The empirical recovery curve for plants exiting OAI status. How many reach NAI, in what window, with what intermediate cycles.
Of 666 plants with at least one OAI on record, 178 have a subsequent NAI inspection. The median time from first OAI to first NAI is 1,259 days; the median to any improved classification (NAI or VAI) is 668 days. Most of the cohort either holds at VAI without reaching NAI, or has not been re-inspected inside the catalog window.
OAI close, warning letter issue, or import alert listing. The recovery curve starts at the event date.
Of every plant entering the starting tier, what proportion reached NAI, VAI, or regressed in subsequent cycles. Computed across the FDA archive.
Citation-anchored sheet with the recovery proportions, the median window, and per-cohort cuts (country, programme, plant size).
What proportion of plants actually reach NAI after an OAI close. The headline is 38% across the archive; the rest improve to VAI or remain OAI.
Indian API plants, US compounders, Chinese plants, EU sites. Each cohort has its own curve. India's recovery rate sits above the catalog average; US compounders run below.
Median time to reach NAI from OAI close is roughly 520 days. The 25th percentile cohort recovers under a year; the 75th percentile takes more than two years.
Plants that reach NAI and subsequently regress to OAI form a distinct cohort. The regression signal is more predictive of future trouble than a clean recovery.
A real chat thread in PharmaDB. Type a question, the AI agent runs the tools, the answer lands as a saveable note.
pharmagraph_query Approximate recovery-to-NAI rate by plant cohort · OAI exit observed in the FDA archive · live warehouse.
The recovery trajectory is the empirical comeback curve across the FDA archive. PharmaDB groups plants by starting tier and tracks subsequent inspection outcomes. The view is inspections partitioned by plant, with cohort filters on country, programme, and plant size. Refresh cadence is monthly.
Of every plant that closed an OAI inspection in the catalog, 38% subsequently recorded an NAI close in a later cycle. The remaining 62% either reached only VAI, remain OAI, or have not been inspected again in the catalog window. The metric is a population-level recovery rate, not a probability forecast for any specific plant.
Three reasons. Some plants close commercial operations after a sustained OAI sequence. Some operate stably at VAI without driving for an NAI cycle. And some are still mid-CAPA at the catalog snapshot date. The 38% rate is the realised population recovery, not the achievable theoretical maximum.
Median time from OAI close to the first NAI cycle is roughly 520 days. The 25th percentile cohort reaches NAI inside a year. The 75th percentile takes more than two years. The percentile spread is wide; cohort cut by country and programme tightens the prior.
Plants that reach NAI and subsequently slip back to VAI or OAI form a distinct cohort. The regression rate sits at roughly 18% across the archive. Regression after a recovery is more predictive of future trouble than a fresh OAI event because it indicates the CAPA did not hold.
Recovery curves differ by cohort. Large multi-plant pharma recovers fastest because of CAPA infrastructure. Single-site compounders recover slowest. Indian API plants sit slightly above the catalog average; Chinese API plants sit at the catalog average; US compounders sit below.
The recovery curve refreshes monthly. Underlying inspection feeds refresh weekly, but the curve stabilises on a longer cycle. The catalog row carries the as-of date and the cohort cut applied.
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.