Dimeski v Victorian WorkCover Authority & Ors [2025] VSC 165

(a) an order in the nature of certiorari quashing the 24 September 2023 opinion ('the opinion') of a medical panel ('the Panel') comprising the second and third defendants; and

(b) an order in the nature of mandamus remitting the medical questions in respect of which the opinion was given to a differently constituted medical panel to be reconsidered according to law.

2 Mr Dimeski worked for South Pacific Tyres between 1973 and 1997. On 18 September 2021 Mr Dimeski made a claim under the Accident Compensation Act 1985 (Vic) ('ACA') for an impairment benefit alleging noise induced hearing loss. The Victorian WorkCover Authority ('VWA') accepted liability for Mr Dimeski's noise induced hearing loss ('NIHL') but assessed his whole person impairment at 0%. Mr Dimeski disputed the assessment of whole person impairment and so medical questions were referred to the Panel.

Question a): What is the worker's degree of permanent whole person impairment resulting from the accepted injury/s assessed in accordance with Division 4 of Part 2?

Answer: The Whole Person Impairment calculated in accordance the Act is 0%. The degree of impairment is permanent.

Question b) Does the worker have an accepted injury, which has resulted in a total loss injury mentioned in the table in Section 221?

4 This application for judicial review boils down to two questions:

(a) Was the Panel required to calculate Mr Dimeski's percentage hearing loss using audiogram results for 2 kHz, 3 kHz and 4 kHz, rather than just 3 kHz and 4 kHz? Mr Dimeski says it was. The VWA says it was not.

(b) Did the Panel give adequate reasons for its determination not to use the 2 kHz audiogram results in its calculation of Mr Dimeski's percentage hearing loss? Mr Dimeski says they did not. The VWA says they did.

5 For the reasons that follow, Mr Dimeski's application for relief should be dismissed.

6 Section 91(4) of the ACA provides that the percentage of diminution of hearing must be assessed in the manner approved by the Minister. On 25 October 2018 the Minister approved a method for the purposes of s 91(4) which relevantly required determination of compensable loss using:

(a) the Guidelines for the Assessment of Noise Induced Hearing Loss (Revised June 2017) by the Victorian Section of the Australian Society of the Otolaryngology Head and Neck Surgery ('ASOHNS Guidelines' or 'the guidelines'); and

(b) the Improved Procedure for Determination of Percentage Loss of Hearing (1988 edition or later prescribed edition) published by the National Acoustic Laboratory ('the NAL Procedure').

7 Section 2 of the ASOHNS Guidelines specifies that the percentage loss of hearing is determined from the NAL Procedure and then converted to a whole person impairment under, relevantly, the ACA.

8 Sections 5 and 6 of the ASOHNS Guidelines are particularly relevant for the determination of Mr Dimeski's claim.

9 Section 5 of the guidelines provides:

5. INTERPRETATION AND PREPARATION OF THE MEDICAL OPINION

The lowest (least hearing loss) reliable thresholds obtained (including CERA) should be used as the basis for compensation.

Noise induced hearing loss (NIHL) is caused by prolonged exposure to loud noise and produces a characteristic bilateral sensorineural (S/N) loss pattern on pure tone audiometry. It is slowly progressive and ceases when the noise exposure stops. The early loss and the fastest rate of deterioration are at 3, 4 & 6 kHz. As the noise induced loss becomes more severe 2 kHz can be involved and after a period of 30 years of exposure, 1500 and even 1000Hz to a small degree as well.

Therefore the amount of loss increases as the frequency rises with a down slope to a maximum at one or more of 3, 4 and sometimes 6 kHz and improves at 8 kHz although this can also deteriorate with increasing exposure. 6 kHz variable, sometimes equal or worse than 4 kHz, sometimes on an upslope to 8 kHz.

The characteristic loss therefore is a reasonably symmetrical bilateral V or U shaped S/N loss maximal at 3-6 kHz, with the greatest loss often being at 4 kHz.

The contribution from various employers and non-occupational noise can be estimated, if required, from serial audiograms or in the absence of these from duration of exposure. It is assumed that the NIHL occurs on an equally cumulative basis over the total period of noise exposure. Occasional non-occupational exposure can be disregarded unless there is an asymmetry, which can be attributed to a predominately one-sided exposure such as sporting shooting.

The maximum loss that can occur with a NIHL at any frequency is 70 dB. This is further discussed under Presbycusis (see 6 a.).

The NIHL pattern of hearing loss can also result from pathology other than noise exposure. The diagnosis therefore cannot be made just from the shape of the audiogram.

10 Section 6 deals with allowances for non-compensable components of hearing loss. It deals with:

(a) age related hearing loss (presbycusis);

(b) where only one ear is affected by NIHL (unilateral compensable hearing loss);

(c) where there is a combination of NIHL and constitutional sensorineural loss; and

(d) where there is compensable sensorineural loss with a non-compensable conductive loss.

11 Only sections 6a and 6c were relevant to Mr Dimeski's claim and there is no suggestion that the Panel made any error in its application of the guidelines in 6a.

12 Section 6a and c of the ASOHNS Guidelines provide:

In many cases allowance will need to be made for non-compensable component(s) of the hearing loss, which may include:

...

In cases where the combined effect of a NIHL and presbycusis exceeds 70 dB at any of the compensable frequencies (see 5 b.), the presbycusis component may contribute to the excess over 70 dB. This can be allowed for by either disregarding the 70 dB limit if the presbycusis correction allows for the excess or by applying the 70 dB limit and not allowing for presbycusis.

This is done by first calculating the total loss over the compensable frequencies and then subtracting the presbycusis correction to get a potential compensable loss. Secondly calculate the loss over the compensable frequencies, limiting the maximum loss at any frequency to 70 dB.

Take the lesser of these 2 amounts as the compensable hearing loss. The second method allows for cases where the excess over 70 dB is greater than that allowed for by presbycusis.

...

c. Combined NIHL and constitutional sensorineural loss

In audiograms where the low frequency component is greater than that consistent with typical NIHL, it must be considered whether the constitutional disorder has overridden any NIHL component.

In other words, would the hearing loss have been the same whether or not there had been any noise exposure? Earlier audiograms may be helpful if they are available.

In cases where there is a definite down slope in the audiogram and a sufficient history of noise exposure to suggest that a component of the loss is noise induced, use either 2, 3 and 4 kHz or 3 and 4 kHz depending on the configuration of the graph to determine the compensable component.

In general use 2, 3 & 4 kHz. However in cases where 2 kHz is similar to 3 & 4, and 1500 Hz is also well down, it is likely that the non-compensable factor is substantially affecting frequencies up to 2 kHz. Only 3 & 4 kHz should then be used for compensation purposes.

In cases where one ear has a typical NIHL pattern, it should be used as the index for compensation for both ears where there is an additional non compensable component in the other ear or the asymmetry at any of 2, 3 and 4 kHz exceeds 20 dB, unless it is definite that one ear is much more exposed than the other, such as in occupational firearm use.

Completely flat, up sloping and mid frequency ("cookie bite") losses are inconsistent with NIHL.

(emphasis added)

13 For ease of reference, I will refer to 'where 2 kHz is similar to 3 & 4' in the second sentence of the bolded portion above as 'the similarity condition' and 'where ... 1500 Hz is ... well down' in the same sentence as the '1.5 kHz condition'. The guidelines and the Reasons sometimes express frequencies in kHz and sometimes in Hz. For the sake of consistency, I will use kHz throughout except where quoting directly from those two sources.

14 The NAL Procedure provides a series of tables at particular frequencies which give values of percentage loss of hearing corresponding to given hearing threshold levels in the better and worse ears. So, for example, at 3 kHz if the hearing threshold level in the better ear is 60 dB and the hearing threshold level in the worse ear is 75 dB (sometimes represented with the shorthand (60/75)), the NAL Procedure table gives a 6.6% loss of hearing.

15 There are also tables for determining hearing loss in one ear only (not relevant here) and a presbycusis correction table which provides that for a male of 72 years (the age and gender of Mr Dimeski when he was tested), the relevant correction is 6.2%.

16 The Panel provided their opinion on 24 September 2023 and provided reasons ('the Reasons').

17 In the Reasons, the Panel summarise Mr Dimeski's employment history, noting that he worked for a total of 25 years in a noisy environment.

18 The Panel conducted hearing tests under appropriate Australian standard conditions. Based on those tests, the Panel found Mr Dimeski 'to have a mild bilateral high frequency sensorineural hearing loss, worse on the left side consistent with constitutional degeneration of the inner ear as well as a high frequency component due to a noise induced hearing loss'.[2]

(a) considered its hearing tests results were accurate and reliable;

(b) said it had used the ASOHNS Guidelines and the NAL procedure in assessing Mr Dimeski's hearing loss;

(c) considered the hearing loss at the frequencies of 3 kHz and 4 kHz represent the loss due to occupational noise exposure; and

(d) had used the lowest (least hearing loss) reliable thresholds as the basis for compensation.

20 Attached to the Reasons is a graphical representation of the results of the Pure Tone Audiogram test on which the Panel relied. It is reproduced below:[3]

21 In order to aid the interpretation of the graphical representation the VWA provided the results of the test in a tabular form. The information in that table is reproduced below:[4]

22 The Reasons discuss why the Panel determined to use only the 3 and 4 kHz frequencies to calculate Mr Dimeski's occupational hearing loss.

In this respect, the Panel considered the instructions in Point 6c of the [ASOHNS Guidelines]: "In audiograms where the low frequency component is greater than that consistent with typical NIHL (noise induced hearing loss), it must be considered whether the constitutional disorder has overridden any NIHL (noise induced hearing loss) component."

The Panel noted the configuration of the hearing loss on the audiogram of 30 August 2023, noting in particular the hearing loss at 1000Hz, 1500Hz and 2000Hz and considered that the worker's hearing loss did comply with the further instruction in Point 6c: "In general use 2, 3 & 4 kHz. However in cases where 2 kHz is similar to 3 & 4, and 1500 Hz is also well down, it is likely that the non-compensable factor is substantially affecting frequencies up to 2 kHz. Only 3 & 4 kHz should then be used for compensation purposes."

The Panel is of the opinion that based on the pattern of the audiogram the constitutional factor has overridden any noise induced hearing loss component at 2000 Hertz and below; as such based its assessment on the frequencies of 3000 Hertz and 4000 Hertz only as the frequencies affected by occupational noise.[5]

23 The Panel then considered age related hearing loss. Utilising the two methods set out in 6a of the ASOHNS Guidelines the Panel:

(a) calculated a hearing loss of 13.3% using the method where the 70 dB limit is applied. In terms of the NAL procedure, the relevant input figures for this calculation were (60/70) at 3 kHz and (65/70) at 4 kHz;

(b) calculated a compensable binaural hearing loss percentage of 7.3% using the method where loss is calculated using the measured results and a presbycusis correction is deducted. In terms of the NAL procedure, the relevant input figures for this calculation were (60/75) at 3 kHz and (65/75) at 4 kHz giving a binaural hearing loss of 13.5% from which was deducted a 6.2% presbycusis correction; and

(c) consistent with the ASOHNS Guidelines, utilised the lower of these two figures.

24 The Panel determined a further 1% impairment for tinnitus should be added to the NAL assessment, bringing the total compensable impairment to 8.3%. Under the ACA this equates to a whole person impairment of 0%.

25 Mr Dimeski says the Panel got it wrong using only 3 and 4 kHz to assess his hearing loss. He says that the 2 kHz results are not 'similar' to the 3 and 4 kHz results and that the conclusion that the loss at 1.5k kHz is 'well down' is not available. In other words, Mr Dimeski contends neither the similarity condition, nor the 1.5 kHz condition are satisfied. Alternatively, he submits that it is sufficient if he shows that one of the similarity conditions or the 1.5 kHz condition is not satisfied. As a result, Mr Dimeski contends the Panel should have used 2, 3 and 4 kHz figures for their calculations.

26 Based on the measurements in the test and the methodology set out in the NAL procedure, taking into account the 2 kHz figures would have resulted in a total compensable hearing loss impairment greater than 10% and, as a result, a non-zero whole person impairment.

27 Mr Dimeski also relied on the shape of the audiogram, in particular its downward sloping nature which was said to be consistent with what section 5b of the AOSNHS Guidelines describes as a typical audiogram showing NIHL. That matter can be disposed of quickly. The AOSNHS Guidelines specifically state that diagnosis cannot be made just from the shape of the audiogram. Whether the audiogram is typical or not of an audiogram showing NIHL (and the VWA did not concede it was typical) does not assist in determining whether the Medical Panel has made an error of law.

28 The VWA says that the Panel was required by the AOSNHS Guidelines to consider whether any constitutional disorder had overridden NIHL. It contends that the Panel made a medical judgement that constitutional factors had overridden NIHL at frequencies of 2 kHz and below. To the extent that the Panel determined that the 2 kHz results were 'similar' to the 3 and 4 kHz figures and that the 1.5 kHz result was 'well down', that determination was a medical judgement for which the Court should not substitute its own view.

29 In the alternative to his main ground for review, Mr Dimeski says the Panel erred in that it failed to give adequate reasons for the conclusion that the 2 kHz result was not to be included in its assessment with no identification that the results for 2 kHz were similar to those for 3 kHz and 4 kHz and no finding that 1.5 kHz was also 'well down'.

30 In relation to this ground, Mr Dimeski also relied on the fact that two other audiologists whose reports were before the Panel, had assessed hearing loss using 2, 3 and 4 kHz. He relied on Wingfoot Australia Partners Pty Ltd v Kocak ('Wingfoot')[6] where it was said that:

The nature of the question referred to a Medical Panel, and the way that question was addressed by other medical practitioners in opinions supplied to a Medical Panel, might allow an inference to be drawn, on the balance of probabilities in a particular case, that the reasoning in fact adopted by a Medical Panel in arriving at its own differing opinion is not adequately reflected in its written statement of reasons. An inference might be drawn, for example, that the reasoning involved one or more steps not reflected in the written statement of reasons either at all or in sufficient detail to allow a court to see whether a Medical Panel made an error of law in those steps.[7]

31 In relation to this inadequacy of reasons ground, VWA said:

(a) that properly analysed, Mr Dimeski's contention regarding the inadequacy of reasons was just a repeat of his primary contention;

(b) there was no obligation on the Panel to explain why it disagreed with another practitioner's assessment relying on the immediately preceding paragraph from Wingfoot:

A Medical Panel explaining in a statement of reasons the path of reasoning by which it arrived at the opinion it formed is under no obligation to explain why it did not reach an opinion it did not form, even if that different opinion is shown by material before it to have been formed by someone else.[8]

(c) that when regard is had to authorities regarding the way in which a panel's reasons are to be assessed, there was nothing inadequate regarding the Reasons. In particular, there was no part of their reasoning which was obscured or not sufficiently explained.

32 The Court has had occasion to recently consider the application of the ASOHNS Guidelines in three cases.[9] All three raised issues relating to the allowance which should be made under the guidelines for tinnitus. No issue regarding tinnitus arises here. In Gray v Victorian WorkCover Authority & Anor ('Gray'),[10] Forbes J set out the accepted approach to interpretation of the guidelines:

The construction and interpretation of the ASOHNS Guidelines and, insofar as it is relevant, Chapter 9 of the AMA Guides does not proceed as if those documents were legislation. The relevant legal question is whether the assessment of hearing loss has been undertaken 'in accordance with' the AMA Guides, approved procedures and guidelines. It is the task of medical practitioners to construe and apply those documents. As was said in Vicinity Centres PM Pty Ltd v Arik, confirming earlier statements in Gamble v Emerald Hill Electrical Pty Ltd:

...the Act requires the Panel to apply the Guides using members' individual and collective skill and expertise as medical practitioners and that their approach to the Guides is not to be supplanted merely because the court comes to a different view as to how it might be interpreted. It is necessary to go further and show that the methodology is not in accordance with the Guides, being a document intended for application by medical practitioners.[11]

33 It is also worth noting that in Gray the Court rejected an argument that the medical panel in that case had impermissibly utilised a mixture of air conduction and bone conduction test results (including utilising different methods for the left and right ear at the same frequency in two instances) in calculating impairment. Her Honour reasoned that the ASOHNS Guidelines referred to the lowest reliable threshold not the lowest reliable method and said that choosing the lowest reliable threshold was a matter of medical judgement.

34 Mr Dimeski says the Panel could not have concluded that the similarity condition and the 1.5 kHz condition were met. It is not apparent however, that even if he establishes that proposition that the ASOHNS Guidelines require the use of the 2 kHz figures.

35 When regard is had to the language of the guidelines and the principles of construction to be applied, it is plain that despite how it is phrased, the following portion of the guidelines:

In general use 2, 3 & 4 kHz. However in cases where 2 kHz is similar to 3 & 4, and 1500 Hz is also well down, it is likely that the non-compensable factor is substantially affecting frequencies up to 2 kHz. Only 3 & 4 kHz should then be used for compensation purposes.

does not create a dichotomy which covers the field of possibilities between cases in which the similarity condition and the 1.5 kHz condition are met and all other cases.

36 Three aspects of the content of the ASOHNS guidelines, in my view lead to the conclusion that a panel is not left with a binary choice between:

(a) where the similarity condition and the 1.5 kHz condition are met, using 3 and 4 kHz; and

(b) in every other case, using 2 kHz, 3 kHz and 4 kHz.

(a) the provisions of 5b of the guidelines make plain that in some circumstances it will be appropriate to take into account hearing loss at 1 kHz and 1.5 kHz frequencies;

(b) the introductory words of 6c state that in cases 'where the low frequency component is greater than that consistent with typical NIHL, it must be considered whether the constitutional disorder has overridden any NIHL component'.[12] This is, it seems, the primary consideration for a panel under 6c; and

(c) the words 'in general' in the passage above are consistent with other possibilities than the use of 2 kHz, 3 kHz and 4 kHz (where at least one of the similarity condition and the 1.5 kHz condition is not met).

38 In other words, circumstances where the similarity condition and the 1.5 kHz condition are both met constitute one exception to the 'general' rule but I am not persuaded that properly construed they are the only exception under the guidelines, in particular:

(a) in certain circumstances, a medical panel may be justified in having regard to 1 kHz or 1.5 kHz results in its calculation; or

(b) there may be circumstances where the panel is satisfied that the constitutional component has overridden any NIHL component, even though one or both of the similarity condition and the 1.5 kHz condition are not met.

39 However here, though in my view they do not say so expressly, the Reasons appear to record that the similarity condition and the 1.5 kHz condition were met. The Panel noted the hearing loss at 1, 1.5 and 2 kHz and said it considered that Mr Dimeski's hearing loss 'did comply with the further instruction in point 6c' before setting out the portion of the guidelines quoted in paragraph 35. This was immediately followed by the conclusion that constitutional factors had overridden NIHL at 2 kHz and below. The VWA says this passage of the Reasons constituted an express finding that the similarity condition and the 1.5 kHz condition had been met. Whether the Reasons constitute an express finding that those conditions were met or it is more accurately described as a necessary implication, it is my view, that a fair reading of the Reasons is that the Panel found that the similarity condition and the 1.5 kHz condition were met.

40 The question whether 1.5 kHz condition has been satisfied must involve a medical judgement. It is not, as Mr Dimeski submits, a straightforward mathematical exercise. What constitutes 'well down' in the context of the testing is a matter where the members of the Panel were required to apply their individual and collective skill and expertise as medical practitioners.

41 Mr Dimeski contends that, whatever view one takes of the judgement to be made regarding the 1.5 kHz condition, the similarity condition involves no particular medical judgement, 'similar' is a word which bears no medical meaning and a judgement can be made as to whether two numerical quantities are similar on a mathematical basis. In his written submissions in reply, Mr Dimeski said that the 2 kHz threshold was 50 dB and the 3 kHz threshold was 65 dB - a difference of 30%, which on no view could result in the two quantities being described as similar. The difference, it was said, was significant.

42 In fact though, the picture as to percentage differences is more complicated. First, it should be noted the figures used are only air conduction results for the right ear and that the 65 dB figure at 3 kHz is not a figure the Panel used. For the left ear, the 3 kHz results used by the committee (75 dB) range between 15% and 36% greater than the 2 kHz results and the 4 kHz results (65 dB) range between 0% and 18% greater than the 2 kHz results. For the right ear, the 3 kHz results used by the committee (60 dB) range between 9% and 20% greater than the 2 kHz results and the 4 kHz results (75 dB) range between 36% and 50% greater than the 2 kHz results. I accept that in some contexts a difference of 30% might be regarded as significant but even accepting that is so, figures like 0%, 15% and 9% may not be.

43 Further, comparing two results to assess similarity without considering the range of results would be to potentially fall into error. Two is 100% greater than one but if the scale is from 0 to 100, a result of two and a result of one may well be described as similar. In a range from 0 to 120 it might be possible to describe numbers between 50 and 75 as mid-range and so similar for that reason. That conclusion might be different if a scale is logarithmic rather than linear.

44 The above shows why the question of whether two quantities are similar, even if approached from a mathematical standpoint, cannot be assumed to be answered by simply expressing one quantity as a percentage of the other. However, the purpose of the assessment of similarity here is ultimately a medical one, it is to determine whether constitutional factors have overridden NIHL at lower frequencies. It is plain, therefore, that similarity cannot be judged just by considering one form of a mathematical relationship between two quantities without taking into account the context.

45 Whether the similarity condition is satisfied, involves a medical judgement. Mr Dimeski does not adduce evidence of the kind referred to by Niall JA (as his Honour then was) in City of Melbourne v Neppessen,[13] to demonstrate that a reasonable medical panel could never have been satisfied that the similarity condition was met. Mr Dimeski has not shown that the Panel was required to use the 2 kHz figures in its assessment of Mr Dimeski's hearing loss.

46 Mr Dimeski's first point regarding the inadequacy of the Reasons is that the Panel failed to give adequate reasons for the conclusion that the 2 kHz result was not to be included in its assessment because it did not identify that the results for 2 kHz were similar to those for 3 kHz and 4 kHz and no finding that 1.5 kHz was also 'well down'.

47 For the reasons I have set out above, I am satisfied that whilst the Reasons do not expressly make those findings, it is a necessary implication arising from the Panel's determination that Mr Dimeski's hearing loss 'did comply with the further instruction in point 6c' where the similarity condition and the 1.5 kHz condition are set out. As a result, I am not satisfied that the Reasons suffer this particular inadequacy for which Mr Dimeski contends.

48 Mr Dimeski points to the fact that other audiologists utilised 2 kHz results in their assessment of Mr Dimeski's hearing loss. He says that that difference of approach allows me to conclude on the balance of probabilities that the Reasons 'involved one or more steps not reflected in the written statement of reasons either at all or in sufficient detail to allow a court to see whether [the Panel] made an error of law in those steps.'[14]

49 It is evident that the measured results of the other audiologists differ in almost every respect from those obtained by the Panel. The conclusions which those other audiologists drew regarding whether the similarity condition and the 1.5 kHz condition were met or not met, therefore, involved comparisons of almost completely different sets of figures. Further, those reports provide no explanation as to why those audiologists regarded their results as not meeting the similarity condition or the 1.5 kHz condition or both. In those circumstances, it is not possible to draw any conclusion of the kind for which Mr Dimeski contends.

50 For the above reasons, I have determined to dismiss Mr Dimeski's claim for relief.

51 The parties should file any consent orders as to costs, or in the event they cannot agree, competing orders within seven days.

[4] Trial Transcript of 31 October 2024 (in S ECI 2023 05300) 31/22-26.

[6] [2013] HCA 43; (2013) 252 CLR 480, 502 [57].

[9] Amcor PLC v Scardamaglia [2023] VSCA 290; Irwin v Victorian WorkCover Authority & Anor [2024] VSC 615; Gray v Victorian WorkCover Authority & Anor [2024] VSC 680.

[10] [2024] VSC 680.

[13] [2019] VSC 84, [84]-[86].

[14] Wingfoot Australia Partners Pty Ltd v Kocak [2013] HCA 43; (2013) 252 CLR 480, 502 [57].