.99,
      "ci": [
        -2.24,
        -1.75
      ],
      "control_mean": 17.3,
      "rel_pct": 11.5
    }
  },
  "conventions": {
    "pct_change": "100*(exp(b)-1)",
    "ci_crit": 1.96,
    "ci_crit_alt": 2.0
  }
}