Courtesy of Kamm & Associates:

Prepared by:     Center for Drugs and Biologics and
 Center for Devices and Radiological Health
Food and Drug Administration

Maintained by:   Division of Manufacturing and Product Quality (HFN-320)
Office of Compliance
Center for Drugs and Biologics
Food and Drug Administration
5600 Fishers Lane
Rockville, Maryland   20857

      This guideline outlines general principles that FDA considers to be
      acceptable elements of process validation for the preparation of
      human and animal drug products and medical devices.

      This guideline is issued under Section 10.90 (21 CFR 10.90) and is
      applicable to the manufacture of pharmaceuticals and medical
      devices.  It states principles and practices of general
      applicability that are not legal requirements but are acceptable to
      the FDA.  A person may rely upon this guideline with the assurance
      of its acceptability to FDA, or may follow different procedures.
      When different procedures are used, a person may, but is not
      required to, discuss the matter in advance with FDA to prevent the
      expenditure of money and effort on activities that may later be
      determined to be unacceptable.  In short, this guideline lists
      principles and practices which are acceptable to the FDA for the
      process validation of drug products and medical devices; it does
      not list the principles and practices that must, in all instances,
      be used to comply with law.


       This guideline may be amended from time to time.  Interested
      persons are invited to submit comments on this document and any
      subsequent revisions.  Written comments should be submitted to the
      Dockets Management Branch (HFA-305), Food and Drug Administration,
      Room 4-62, 5600 Fishers Lane, Rockville, Maryland 20857.  Received
      comments may be seen in that office between 9\a.m. and 4\p.m.,
      Monday through Friday.

      Process validation is a requirement of the Current Good
      Manufacturing Practices Regulations for Finished Pharmaceuticals,
      21 CFR Parts 210 and 211, and of the Good Manufacturing Practice
      Regulations for Medical Devices, 21 CFR Part 820, and therefore, is
      applicable to the manufacture of pharamaceuticals and medical

      Several firms have asked FDA for specific guidance on what FDA
      expects firms to do to assure compliance with the requirements for
      process validation.  This guideline discusses process validation
      elements and concepts that are considered by FDA as acceptable
      parts of a validation program.  The constituents of validation
      presented in this document are not intended to be all-inclusive.
      FDA recognizes that, because of the great variety of medical
      products (drug products and medical devices), processes and

      manufacturing facilities, it is not possible to state in one
      document all of the specific validation elements that are
      applicable.  Several broad concepts, however, have general
      applicability which manufacturers can use successfully as a guide
      in validating a manufacturing process.  Although the particular
      requirements of process validation will vary according to such
      factors as the nature of the medical product (e.g., sterile vs
      non-sterile) and the complexity of the process, the broad concepts
      stated in this document have general applicability and provide an
      acceptable framework for building a comprehensive approach to
      process validation.

      Installation qualification - Establishing confidence that process
      equipment and ancillary systems are capable of consistently
      operating within established limits and tolerances.

      Process performance qualification - Establishing confidence that
      the process is effective and reproducible.

      Product performance qualification - Establishing confidence through
      appropriate testing that the finished product produced by a
      specified process meets all release requirements for functionality
      and safety.

      Prospective validation - Validation conducted prior to the
      distribution of either a new product, or product made under a
      revised manufacturing process, where the revisions may affect the
      product's characteristics.

      Retrospective validation - Validation of a process for a product
      already in distribution based upon accumulated production, testing
      and control data.

      Validation - Establishing documented evidence which provides a high
      degree of assurance that a specific process will consistently
      produce a product meeting its pre-determined specifications and
      quality attributes.

      Validation protocol - A written plan stating how validation will be
      conducted, including test parameters, product characteristics,
      production equipment, and decision points on what constitutes
      acceptable test results.

      Worst case - A set of conditions encompassing upper and lower
      processing limits and circumstances, including those within
      standard operating procedures, which pose the greatest chance of
      process or product failure when compared to ideal conditions.  Such
      conditions do not necessarily induce product or process failure.

      Assurance of product quality is derived from careful attention to a
      number of factors including selection of quality parts and
      materials, adequate product and process design, control of the
      process, and in-process and end-product testing.  Due to the
      complexity of today's medical products, routine end-product testing
      alone often is not sufficient to assure product quality for several
      reasons.  Some end-product tests have limited sensitivity.1  In
      some cases, destructive testing would be required to show that the
      manufacturing process was adequate, and in other situations
      end-product testing does not reveal all variations that may occur
      in the product that may impact on safety and effectiveness.2

      The basic principles of quality assurance have as their goal the
      production of articles that are fit for their intended use.  These

  1   For example, USP XXI states:  "No sampling plan for applying
      sterility tests to a specified proportion of discrete units
      selected from a sterilization load is capable of demonstrating with
      complete assurance that all of the untested units are in fact

  2   As an example, in one instance a visual inspection failed to detect
      a defective structural weld which resulted in the failure of an
      infant warmer.  The defect could only have been detected by using
      destructive testing or expensive test equipment.


      principles may be stated as follows:  (1) quality, safety, and
      effectiveness must be designed and built into the product; (2)
      quality cannot be inspected or tested into the finished product;
      and (3) each step of the manufacturing process must be controlled
      to maximize the probability that the finished product meets all
      quality and design specifications.  Process validation is a key
      element in assuring that these quality assurance goals are met.

      It is through careful design and validation of both the process and
      process controls that a manufacturer can establish a high degree of
      confidence that all manufactured units from successive lots will be
      acceptable.  Successfully validating a process may reduce the
      dependence upon intensive in-process and finished product testing.
      It should be noted that in most all cases, end-product testing
      plays a major role in assuring that quality assurance goals are
      met; i.e., validation and end-product testing are not mutually

      The FDA defines process validation as follows:
         Process validation is establishing documented evidence which
         provides a high degree of assurance that a specific process will
         consistently produce a product meeting its pre-determined
         specifications and quality characteristics.


      It is important that the manufacturer prepare a written validation
      protocol which specifies the procedures (and tests) to be conducted
      and the data to be collected.  The purpose for which data are
      collected must be clear, the data must reflect facts and be
      collected carefully and accurately.  The protocol should specify a
      sufficient number of replicate process runs to demonstrate
      reproducibility and provide an accurate measure of variability
      among successive runs.  The test conditions for these runs should
      encompass upper and lower processing limits and circumstances,
      including those within standard operating procedures, which pose
      the greatest chance of process or product failure compared to ideal
      conditions; such conditions have become widely known as "worst
      case" conditions.  (They are sometimes called "most appropriate
      challenge" conditions.)  Validation documentation should include
      evidence of the suitability of materials and the performance and
      reliability of equipment and systems.

      Key process variables should be monitored and documented.  Analysis
      of the data collected from monitoring will establish the
      variability of process parameters for individual runs and will
      establish whether or not the equipment and process controls are
      adequate to assure that product specifications are met.


      Finished product and in-process test data can be of value in
      process validation, particularly in those situations where quality
      attributes and variabilities can be readily measured.  Where
      finished (or in-process) testing cannot adequately measure certain
      attributes, process validation should be derived primarily from
      qualification of each system used in production and from
      consideration of the interaction of the various systems.

      Process validation is required, in both general and specific terms,
      by the Current Good Manufacturing Practice Regulations for Finished
      Pharmaceuticals, 21 CFR Parts 210 and 211.  Examples of such
      requirements are listed below for informational purposes, and are
      not all-inclusive.

      A requirement for process validation is set forth in general terms
      in section\211.100 -- Written procedures; deviations -- which
      states, in part:
         "There shall be written procedures for production and process
         control designed to assure that the drug products have the
         identity, strength, quality, and purity they purport or are
         represented to possess."


         Several sections of the CGMP regulations state validation
         requirements in more specific terms.  Excerpts from some of
         these sections are:
              Section 211.110, Sampling and testing of in-process
              materials and drug products.

         (a)  "....control procedures shall be established to monitor the
         output and VALIDATE the performance of those manufacturing
         processes that may be responsible for causing variability in the
         characteristics of in-process material and the drug product."
         (emphasis added)

         Section 211.113, Control of Microbiological Contamination.

         (b)  "Appropriate written procedures, designed to prevent
         microbiological contamination of drug products purporting to be
         sterile, shall be established and followed.  Such procedures
         shall include VALIDATION of any sterilization process."
         (emphasis added)

      Process validation is required by the medical device GMP
      Regulations, 21 CFR Part\820.  Section 820.5 requires every
      finished device manufacturer to:
         "...prepare and implement a quality assurance program that is
         appropriate to the specific device manufactured..."

      Section 820.3(n) defines quality assurance as:
         "...all activities necessary to verify confidence in the quality
         of the process used to manufacture a finished device."

      When applicable to a specific process, process validation is an
      essential element in establishing confidence that a process will
      consistently produce a product meeting the designed quality

      A generally stated requirement for process validation is contained
      in section\820.100:
         "Written manufacturing specifications and processing procedures
         shall be established, implemented, and controlled to assure that
         the device conforms to its original design or any approved
         changes in that design."

      Validation is an essential element in the establishment and
      implementation of a process procedure, as well as in determining
      what process controls are required in order to assure conformance
      to specifications.

      Section 820.100(a)(1) states:
         "...control measures shall be established to assure that the
         design basis for the device, components and packaging is
         correctly translated into approved specifications."

      Validation is an essential control for assuring that the
      specifications for the device and manufacturing process are
      adequate to produce a device that will conform to the approved
      design characteristics.

      A manufacturer should evaluate all factors that affect product
      quality when designing and undertaking a process validation study.
      These factors may vary considerably among different products and
      manufacturing technologies and could include, for example,
      component specifications, air and water handling systems,
      environmental controls, equipment functions, and process control
      operations.  No single approach to process validation will be
      appropriate and complete in all cases; however, the following
      quality activities should be undertaken in most situations.

      During the research and development (R&D) phase, the desired
      product should be carefully defined in terms of its
      characteristics, such as physical, chemical, electrical and


      performance characteristics.3  It is important to translate the
      product characteristics into specifications as a basis for
      description and control of the product.

      Documentation of changes made during development provide
      traceability which can later be used to pinpoint solutions to
      future problems.

      The product's end use should be a determining factor in the
      development of product (and component) characteristics and
      specifications.  All pertinent aspects of the product which impact
      on safety and effectiveness should be considered.  These aspects

  3   For example, in the case of a compressed tablet, physical
      characteristics would include size, weight, hardness, and freedom
      from defects, such as capping and splitting.  Chemical
      characteristics would include quantitative formulation/potency;
      performance characteristics may include bioavailability (reflected
      by disintegration and dissolution).  In the case of blood tubing,
      physical attributes would include internal and external diameters,
      length and color.  Chemical characteristics would include raw
      material formulation.  Mechanical properties would include hardness
      and tensile strength; performance characteristics would include
      biocompatibility and durability.


      include performance, reliability and stability.  Acceptable ranges
      or limits should be established for each characteristic to set up
      allowable variations.4  These ranges should be expressed in
      readily measurable terms.

      The validity of acceptance specifications should be verified
      through testing and challenge of the product on a sound scientific
      basis during the initial development and production phase.

      Once a specification is demonstrated as acceptable it is important
      that any changes to the specification be made in accordance with
      documented change control procedures.

      A. Prospective Validation
      Prospective validation includes those considerations that should be
      made before an entirely new product is introduced by a firm or when
      there is a change in the manufacturing process which may affect the
      product's characteristics, such as uniformity and identity.  The
      following are considered as key elements of prospective validation.

4     For example, in order to assure that an oral, ophthalmic, or
      parenteral solution has an acceptable pH, a specification may be
      established by which a lot is released only if it has been shown to
      have a pH within a narrow established range.  For a device, a
      specification for the electrical resistance of a pacemaker lead
      would be established so that the lead would be acceptable only if
      the resistance was within a specified range.

         1.  Equipment and Process
         The equipment and process(es) should be designed and/or selected
         so that product specifications are consistently achieved.  This
         should be done with the participation of all appropriate groups
         that are concerned with assuring a quality product, e.g.,
         engineering design, production operations, and quality assurance
             a.  Equipment:  Installation Qualification
             Installation qualification studies establish confidence that
             the process equipment and ancillary systems are capable of
             consistently operating within established limits and
             tolerances.  After process equipment is designed or
             selected, it should be evaluated and tested to verify that
             it is capable of operating satisfactorily within the
             operating limits required by the process.5  This phase of
             validation includes examination of equipment design;
             determination of calibration, maintenance, and adjustment
             requirements; and identifying critical equipment features
             that could affect the process and product.  Information
             obtained from these studies should be used to establish
             written procedures covering equipment calibration,
             maintenance, monitoring, and control.

         5   Examples of equipment performance characteristics which may
             be measured include temperature and pressure of injection
             molding machines, uniformity of speed for mixers,
             temperature, speed and pressure for packaging machines, and
             temperature and pressure of sterilization chambers.

             In assessing the suitability of a given piece of equipment,
             it is usually insufficient to rely solely upon the
             representations of the equipment supplier, or upon
             experience in producing some other product.6  Sound
             theoretical and practical engineering principles and
             considerations are a first step in the assessment.

             It is important that equipment qualification simulate actual
             production conditions, including those which are "worst
             case" situations.

         6   The importance of assessing equipment suitability based upon
             how it will be used to attain desired product attributes is
             illustrated in the case of deionizers used to produce
             Purified Water, USP.  In one case, a firm used such water to
             make a topical drug product solution which, in view of its
             intended use, should have been free from objectionable
             microorganisms.  However, the product was found to be
             contaminated with a pathogenic microorganism.  The apparent
             cause of the problem was failure to assess the performance
             of the deionizer from a microbiological standpoint.  It is
             fairly well recognized that the deionizers are prone to
             build-up of microorganisms--especially if the flow rates are
             low and the deionizers are not recharged and sanitized at
             suitable intervals.  Therefore, these factors should have
             been considered.  In this case, however, the firm relied
             upon the representations of the equipment itself, namely the
             "recharge" (i.e., conductivity) indicator, to signal the
             time for regeneration and cleaning.  Considering the desired
             product characteristics, the firm should have determined the
             need for such procedures based upon pre-use testing, taking
             into account such factors as the length of time the
             equipment could produce deionized water of acceptable
             quality, flow rate, temperature, raw water quality,
             frequency of use, and surface area of deionizing resins.

             Tests and challenges should be repeated a sufficient number
             of times to assure reliable and meaningful results.  All
             acceptance criteria must be met during the test or
             challenge.  If any test or challenge shows that the
             equipment does not perform within its specifications, an
             evaluation should be performed to identify the cause of the
             failure.  Corrections should be made and additional test
             runs performed, as needed, to verify that the equipment
             performs within specifications.  The observed variability of
             the equipment between and within runs can be used as a basis
             for determining the total number of trials selected for the
             subsequent performance qualification studies of the

             Once the equipment configuration and performance
             characteristics are established and qualified, they should
             be documented.  The installation qualification should
             include a review of pertinent maintenance procedures, repair
             parts lists, and calibration methods for each piece of
             equipment.  The objective is to assure that all repairs can
             be performed in such a way that will not affect the

         7   For example, the AAMI Guideline for Industrial Ethylene
             Oxide Sterilization of Medical Devices approved 2 December
             1981, states:  "The performance qualification should include
             a minimum of 3 successful, planned qualification runs, in
             which all of the acceptance criteria are met.....(

             characteristics of material processed after the repair.  In
             addition, special post-repair cleaning and calibration
             requirements should be developed to prevent inadvertent
             manufacture a of non-conforming product.  Planning during
             the qualification phase can prevent confusion during
             emergency repairs which could lead to use of the wrong
             replacement part.

             b.  Process:  Performance Qualification
             The purpose of performance qualification is to provide
             rigorous testing to demonstrate the effectiveness and
             reproducibility of the process.  In entering the performance
             qualification phase of validation, it is understood that the
             process specifications have been established and essentially
             proven acceptable through laboratory or other trial methods
             and that the equipment has been judged acceptable on the
             basis of suitable installation studies.

             Each process should be defined and described with sufficient
             specificity so that employees understand what is required.


             Parts of the process which may vary so as to affect
             important product quality should be challenged.8

             In challenging a process to assess its adequacy, it is
             important that challenge conditions simulate those that will
             be encountered during actual production, including "worst
             case" conditions.  The challenges should be repeated enough
             times to assure that the results are meaningful and

         8   For example, in electroplating the metal case of an
             implantable pacemaker, the significant process steps to
             define, describe, and challenge include establishment and
             control of current density and temperature values for
             assuring adequate composition of electrolyte and for
             assuring cleanliness of the metal to be plated.  In the
             production of parenteral solutions by aseptic filling, the
             significant aseptic filling process steps to define and
             challenge should include the sterilization and
             depyrogenation of containers/closures, sterilization of
             solutions, filling equipment and product contact surfaces,
             and the filling and closing of containers.


             Each specific manufacturing process should be appropriately
             qualified and validated.  There is an inherent danger in
             relying on what are perceived to be similarities between
             products, processes, and equipment without appropriate

             c.  Product: Performance Qualification
             For purposes of this guideline, product performance
             qualification activities apply only to medical devices.
             These steps should be viewed as pre-production quality
             assurance activities.

         9   For example, in the production of a compressed tablet, a
             firm may switch from one type of granulation blender to
             another with the erroneous assumption that both types have
             similar performance characteristics, and, therefore,
             granulation mixing times and procedures need not be
             altered.  However, if the blenders are substantially
             different, use of the new blender with procedures used for
             the previous blender may result in a granulation with poor
             content uniformity.  This, in turn, may lead to tablets
             having significantly differing potencies.  This situation
             may be averted if the quality assurance system detects the
             equipment change in the first place, challenges the blender
             performance, precipitates a revalidation of the process, and
             initiates appropriate changes.  In this example,
             revalidation comprises installation qualification of the new
             equipment and performance qualification of the process
             intended for use in the new blender.


             Before reaching the conclusion that a process has been
             successfully validated, it is necessary to demonstrate that
             the specified process has not adversely affected the
             finished product.  Where possible, product performance
             qualification testing should include performance testing
             under conditions that simulate actual use.  Product
             performance qualification testing should be conducted using
             product manufactured from the same type of production
             equipment, methods and procedures that will be used for
             routine production.  Otherwise, the qualified product may
             not be representative of production units and cannot be used
             as evidence that the manufacturing process will produce a
             product that meets the pre-determined specifications and
             quality attributes.10

        10   For example, a manufacturer of heart valves received
             complaints that the valve-support structure was fracturing
             under use.  Investigation by the manufacturer revealed that
             all material and dimensional specifications had been met but
             the production machining process created microscopic
             scratches on the valve supporting wireform.  These scratches
             caused metal fatigue and subsequent fracture.  Comprehensive
             fatigue testing of production units under simulated use
             conditions could have detected the process deficiency.

             In another example, a manufacturer recalled insulin syringes
             because of complaints that the needles were clogged.
             Investigation revealed that the needles were clogged by
             silicone oil which was employed as a lubricant during
             manufacturing.  Investigation further revealed that the
             method used to extract the silicone oil was only partially
             effective.  Although visual inspection of the syringes
             seemed to support that the cleaning method was effective,
             actual use proved otherwise.

         After actual production units have sucessfully passed product
         performance qualification, a formal technical review should be
         conducted and should include:

         o   Comparison of the approved product specifications and the
             actual qualified product.

         o   Determination of the validity of test methods used to
             determine compliance with the approved specifications.

         o   Determination of the adequacy of the specification change
             control program.

         2.  System to Assure Timely Revalidation
         There should be a quality assurance system in place which
         requires revalidation whenever there are changes in packaging,
         formulation, equipment, or processes which could impact on
         product effectiveness or product characteristics, and whenever
         there are changes in product characteristics.  Furthermore, when
         a change is made in raw material supplier, the manufacturer
         should consider subtle, potentially adverse differences in the
         raw material characteristics.  A determination of adverse
         differences in raw material indicates a need to revalidate the

         One way of detecting the kind of changes that should initiate
         revalidation is the use of tests and methods of analysis which
         are capable of measuring characteristics which may vary.  Such
         tests and methods usually yield specific results which go beyond
         the mere pass/fail basis, thereby detecting variations within
         product and process specifications and allowing determination of
         whether a process is slipping out of control.

         The quality assurance procedures should establish the
         circumstances under which revalidation is required.  These may
         be based upon equipment, process, and product performance
         observed during the initial validation challenge studies.  It is
         desirable to designate individuals who have the responsibility
         to review product, process, equipment and personnel changes to
         determine if and when revalidation is warranted.

         The extent of revalidation will depend upon the nature of the
         changes and how they impact upon different aspects of production
         that had previously been validated.  It may not be necessary to
         revalidate a process from scratch merely because a given
         circumstance has changed.  However, it is important to carefully
         assess the nature of the change to determine potential ripple
         effects and what needs to be considered as part of revalidation.


         3.  Documentation
         It is essential that the validation program is documented and
         that the documentation is properly maintained.  Approval and
         release of the process for use in routine manufacturing should
         be based upon a review of all the validation documentation,
         including data from the equipment qualification, process
         performance qualification, and product/package testing to ensure
         compatibility with the process.

         For routine production, it is important to adequately record
         process details (e.g., time, temperature, equipment used) and to
         record any changes which have occurred.  A maintenance log can
         be useful in performing failure investigations concerning a
         specific manufacturing lot.  Validation data (along with
         specific test data) may also determine expected variance in
         product or equipment characteristics.

      B. Retrospective Process Validation
      In some cases a product may have been on the market without
      sufficient premarket process validation.  In these cases, it may be
      possible to validate, in some measure, the adequacy of the process
      by examination of accumulated test data on the product and records
      of the manufacturing procedures used.

      Retrospective validation can also be useful to augment initial
      premarket prospective validation for new products or changed
      processes.  In such cases, preliminary prospective validation

      should have been sufficient to warrant product marketing.  As
      additional data is gathered on production lots, such data can be
      used to build confidence in the adequacy of the process.
      Conversely, such data may indicate a declining confidence in the
      process and a commensurate need for corrective changes.

      Test data may be useful only if the methods and results are
      adequately specific.  As with prospective validation, it may be
      insufficient to assess the process solely on the basis of lot by
      lot conformance to specifications if test results are merely
      expressed in terms of pass/fail.  Specific results, on the other
      hand, can be statistically analyzed and a determination can be made
      of what variance in data can be expected.  It is important to
      maintain records which describe the operating characteristics of
      the process, e.g., time, temperature, humidity, and equipment
      settings.11  Whenever test data are used to demonstrate
      conformance to specifications, it is important that the test
      methodology be qualified to assure that test results are objective
      and accurate.

 11   For example, sterilizer time and temperature data collected on
      recording equipment found to be accurate and precise could
      establish that process parameters had been reliably delivered to
      previously processed loads.  A retrospective qualification of the
      equipment could be performed to demonstrate that the recorded data
      represented conditions that were uniform throughout the chamber and
      that product load configurations, personnel practices, initial
      temperature, and other variables had been adequately controlled
      during the earlier runs.

      In some cases, a drug product or medical device may be manufactured
      individually or on a one-time basis.  The concept of prospective or
      retrospective validation as it relates to those situations may have
      limited applicability, and data obtained during the manufacturing
      and assembly process may be used in conjunction with product
      testing to demonstrate that the instant run yielded a finished
      product meeting all of its specifications and quality
      characteristics.  Such evaluation of data and product testing would
      be expected to be much more extensive than the usual situation
      where more reliance would be placed on prospective validation.