# Data audit (2026-10-30)
| file | sha256 | wc -l | notes |
|---|---|---|---|
| data/assertions.json | 2965ab3acc6f326653169b83a5345698b51fac88a556de65faf89210aea0cad4 | 6 items (JSON list) | 6 assertions, fields id/cites/text, no duplicates, ids A1–A6; A1–A5 cite arxiv-2609.27732v1, A6 cites arxiv-2606.14621v1 |
| data/README.md | 5348bd49912099735108dbbe19ae09fa72bd600cb3d12b878658e6b9806198a7 | 14 | provenance |
| data/sources/arxiv-2606.14621v1.txt | 27c463c0a144d9d5a3916dc641f15b6029f6c68b5e574ad9074e7df1aeb0a238 | 58 | matches README hash |
| data/sources/arxiv-2609.27732v1.txt | ee2d444d2f78f2af2ca2cf69b59a61fcd7ac5e10ea29e8ddcd67917bf07d6a8b | 134 | matches README hash |
assertions.json sha256: 2965ab3acc6f326653169b83a5345698b51fac88a556de65faf89210aea0cad4 (matches bundle.json).

Not tabular data: "rows" = assertions; no missing values; no exclusions (all 6 assessed).

## Anomalies found in sources
1. DCR Table A19 CS-DR column prints SE (0.218) for the same estimate -0.431 that Table 6 prints with SE (0.210).
2. DCR Table A19 note says 640 never-treated comparison regions; Table 6/A19 body say 608.
3. DCR abstract says fatal outcomes "fall by 35%"; Table 7 5 km fatal ATT -0.354 converts to -29.8% (Section 6.1 says ~30%). Abstract appears to read log points as percent.
4. DCR text says HIV estimates "close to zero" although 5 km HIV ATT is -0.340 (insignificant).
5. RESEA text gives CIs for reductions as positive magnitudes ([1.75, 2.24]) while Table 2 prints [-2.24, -1.75]; "p ¡ 0.001" rendering defect.
None violates the availability rules; all inputs admissible. Handling: decide verdicts on the table cells; record issues in data_issues.