New unit proposal: dB(HL)

[Swapna Abhyankar]

I would like to propose a new unit of measure for audiogram results that are typically reported as decibels hearing level, or dB HL. We are creating a set of LOINC codes for audiometry measurements and would like to include dB(HL) as the example UCUM units of measure.

Thank you!

[Gunther Schadow]

We currently have

• dB(SPL) - "sound pressure level"
• db(A) - using the A-scale of loudness - correction: we don't have that, we only have it mentioned in the text as an open question.

how is dB(HL) defined and defined differently to them?

Compare and contrast.

PS: we should elicit more such details from the submitters, who usually should be more expert in the field they are proposing. In this case submitter might not know. But we need more info about it.

I could find ​http://www.thieme.com/media/samples/pubid1260387312.pdf stating:

"Decibels in hearing level, or dB HL, is commonly used in audiology because it refers to the decibel level on the audiometer. The reference level for dB HL is “0,” which is related to the average threshold in decibels sound pressure level (dB SPL) for the average, normal-hearing listener."

I remember this now darkly from physiology class. There was something like Stevens scales. And what had impressed me at the time was that the perception threshold was used sort of as the unit. But then there was the other unit, which was the discrimination threshold, i.e., the smallest difference in objective physical quantity which the humans are able to discern.

That article mentions the following

• dB(SPL)
• dB(IL)
• dB(HL)

---

More text from that PDF:

What Is the Difference between dB HL and dB SPL?

The decibel ( dB ) is a logarithmic unit of measurement used to express the magnitude of a sound relative to some reference level. Decibels in hearing level, or dB HL , is commonly used in audiology because it refers to the decibel level on the audiometer. The reference level for dB HL is “0,” which is related to the average threshold in decibels sound pressure level (dB SPL) for the average, normal-hearing listener. In Fig. 1.1, the solid black line represents the average auditory threshold in dB SPL at each audiometric frequency. Each threshold in dB SPL, which is noted in the table at the bottom of Fig. 1.1, is equal to 0 dB HL on the audiometer for the corresponding frequency.

Decibels in sound pressure level, or dB SPL , refers to the magnitude of the displacement of molecules in the air. The reference for dB SPL is 20 micropascals (20 μPa) or 0.0002 dynes/cm 2 . Because it is easy to measure dB SPL with a condenser or free-field microphone coupled to a sound level meter, sound measurements are often expressed in dB SPL.

What Is the Difference between dB IL and dB SPL?

As stated in the previous section, a common method to express the magnitude of a sound is in decibels sound pressure level or dB SPL, for which the reference is 0.0002 dynes/cm 2. A less common method to express sound intensity is in acoustic power by using decibels intensity level, or dB IL . The reference for dB IL is 10 −16 W/cm 2 . According to Gulick, Gescheider, and Frisina (1989) , “a dangerously intense sound would represent a power of only about 0.0024 watt” (p. 46). As a result, it is cumbersome to measure the incredibly small amounts of power associated with the enormous range of intensities that the human ear can hear. Note in Table 1.1 that the dB values are the same regardless of whether they were calculated from intensity (power) or pressure; however, a 100-fold increase in intensity is equal to a 10-fold increase in pressure.

---

So, given how we define the various kinds of dB scales, here is where we are:

• we specify the bel B as the base unit (and magically dB will behave like dm to m)
• we only specified various bel units based on existing units, in case of the dB(SPL) it is:
  • bel sound pressure
  • B(SPL)
  • defined as 2lg(2 10*-5.Pa)

where "2lg" is the decadic logarithm times two.

I confess the whole dB matter always confuses me. Perhaps Christof should look this over and check my work.

But the bottom line so far is: we have no precedence of adding any sort of correction curves here, likewise, we don't have things like dB_i vs. dB_d for the comparison with isotropic antenna vs. dipole.

The issue with all dB scales is that they are some kind of comparison with something else. If that something else is a constant linear value, like a millivolt, then there is no problem. But if that something is some kind of curved empirically established phenomenon, then it gets more complicated, and we have just not addressed that at all yet.

I propose to address it, we should create a sub-group to study how to do it.

[Brenée Mitchell]

From an audiologist's perspective, dB HL (used with standard air/bone conduction audiometry) and dB nHL (used with electrophysiologic measures) are both commonplace clinically.

dB SPL is frequently used in industrial/occupational audiology to assess the noise levels in a given environment.

Here is a simplified conversion chart from SPL to HL ​https://ec.europa.eu/health/scientific_committees/opinions_layman/en/hearing-loss-personal-music-player-mp3/figtableboxes/table-2.htm

Another resource that explains the different applications of the three reference scales (SPL, HL, nHL). ​https://www.interacoustics.com/guides/basics/the-variety-of-decibel-basics

Ultimately, it all points back to the Bel. If we have SPL, then it would seem prudent to add HL and nHL, even though all three are reference scales, rather than distinct units of measure.