What happened
Yuelin Han, a Chinese citizen born in 1984, arrived in Australia in 2007 on a student visa. She lodged an application for a Skilled Independent (Subclass 885) visa on 21 December 2009. In 2011 she suffered acute liver failure and received an emergency liver transplant at Royal Prince Alfred Hospital. The transplant was successful, but it engaged public interest criterion 4005 (PIC 4005) in Schedule 4 to the Migration Regulations 1994 (Cth).
PIC 4005(1)(c)(ii)(A) required that Ms Han be free from a disease or condition where the likely provision of health care or community services would result in a significant cost to the Australian community. Three successive Medical Officers of the Commonwealth (MOCs) formed the opinion that a hypothetical person with Ms Han's condition—an asymptomatic previous liver transplant that was permanent and required ongoing specialist medical review and immunosuppressive medication—would be likely to incur such significant costs. The MOCs had regard to detailed reports from Ms Han's treating transplant specialist, Associate Professor Simone Strasser, which spoke of excellent recovery, low-dose generic immunosuppressants available on the Pharmaceutical Benefits Scheme, quarterly reviews at Medicare rates, and an overall view that the costs would be relatively low and Ms Han would be a fully contributing member of society.
A delegate refused the visa on 4 March 2015. Ms Han sought merits review. After an earlier Tribunal decision was set aside by consent, a second Tribunal affirmed the refusal on 6 December 2017, stating that reg 2.25A(3) required it to take the final MOC opinion as correct. Judicial review in the Federal Circuit Court was dismissed on 15 August 2018. The sole ground advanced in that Court was that the words "significant cost to the Australian community" in PIC 4005 were incapable of objective ascertainment, rendering the criterion invalid. That ground was abandoned on appeal.
In the Federal Court, Ms Han sought leave to rely on an entirely new ground: that the Tribunal had constructively failed to exercise jurisdiction because it relied on an MOC opinion not formed in accordance with law, either because the MOC misunderstood its statutory task by assessing at too high a level of generality rather than by reference to her actual condition, or because the MOC failed to give proper, genuine and realistic consideration to the merits by not explaining why it disagreed with Associate Professor Strasser's reports. Bromwich J refused leave and, for completeness, held that the proposed ground would have failed on the merits in any event. The appeal was dismissed with costs.
Why the court decided this way
Bromwich J approached the application for leave by reference to well-established High Court and Full Court authority. His Honour began with the elementary proposition from University of Wollongong v Metwally [1985] HCA 28 at 71 that a party is bound by the conduct of its case except in the most exceptional circumstances. This was reinforced by the observations in Coulton v Holcombe (1986) 162 CLR 1 at 7-8 that appeals by way of rehearing do not permit the main arena for dispute resolution to shift from the court of first instance to the appellate court.
The judgment placed particular weight on VUAX v Minister for Immigration & Multicultural & Indigenous Affairs [2004] FCAFC 158 at [46]-[48], which itself drew on O'Brien v Komesaroff and Branir Pty Ltd v Owston Nominees. VUAX makes clear that leave is granted only where expedient in the interests of justice. Merit is necessary but not sufficient. Other factors—absence of an acceptable explanation, prejudice to the respondent, and the risk of undermining the statutory scheme—are weighed. In migration matters the prejudice analysis is sharpened by the evident purpose of Pt 8 of the Migration Act 1958 (Cth) and s 476A, which together contemplate one substantive trial in the Federal Circuit Court followed by one substantive appeal to the Federal Court, with only limited further review by the High Court on special leave.
In this case no explanation was proffered for the failure to run the MOC ground below. Counsel at first instance had expressly conceded that if the validity argument failed the judicial review application must be dismissed and had disavowed any contention that the MOC opinion was invalid or that the Tribunal had otherwise fallen into jurisdictional error. Bromwich J inferred a forensic choice. That choice could not be revisited on appeal. Allowing the new ground would have converted the appeal into a de facto trial on an entirely new factual and legal basis, denying the Minister any practical right of appeal from an adverse first-instance determination in the Federal Court. Such an outcome would thwart the legislative design described in AAM15 v Minister for Immigration and Border Protection [2015] FCA 804 at [14] and endorsed in BZD17 v Minister for Immigration and Border Protection [2018] FCAFC 94 at [28]-[29].
His Honour also noted that the new ground was not of broad legal significance; it turned on the particular wording of one MOC report and the application of existing authority to its facts. The absence of any need for fresh evidence was acknowledged but given little weight in the overall discretionary balance.
For completeness, Bromwich J examined the merits. The final MOC report of 21 April 2016 identified the condition with specificity ("asymptomatic previous liver transplant"), recorded that the transplant occurred in 2011 with excellent long-term prognosis, stated that the condition was permanent, and identified the services as "ongoing specialist medical review and immunosuppressive medication". It expressly applied the hypothetical-person test required by Robinson v Minister for Immigration and Multicultural and Indigenous Affairs [2005] FCA 1626 at [43] and concluded that provision of those services would result in significant cost. The report recorded that regard had been had to all of Associate Professor Strasser's reports.
Drawing on Minister for Immigration and Multicultural Affairs v Seligman (1999) 85 FCR 115 at [66] and Perez v Minister for Immigration and Border Protection [2017] FCAFC 180 at [9], the Court held that an opinion is "formed in accordance with law" if it is an opinion of the kind the regulations authorise. Blair v Minister for Immigration and Multicultural Affairs [2001] FCA 1014 at [33] and [38] confirmed that an MOC is entitled to form an opinion even if it conflicts with evidence submitted by the applicant and is not required to give reasons for rejecting that evidence or to specify a precise monetary figure before characterising a cost as significant. The conflict with Associate Professor Strasser's lower-cost assessment went only to the merits of the medical judgment, not to jurisdiction. The MOC had not asked itself the wrong question, had not assessed at an impermissibly generic level, and had not failed to engage in an active intellectual process. The challenge was therefore impermissible merits review. Had leave been granted the ground would have failed.
Before and after state of the law
Prior to this judgment the law on leave to raise new grounds was settled by VUAX, Metwally and Coulton v Holcombe. Full Court decisions such as Sun v Minister for Immigration and Border Protection [2016] FCAFC 52, Haritos v Commissioner of Taxation [2015] FCAFC 92 and Murad v Assistant Minister for Immigration and Border Protection [2017] FCAFC 73 had reiterated that merit is necessary but not sufficient and that the statutory scheme in Pt 8 of the Migration Act is a powerful discretionary factor against allowing what is in substance a second trial.
The law on MOC opinions under PIC 4005 was also settled. Seligman established that a decision-maker takes an MOC opinion as correct only if it is an opinion of a kind authorised by the regulations and formed according to law. Robinson clarified that the MOC must assess the applicant's actual severity and then apply the criteria to a hypothetical person with that condition at that severity; assessment at a higher level of generality is impermissible. Blair confirmed that the MOC is not required to reconcile its opinion with contrary medical evidence or to quantify costs. Haque & Ors v Minister for Immigration and Anor [2015] FCCA 1765 illustrated the type of fundamental error (misstatement of the level of disability leading to the wrong hypothetical) that can vitiate an opinion; it did not assist Ms Han because her case involved no such foundational misstatement.
This judgment does not change those principles. It applies them in a context where the new ground effected a complete change of case theory and where the MOC report satisfied the Robinson and Seligman requirements. The decision reinforces that mere conflict with a treating doctor's opinion, or brevity of expression, does not equate to jurisdictional error. The "before and after" position is therefore one of continuity: the discretionary bar to new grounds remains high in migration appeals, and the threshold for impugning an MOC opinion remains anchored in whether the opinion is of the authorised kind rather than whether it is medically persuasive.
Key passages with plain-English translation
At [11] the judgment quotes the High Court in Metwally: "It is elementary that a party is bound by the conduct of his case. Except in the most exceptional circumstances, it would be contrary to all principle to allow a party, after a case had been decided against him, to raise a new argument which, whether deliberately or by inadvertence, he failed to put during the hearing when he had an opportunity to do so." In plain English: you run the case you choose at first instance; you cannot usually get a second bite at the cherry on appeal simply because the first choice did not succeed.
At [14], citing VUAX: "Where, however, there is no adequate explanation for the failure to take the point, and it seems to be of doubtful merit, leave should generally be refused." Translation: if you cannot explain why you did not run the argument below and the argument looks weak, you will almost certainly be refused permission to run it for the first time on appeal.
At [38], quoting Seligman at [66]: "If the opinion is vitiated by error of law the delegate errs in taking it as correct." Translation: the Tribunal can only treat the MOC opinion as binding if the MOC asked itself the right legal question and stayed within the limits of its statutory task. If the MOC made a legal mistake, the Tribunal makes the same mistake by treating the opinion as conclusive.
At [41], summarising Robinson at [43]: the MOC "must ascertain the form or level of condition suffered by the applicant in question and then … apply the statutory criteria by reference to a hypothetical person who suffers from that form or level of the condition. It is not the case that the MOC is to proceed to make the assessment at a higher level of generality by reference to a generic form of the condition." Translation: the doctor must look at how bad this person's condition actually is, then imagine another person with exactly the same severity, and ask whether that imaginary person would cost the community a significant amount. The doctor cannot simply say "liver transplants in general are expensive".
At [60], discussing Haque: "even though this opinion is expressed briefly, it is set out in plain terms and there is no room to find that the phrase in question means anything other than what it actually says." Translation: courts should read MOC reports sensibly and beneficially, but they cannot rewrite them to fix factual mistakes that go to the heart of the hypothetical assessment.
What fact patterns trigger this precedent
This precedent is triggered in migration appeals to the Federal Court where an appellant seeks leave to advance a ground of review that was not run in the Federal Circuit Court, particularly where (a) the new ground effects a wholesale change in the nature of the case (for example, from a pure legal validity challenge to a fact-sensitive jurisdictional-error challenge to an MOC opinion), (b) no acceptable explanation is offered and a forensic choice can be inferred, (c) the appellant was represented by experienced migration counsel at first instance, and (d) the proposed ground would require the Federal Court to determine contested factual matters about the adequacy of an administrative medical assessment as though it were the court of first instance.
It is also engaged where a visa applicant attacks an MOC opinion under PIC 4005 (or analogous health criteria) on the basis that the opinion is "sparse", fails to quantify costs, or reaches a different conclusion from a treating specialist. The precedent confirms that such an attack will fail unless the opinion either (i) uses the wrong legal test (for example, assessing a generic form of the condition rather than the applicant's actual severity), (ii) contains a fundamental factual misstatement that taints the hypothetical exercise (as in Haque), or (iii) demonstrates that the MOC asked itself the wrong question. Mere brevity, absence of a monetary figure, or failure to explain rejection of the treating doctor's opinion does not constitute jurisdictional error.
The decision is likely to be cited whenever a party argues that the "proper, genuine and realistic consideration" line of cases (Carrascalao, Nguyen, Malek Fahd) applies to the formation of an MOC opinion. Bromwich J's analysis shows that those principles must be applied with caution to highly prescriptive criteria that expressly require deference to an independent medical expert.
How later courts have treated it
Although the judgment is relatively recent (11 March 2019), it has been cited in subsequent Federal Court and Federal Circuit Court decisions as authority for the strict approach to leave in migration appeals. It has been treated as an orthodox application of VUAX, AAM15 and BZD17 rather than as creating new doctrine. Later courts have cited the paragraphs dealing with the Pt 8 statutory scheme ([17]-[18]) when refusing leave to run points that could have been, but were not, run below.
On the MOC opinion issue, the decision has been followed for the proposition that an opinion which identifies the condition with specificity, uses the hypothetical framework, and applies the significant-cost criterion is not vitiated merely because it differs from a treating doctor's more optimistic cost assessment. It has been distinguished in cases where the MOC report contained an obvious factual error about the severity of the condition that infected the hypothetical exercise. No court has suggested that Bromwich J misstated the law in Seligman, Robinson or Blair. The decision therefore sits comfortably within the existing stream of authority and is likely to be cited as an example of the high bar an applicant must clear before an MOC opinion can be impugned on judicial review.
Still-open questions
The judgment leaves open the precise boundaries of what constitutes an "acceptable explanation" for not raising a point below in circumstances where fresh evidence emerges after the Federal Circuit Court hearing that could not reasonably have been known. It also leaves open whether, in a case where the new ground raises a pure question of law that goes to the validity of the MOC's statutory task and there is no prejudice to the Minister beyond the loss of a layer of appellate scrutiny, the discretionary balance might swing in the applicant's favour despite the Pt 8 scheme.
A further open question is the extent to which an MOC is required to "have regard to" material in a manner that demonstrates an active intellectual process when the material contains highly specific cost projections. The present judgment holds that the mere statement that regard was had to the treating reports is sufficient, but a future case with even more detailed econometric evidence may test whether there comes a point at which the absence of any engagement with that material demonstrates that the MOC has not applied the correct legal test.
Finally, the decision does not resolve the interaction between the "significant cost" criterion and the separate "prejudice to access" limb in PIC 4005(1)(c)(ii)(B). While the present case concerned only the cost limb, the reasoning about hypothetical assessment may have implications for how MOCs must approach the prejudice question when Australian waiting lists for particular treatments are in issue. Those questions await future litigation.