General Health

[caitlink12]

Included ICES survey cycles to existing general health variables GEN_07 and GEN_10 to variable and variable_details.

[DougManuel]

Code review

Reviewed 2 variables (GEN_07, GEN_10) for PUMF and Master across 2001–2017-2018.

L6 integration test: cross-cycle prevalence

rec_with_table() ran successfully for all 9 PUMF cycles. Since this PR only changes master (_m) mappings, PUMF results confirm the full-file rewrite didn't break existing functionality. Both variables show 100% valid across all cycles with stable category distributions — no era boundary issues.

Cycle	GEN_07 valid %	GEN_10 valid %
cchs2001_p	100%	100%
cchs2003_p	100%	100%
cchs2005_p	100%	100%
cchs2007_2008_p	100%	100%
cchs2009_2010_p	100%	100%
cchs2011_2012_p	100%	100%
cchs2013_2014_p	100%	100%
cchs2015_2016_p	100%	100%
cchs2017_2018_p	100%	100%

Master (_m) mappings validated by worksheet checks only — no runtime data available for L6 testing.

Issues found

1 issue (P0)

1. GEN_10 wrong master source variable names (P0, L1 concordance / Check 3, confidence: 100)

   All 7 GEN_10 rows map pre-2007 master databases to the wrong CCHS source variable:

   • Current: cchs2001_m::GENA_01, cchs2003_m::GENC_01, cchs2005_m::GENE_01
   • Correct: cchs2001_m::GENA_10, cchs2003_m::GENC_10, cchs2005_m::GENE_10

   GENA_01 = "Self-perceived health" (this is GEN_07's source variable)
   GENA_10 = "Sense of belonging to local community" (the correct source for GEN_10)

   Confirmed against cchsflow-docs data dictionary. This bug was pre-existing on the target branch (with _i suffix) — the PR converted _i → _m without correcting the source name. At runtime, GEN_10 would return self-perceived health data instead of sense-of-belonging data for 2001, 2003, and 2005 master cycles.

   Fix: Replace GENA_01 → GENA_10, GENC_01 → GENC_10, GENE_01 → GENE_10 in all 7 GEN_10 rows' variableStart field.

Pre-existing issues (not introduced by this PR)

• Space after :: in GEN_07: cchs2017_2018_p:: GEN_020 (all 8 rows)
• _s/_m mismatch: variable_details.csv uses cchs2009_s/cchs2010_s/cchs2012_s, variables.csv uses _m
• dummyVariable colon pattern: GEN_07_cat5_NA::a instead of _NAa

CEP: ceps/cep-010-gen-health/

Generated with Claude Code

[DougManuel]

Additional review finding: _s → _m conversion needed

Both GEN_07 and GEN_10 still have deprecated _s (share file) databases in variable_details.csv that should be converted to _m:

• cchs2009_s → cchs2009_m
• cchs2010_s → cchs2010_m
• cchs2012_s → cchs2012_m

This affects all 15 rows (8 for GEN_07, 7 for GEN_10). variables.csv already uses _m for these databases, so the conversion also resolves the databaseStart consistency mismatch between the two files.

Since the PR is already converting _i → _m in these rows, this is a natural cleanup to include.

Summary of recommended fixes for this PR

1. P0: Fix GEN_10 source variable names: GENA_01 → GENA_10, GENC_01 → GENC_10, GENE_01 → GENE_10 (all 7 rows)
2. P2: Convert _s → _m for cchs2009_s, cchs2010_s, cchs2012_s in both GEN_07 and GEN_10 (all 15 rows)