Unrealistic Packaging Validation Sample Size

indubioush

Quite Involved in Discussions
#11
Bev, thanks for clarifying that. I see there may be cases where n=1 is acceptable; however, I believe it would be very difficult to justify an N of one for a high risk, sterile Class III device.

This is the feedback my company received years ago from US FDA regarding a Class III sterile device:

"In xxx Packaging Test Report – ....
We noted in several places of this report an unusually low sample size (n=10). Please be reminded that a risk analysis should be performed to determine the number of samples to be used during your studies. ....we recommend 95%/90% confidence and reliability."
 
Elsmar Forum Sponsor

Bev D

Heretical Statistician
Staff member
Super Moderator
#13
In validation the 'packaging error' is not tested for. Validation is for the 'design'. there is is simply no meaningful (predictive) way to test for process errors. Instead the intent is to put in mistake proofing, inspection and other controls because errors can happen randomly at any time. What we can validate is that any mistake proofing or inspection are effective through appropriate MSAs.

In response to indubioush's case I really don't understand what people mean when they say 'risk' or 'based on risk'. I know what I mean by risk (high severity NOT probability) and I know how I handle it (well characterized designs with sufficient margin and process controls). I also know that some statisticians and many non-statistician reviewers and auditors have NO idea how to apply statistics in a meaningful way to manufacturing.
 

Ed Panek

QA RA Small Med Dev Company
Trusted Information Resource
#14
What value does your packaging add might be a better question? In some products, the packaging is also functional.
 

Wolf.K

Quite Involved in Discussions
#15
Of course you can have statistical significance (whatever you think that is) with a sample size of 1. (...) There is more than one kind of statistics in the world...
BevD, I wonder what number of tests you would recommend for this situation:

Class III medical device, sterile in aluminium pouch bag (first barrier) and Tyvek (second barrier). 10 of these are placed within a collapsible box and six of the collapsible boxes are shipped within an outer box/transportation box (type: FEFCO ESBO 0201). That means, within one transport box there are 60 medical devices. This packaging combination has been used already for shipment of >100.000 units (= historic data), without any complaints by the customers. For the registration in the USA, we need to present packaging validation data and transport validation data. What number of packages do we have to test? 29 or 60 outer boxes (including 3600 medical devices), or one outer box with 29 or 60 medical devices?

Truly yours,
Wolf
 

ChrisW

Starting to get Involved
#16
you are validating the individual unit shipped in worst case conditions so therefore you only need to test 30 or 60 individual devices.

Chris
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#17
Wolf - first I have to say that I cannot over rule your reviewer/statistician at the FDA. You can make a recommendation and provide the rationalization. First you are not validating the medical devices. The unit of test is the packaging configuration. If you will be shipping 60 devices in a single package then you have 60 devices in a single sample (n=1) of a packaging configuration. When a reviewer/statistician requests that you test more than one sample of a packaging configuration it is an attempt to randomly subject the configuration to a variety of shipping conditions. The routes you choose are more important than the example size. You need your packaging to survive the worst case conditions (short of a truck crash or deliberate attempts to destroy packages) that would deliver your product to the end user. The more deliberate you can be in selecting the worst case conditions the lower your sample size (1 package per condition).

There are other concerns that I cannot predict with your reviewer. If they consider this actual shipment study to be a PQ validation, then they may want to see a larger sample size and that might be statistically based or just “10 is bigger than 1” type of choice.
 

Wolf.K

Quite Involved in Discussions
#18
You need your packaging to survive the worst case conditions (short of a truck crash or deliberate attempts to destroy packages) that would deliver your product to the end user. The more deliberate you can be in selecting the worst case conditions the lower your sample size (1 package per condition).

There are other concerns that I cannot predict with your reviewer. If they consider this actual shipment study to be a PQ validation, then they may want to see a larger sample size and that might be statistically based or just “10 is bigger than 1” type of choice.
I was considering a testing according to ISTA 3A or ASTM D4169. For ISTA 3A, an atmospheric conditioning test block 1 (-29°C uncontrolled RH; 72h) and an atmospheric conditioning test block 2 (+38°C /85% RH,72h, 60°C / 30% RH, 6h) followed by drop/vibration with and without topload, random vibration air, low pressure test and final drop test would be reasonable, as it covers the atmospheric conditions which are important according to our risk assessment. I would test the devices within the boxes (seal strength, dye penetration, sterility and so on) according to ISO 2859 requirements. What I wonder is, how many packages do we have to test? The German DHL shipping company uses three boxes for their in-house certification scheme. Do you think, that 3 boxes including 3x60= 180 medical devices would satisfy the requirements? I would test 59 of the 180 devices for sterility (95%/90% confidence/reliability), and 29 of the devices for each of all other parameters (95%/90%). Based on our past experience with >100.000 units shipped without any issues...

Unfortuantely, I don't know any other quality manager with experience in transport validation. Somehow people seem to "ignore" this issue...

Thanks in advance to any recommendation! :) Have a nice weekend!
 

Wolf.K

Quite Involved in Discussions
#19
you are validating the individual unit shipped in worst case conditions so therefore you only need to test 30 or 60 individual devices.

Chris
Ok, that is, I expose one shipping box including 60 medical devices to the extreme conditions, and then I test all of these devices? But don't I have to test if the shipping box does survive the transports?
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#20
Ok, that is, I expose one shipping box including 60 medical devices to the extreme conditions, and then I test all of these devices? But don't I have to test if the shipping box does survive the transports?
using a standard for testing is more like OQ validation. since you are selecting the test conditions (rather than actual shipping and allowing 'nature' to take it's course). You must test all devices in the package. The individual devices are NOT part of the confidence/reliability sampling plan as they are not being packaged individually for shipment; they are not independent because they are all being subjected to the same stresses/conditions in a single box. There must be no device failures (due to 'damage' from the stresses). The sample size is the number of 'outer boxes'. In this case 1 package of devices is sufficient (n=1), but some people feel more comfortable with a few more packages but this is not statistical it is 'feels better'.
 
Thread starter Similar threads Forum Replies Date
M Is MDR Article 61(5) Unrealistic? EU Medical Device Regulations 11
M Problems with Unrealistic Customer Expectations Customer Complaints 12
K Unrealistic Goals ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 6
M Packaging standards - ASTM vs ISTA Other Medical Device Related Standards 0
M Packaging studies - ISTA vs ASTM CE Marking (Conformité Européene) / CB Scheme 0
F UDI-PI required on packaging (MDR) EU Medical Device Regulations 2
C Packaging Long Parts Manufacturing and Related Processes 3
H Software Validation for FFS Packaging Machine Qualification and Validation (including 21 CFR Part 11) 1
G Special 510K possible? Only the packaging has changed Other US Medical Device Regulations 0
B Setting flexible packaging specification limits ISO 14971 - Medical Device Risk Management 4
V Packaging in Distributor Practices EU Medical Device Regulations 0
R Green dot required on packaging? Medical Device and FDA Regulations and Standards News 2
shimonv Single lot release for sterile packaging 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 3
R Applicable ISO standards - packaging and others Medical Device and FDA Regulations and Standards News 4
A FDA guidance on non-sterile Medical Device Packaging Medical Device and FDA Regulations and Standards News 7
C 21 CFR 803 malfunction detected during packaging - report? Other US Medical Device Regulations 1
E Latex in packaging Other Medical Device Related Standards 4
H Expired packaging material - Expired unused Tyvek pouches Other Medical Device and Orthopedic Related Topics 2
A GUDID - Packaging Levels 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 8
A Packaging device with accessories already on the market Other Medical Device Regulations World-Wide 0
T Root Cause Failure Analysis - Not following Customer packaging Specification Problem Solving, Root Cause Fault and Failure Analysis 9
Q EU Rep for Packaging waste, WEEE and Battery disposal CE Marking (Conformité Européene) / CB Scheme 0
L AS5553 Clause 3.1.7 e "Control packaging material ..................reused" AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 0
L Packaging unsterile medical devices ISO 13485:2016 - Medical Device Quality Management Systems 6
M MDR - Packaging - Secondary Medical Device Boxes EU Medical Device Regulations 2
Y Packaging validation for non-sterile Medical Equipment Other Medical Device Related Standards 1
M UDI size on packaging EU Medical Device Regulations 3
M UDI placement - Packaging levels EU Medical Device Regulations 2
E Unit Dose Packaging Pharmaceuticals (21 CFR Part 210, 21 CFR Part 211 and related Regulations) 0
K Packaging and Damaged Parts returned from a Service Provider (Anodize) - Who is Responsible? Manufacturing and Related Processes 20
B 510(k) clearance claim on medical device product packaging Other US Medical Device Regulations 6
E Packaging Specifications Mercedes-Benz U.S. International IATF 16949 - Automotive Quality Systems Standard 4
T Colour of symbols on Medical Device labelling and packaging Other Medical Device Related Standards 1
R Labels on sterile packaging EU Medical Device Regulations 5
A Interpretation of Article 16 (2b) - Packaging, samples and Certificate EU Medical Device Regulations 10
I Medical Device Packaging Multilingual Labeling EU Medical Device Regulations 3
N Process Flow diagram steps for inspection and packaging APQP and PPAP 4
C ISO 7010 m002 mandatory action - Color requirements on Packaging IEC 60601 - Medical Electrical Equipment Safety Standards Series 1
shimonv GTIN Number on the Packaging of a Serviced Medical Device EU Medical Device Regulations 0
shimonv Is UDI required on the Packaging of a Serviced/Repaired Medical Device? Other US Medical Device Regulations 3
C QR Code on Medical Device Packaging and Notification to Notified Body EU Medical Device Regulations 1
TheMightyWife Address Change - Medical Device Packaging Labelling Requirements EU Medical Device Regulations 7
E Use of FDA Cleared logo on packaging US Food and Drug Administration (FDA) 8
V Tyvek packaging for Medical device process validation Qualification and Validation (including 21 CFR Part 11) 2
S Symbols - Device label versus Packaging label ISO 13485:2016 - Medical Device Quality Management Systems 1
Y Sterile device packaging under MDR EU Medical Device Regulations 3
A Packaging seal inspection Inspection, Prints (Drawings), Testing, Sampling and Related Topics 2
A Capability Analysis for Packaging Seal Strength with spec. >0.1 Kgf using Minitab Using Minitab Software 6
A Medical Device Class 2A, non sterile bodyworn packaging requirements CE Marking (Conformité Européene) / CB Scheme 7
N Medical device non sterile packaging requirement CE Marking (Conformité Européene) / CB Scheme 1

Similar threads

Top Bottom