# Cohort Flow Counts Source

`cohort_flow_counts.csv` is a frozen, non-patient-level aggregate cohort-construction output from the dev-side extraction stage. It is intentionally included in the clean submission package because Figure 1 requires intermediate cohort-construction counts that cannot be recomputed from the three final derived cohort CSV files alone.

The analysis R pipeline treats this file as an aggregate input source:

- `01_data_prep_table1.R` reads `cohort_flow_counts.csv` and verifies that `final_cohort` matches the locally supplied derived cohort CSV row counts.
- `05_figures.R` reads `cohort_flow_counts.csv` to render Figure 1.

This file contains only aggregate counts and does not contain patient-level raw data or patient-level derived data.