Many Roads Lead to Rome
FDA is Not the Sole Arbiter of what Quality System Regulation (QSR) Compliance Requires
United States of America vs. Utah Medical Products, Inc.

Briefed by: Dana Ritchie, Esq., April 2006

Abstract

The U.S. District Court for the District of Utah, Central Division examined the Quality Systems Regulations (QSR) at 21 CFR Part 820 (2004) and found that validation and good manufacturing processes chosen by a company, although different from what the FDA thought appropriate, are not wrong or in violation of the regulations.

History

Utah Medical (UM) is a Utah Company. Defendant Cornwell is the CEO and Defendant Shirley is the VP of QA and Product Development. UM develops, manufactures, and markets, among other things, components for medical devices employing two methods of manufacturing: extrusion and injection molding. Beginning in 2001, the FDA inspected UM and issued written observations on Forms FDA-483. Although,safety was never a concern, the Government alleged that UM failed to comply with applicable QSRs and sought to stop UM from selling product until UM complied with the regulations “and in a manner that has been found acceptable to FDA” (assuming that they are one and the same). During the pendency of the trial, the Government softened the relief it sought to simple regulatory compliance. The Court examined whether UM was currently in compliance (rather than looking at past compliance) and to that end, addressed three issues:

  1. Did UM properly validate its extrusion and injection molding methods? Yes.
    The Government contended that UM had insufficient documentation for validation of its processes and that industry standards required installation qualification, operational qualification and performance qualification. But, it was the company who had to establish and maintain the quality system. The court held that:

    • There was proper validation of UM’s installation of its equipment as the machines had been in place and producing product for many years (UM was founded in 1978, so the equipment was installed sometime thereafter) and manuals and drawings were on hand. In 2004, UM used its own engineers to go through a documented testing process over a period of months to verify the validity of what was done at installation.
    • There was proper calibration of the extrusion and injection molding methods by UM’s outside vendors on a regular, periodic basis, following established, documented procedures.
    • There was proper establishment of process parameters during the design phase of manufacture, adequate monitoring of critical dimensions of products and adequate inspection and testing by which samples of component parts were inspected at regular intervals. All steps were documented.
    • Although the Government asserted that process validation was deficient due to an absence of “edge of failure”, there was no need for UM to establish “edge of failure” testing to demonstrate what does not work, when: engineers had specified what does work by providing design and operational specifications (with operation sheets provided to machine operators with specification limits presented by the engineer, history sheets which were kept, specification sheets, designated temperature ranges, speed of a particular run, pressure, stated directions from suppliers of raw material); measurement by laser beam was available by tubing; and assembled products were visually inspected often more than once before shipping to customers.
  2. Did UM properly validate its software programs used as part of its production or quality system? Yes. The Government asserted that UM failed to document intended uses for each software program, exercised improper sequencing of test protocols and test reports, and the existence of a “Y2K” problem. The Court held that UM presented substantial testimony and documentary evidence that intended use of the software was adequately documented and properly tested. Date discrepancies and Y2K matters were explained by UM.
  3. Did UM properly process complaints with regard to lookback and failure codes? Yes.
    The Government asserted that UM did not assign “failure codes” to complaints in a uniform manner and that UM’s lookback analysis were not performed in a uniform manner. The Court held that UM had established uniform procedures for implementing corrective and preventive action by employing a detailed uniform procedure that it followed for receiving, reviewing, and evaluating complaints. UM used “failure codes” as an internal general trending tool, but grouped complaints by a corrective action number, allowing people to review prior complaints which were the responsibility of one individual. UM held regular weekly meetings to discuss complaints which were periodically reviewed by the CEO.

 

Findings

The District Court rejected the Government’s arguments, denied the Government’s petition for injunction and dismissed the case against UM. The Department of Justice will not appeal the decision.

Conclusions

The FDA attempted to incorporate The Quality Management Systems-Process Validation Guidance and The Quality System Compendium into the regulations, asserting that they are the best sources of information as to current good manufacturing practice. The Court stated that industry publications were not incorporated into the regulations and therefore, other process and validation methods may be used by companies and still comply with the regulations.