Applicant v Respondent

1 When Steven Elsdon, the applicant, was unloading a heavy bed from a furniture removal truck in the course of his employment on 10 June 2008, the bed fell on him causing two fractures to his lumbosacral spine: an undisplaced fracture to the right transverse process of L1; and a fracture to the superior end plate of L2. The dispute in this case is whether those two fractures, without more, should have entitled Mr Elsdon to an impairment assessment under the Accident Compensation Act 1985 (Vic) of 20 percent, as Mr Elsdon contends, or only five percent as was determined by the medical panel (comprising the second to fourth respondents) convened under the Act.

2 Mr Elsdon seeks judicial review, pursuant to s 3 of the Administrative Law Act 1978 (Vic), of the opinion of the medical panel given by a certificate of opinion dated 2 August 2011.[1]

3 Mr Elsdon says that once it was determined (as it was) that he had two fractures in adjacent vertebral bodies in his lumbosacral spine, the panel was obliged to assess his impairment as falling within lumbosacral impairment Category IV of the American Medical Association's Guides to the Evaluation of Permanent Impairment (4th Edition) (the Guides).[2] That categorisation mandated a 20% whole person impairment assessment. That conclusion was required, he contends, because in Transport Accident Commission v Serwylo[3] Kaye J authoritatively determined the proper construction of the criteria for that impairment category in such a way as to dictate that fractures of two adjacent vertebral bodies, without more, fall within it.

4 The panel neither referred to, nor appeared to apply, the principles in Serwylo. By not applying the Serwylo decision, Mr Elsdon argues that the panel impermissibly concluded that his impairment fell within lumbosacral impairment Category II, resulting in a lower impairment assessment than his lawful entitlement. On his analysis, the Serwylo decision was (and remains) the applicable law at the time of the medical panel decision. The panel was therefore obliged to apply it, so, in failing to do so, it engaged in jurisdictional error. Alternatively, because the panel did not refer to the decision in its reasons for opinion, the panel's reasons were defective and disclosed error on the face of the record.

5 The Victorian Workcover Authority (VWA) (substituted as the first respondent in place of the employer, RTS Transport Services Pty Ltd) argued that the panel correctly declined to find that Mr Elsdon's impairment fell within lumbosacral impairment Category IV, when the criteria for that category are correctly construed. On its case, Serwylo did not decide the issue that is critical for the resolution of this matter and, so, the panel did not engage in jurisdictional error by not applying it. Alternatively, if the decision did apply, the VWA invites me not to follow the decision.

6 Accordingly, the questions for decision, stated broadly are:

(1) did the medical panel engage in jurisdictional error by failing to apply the legal principles in Serwylo in evaluating Mr Elsdon's injury and impairment?

(2) do the medical panel's reasons for opinion disclose error on the face of the record in that they do not refer to Serwylo?

7 It was stated in Craig v South Australia[4] that if an administrative tribunal:

falls into an error of law which causes it to identify a wrong issue, to ask itself a wrong question, to ignore relevant material, to rely on irrelevant material or, at least in some circumstances, to make an erroneous finding or to reach a mistaken conclusion, and the tribunal's exercise or purported exercise of power is thereby affected, it exceeds its authority or powers. Such an error of law is jurisdictional error which will invalidate any order or decision of the tribunal which reflects it.

8 There is no dispute that the panel did not apply, or even refer to, Serwylo. Mr Elsdon's argument is, in effect, that the panel made a legal error and identified the wrong issue or asked itself the wrong question. It failed to ask itself the (correct) question, namely whether, because Mr Elsdon had lumbosacral spine fractures at more than one level, he suffered 'multilevel structural compromise of the spine' in accordance with Serwylo's case. In doing so, he says the panel engaged in jurisdictional error.

9 Although the second issue, that of error on the face of the record, was not abandoned, it was clear that the first question concerning jurisdictional error raised the real substance of the dispute.

10 The dispute can be resolved by answering the following questions: how is impairment to be determined under the Act; how did the medical panel determine Mr Elsdon's impairment; what did the Serwylo case say about the task; and, should the panel have applied it?

How is impairment to be determined under the Act?

11 An assessment of impairment under s 91 of the Act is to be undertaken in accordance with the Guides, as modified by the Act.[5]

12 In the Guides, spinal impairment categories ranging from Category I to Category VIII are assigned to particular impairment conditions identified in Table 70. The Guides require the physician to start with Table 70 as a guide toward the appropriate category for the spine impairment. Table 70 lists, in one column, short descriptions of the 'Patient's Condition' and, in adjacent columns, the applicable category level for that condition. In a number of cases there is a range of category levels assigned to the particular condition.

13 Relevant to Mr Elsdon, Table 70 lists the following conditions and associated categories:

Posterior element fracture, healed, stable, no dislocation or radiculopathy

Transverse or spinus process fracture with dislocation of fragment, healed, stable

Vertebral body fracture without loss of motion segment integrity or radiculopathy

Vertebral body fracture with loss of motion segment integrity or radiculopathy

14 More detailed explanations of those impairment conditions, relevant to the lumbosacral spine, are set out in section 3.3(g) of the Guides. It is necessary to refer to those more detailed descriptions to locate the particular patient in the correct impairment category.

15 In that section of the Guides, spinal impairment categories are described by means of a diagnosis related estimate (DRE) category level, and a short descriptive label. For example, for the lumbosacral spine, the first category is DRE Category I, labelled 'Complaints or Symptoms'. The second is DRE Category II, labelled 'Minor Impairment'. The categories ascend in severity of condition and attract increasing measures of whole person impairment, ranging from nil percent for DRE Category I (Complaints and Symptoms) to 75 percent for DRE Category VIII (Paraplegia, Total Loss of Lumbosacral Spinal Cord Function).

16 Each category has three parts. The first, titled 'Description and Verification', contains a brief description of the medical indicia of the impairment to qualify for inclusion in that category. The second part, 'Structural Inclusions' describes conditions which, if suffered by the patient, automatically qualify the patient for inclusion in the category so that the examining physician need not determine if other criteria contained in the 'Description and Verification' part are present. If a patient demonstrates the Structural Inclusions of two categories, the physician should place the patient in the category with the higher impairment percentage.

17 I will set out in full the relevant parts of two impairment categories; first, the one in which the panel assessed Mr Elsdon to fall (Category II), and, secondly, one to which Mr Elsdon says he properly belongs (Category IV). On the facts of this case, it is the Structural Inclusions of each category which fall for particular scrutiny.

DRE Lumbosacral Category II: Minor Impairment

Description and Verification: The clinical history and examination findings are compatible with a specific injury or illness. The findings may include significant intermittent or continuous muscle guarding that has been observed and documented by a physician, nonuniform loss of range of motion ..., or nonverifiable radicular complaints. There is no objective sign of radiculopathy and no loss of structural integrity.

...

Structural Inclusions: (1) Less than 25% compression of one vertebral body; (2) posterior element fracture without dislocation (not developmental spondylolysis); the fracture is healed, and there is no loss of motion segment integrity.

A spinous or transverse process fracture with displacement without a vertebral body fracture is a Category II impairment because it does not disrupt the spinal canal.

Impairment: 5% whole-person impairment.

...

DRE Lumbosacral Category IV: Loss of Motion Segment Integrity

Description and Verification: The patient has loss of motion segment integrity (differentiator 5, Table 71 p. 109). Loss of motion segment or structural integrity is defined as at least 5 mm of translation of one vertebra on another, or angular motion at the involved motion segment that is 11° more than that at an adjacent motion segment (Figs. 62 and 63, p. 98). Loss of structural integrity at the lumbosacral joint is defined as at least 15° more angular motion than at the L 4 and L 5 motion segment.

A documented history of muscle guarding and pain is present. Neurologic abnormalities need not be present. If they are present, the examiner should consider using Category V.

Structural Inclusions: (1) Greater than 50% compression of one vertebral body without residual neurologic compromise; (2) multilevel spine segment structural compromise, as with fractures or dislocations, without residual neurologic motor compromise.

Impairment: 20% whole-person impairment.

How did the medical panel determine Mr Elsdon's impairment?

18 After noting the clinical presentation of Mr Elsdon, and concluding that there was no objective clinical evidence of radiculopathy, the panel reviewed the results of three diagnostic images and reports of Mr Elsdon's lumbosacral spine:

... A fracture line extending through the anterior aspect of the superior end plate of L2 without significant displacement. An undisplaced fracture of the right L1 transverse process was noted. There was no definite vertebral canal compromise seen ... .

... anterior wedging of the L2 vertebral body consistent with the worker's history of injury. The Panel measured 10% loss of vertebral body height anteriorally.

... healed compression fracture of L2 with slight disc space narrowing at L1-2. There was also a mild right predominant non-neurocompressive disc bulge at L4-5. The Panel were unable to identify any fracture line at the right transverse process of L1 and considered that the fracture was undisplaced and has healed.

19 From those results, and his clinical presentation, the panel concluded that Mr Elsdon was suffering residual dysfunction of his lumbosacral spine following -

• a compression fracture of L2 with 10% loss of anterior vertebral body height, and
• an undisplaced fracture of the right transverse process of L1,

20 Acknowledging its obligation to assess Mr Elsdon's impairment in accordance with the relevant parts of the Guides, the panel stated its conclusion in this way:

The Panel assessed the worker's back in accordance with Table 70 of Chapter Three and concluded that there are clinical signs of lumbar spine injury with evidence of L 2 vertebral body compression fracture with 10% loss of anterior vertebral body height at [sic] without clinical evidence of radiculopathy. The Panel therefore concluded that the appropriate impairment category for the lumbosacral spine is DRE Category II (structural inclusions) pursuant to Table 72 of Chapter Three, resulting in a whole person impairment of 5%. The degree of impairment is permanent.

21 The features of Mr Elsdon's observed condition that conform to the Category II criteria include:

• L2 vertebral body compression fracture with 10% loss of body height - matching the first structural inclusion; and
• No clinical evidence of radiculopathy - equating to the condition set out in the 'Description and Verification'.

22 Significantly, no mention is made in the panel's final conclusion, extracted above, of the transverse process fracture at L1. The panel had previously noted that the fracture was undisplaced and had, by 25 August 2009, healed and was not even detectable on CT scan. Such a fracture, on its own, would not have qualified as a structural inclusion within Category II. Even had a transverse process fracture featured displacement, alone it still would not qualify as a structural inclusion within Category II unless also associated with 'vertebral body fracture'. That is because, according to the terms of the structural inclusion, a transverse process fracture 'does not disrupt the spinal canal'.

23 But, most critically for this case, the panel did not expressly consider the significance (if any) of there being two fractures of adjacent bodies, and the possible application of the second structural inclusion under Category IV; that is, 'multilevel spine segment compromise, as with fractures', as that expression was interpreted in Serwylo. It is that failure which Mr Eldson identifies as the jurisdictional error. Since that is the identified jurisdictional error, it is necessary to identify exactly what it was that Serwylo decided.

24 Ms Serwylo sustained fractures of the superior end plate of three lumbar vertebrae, at levels L2, L3 and L4, the most serious of which was a 10% compression fracture at L3 (the other two at lower percentages). A member of the Victorian Civil and Administrative Tribunal (VCAT), after hearing opposing evidence from three orthopaedic surgeons, determined that Ms Serwylo's lumbosacral impairment fell within DRE Category IV. Kaye J upheld that determination, and leave to appeal to the Court of Appeal was refused.

25 All three orthopaedic surgeons had agreed that Ms Serwylo's condition did not fit within the criteria under 'Description and Verification' for Category IV. That is, there was no basis to conclude that she had 'loss of motion segment or structural integrity' as defined in the criteria. But two doctors considered she qualified for Category IV, under Structural Inclusion (2), because she had multilevel fractures of the vertebrae, and because multilevel fractures imply a greater level of force and thus a greater level of damage than a single level fracture.

26 Category II was not appropriate, it was said, because the pertinent structural inclusion within that category focused only upon a single vertebral fracture.

27 The opposing medical view was that the particular fractures of the superior end plates, albeit at three levels, were minor and involved no structural compromise to the spine. Structural compromise to the spine, it was said, required something which disrupts, or had real capacity to disrupt, the spinal canal, or the ability of the spine to provide postural support. Because no such disruption was evident, the view was that there was in fact no spine segment structural compromise. So, Ms Serwylo's condition did not attract Structural Inclusion (2) of Category IV.

28 It is important to understand the arguments that were put to the court in order to discern the scope of the decision that was reached.

29 The appellant (in that case, the Transport Accident Commission (TAC)) submitted that, in Structural Inclusion (2) of Category IV, the words 'as with', appearing in the phrase 'as with fractures', were intended only to illustrate an association, conjunction or combination. They did not operate to deem that two or more fractures of the lumbosacral spine, however minor, constituted 'multilevel spine segment structural compromise'. It was multilevel spine segment structural compromise which was the controlling condition to engage the structural inclusion. The phrase 'as with fractures' was merely an example of circumstances in which such a compromise of the spinal structure might be found to have occurred.

30 Against this, it was put on behalf of Ms Serwylo that the phrase 'as with fractures' in Structural Inclusion (2) meant that injuries, consisting of fractures, in more than one lumbar vertebrae, by definition constituted multilevel spine segment structural compromise, thereby bringing the condition within Category IV.

31 Those being the arguments, his Honour identified the issue in Serwylo in this way:

The issue which I need to determine on this appeal is quite narrow. In essence, it turns on the meaning of the phrase 'as with fractures' in Structural Inclusion (2) of Category IV. The critical question is whether (as contended by the respondent) that phrase denotes instances of conditions which are cases of multilevel spine segment structural compromise, or whether (as contended by the appellant) it denotes instances of conditions which might constitute multilevel spine segment structural compromise.[6] [underlining added]

32 His Honour referred to the relevant principles governing the construction of the Guides, adopting what had been said by Kyrou J in HJ Heinz Company Australia Ltd & Anor v Kotzman & Ors[7], and emphasised that the construction of the Guides is a question of law. His Honour acknowledged that the task for him was that of construing the Guides, and therefore a question of law. He regarded the competing medical views as largely mirroring the legal debate concerning the proper construction of Structural Inclusion (2).

33 The conclusion which Kaye J ultimately reached was that, on the proper construction of Structural Inclusion (2), fractures or dislocations at multilevels of the lumbar spine are, per se, instances of multilevel spine segment structural compromise for the purposes of that structural inclusion.[8]

34 Returning to the present case for a moment, the argument put by the VWA is that Serwylo only decided the meaning of the phrase 'as with', determining that it deems relevant fractures to constitute multilevel spine segment compromise rather than merely illustrate instances which might constitute such compromise. However, so the argument is put, Serwylo did not conclusively determine which fractures actually have that effect. Not every fracture of a vertebral element qualifies for Structural Inclusion (2), it argued. At the least, it was contended that healed, non-displaced fractures of the transverse process are not to be understood as embraced by the phrase 'as with fractures' when that phrase is properly construed.

35 Bearing that in mind, I now return to the reasons which Kaye J gave for arriving at his conclusion. Essentially there were three.

36 The first was that the language 'as with' was, as a matter of plain English, more appropriately understood to denote instances which do - and not just which might - constitute instances of multilevel spine segment structural compromise.

37 Secondly, the construction of Structural Inclusion (2) put forward by the TAC was not likely to be correct because medical practitioners, to whom the Guides were directed, would well understand that fractures or dislocations might constitute multilevel spine segment structural compromise. Therefore, there being no reason to tell medical practitioners something they already knew, it was more likely that the phrase 'as with fractures' was intended to specify, for the users of the Guides, a condition which did, as distinct from might, constitute multilevel spine segment structural compromise.

38 Thirdly, because the phrase 'multilevel spine segment structural compromise', at least arguably, lacks a degree of precision, it was understandable that the authors of the Guides would see fit to include in Structural Inclusion (2) specific examples of conditions which do - and not just might - constitute such a structural compromise.

39 His Honour was fortified in his conclusion by observing the differences in the structural inclusions in Categories II, III and IV, noting the ascending gradation of conditions specified that were necessary to qualify for each, more serious category. His Honour noted that it was the element of fractures or dislocations at more than one level that differentiated Structural Inclusion (2) of Category IV from the corresponding structural inclusions in Categories II and III, which were explicitly only concerned with fractures at one level.

40 That observation, his Honour believed, supported his construction of Structural Inclusion (2) of Category IV 'that fractures or dislocations of more than one spinal segment are, by their nature, instances of multilevel spine segment structural compromise'.[9]

41 It is important to bear in mind that his Honour's conclusion was reached in the face of an argument, that:

• not just any fractures, however minor, denoted 'multilevel spine segment structural compromise', and
• in particular, because fractures of the superior end plate were minor and did not have the capacity to disrupt the spinal canal or to affect the spine's ability to provide postural support, they did not fall for inclusion within Category IV.

42 So, the conclusion that fractures at more than one level, by definition, satisfy the criteria of Structural Inclusion (2), was a conclusion that was reached regardless of the actual severity of the fracture, or its capacity to disrupt the spinal canal or otherwise compromise the structural integrity of the spine. And it was reached for the three reasons his Honour explained.

43 It is true that his Honour gained some comfort for his conclusion from the evidence which tended to support the proposition that multilevel fractures, per se, involved some degree of compromise of the spine.[10] But he was careful to explain that the conclusion on construction, which ultimately was determinative, was reached independently, saying, '...the [medical] evidence to which I have referred is, at least, consistent with the conclusion [on construction] which I have otherwise reached' (underlining added).[11]

44 The Court of Appeal (Ashley and Harper JJA, and Ross AJA) refused leave to appeal from Kaye J's decision, stating:

We consider that the judge's full and careful reasons - the decisive aspects of which, counsel for the applicant accepted, were valid points to weigh in determination of the competing constructions - express and explain the preferable construction of that part of Chapter 3 of the AMA Guides which was in issue.[12]

45 As well as upholding the trial judge's principal conclusion on construction, the Court of Appeal similarly found subsidiary support for the ultimate result from the medical evidence:

We further consider that the medical opinions, in which we include the viva voce evidence given by Messrs Doig, Moran and Shannon, supported a conclusion, even assuming the applicant's construction argument to be correct, that the respondent suffered multilevel spine segment structural compromise - a concept unattended, in the language of Category IV of DRE Lumbosacral Spine Impairment, by characterisations 'significant' or 'very significant'.[13]

Should the medical panel have applied Serwylo?

46 The VWA contends that, having regard to certain medical opinion evidence it sought to tender, the proper construction of the word 'fractures' in the subsection 'Structural Inclusions' for DRE Category IV excludes a fracture of the transverse process. If that is correct, it argues there are no multilevel fractures (plural) to bring Mr Elsdon's condition within Category IV. Relevantly, there was only one fracture of significance, the compression fracture to the superior endplate at L2. So, the medical panel was correct to assess his condition as falling within Category II only.

47 The VWA's construction argument relies, in part, upon an argument that the Guides were written against a particular factual context. It seeks to establish that factual context by the tender of a medical opinion, a report of Mr Wilde. The medical opinion it seeks to adduce is to the effect that under no circumstance can a fracture of the transverse process (still less, an undisplaced fracture) ever compromise the structural integrity of the spine. In that alleged factual context it urges me to construe the meaning of 'fracture' within the critical phrase as excluding fractures of the transverse process.

48 Also, the VWA relies upon other contextual indicia within the Guides themselves. For example, it relies on the explanation given as to why a spinous or transverse process fracture, without vertebral body fracture, does not qualify for inclusion in Category II; namely, because they do not disrupt the spinal canal. It also points to the distinction in severity of impairment, revealed in Table 70, between a 'vertebral body fracture' and a 'transverse or spinous process fracture',[14] an anatomical difference said to be supported by the proposed medical opinion.

49 Mr Elsdon objected to the admissibility of such opinion evidence. Nevertheless, if I decide to admit VWA's evidence, he wishes to rely on some of his own (a report of Mr Brownbill). I received the opinion evidence tendered by both sides, subject to objection,[15] with a view to determining the admissibility question along with the substantive issues.

50 Because of the arguments put forward by the TAC in Serwylo, and the construction finding which was the principal reason for rejecting those arguments (both at first instance, and on appeal), I do not believe it remains open to argue that 'as with fractures' in Category IV excludes particular types of (minor) fractures which have no capacity, in fact, to compromise the spinal structure. That, essentially, was the same argument put, and rejected, in Serwylo.

51 True, the Serwylo case did not have the same focus, in particular, on a fracture of the transverse process. But, nevertheless, it concerned an argument put in relation to a species of fracture that was minor and had no capacity to disrupt the spinal canal, or compromise structural support. In this case the VWA wishes to rely upon the same characterisation of a (healed, undisplaced) transverse process fracture, allegedly known both to the authors of the Guides and those for whom it was written, as the foundation for a construction argument that would reach a result that is inconsistent with the legal construction already decided in Victoria.

52 The contention that there is a difference between the issue of construction decided in Serwylo, and the issue of construction to be determined in this case, relies upon a difference that is illusory. And, I would not adopt the VWA's suggestion that, should I come to that conclusion, I decline to follow that authority.

53 In my view the VWA's construction argument is foreclosed by the Serwylo decision. As I have shown, that decision was reached, on its primary foundation, independently of any medical evidence, and regardless of whether the particular condition was actually capable (or known to be capable) of compromising the spinal structure. It follows that the medical evidence each side seeks to adduce in this case is irrelevant and, therefore, inadmissible. I refuse the proposed tender of the (partly redacted) reports of Mr Wilde and Mr Brownbill.

54 I conclude that the medical panel was obliged to apply the Serwylo decision to the circumstances of Mr Elsdon's case. It did not do so, and thereby fell into jurisdictional error. That is, it failed to take into account relevant material, namely the separate fractures at two levels of the lumbosacral spine. Alternatively it failed to apply the applicable law which says that multilevel fractures, per se, constitute 'multilevel spine segment structural compromise' regardless of the capacity of each fracture to disrupt the spinal canal or affect the spine's ability to provide postural support.

55 Whether or not the panel's reasons disclosed error on the face of the record - because there was no reference to Serwylo, explicit application of it to the facts or explanation why it did not apply - is unnecessary to decide. Indeed, both parties accepted that it would be unnecessary for me to decide that issue were I to decide, as I do, that the panel engaged in jurisdictional error.

56 Subject to further submissions, I propose to make orders that the opinion of the medical panel dated 2 August 2011 be quashed, and that the medical questions be remitted to a medical panel to be convened in accordance with the Act. Whether or not that should be a differently constituted medical panel[16] is one of the matters upon which I will hear from the parties.

[1] Given in response to questions referred to the panel under s 104B(9) of the Act. The medical panel was asked to assess Mr Elsdon's degree of whole person impairment in accordance with s 91 of the Act. That assessment, in turn, was required to calculate Mr Elsdon's entitlement, if any, to compensation for non-economic loss under s 98C of the Act.

[2] Section 91(1)(a(ii) of the Act requires that an impairment assessment be undertaken in accordance with the Guides, as modified by the Act.

[3] Transport Accident Commission v Serwylo [2010] VSC 421.

[4] Craig v South Australia [1995] HCA 58; (1995) 184 CLR 163, 179.

[5] Section 91(1)(a)(ii) and s 5.

[7] Heinz Company Australia Ltd & Anor v Kotzman & Ors [2009] VSC 311 [24]-[28].

[12] Transport Accident Commission v Serwylo [2011] VSCA 305 [3].

[14] See those parts of the table extracted at [13] above.

[15] Neither party sought to call their witness, nor cross examine the other's, and both agreed that certain portions of each report should be excluded whichever way I decided the general admissibility issue.

[16] As to which, see, for example Abbott v Eptec Victoria Pty Ltd & Ors [2011] VSC 267 [41].