Eight OOS Results, Zero Root Causes: Lessons from Glenmark's FDA 483
Eight out-of-specification results for the same test on the same product. Three different analysts, three different days, same failure mode. Root causes attributed to 'analyst error' with no supporting evidence. Batches released to the US market without re-validating the suspected root cause. The Glenmark 483 is a case study in how OOS investigation programmes fail when the system is designed to explain away failures rather than find them.
On February 14, 2025, the FDA issued a 483 observation to Glenmark Pharmaceuticals Ltd. at their Pithampur, Madhya Pradesh facility in India. The finding cited a failure under 21 CFR 211.192: “There is a failure to thoroughly review any unexplained discrepancy and the failure of a batch or any of its components to meet any of its specifications whether or not the batch has been already distributed.”
The specifics reveal an OOS investigation programme that had been producing the same result for the same product since 2022 — and consistently failing to identify why. Eight OOS investigations for dissolution testing on a single capsule product since April 2024 alone. Multiple OOS and OOT results dating back to 2022. No specific root causes identified. Annual stability batches failing at various time points. And through all of this, batches continued to be released to the US market, manufactured using the same validated process that the firm’s own data suggested was the source of the problem.
Three of those OOS investigations, opened between September and November 2024, illustrate the pattern precisely. Three different analysts, three different days, each producing failing dissolution results. The root cause assigned to all three: improper analyst technique. The evidence supporting that conclusion: after-failure interviews. The documentation of sample preparation specifics — the types and numbers of implements used — that might have confirmed or refuted analyst error: not recorded. The original OOS results were invalidated, retested with new samples, and the batches released. The dissolution failures stopped being a quality signal. They became a paperwork exercise.
Eight OOS investigations for the same test on the same product. Zero specific root causes identified. Batches released to the US market using a process the firm’s own data had called into question since 2022. The Glenmark 483 is not about one bad investigation — it is about an OOS programme that functions as a path to release rather than a path to understanding.
What the FDA Found
Three distinct failures, all pointing to the same systemic problem: an OOS investigation framework that attributes failures without explaining them, and a batch disposition process that continues releasing product while the root cause remains open.
Recurring OOS and OOT results with no root cause, batches released anyway. Multiple OOS and OOT results for capsule dissolution testing had been documented since 2022. The firm’s investigation attributed these failures to variation in previously validated equipment critical process parameters — but did not re-validate those parameters. An annual stability batch in 2023 produced OOS results at multiple time points. The impact assessment tested only reserve samples from batches still in the market at the time of the finding. One of those retested batches produced an additional OOS result and was recalled. The remaining untested batches had no adequate supporting stability data. Despite all of this, the firm continued manufacturing and releasing batches of the same product to the US market using the same process, without re-validating the equipment critical process parameters that had been identified as the suspected root cause.
OOS results invalidated on the basis of after-failure interviews, not evidence. For dissolution OOS investigations on the same capsule product, the root cause conclusion was analyst error — specifically, improper technique during sample preparation. But the specific types and number of implements used during sample preparation were not documented at the time of the test. The root cause was reconstructed through after-failure interviews rather than contemporaneous records. All three initial OOS results — obtained by three different analysts on three different days — were invalidated and retested with new samples. The retests passed. The batches were released to the US market. The investigation never adequately explained how the passing results from the remainder of the capsules in the same sample sets were obtained if the analysts’ technique was indeed the cause of failure.
Eight OOS investigations, one product, one test — since April 2024. 21 CFR 211.192 requires thorough review of any unexplained discrepancy and the failure of a batch to meet specifications. Eight dissolution OOS investigations for the same capsule product in less than a year is not a series of isolated events. It is a pattern. And when a pattern of this magnitude produces no specific root cause, the investigation programme itself has become the failure — not the dissolution test and not the analysts performing it.
8 OOS
Investigations Since April 2024
Eight out-of-specification results for dissolution testing on a single capsule product in less than a year — all investigated, none with a specific root cause that addressed the systemic pattern.
0
Root Causes Identified
No specific root causes were identified for the recurring OOS and OOT results. The firm attributed failures to equipment process parameter variation but did not re-validate the suspected parameters.
3 analysts
Same Failure, Different People
Three OOS results in Sept–Nov 2024 from three different analysts on three different days, all attributed to 'improper analyst technique' based on after-failure interviews — not contemporaneous documentation.
Why This Keeps Happening
Glenmark's OOS investigation failures are not unique. They reflect a structural pattern in how pharmaceutical quality systems handle out-of-specification results — investigating to disposition rather than investigating to understand.
The root cause is not negligent quality assurance staff or incompetent analysts. It is an investigation framework where the incentive structure, documentation practices, and decision-making architecture are aligned toward explaining away OOS results rather than identifying their true origin. When every investigation ends with “analyst error” and every batch gets released, the system is working as designed — just not as regulated.
Investigations default to analyst error because it is the easiest root cause to assign.
Analyst error requires the least systemic follow-up. It does not trigger process re-validation, equipment qualification, or formulation review. It requires only retraining — a corrective action that can be completed in a day and documented in a paragraph. At Glenmark, three different analysts on three different days produced the same dissolution failure, and all three were attributed to improper technique. When the same root cause explains every failure regardless of who performed the test, it is not a root cause — it is a default assignment.
Sample preparation details are not documented contemporaneously.
The FDA noted that the specific types and number of implements used during sample preparation were not documented at the time of testing. Without contemporaneous records, the only way to reconstruct what happened is through after-failure interviews — asking analysts to remember what they did, after they know their results failed. This is not an investigation methodology. It is confirmation bias with a signature line. Contemporaneous documentation of sample preparation is a basic requirement, and its absence removes the evidentiary foundation for any root cause conclusion.
OOS results are invalidated and retested without resolving the original failure.
The Glenmark investigation pattern follows a common sequence: initial test fails, root cause is assigned to the analyst, original result is invalidated, new samples are pulled, retest passes, batch is released. The original failing result — the data that triggered the investigation — is effectively discarded. But 21 CFR 211.192 does not permit invalidation of OOS results without scientifically sound justification. And retesting with new samples after attributing the failure to analyst error, without contemporaneous evidence of that error, does not meet that standard.
Trend data is acknowledged but not acted upon.
The firm's own investigation identified equipment critical process parameters as a suspected source of variation. Annual stability data confirmed ongoing failures. But the response was limited to testing reserve samples — not re-validating the process, not qualifying the equipment, not reformulating the product. When a firm identifies a suspected root cause and then continues manufacturing without addressing it, the investigation has documented the problem while the quality system has chosen to accept it. That is precisely the gap 21 CFR 211.192 exists to close.
The question is not whether OOS results can be attributed to analyst error. The question is whether your investigation system can distinguish genuine analyst error from a systemic process failure — and whether it documents the evidence needed to tell the difference, or reconstructs narratives after the fact.
Manual OOS Investigations vs System-Enforced Quality
The difference between the Glenmark investigation pattern and a compliant OOS programme is not investigator skill — it is whether the system captures the data needed for a real investigation or leaves the conclusion to be reconstructed from memory.
Each comparison below addresses a specific failure documented in the Glenmark 483. The system-enforced approach does not add review layers to an already strained investigation workflow. It captures the data at the point of execution so that investigations have evidence to work with rather than interviews to rely on.
Sample Preparation Documentation
Sample preparation steps, implement types, and quantities are not documented contemporaneously. When an OOS result occurs, the investigation relies on after-failure interviews to reconstruct what happened — days or weeks after the test. The analyst's recollection, influenced by knowledge of the failing result, becomes the primary evidence for the root cause conclusion.
Result: Root causes based on memory, not evidence
Electronic batch records and lab execution systems capture sample preparation details at the point of execution — implement types, quantities, preparation sequence, and timestamps. When an OOS occurs, the investigation starts with contemporaneous, attributable records of exactly what was done. No reconstruction required. No interviews needed to establish basic facts.
Result: Investigations built on contemporaneous data
OOS Trend Detection
Each OOS is investigated as an isolated event. Trend analysis, if performed, is periodic — quarterly or annual. At Glenmark, eight dissolution OOS results for the same product accumulated over ten months. The pattern was visible, but the investigation framework treated each occurrence as a standalone failure, assigning individual root causes without addressing the systemic signal.
Result: Patterns visible only in retrospect
Automated trend analysis flags recurring OOS patterns in real time — by product, by test, by equipment, by analyst. The second or third dissolution failure for the same product automatically escalates to a systemic investigation rather than defaulting to individual analyst error. Trending data is presented to investigators before they assign a root cause, not compiled after disposition.
Result: Patterns detected and escalated automatically
Batch Disposition Decisions
Batches are released based on passing retest results after the original OOS is invalidated. The decision to release is made within the individual OOS investigation, without systematic visibility into the broader pattern. At Glenmark, batches continued to be released using an unre-validated process, even after the firm's own data identified the process parameters as a suspected failure source.
Result: Disposition disconnected from trend data
Batch disposition workflows incorporate OOS history, trending data, and open investigation status for the product and process. The system surfaces relevant OOS history and stability data before the disposition decision is made. Release cannot proceed without documented acknowledgement of unresolved trends. Process re-validation requirements are triggered automatically when OOS frequency exceeds defined thresholds.
Result: Disposition informed by full quality context
What a Modern System Must Do
Preventing the Glenmark pattern requires a quality system that captures execution data contemporaneously, detects OOS trends automatically, and enforces investigation rigour before batches can be dispositioned.
The three capabilities below directly address the root causes behind the FDA’s findings. They work because they shift the burden of evidence from post-hoc reconstruction to real-time capture — making it structurally difficult to investigate without evidence or release without resolution.
Contemporaneous Execution Capture
Every step of laboratory and manufacturing execution is captured at the point it occurs — sample preparation details, equipment parameters, analyst actions, timestamps. When an OOS result triggers an investigation, the system provides a complete, timestamped record of exactly what happened during the test. Root cause conclusions must be supported by this contemporaneous data, not reconstructed from after-failure interviews.
Automated OOS Trending and Escalation
The system continuously monitors OOS and OOT results by product, test method, equipment, and analyst. Recurring patterns trigger automatic escalation from individual investigation to systemic review. The second dissolution failure for the same product does not get the same investigation template as the first — it gets a broader scope that includes process capability, equipment qualification, and method suitability. Eight OOS results for one product in ten months cannot accumulate without triggering systemic action.
Integrated Batch Disposition
Batch release decisions are connected to the full quality data set — OOS history, stability trending, open investigations, and process validation status. The system prevents release of batches manufactured using a process that is under investigation for systemic OOS patterns. When a firm identifies equipment critical process parameters as a suspected root cause, the disposition workflow requires documented evidence that those parameters have been re-validated before additional batches are released.
20→1 days
Batch Review Time
Valent BioSciences reduced batch review from 20 days to 1 day with electronic batch records — because reviewers have immediate access to complete execution data, OOS history, and trending information without manual compilation.
2,700 hrs
Saved Annually
Annual hours eliminated across Valent BioSciences' operations by digitising execution workflows — hours previously spent on manual documentation, retrospective data compilation, and paper-based investigation support.
30 facilities
Cipla: Enterprise Scale
Cipla operates 2,500+ concurrent users across 30 facilities on a unified platform — with consistent OOS investigation workflows, automated trending, and integrated batch disposition across every site.
From Gap to Prevention
Three phases to transform an OOS investigation programme from a path-to-release workflow into a system that identifies real root causes and prevents recurrence.
The objective is not to add more review steps to existing OOS procedures. It is to redesign the data capture and investigation architecture so that the evidence needed for genuine root cause analysis exists before an OOS occurs — and so that trending, escalation, and disposition decisions are informed by the full quality data set rather than the individual investigation in isolation.
Phase 1: Audit OOS investigation outcomes for systemic patterns.
Review the last 24 months of OOS investigations. For each closure, assess: Was the root cause supported by contemporaneous evidence or reconstructed through interviews? Were sample preparation details documented at the time of testing? How many OOS results were attributed to analyst error, and how many of those involved different analysts? Were retest results from new samples used to override original OOS data? Map the frequency of OOS results by product, test, and equipment to identify the patterns that your current investigation framework is missing. The Glenmark pattern — eight dissolution OOS results for one product with no systemic root cause — is only visible when investigations are viewed collectively rather than individually.
Phase 2: Implement contemporaneous capture and automated trending.
Deploy electronic execution systems that capture sample preparation details, equipment parameters, and analyst actions at the point of testing. Configure automated OOS trending that monitors results by product, test method, equipment, and analyst — with escalation rules that trigger systemic investigation when patterns emerge. Integrate OOS data with batch disposition workflows so that release decisions are informed by the full quality context. Establish process re-validation triggers linked to OOS frequency thresholds so that a suspected root cause cannot remain unaddressed while production continues.
Phase 3: Validate investigation rigour through metrics.
Measure the health of your OOS investigation programme with metrics that expose the Glenmark pattern: percentage of OOS results attributed to analyst error, percentage of investigations closed without contemporaneous evidence, average time from OOS detection to systemic root cause identification, and number of batches released while systemic investigations remain open. These metrics should be reported to site quality leadership monthly and included in management review. Within 90 days, the data should demonstrate that OOS investigations are producing genuine root causes — not default assignments that clear the path to release.
Glenmark’s facility produced eight dissolution OOS results for one capsule product in ten months, attributed failures to analyst error based on after-failure interviews, invalidated original results, and continued releasing batches using an unre-validated process. The FDA found every element of this pattern under a single 211.192 observation. The question for every pharmaceutical quality leader is whether your OOS investigation programme would produce a different outcome — or the same one.
The pharmaceutical industry’s OOS investigation failures follow a predictable pattern. A test fails. The investigation opens. The root cause defaults to something attributable to the individual — analyst error, sample preparation technique, instrument handling — rather than something that implicates the process, the equipment, or the formulation. The original result is invalidated. A retest passes. The batch is released. And the next time the same product fails the same test, the cycle repeats.
The Glenmark 483 is a precise documentation of this pattern operating at scale. Eight dissolution OOS results for a single capsule product. Root causes attributed to analyst error without contemporaneous evidence of what the analysts actually did. After-failure interviews used as the primary investigative tool. Three different analysts producing the same failure on three different days, with no adequate explanation of how passing results were obtained from the same sample sets. And throughout, batches released to the US market using a process that the firm’s own investigation had identified as potentially compromised — without re-validating the suspected parameters.
21 CFR 211.192 exists because this pattern is dangerous. When OOS results are systematically explained away rather than investigated, the quality system loses its ability to detect real problems. Every invalidated result is a data point removed from the firm’s understanding of its own process. Every batch released without resolution is a product on the market whose quality is supported by an investigation conclusion rather than by evidence. Modern quality systems prevent this not by making investigators work harder, but by capturing execution data contemporaneously, detecting OOS trends automatically, and connecting batch disposition decisions to the full quality data set — so that eight dissolution failures for one product in ten months cannot be investigated, explained away, and released in isolation. The architecture must make the Glenmark pattern structurally impossible, because procedure alone has proven insufficient to prevent it.
Related Articles
Three Facilities, Three FDA Actions, Five Architectural Gaps: How AI Agents Address Cipla's Regulatory Exposure
Between 2023 and 2026, three Cipla facilities — Pithampur, Raigad, and Pharmathen Greece — received FDA enforcement actions documenting the same five systemic failures: complaint investigation, CAPA effectiveness, electronic data review, contamination control, and QC oversight. LeucineOS AI agents map directly to each gap.
35% OOS Invalidations, Zero Scientific Justification: Lessons from Aurobindo Pharma's FDA 483
A February 2026 FDA 483 at Aurobindo Pharma's Unit-III found 35% of OOS invalidations in the QC Chemistry lab — with 57% blamed on analyst error and 18% on equipment, none supported by adequate scientific justification. Batches shipped to the US after unresolved Grade A maintenance interventions.
Equipment Swapped, Cleaning Not Revalidated, OOS Dissolved Away: Lessons from Dr. Reddy's FDA 483
A December 2025 FDA 483 at Dr. Reddy's FTO-SEZ facility in Srikakulam found cleaning validation not performed after equipment replacement, OOS dissolution results invalidated despite contradictory evidence, and process qualification gaps — all traceable to a single uncontrolled equipment change eighteen months earlier.
Newsletter
Stay ahead in the Industry
Regulatory updates, pharma quality insights, and AI in manufacturing — written for quality leaders, not marketers.
Please use your official work email. Personal email addresses (Gmail, Yahoo, etc.) will not receive the newsletter. No spam. Unsubscribe anytime.
Ready to see what an AI-native quality platform looks like? Leucine unifies quality management, regulatory compliance, and production operations into one intelligent system.
