Identifying and dealing with difficult issues in NIHL cases.

Introduction 

The diagnosis and quantification of NIHL is affected by innumerable confounding factors, which include: 

(i) Constitutional issues, such as unrelated third pathologies, which can

‘replicate’ the pattern of threshold elevation as appears in NIHL cases;

(ii) Personal susceptibility to hearing damage: ‘soft and hard ears’;

(iii) The actual threshold at birth or before noise exposure, which means assumptions must be made regarding the extent of any allegedly raised threshold;

(iv) Age. Particularly how the effects of age are to be calculated and the assumptions which are valid in arriving at an approved or reliable AAHL table of estimates; 

Very often these factors, as well as their impact upon the shape of the audiogram, cannot be readily ruled in or ruled out when looking at causes of hearing ‘loss’, as distinct from whether there is in fact hearing loss. 

Further complexities are introduced when we consider 

(i) The point at which audiometry takes place; often many years after noise exposure has ceased. This is a particular complexity in the medico-legal setting;

(ii) Audiometric variability: being the differences which emerge on serial testing,

even in the same sitting, for reasons which are not always clear;

(iii) Difficulty in testing certain frequencies. This can occur particularly at 6kHz as the corrections for TDH-39 headphones illustrate. Also, 6kHz tends to be a difficult frequency for which to calibrate and is affected by socioacusis; and

(iv) The prospects that a person may, pure and simple, be geared up to exaggerate his ‘loss’.

THE LOW FENCE: 

The Limitation Argument. 

• Appears at the limitation stage.
• Polar opposite of de minimis arguments,
• Can conflict with the Defendant’s interests in special damages claims 

The contention 

• End of exposure end of damage.
• A low fence = point the ‘loss’ present is so great X is noticeably affected
• Crossed at end of exposure ceases or later: AAHL is added;
• Usually defined by binaural across 1-3kHz or 1-4kHz;
• Expert could consider elevation at isolated frequencies. 

Where is the Low Fence? 

Parkes the experts were asked about this issue. 

“…..low fences have been taken at an average loss of 30dB, or 27dB. In this case Mr Parker suggested 25dB; Mr McCombe 20dB, though some disability in some individuals occurred at lower levels. Mr Jones favoured a low fence, though accepted that at the extreme edge of the normal population some disability might show itself in the area 15-19dB.” 

Getting the Expert on Board 

• The expert must be on board; because
• “marked constitutional variation in hearing has been demonstrated in normal subjects. There is evidence of up to 30 dB variation at 2 kHz in young otologically normal adults match for age and sex”.
• No idea where normal begins and ends. 
• The expert to agree with the underlying contention;
• done by reference to the PCT prescribing levels or by clinical data sets. 

Turning the Tables on the Claimant 

Claimant’s own arguments against de minimis playing a part to the defendant’s benefit? Claimants, say: 

• Temporarily induced NIHL [TTS] still causes some actual permanent auditory
• Commonly HF damage caused: undisclosed impairment not present on audiogram;
• 12% of persons reporting hearing deficit had normal audiograms for age;
• Anianson, considering factors affecting speech discrimination a bilateral loss of 50dB at 4k, persons would grasp 70% of a conversation;
• Ladefoged & Disner : fricatives most significant at and above. 5000 words in the English language using.
• Smoorenberg 1994: 4k is important for distinguishing frequencies and consonants.
• Only succeed if court considers the claimant unreliable and, better still, dishonest.
• Both elements of expert assistance and witness unreliability are necessary in order to succeed. 

DE MINIMIS NON CURAT LEX

THE LAW IS NOT CONCERNED WITH TRIFLES

• de minimis sits at the opposite end of the spectrum to the low fence.
• C to show that breach caused or materially contributed to actionable damage.
• Discussing the extent of the injury itself

What is de minimis and where do the boundaries lie?

Does the Wider Jurisprudence Assist?

• Is the injury so trifling, so insignificant, that it can be described as “de minimis”?
• Not saying no compensatable injury at all – that would be a situation where liability cannot be established.
• “de minimis” between;

o   no contribution and material contribution; and/or
o   no harm and actionable harm.

• Cartledge v Jopling [1963] 1 All ER 341

o   Lord Pearce set out[1]:-
o   “Although those words were spoken with reference to the emission of the harmful dust, they are equally applicable to the injuries caused by it. It is a question of fact in each case whether a man has suffered material damage by any physical changes in his body.  Evidence that those changes are not felt by him and may never be felt tells in favour of the damage coming within the principle of de minimis non curat lex. On the other hand, evidence that in unusual exertion or at the onset of disease he may suffer from his hidden impairment tells in favour of the damage being substantial”.

• In Johnston v NEI [2007] UKHL 39[2], Lord Hoffman[3]:-

o   “…. [A] claim in tort based on negligence is incomplete without proof of damage…..Damage in this sense is an abstract concept of being worse off, physically or economically, so that compensation is an appropriate remedy..How much worse off must one be? An action for compensation should not be set in motion on account of a trivial injury. De minimis non curat lex. But whether an injury is sufficiently serious to found a claim for compensation or too trivial to justify a remedy is a question of degree. ….The question is whether the claimant has suffered damage. That means: is he appreciably worse off on account of having plaques?” 

• Defendants say that the test for de minimis is “appreciably worse off”.
• The threshold imposed was that between “trivial” and “appreciably worse off”.
• There is no margin between those two concepts 

• Sienkiewicz v Greif [2011] UKSC 10 Lord Phillips said this[4]:

o   “I doubt whether it is ever possible to define, in quantitative terms, what for the purposes of the application of any principle of law, is de minimis. This must be a question for the judge on the facts of the particular case.”

THE NIHL EXPERIENCE OF DE MINIMIS

• It is simply impossible to predict what a Court may consider to fall beyond the de minimis line and/or what it will come to regard as significant or appreciable in terms of its impact on the claimant. Any given case is subject to its own facts and, as such, reliance on previously decided cases may not be the best way forward.

Just Noticeable Difference?

The Binaural Average Aspect of de minimis 

• David McShefferty[1]

o       SNR [speech noise ratio] relevant to ability to hear in background noise.[2]
o       Higher SNR is required for people with hearing impairment.
o       What extent of increase in the SNR is required to render clarity ‘noticeable’ or ‘meaningful’. [the JND] 

• Testing on screened subjects showed a 3dB SNR was necessary There are however criticisms of the paper: 

Criteria for Running de minimis? 

• The case table suggests

  o       Putting Part 35 questions to the Claimant’s expert;
  o       Apply for own medical evidence and instruct that expert correctly and thoroughly;
  o       Develop that evidence by relying on and citing key reports and data;
  o       Challenge the Claimant’s evidence, both medical and lay, at trial on key issues;
  o       Call the expert irrespective perhaps of what Claimant’s expert has said. 

• If a case is selected for such treatment the following arbitrary lines may assist:

o       1-3kHz are relatively unaffected: setting perhaps a maximum of 5dB, although 3dB will be a more comfortable tariff [see McShefferty]
o       NIHL at 4kHz or 6kHz of not > than say 10-15dB.
o       No tinnitus is present.
o       No advanced needs for aiding;
o       A case with a very high level of elevation 

DE MINIMIS AND THE ELEVATION AT 4KHZ 

Should 4k be used in the assessment of HL/NIHL in any event?         

• Use of 1-3kHz is the commonly adopted range of frequencies for assessment of ‘loss’,  but does not mean there is no disability in some cases beyond this range.
•  4Khz is the frequency most susceptible to noise damage. [3] However, that does not mean it is relevant to disability solely as it is most affected by noise. 3Khz:

o       Most common for telecoms broadcasting
o       Frequencies beyond 0.3-3 do not add to the ability to perceive speech
o       Use of 4kHz in perception of certain consonants does not mean they cannot be heard if the thresholds there are elevated.
o       Degree of ‘redundancy’ built
o       It is more logical, when choosing between 3 & 4 kHz, to use the better frequency
o       To choose between 3 or 4, depending upon the site of the loss, measures impairment not disability 

What is 4kHz’s relevance to ‘disability’? 

• Human ears capable of detecting sound between 20Hz and 20kHz
• The voice can produce sound from 60Hz to c. 7kHz: speech falls mainly range 250Hz to 3kHz.
• 4k has its part to play.

o Sounds most affected by a deficit at 4kHz [usually “f”, “th” & “s”] The ‘speech banana’ model is sometimes used. In Hinchliffe Mr. Jones contended that the concept of a banana was nonsense.
o Built-in redundancy to recover some of the energy from such sounds is debatable,
o How typically those sounds appear in human speech and the effect of that loss of perception on whole sentence understanding has been the subject of much study. 

If 4kHz is relevant, at what point does damage there become disabling? 

• How substantial an added elevation must be to sound in damages at 4kHz [against a background of low level loss at 1-3k] is just as much in doubt as any other de minimis argument.
  o Most loss assessment models agree disability at about 25dB of loss. BUT measurements apply to average PT thresholds and not 4kHz
• Smoorenburg, predicted deficiency in speech recognition in noise by considering the degree to which loss at 2 & 4kHz exceeds 12dB.
  o There are no perfect studies but there is reasonable research to suggest 4k sound makes a contribution to speech intelligibility.
• Remember CLB 2000 at [3.2] 

ASYMMETRY IN THE CONTEXT OF NIHL 

• Occupational noise = bilateral &  symmetrical loss:
  o “..is almost always bilateral [and] audiometric patterns are usually similar bilaterally”.[4]
• Asymmetry can ‘legitimately’ arise in the context of OE.:
  o Genuinely asymmetrical noise exposure such as from gun-fire or working with one ear close to a very noisy source, such as an extractor fan;
  o Badly fitting PPE;
  o One-sided conductive loss [the protective effect of an already damaged ear]; and
  o Asymmetric AAHL [5] 

Asymmetry in the Population etc 

• An asymmetrical audiogram = a difficult confounding features.
  o very common feature
  o Normal males have a 5dB difference at 4kHz between left and right ears.
  o Hand dominance, can explain t asymmetry.[6] 

• Alberti[7] incidence: of 10% in those excessively exposed to noise. 

• 2014 Dobie: is asymmetry increased by ON exposure.[8] 

o Increase with greater total ‘loss’ [and thus also age]
o with ON exposure no more asymmetry when compared to men with no ON exposure. 

• The 2016 LCB Guide states: 

“unexplained SNHL asymmetries occur commonly in the general population’

The addition of symmetrical noise will result in an asymmetrical audiogram. 

What Qualifies as Asymmetry? 

• Audiometric measurement moves in 5dB steps: thus +/- 5dB differences is within the range of properly conducted testing. Could be +/- 10dB from that same process: thus 4k = 10 dB on the right : 20dB on the left but still be an equal measurement. [9]

o Lutman and Coles 2009[10], 15dB [at 4kHz] required where the test moved in 5dB stages.[11] 

Note 11 CLB 2000

• Note 11 of Coles deals with asymmetry:

o       compatible with noise but with differing degrees of probability.
o       Parts 1-2 if the CLB 10dB-shift criteria are not met on one ear, then the worse ear can still drag the diagnosis over the line.[12]

• Problems arise over:

o       Which is or what constitutes ‘a better ear’;
o        Whether it is the ear meeting R3(a) or (b)
o       What is a “smaller notch or bulge” and in which ear must it appear
o       Which frequencies may be taken as the ‘high frequencies’.
o       What is meant by “little or no NIHL”: how much is a “little”
o       ? “severe noise exposure”. 

Parameters of Asymmetry:

Often Forgotten 

• There are also limits or parameters within which the asymmetry must fit:

o The Robinson parameters [used by Coles] are at the 95^th and 98^th centile.
o  Anything worse than the expected asymmetry at those centiles will be beyond the parameters.

• In such cases, the difficulty for the Claimant must not be overcome by ‘discarding’ the worst ear as Coles et al do, and NIHL cannot be shown by concentrating on the worst ear either, which some experts will attempt. 

Fernandes & Fernandes 

• Fernandes[13] explains asymmetry and noise exposure.

o But confuses incidence and causation.
o incidence of asymmetry is lower in the non-exposed population than the exposed ergo noise causes asymmetrical loss;
o 208 people were studied.[14] 

• Cran v Perkins Engines Co Ltd[15]

o Lancer –v- Parker.
o Asymmetry likely N11 cat 4;

• Lancer says N11 is satisfied alternatively Fernandes;
• Parker says

o asymmetry is not noise;
o Moreover, the one-sided notch grew worse over time whereas on the better side, nothing seemed to occur;
o Fernandes’ definition of asymmetry was too low a bar
o Fernandes used a very select group who were known to suffer asymmetrical loss; and
o Their use of 200 cases was on the basis the authors decided that 200 had been exposed;
o Their cohort compared itself with the L&C 2009 cohort but they were incompatible due to the definitions of asymmetry being different

• Dt wins 

• In Aldred v Courtaulds Northern Textiles Ltd[16].

o several audiograms AHL.;
o variable PTA resulted in CERA = one-sided bulge at 4kHz.
o averaging permitted, bulge in the better ear increases: diagnosis under N11 met.[17] 

• Both Aldred and Briggs were cases where the degree of asymmetry was of modest degree, making them susceptible to fall in N11 1-2, as audiometric variability can account then for the differences.
• Note: distinction between cases which appear asymmetric but fall within the boundaries of test error, and those where there is true asymmetry
• Once the claimant finds himself in N11 territory, it is thought that his battle is almost won. 

THE LCB GUIDELINES 2016[18] 

• The CLB 2000  for diagnosis of NIHL:

o replaced subjective opinions from experts:
o approved for use, in the Nottinghamshire Textile Litigation, by HHJ Inglis.
o Consolidation of principle and a degree of uniformity
o But unsuitable for quantification because of the +/- 10 year, ‘best fit’ approach to the impact of AAHL
o Account for an ‘average rate of AAHL’,
o ISO 7029 : population study not a bespoke assessment 

• LCB pick up where the CLB left off.

o Mod 7029 table in CLB, adopting anchor points of typically 1 & 8kHz, overstates AAHL / understates NIHL.
o Noise has impact at those markers.
o A controversial assertion: such frequencies affected with high [>95dB Lepd] exposure for at least 10 years. 40 years at 85-90dB Lepd, is considered unlikely to affect 1kHz. 

• Mark Lutman: former paper , diagnosis, latter quantification 

• LCB offers 3 modifications to use of CLB: 

o Interpolation of misfit values:
o Selection of AAHL statistics: ISO 7029 (2000) and not 1984; and
o 6kHz is no longer recommended for use as anchor 

• A 2 pass method is taken.

o Pass one bulge analysis compares tested HTL with the ‘average pattern’ of AAHL.
o Pass two compares tested HTL compares with modified AAHL data and modified thresholds at anchors 1 & 8kHz.  

• LCB focuses on 4kHz. Unless there is substantial loss at that frequency, then overall loss will probably be lower using the new Guidelines. [19]

o Focus is now on the bulge, which is thought to be a truer representation of NIHL than merely the notch.. 

Simple or Shortcut Method 

(i) Perform a row (f) bluge calculation as per CLB 2000
(ii) Set any negative values at row (g), 1-2-3-kHz to zero
(iii) Calculate the 1-2-3-kHz average from the bulge row (f) in as per CLB
(iv) Then multiplying the averages by 1.32 or ‘adding one third’.
(v) Perform DSS calculation [4 x better ear ave + 1 x better ear ave /5] 

• LCB say:

o “deep notch” [not defined] at 4kHz : use full method should be used and AAHL plotted on the audiogram.
o No or minimal loss at 4kHz then short-cut method may be used;
o Maximal 3kHz NIHL cases, the short cut method is not to be used. 

Full Method 

(i) Carry out the row (f) bulge calculation as per CLB but modified as follows:
(ii) Estimate the NIHL at 1 & 8kHz [anchor points]

a. At 1k: NIHL = row (g) at 4k x 0.15
b. At 8k: NIHL = row (g) at 4k x 0.4
c. Correct –ve values to zero

(iii) Subtract the above calculated NIHL damage from the anchor point values to give a non-noise affected anchor at 1 & 8k [i.e. AAHL affected alone]

(iv) Do the CLB analysis again with the modified anchor

(v) Apply the new bulge effect at 1-2-3kHz and use the DSS formula 

“the LCB have the impact of reducing disability assessment compared to the conventional methods:

-the LCB full reduces disability more than the shortcut method  [odd as this may imply more noise decreases deafness]

-Using the full method, 35% of claimants fall in the de minimis category [<4dB] and 40% have minimal disability

-Using the conventional methods 7% have disability >30dB, which claims virtually disappear under the LCB”

• First impressions

o Reduced disagreement on how to use CLB 2000:
o Black Book methodology improved as it assumes all persons lie on the 50^th centile
o The bulge method is now the preferred method of analysis with the LCB allowing greater customization of the AAHL curves, especially with substantial HF loss
o Third pathologies may become more notable in low noise 

Criticisms 

• Evans v DECC[20], a de minimis case. Lutman –v- Singh. 

• R1, R2 & R3 were satisfied: bulges at 3 & 4kHz were noted on CLB 2000;
• Should 4kHz be included in assessment or, as Lutman argued, just 1-2-3kHz;
• 1-2-3kHz was measured at 1.1dB, using LCB 2016, by Lutman;
• Those frequencies were measured by Singh as 2.3dB , who accepted this would make no appreciable difference; however
• Singh measures 3-4kHz using Black Book methodology;
•  4kHz was where the greatest elevation was to be found, which Singh says would be unfair if not taken into consideration;
• Singh relied on the Prof Brian Moore paper[21]:
• Lutman says that work before  the Moore paper, Moore himself found a different result. 

logic 

• Singh:

o LCB as “illogical and unfair”.
o Lutman previously said, that a quantification over 2-4Khz could be appropriate in some cases.
o Would deal with the illogicality in the LCB.
o Use of 1k as the low anchor [in that case] would distort the estimation of NIHL.
o In certain cases, a diagnosis is effected by CLB 2000 criteria but which LCB 2016 reduces to zero dB. 

[**suggesting that LCB should not be used [without modification] if 2kHz is better than 1kHz.[22]] 

unfairness 

• ‘elevation’ at 4kHz turned the case. 

o “the audiometric threshold at 4kHz and possibly also at 6kHz, should be taken into account when considering compensation for NIHL in a medico-legal setting….A good predictor of the ability to understand speech in noise for people with NIHL is the average audiometric threshold at 2 and 4Khz”. 

• C wins:

o largely on the basis of Lutman’s concessions regarding his earlier expressed views; and
o rliance on earlier and unpublished studies with Prof Moore. 

Criticisms by Reporting Experts 

• Using Pass one for diagnosis and two for quantification is illogical.

o LCB 2016 acknowledges that Pass 1 data are average data and amount to an “abstract notion that does not apply to many individuals”.
o Modifying anchor at 1 & 8kHz using pass 2 data, means really should used for DSS and LCB quantum assessments.
o There are some situations where, despite a notch at 3/4/6k, 1-2-3 average is improved even in the face of noise exposure. The paper adjusts the –ve bulge values to zero showing no loss rather than improvement.
o The calculation methodology can, in cases, underestimate the loss. The best ear is used to determine noise loss, not the worst. There are cases where more weight is attached to a small loss in the worst ear or the loss is reduced in the better ear. This can the serve to diminish NIHL as it cannot be the case that an addition of 1/5 of NIHL from the worst ear makes noise damage overall better.
o If 1kHz is worse than 2kHz a bulge is created at 1-2-3k, which would not be representative of hearing loss. 

• These matters suggest the audiogram should be considered carefully in each case. The Guidelines are, again, just that. Illogical outcomes must be avoided where possible. 

THE 6KHZ NOTCH & 6K’S ROLE IN SPEECH PERCEPTION 

• Noise’s impact is greater at HF than LF. 

o PTS starts at around 4kHz and most damaging there  in 1^st 10-15 years of exposure, all things being equal.
o A notch is produced as the threshold is permanently altered;
o notch deepens for up to about the first 10 years’:
o CLB : notch can occur anywhere between 3-6kHz;
o Burns & Robinson 1963-68: invariably the ‘noise feature’ was located at 4K in ‘all but 2%’ of the substantial survey.[23] 

• Presence of a notch at 3-6Khz is not diagnostic of NIHL;
• Working in noise does not protect the ears from all other causes of loss.

o A CLB compliant audiogram does not mean that the loss is noise induced. 

• 6kHz notches present interesting issues:

o ‘the 6kHz notch’ characterizes the general population in the UK.[24]
o Socioacusis
o Calibration difficulties
o Baseline standardisations cause diagnosis issues with 6k 

• 1981 NPL:

o “the standard for audiometric zero was set too low at 6k with the error amounting to as much as 9dB”
o BUT CLB assumes that a 10dB change between 3-6 satisfactory in diagnosis.
o 6k is not 4k! It is a variable frequency, calibration is often not as strict and such notches at that frequency are not strongly associated with noise [Lutman, McBride & Wiliams, Colvin & Luxon[25]] 

• Pascoe v MoD: notching at 6 kHz but no evidence of deterioration at 4 kHz.

o claim dismissed

• The guidelines are merely that
• 4kHz is the most noise sensitive frequency;
• Notching at 6kHz without involvement at 4kHz was neither usual nor typical of noise type loss.
• Arguably less information or support for the relevance of 6kHz in speech perception:

o The Irish Expert Hearing “frequencies of 6000 Hz and 8000 Hz carry no information for speech comprehension”. 

• Studies such as Kujawa  & Liberman [2009], Stamper & Johnson [2015] & Tremblay et al [2015] suggest that noise has an impact of nerve structures such that even in the face of an apparently normal audiogram, speech intelligibility will be affected. 

• Evans: referred to the relevance of 6k for perception, relying, inter alia on the Moore paper. To say that this frequency has no relevance to perception is, however, an expert matter if that point is to be run. [see HHJ Keyser in Roberts v Prysmian Cables]. 

Michael Ditchfield

[1]“The Just Noticeable Difference in SNR” Trends Hear 2015 Feb  [there is a later 2016 paper also]

[2] SNR is the level of speech relative to noise ratio. Speech at 72dB against noise of 70dB = SNR 2dB

[3]see MRC National Study of Hearing

[4]American College of Occupational Medicine

[5]see N11 CLB 2000

[6] Masterton et al “Asymmetrical Hearing Loss in Cases of Industrial Noise Exposure…” Otol Neurotol: Americam Otological Society (2016) 37 998. See also Chung “SNHL & Workers’ Compensation” J Occ Med 23 418 (1981), Barrs “Asymmetry in NIHL” Noise Health 16 102 (2014) & Alberti “Occ HL : the significance of asymmetrical hearing thresholds” Acta Otolaryngol 87 255 (1979)

[7]  Noise and the Ear: Scott-Brown’s Otolaryngology Ch 18

[8] “Does ON Cause Asymmetric Hearing Loss” Ear Hear 35 577 (2014)

[9]one of the ‘deficiencies’ in the concept of averaging is explained by this phenomenon. AG error tends to be in favour of worse/higher thresholds. This is systemic error which cannot be corrected by averaging.

[10] “Asymmetric SNH Thresholds in the Non-Noise Exposed UK Population” Clin. Otolaryngology 34 316-321

[11] noting also that 1% of 18-80 year olds have 15dB or more of asymmetry [non exposed]

[12]Note that L & C in the 2009 study required a 15dB difference to count as significant, yet diagnose NIHL on a 10dB shift!

[13]Fernandes S. & Fernandes C “Medicolegal significance of asymmetrical hearing loss in cases of industrial noise exposure”. Jour of Laryngology Amp Otol 124 1051-55 (2010)

[14]they defined asymmetry as ‘either 10db at 2 x consecutive frequencies or 15dB at any between 250Hz and 6kHz’

[15]Norwich CC 14.12.12

[16]Liverpool CC 2.11.12

[17]see also Briggs v RHM Frozen Food [HHJ Coe] 2015: C wins. Jones for D & Homer for C. Disagreement as to what is significant asymmetry : 15dB at 2 or > contiguous frequencies not significant  says Homer: 15dB significant says Jones [ too significant to be connected to noise]

[18]Clin Otolaryngology 2016 41, 347-357

[19] consider the potential impact on a case like Pascoe v MOD [2013] Torquay CC. CLB was satisfied but only a 6kHz notch was present with no loss at 4kHz. D wins.

[20] [2017] Cardiff CC. See also Dudhill v DECC [2018] Sheffield CC DJ Baddeley

[21] “Review of the Perceptual Effects of Hearing Loss For Frequencies above 3kHz” July 2016: International Journal of Audiology

[22]Lutman recalculated. He made the adjustment suggested in LCB due to a precipitous fall at 8. The 15dB value at 1k was replaced with one of 7dB being the 50^th centile AAHL at 1 for that claimant = 2.3dB 1-2-3-k

[23]A 2008 study of Indian Air Force Personnel [Ind J Aerospace Med 52(2)] found of 229 personnel showed a significantly higher proportion of people with 6k affected as opposed to 4k. see also ‘Int Symposium on Performance Science’ Bradford et al 2009., which reached the same conclusion in musicians.

[24]Lutman & Qasem (1998)

[25]Clinical Diagnosis of NIHL: Chap 8 Luxon: Advances in Noise Research V1. Prasher & Luxon

[1] at 349E

[2] sub nom. Rothwell v Chemical and Insulating Company and Grieves v FT Everard

[3] at paragraphs 7 and 19

[4] At para 108