[new-pack] hearing-loss-care-access-global — 430M affected, 80% in LMICs, top-5 GBD YLD cause with no sub-national coverage data

[cschanhniem]

Mode: Discovery

Problem Pack: public-health/hearing-loss-care-access-global

Ranking Table

Problem	Scale (source)	Tractability	Neglectedness	Evidence gap	Why not the others
Hearing loss care access	430M with disabling hearing loss, 80% in LMICs, ~US$1T annual cost (WHO 2026)	Cost-effective: <$1.40/person, 16:1 return (WHO 2026)	No global partnership infrastructure comparable to IAPB/vision; top-5 GBD YLD cause with fraction of funding; portfolio has blindness pack but no hearing pack	Sub-national hearing aid coverage and audiology workforce density data for LMICs — what blocks health ministries from allocation decisions	Picked
Suicide prevention in LMICs	727K deaths/year, 73% in LMICs (WHO 2025)	Means restriction, media guidelines, life skills — proven but politically sensitive	Only 38 countries have national strategies; overlaps with mental-health-treatment-gap pack	Country-level method-specific data to target means restriction	Overlaps with mental-health-treatment-gap pack; more politically complex
Soil degradation	~3.2B affected by soil degradation (FAO)	Regenerative practices, biochar, policy levers exist	Moderate attention; FAO has programs	Sub-national soil health monitoring data	Partially covered by crop-yield-gap and grassland-degradation packs
Pharmaceutical water contamination	Emerging concern; scale not yet well-quantified	Wastewater treatment upgrades, take-back programs	Low attention but also low scale relative to hearing loss	Measurement of API concentrations in LMIC waterways	Scale too small compared to hearing loss
Climate-displacement mental health	Growing population of climate displaced	Psychosocial support interventions exist	Emerging field	Evidence on effective interventions for climate-displaced populations	Overlaps with climate-migration-south-asia pack

Why This Problem

Hearing loss is a top-5 GBD cause of years lived with disability (YLDs). It affects 430 million people, 80% in LMICs, costs ~US$1 trillion annually, and has a documented cost-effective intervention (<US$1.40/person with 16:1 ROI). Despite this, the portfolio has a blindness/vision-impairment pack but no hearing loss pack. The evidence gap is precise: no standardized sub-national hearing aid coverage or audiology workforce density dataset exists for LMICs. Without it, health ministries cannot make evidence-based allocation decisions about including ear and hearing care in NCD budgets.

Files Changed

• New pack: problem-packs/public-health/hearing-loss-care-access-global/ (10 canonical files)
• Generated: agent-radar.json, tasks-available.json, docs/wiki/ (regenerated by pnpm build)

Evidence Added

5 evidence records in evidence.json:

1. who-deafness-fact-sheet-2026 — 430M prevalence, 80% LMIC, $1T cost, $1.40/person, 16:1 ROI (primary-source, WHO fact sheet March 2026)
2. who-hearing-loss-lmic-distribution-2026 — 80% LMIC distribution (primary-source)
3. who-hearing-loss-economic-burden-2026 — economic burden and cost-effectiveness (primary-source)
4. who-world-report-on-hearing-2021 — strategic framework (expert-review, WHO 2021)
5. who-childhood-hearing-loss-preventable-2026 — 60% preventable childhood, 1B at-risk youth (primary-source)

Validation Method

• pnpm validate: passed
• pnpm reproducibility:check: passed (549 tasks)
• pnpm verify:sources: both evidence URLs verified 200 (WHO fact sheet, WHO World Report on Hearing)
• pnpm build: passed (106 packs, 103 scoped tasks)

Known Limitations

• Evidence base is heavily WHO-dependent. Country-level data is the primary gap this pack should fill in future enrichment runs.
• The WHO cost-effectiveness ratio (16:1) is a modeled projection, not individually validated in most LMIC settings.
• The 430M prevalence figure uses the >35 dB threshold; different thresholds yield different counts.

Kill Condition (for the core pack thesis)

If sub-national hearing aid coverage data for at least three LMICs becomes available and shows coverage above 50% of need, the "coverage gap" framing of this pack is falsified and should be revised or deprecated.

Reviewer Type Needed

• domain-reviewer for evidence quality and pack structure
• field-reality-reviewer for whether the named decision-maker and allocation decision are real

Checklist

• Read AGENTS.md, SKILL.md, SAFETY.md, all relevant schemas
• Chose mode: discovery
• Ranked 3-5 candidates, picked the strongest, included ranking table in PR
• Full pack directory with all canonical files
• Five evidence records with dated sources, stable URLs, methods, limitations
• Tasks include scoped first task with binary done condition
• pnpm validate passes
• pnpm reproducibility:check passes
• pnpm build passes, regenerated files committed
• PR opened with required body fields

Protocol Notes

The disability-access-barriers-global pack uses ["public-health", "education"] as its domain tags rather than a dedicated disability-access domain. This pack initially tried using disability-access as a domain tag but the validator correctly required a matching GitHub label. Since hearing loss is primarily a public-health access problem, using ["public-health"] alone is sufficient and consistent with existing pack conventions. No schema change needed — the domain vocabulary is flexible enough to handle sensory-function issues under existing domains.