Device functionality over service life - Objective evidence required?

[david316]

Hello,

If a medical device has key performance attributes is there any requirement that the manufacturer must prove the device is capable of meeting said performance limits throughout the devices service life? My understanding is that a manufacturer must prove that essential performance and basic safety must be maintained over the service life but I am unsure about other performance limits (e.g. limits documented in the manual) which are narrower than the essential performance limits?

e.g.
Manual states oxygen is controlled to +/- 1 lpm but essential performance is determined to be +/- 5 lpm.

Thanks for any guidance.

 

[indubioush]

Often times this type of testing is dictated by the standard that is specific to that type of medical device. Is there a standard that is specific to your type of device?

 

[david316]

The general standard IEC 60601-1 only states the the expected service must be specified by the manufacturer and that during this time the device must remain safe (i.e. essential performance and basic safety is maintained). The particular standard then clarifies the essential performance requirements. There is nothing I can find that covers off whether a manufacturer needs to prove a device performs to its stated specifications (which may be tighter/different to the essential performance limits) over the service life.

 

[indubioush]

It is all about risk. You must determine what actions are needed to mitigate risk associated with poor reliability.
Also read this:
Exploring Medical Device Reliability and Its Relationship to Safety and Effectiveness - IEEE Journals & Magazine

 

[david316]

So if via the risk assessment you document that there are no unacceptable risks related to reliability and that the residual risk is acceptable you don't need to prove the product functions as intended over the service life?

 

[david316]

Any other input would be very much appreciated.

Thanks

 

[indubioush]

There is not enough information to delve in fully. We don't know what type of device you have; we don't know all the product requirements; we don't know if you plan on servicing your device, etc. If you are the manufacturer, you are in the best position to determine what testing needs to be done to ensure device reliability and safety.

 

[david316]

I guess I should have asked if there are any regulations or standards that require that a manufacturer prove that an electrical medical device performs as intended over it's service life. My understanding is 60601-1 requires a manufacturer to state the devices service life over which essential performance and basic safety is expected to be maintained. I would expect accelerated life testing to prove this. But as far as I know there is no requirement to prove a medical device operates as per it's specs with accelerated life testing or that it even needs to be justified. You need to conduct risk management as per 14971 and identify any risks associated with the device which would include the device falling to operate within its spec and make the risks acceptable. But this does not mandate that testing must be conducted to prove a device performs as intended over it's life. Servicable or not.

So the questions is...is it acceptable to prove as per 60601-1 that a medical device maintains essential performance and basic safety over it's service life as well as conducting risk management as per 14971 or are there regulations or expectations from regulatory bodies that additional testing be conducted?

 

[Tidge]

In my experience (with long-life medical devices that receive service visits) 'accelerated life testing' is not sufficient to speak to "intended service life". Accelerated life testing can inform activities necessary to establish for the support a device over its expected service life.

It is necessary that a medical device remain safe over its entire expected service life, which is where the terminology about maintaining essential performance and basic safety comes from. I'm ignorant of any particular regulatory expectations, but manufacturers of certain devices may have 'regulatory attention' (e.g. transmission of pathogens from Endoscopes and Cooler-Heaters used in surgical environments... the sterilization/cleaning of those devices may be considered a 'service life' dimension of those products).

That being said; a regulatory body may look for risk analysis around certain elements of any particular design with an eye towards expected service life. For example: if a device includes batteries which must be serviced it seems appropriate to me that a regulatory authority may want to examine the data used to establish the expected lifetime of the batteries as well as the service procedures for battery replacement.

 

[david316]

Thanks. Assume the device doesn't require routine service. Maybe an example. Consider a Forehead thermometer. The manual states that the accuracy is +/- 0.5 Fahrenheit (+/- 0.3 degrees C). When determining essential performance as per 60601-1 the essential performance is determined to be +/- 1.0 Fahrenheit. As part of V&V, life testing is conducted to prove that the device meets essential performance and basic safety over its service life. Prior to the test the device is accurate to +/-0.2 Fahrenheit. At the end of life testing the device operates as per normal but is only accurate to +/- 0.8 Fahrenheit. Hence the device still meets essential performance and by definition the risk is acceptable. Is this acceptable?

Also, assume there is no instructions in the user manual to check or do periodic calibration.