v1.3.0 Update to TreeScan v2.4.1

• Added compatibility with new TreeScan v2.4.1. Parameter files updated to reflect this update
• Adds 2027 tree file. Several new ICD-10 codes will not be evaluated until 90 days of existence starting Oct. 1
• The do not evaluate nodes file is now a column within the new tree file
• Odds ratio and excess case calculations fixed to replace previous relative risk column, now includes day of week effect for ICD-10 codes impacted by this
• Now allows a new cadence of running TreeScan with updated recurrence intervals, e.g. 7 days vs daily
• Running script 5 TS analysis much faster

Update Scripts 6 & 8

• bug fix script 6
• update script 8

Script 7.1 & 8 Bug Fixes

• Script 8 has been updated to fully remove duplicates from the unique signals file and is much more streamlined. It has less tabs so just the one main one.
• Script 8 now allows you to set final_date <- the date you want to fix, then run 7.1_create_signal_report_standalone.R with that final_date and it will give you the original correct spreadsheet
• Script 7.1 bug fix of sigs_maxout <- character()

New Signal Review/Action Column

1. Added 8_WC_signal_review.R, a new script that:

• Processes the results and signal report files.
• Generates a deduplicated list of all new World Cup signals identified to date.
• Creates an output spreadsheet with a dedicated column for analysts to record signal classifications and review outcomes. Goal is to keep notes as you go for a final summary report of how TreeScan was used and impacted WC PH response.

2. Updated run_full_pipeline.R to automatically execute 8_WC_signal_review.R as part of the standard workflow.

• The signal review spreadsheet is now regenerated and updated automatically whenever new signals requiring investigation are identified during daily analyses.
• Based on original script by Alison Levin-Rector (NYC DOHMH)

Improve safety & TreeScan 2.4 usage

• Make script 2 download safer
• Make script 6 subregion use safer
• Tidy up percent vs proportion in script 7
• Correctly get Monte Carlo simulations for TS 2.4 in script 7

v1.1.0 Update to TreeScan 2.4

• Update tree file, do not evaluate, and common cause data files with version 2 - name remains the same
• Remove old tree scan files from 2025 and pre-print
• Signal interpretation files now save with date being the final_date set in run_full_pipeline
• Update prm file for TreeScan 2.4
• Fix 6 so maps are correct geometry and also have hospital + patient zip plots
• Separate Out Urgent Care Facilities for 6.1
• Fix lag changing in 2.2
• 6 now pulls in data for the signals that are phrases instead of ICD10 codes (e.g., Other Enterovirus)

v1.0.1

Bug fixes only:

• Maximum outlier problem
• Fix missing base_rows issue #17
• Require "tigris" as packages installed addressing issue #18
• Separate out urgent care facilities, resolving issue #19

v1.0.0

TreeScan™ v2.3 for use by public health officials across the US primarily with ESSENCE-based ED data on ICD-10 codes to detect deviations from a baseline 90 day period. Goal is to strengthen capacity to detect health issues not captured by syndromic surveillance for the US 2026 World Cup. Initial analytic pipeline including R code and supporting files from the NYC Health Dept.'s pre-print (Greene SK, Levin-Rector A, Kulldorff M, and Lall R). The authors' advice has allowed for the code expansion by Nyall Jamieson within epiENGAGE.