Median time from a warning letter to the next NAI inspection at the same plant. The empirical recovery clock across the FDA enforcement archive.
Across the catalog, exactly one plant has both a warning letter and a subsequent NAI inspection on record. The clock from event to recovery on that case is 403 days. This is the recovery shape, not yet the population median. The catalog needs more paired observations before the half-life can be quoted authoritatively.
Warning letter, OAI close, or import alert listing. The recovery clock starts at the event date.
Median days from event to the next NAI inspection at the same plant. Computed across the historical archive; the percentile distribution surfaces alongside.
Citation-anchored sheet with the empirical half-life, the percentile distribution, and a recommendation for distress-asset BD timing or supplier-risk reserve sizing.
Recovery time is empirical across the entire archive of paired events. The median is the planning number; the percentile range tells you how much variance to budget.
Warning letter, OAI close, and import alert each carry their own half-life curve. Import alerts run longest because they require petition-based delisting; OAI takes longer than WL because the cycle gap is wider.
BD teams use the half-life to time distressed-asset conversations. A plant 12 months into a warning letter is approaching the median recovery point; a plant past 24 months is increasingly unlikely to recover this cycle.
Finance teams use the half-life as an empirical supply-risk reserve input. A supplier with an active warning letter carries roughly 13 months of expected disruption from the event date.
A real chat thread in PharmaDB. Type a question, the AI agent runs the tools, the answer lands as a saveable note.
pharmagraph_query Median recovery days by event tier · the empirical recovery clock across the FDA archive.
The compliance half-life is the empirical recovery clock across the FDA enforcement archive. PharmaDB pairs each event to the next inspection at the same plant and computes the median time-to-recovery by tier. The view is compliance_events joined to subsequent inspections filtered to improving classification. Refresh cadence is monthly.
Recovery times are right-skewed. A small number of multi-year recoveries pull the average up; the median is the better planning number. The percentile distribution captures the full shape.
It is a population-level prior, not a plant-level forecast. Specific plant outcomes depend on CAPA quality, regulator scrutiny, and corporate response. The half-life is the right number for portfolio-level reserve sizing and BD timing, not for predicting any single plant's clock.
Roughly 32% of plants with a warning letter do not record a subsequent NAI in the catalog window. Some close commercial operations; some continue at VAI without reaching NAI; some are still recovering when the catalog snapshot ends. The non-reaching cohort is itself a signal.
Import alerts run roughly two years median active. The mechanism is petition-based delisting, which is slower than passing a follow-up inspection. The 66-41 tier (unapproved new drug) runs longer than 66-40 (GMP failure) because the underlying issue is harder to remediate.
The half-life refreshes monthly. The underlying inspection and enforcement feeds refresh weekly, but the percentile curves stabilise on a longer cycle. The catalog row carries the as-of date.
EMA non-compliance reports and EDQM CEP suspensions are tracked separately. The cross-regulator half-life view is a separate analysis because each agency's classification taxonomy and recovery mechanism differs.
Bring the molecule, the lane, or the supplier you're sourcing this week. The AI agent runs it on PharmaDB in 30 minutes. You keep the brief.