Summary

Sam selected option B: build a conflict report comparing Follow_up_Treatment_from_Intake__c against later treatment sources before using that field in automation.

I generated a live read-only report from Litify using open matters with intake follow-up text, later Treatment_Update__c rows, medical Requests-tab rows, and current Account provider names.

Results

• Reviewed open matters with intake follow-up text: 1,037
• Confirmed by later source: 452
• Conflict: 46
• Intake-only provider clue: 21
• Intake-only low confidence: 4
• No provider clue: 514

Important QA correction

The first script pass was too noisy because it treated hospitals, EMS, and imaging request facilities as competing later providers. I corrected the matcher before finalizing the report. The final conflict classification uses stronger follow-up-provider candidates and does not treat ER/EMS/diagnostic-only later sources as automatic conflicts.

Files

• Full report: /Users/samaguiar/Documents/Projects/admin/session_logs/litify-intake-followup-conflict-report-2026-05-28/intake-followup-conflict-report.csv
• Conflicts only: /Users/samaguiar/Documents/Projects/admin/session_logs/litify-intake-followup-conflict-report-2026-05-28/intake-followup-conflicts-only.csv
• Summary: /Users/samaguiar/Documents/Projects/admin/session_logs/litify-intake-followup-conflict-report-2026-05-28/intake-followup-conflict-summary.json
• README: /Users/samaguiar/Documents/Projects/admin/session_logs/litify-intake-followup-conflict-report-2026-05-28/README.md
• Script: /Users/samaguiar/Documents/Projects/admin/scripts/litify_intake_followup_conflict_report.py

Handoff

The report is complete and read-only. No Litify records or configuration were changed. A future agent can use the conflicts-only CSV for manual review before building treatment-roadmap automation. The 452 confirmed rows can support confidence, but later sources should remain authoritative. The 21 intake-only provider clues should not become automated provider roadmap rows without review.