Statistical Techniques: Zero for 11
Juran and Deming would be rolling over in their graves if they could read this week’s Devine Guidance. Dr. D came across an offending device establishment that managed to rack up 11 Form 483 observations, including one for statistical techniques. Now for those readers who peruse through the FDA-issued warning letters each week, finding an establishment that is cited for failure to establish statistical techniques is a rare event indeed. Most Chief Jailable Officers (CJO) are increasingly omniscient (look-it-up) about all of the quality, regulatory and statutory requirements needed to successfully play in the U.S. medical device industry sandbox. Unfortunately, some individuals just do not seem to understand one salient word: Compliance (look-it-up). Enjoy!
Warning Letter – March 4, 2016
As mentioned in the opening paragraph, the offending establishment targeted in this week’s guidance was able to collect 11 Form 483 observations during a brief four-day inspection in October 2015. That total, in and of itself, is a pretty impressive feat for any device establishment. Additionally, the establishment wrote 11 responses (one for each observation) to FDA, and FDA responded with 11: “We reviewed your firm’s response and conclude it is not adequate.” Wow, 0 for 11 is bad in baseball, bad from the free-throw line in basketball, and an unmitigated disaster in the device industry. However, the proverbial topping on the Form 483 observation cake, for this offending establishment, was Form 483 Observation Seven for statistical techniques. Failure to comply with 21 CFR, Part 820.250 is almost never cited. Why? Because every single establishment in the world (device industry or any other industry) realizes the value of employing statistics to assist in running businesses.
Warning Letter Excerpt
Observation Seven (7) – “Failure to establish and maintain procedures for identifying valid statistical techniques required for establishing, controlling, and verifying the acceptability of process capability and product characteristics, as required by 21 CFR 820.250. For example:
a) No procedures were established for identifying valid statistical techniques for establishing, controlling, and verifying, the acceptability of process capability and product characteristics.
b) No procedures were established to ensure sampling plans are based on a valid statistical rationale.
We reviewed your firm’s response and conclude it is not adequate. The response indicated your firm created “Measurement and Analysis,” SOP–(b)(4). Additionally, the response indicated your firm’s management representative has been trained on SOP–(b)(4). However, the response did not indicate how your firm plans to identify and train all appropriate staff.”
21 CFR, Part 820.250 – Statistical Techniques
(a) Where appropriate, each manufacturer shall establish and maintain procedures for identifying valid statistical techniques required for establishing, controlling, and verifying the acceptability of process capability and product characteristics.
(b) Sampling plans, when used, shall be written and based on a valid statistical rationale. Each manufacturer shall establish and maintain procedures to ensure that sampling methods are adequate for their intended use and to ensure that when changes occur the sampling plans are reviewed. These activities shall be documented.
Compliance for Dummies
Folks, it is time for a little broken record time from the doctor. As with any quality system regulation (QSR) requirement, the term “establish” (define, document and implement) is a prerequisite for §820.250 that is paramount for complying with any section of the QSR. A well thought out and scripted procedure is mandated by the QSR in support of statistical techniques, including sampling plans. First off, please do not reinvent the wheel, some of the tools that can be employed to support the scripting of a statistical techniques procedure are:
- ANSI/ASQ Z1.4 and Z1.9: Sampling Procedures and Tables Package
- Juran’s Quality Handbook
Essentially, the FDA likes to see that device establishments have employed or considered: (a) sampling plans (e.g., C=0), (b) reliability and confidence intervals, (c) probability, (d) descriptive or inferential statistics, (e) Anderson-Darling (and other tests), (f) tightening and loosening of sample plans premised on previous performance, etc.
Now granted, it is up to each establishment to decide how the stats program is going to be implemented. However, please keep in mind, statistical techniques are not just for the quality inspection areas. The FDA expects statistical techniques (including sample-size rationales) to be clearly defined in design and process verification and validation testing. Simply stated, evidence of statistical methodologies employed better be clearly defined in written test protocols.
Takeaways
For this week’s guidance, the doctor will leave the readers with two takeaways. One: Statistical techniques, when properly applied, are really an establishment’s best friend. The employment of sound statistical concepts can be used to drive all aspects of an establishment’s operation. Two: It is a misnomer to believe applied statistics is just a quality tool to be employed by quality inspection (e.g., C=0 sampling plans). The FDA requires statistics to be used in assessing the capability of processes and product characteristics, too. Remember, documented evidence of compliance will always be a CJO’s best friend during an agency inspection. In closing, thank you again for joining Dr. D, and I hope you find value in the guidance provided. Until the next installment of DG, cheers from Dr. D., and best wishes for continued professional success.
References
- Code of Federal Regulation. (April 2015). Title 21 Part 820: Quality system regulation. Washington, D.C.: U.S. Government Printing Office.
- Devine, C. (2011). Devine guidance for complying with the FDA’s quality system regulation – 21 CFR, Part 820. Charleston, SC: Amazon.
- Devine, C. (2013). Devine guidance for managing key attributes of a FDA-compliant quality management system – 21 CFR, Part 820 Compliance. Charleston, SC: Amazon.
- FDA. (March 2016). Inspections, Compliance, Enforcement, and Criminal Investigations. DMP ,Ltd. Accessed April 13, 2016. Retrieved from http://www.fda.gov/ICECI/EnforcementActions/WarningLetters/2016/ucm497069.htm.
Related Articles
-
Holmes, who was found guilty in January on three counts of investor fraud and one count of conspiracy to commit fraud related to her blood-testing startup, faced up to 20 years in prison for each guilty count.
-
The new guidance is intended to establish confidence in automation used for production or quality assurance systems and describe various methods and testing activities that may be applied to establish computer software assurance and meet regulatory software validation requirements.
-
The updated guidance document clarifies what constitutes a statement of the basis for the deficiency and includes examples of well-constructed deficiencies and industry responses to facilitate a more efficient review process.
-
The revised cybersecurity draft publication is not intended to be a checklist for healthcare organizations to follow, but rather a guide to help them comply with the HIPAA Security Rule.
