IEC 60601-1-8 auditory ALARM SIGNALS requirements, are these requirements mandatory

[errethakbe]

In the IEC 60601-1-8, it can be understand in a way that the requirements are mandatory.
However, in the Subclause 6.3.3.1 of the same standart it is clearly stated that "The MANUFACTURER can provide more than one set of auditory ALARM SIGNALS. VALIDATION by USABILITY testing is not required if each set complies with Table 3 and Table 4 (or Annex F). If additional non-standard auditory ALARM SIGNAL sets (i.e., those that do not comply with Table 3 and Table 4 or Annex F) are provided, they require clinical VALIDATION to ensure that they provide at least an equivalent degree of safety as the standard sounds. Permission to provide non-standard sounds is intended to allow a RESPONSIBLE ORGANIZATION to continue to use non-standard but “historically validated” sound sets that have been successfully used for significant periods of time in their PATIENT care areas, and to ensure that this collateral standard is not excessively design restrictive."

I understand that if following the standart is mandatory unless the auditory alarm is clinically validated and tested for usability.

Could anyone please enlighten me?

 

[Al_Z1]

First of all, usage of alarm signals, compliant to IEC 60601-1-8 is mandatory only, if you use alarm signals to reduce risks, which is stated in IEC 60601-1.
Secondly, yes, last version of IEC 60601-1-8 sets requirements for alarm signals, as i remember, forms of audio signal are divided through risk levels + standard implicates forms for specific alarms, connected with IVD, life support systmes etc. These forms were chosen via clinical investigations and also are "historically validated". But for base auditory signals standards defines only pulse, harmonic frequencies and some other settings, so it is possible to vary these signals in a way. Also there is link in the standard to signal samples, which are fully compliant with it. Link was broken about a year ago, when i tried it, but I found it somwhere in additional materials on TC 62 IEC commitee page, but it was not easy. Archive is attached.
Also, requirements to signal form where changed significantly when IEC 60601-1-8 Amd2:2020 was implemented, and these signals are compliant with it.
After quick review I can add some more info: working link to samples ISO/TC 121/SC 3 - Respiratory devices and related equipment used for patient care - from appendix G of the standard.
New signals creation with examples of validation etc. is described in appendix H of the standard.

 

[yodon]

Effectively, yes (mandatory). @Al_Z1 provides an excellent description. Regulatory bodies may not say that compliance to a standard is mandatory but they have time on their side. If you aren't compliant, they can repeatedly ask for more data until they are satisfied (which usually isn't until you've actually done the testing to show you're compliant).

 

[Tidge]

I have no issues with the principles of 60601-1-8, but I have some personal opinions about the attempt to standardize the ALARM SIGNALS; my sensitivity manifests for ME equipment that will be used in conjunction with other ME equipment. Personally: I'm less motivated to try to specifically use standardized ALARM SIGNALS in use cases where it could be hard for the users to tell which piece of equipment is signaling an ALARM CONDITION.

I haven't worked with a NRTL that is especially comfortable when an ALARM SIGNAL doesn't precisely satisfy the standard... and this gets complicated when a device includes "alarms" (as described in the IFU) but don't actually indicate an ALARM CONDITION.

This can be really quite tricky for a monitored physiological parameter for which there is no absolutely recognized value of unacceptable risk for the parameter (because patients vary, or user preferences during procedures (*1)), yet manufacturers often include functionality to notify users if a parameter crosses a certain threshold... and calls this sort of feature an "alarm". I believe the collateral as written can handle this, but the text of the standard and the vernacular usage (especially in marketing) can lead to some confusion.

(*1) I think there is a subtle point about Physiological Alarm Conditions, although I recognize that I may be making too much about them. My thinking is: for a monitored physiological alarm condition for which there exists unacceptable risk, the risk management file is obligated (6.1.2) to detail the specific/precise levels of risk for specific physiological parameters and mandates the implementation of appropriate alarm signals. Yet doing this would remove the freedom for clinically trained users to vary the notification limits, which seems less than ideal. For example: there may be procedures where the human body temperature is lowered below a value that would be considered safe for some circumstances but is clinically appropriate for some fraction of the procedure duration. It would be a real PITA for a surgical team to have to keep muting an alarm signal while the patient is cold, in contrast it seems totally natural to allow the team to set notification limits on body temperature for whatever reason. There is a subtle interplay between what the manufacturer can design (for an alarm system) per their risk assessment and what a clinical user can decide as far as her own risk assessment; the standard pretty much puts the entire burden for compliance on the manufacturer... at least per my experience with NRTLs.