The problem this diversity case presents is whether the doctrine of "informed consent"1 in medical malpractice cases is a part of Oklahoma law. If it is, we must decide whether the trial court erred in not instructing the jury on "informed consent" in light of the evidence introduced at trial. It is clear that the law of Oklahoma is in flux and that we have no unambiguous guidance in Oklahoma case law on whether the doctrine is recognized by the courts of that state. This is nowhere more clearly illustrated than in the reliance by both parties on the same case, Martin v. Stratton, 515 P.2d 1366 (Okl.1973), as settling Oklahoma law in their favor.

We agree that Martin is the key to the existing but perhaps undeclared law of Oklahoma. By using that key, we conclude that informed consent is a part of Oklahoma law although its dimensions have not yet been fully developed. Martin was a case brought partly on an informed consent theory. The Oklahoma Supreme Court had an opportunity to affirm the trial court's grant of defendant's demurrer by simply declaring the doctrine was not a part of Oklahoma law. Instead, a unanimous2 court elected to discuss the doctrine extensively, concluding:

(I)f the theory of liability referred to as "informed consent" is ever adopted by this Court the plaintiff will have the burden to either introduce evidence from which the jury could reasonably infer that the defendant failed to disclose to plaintiff what a reasonably prudent physician in the medical community in the exercise of reasonable care would have disclosed to his patient, or evidence from which the jury could reasonably infer that material risks were inherent in the proposed medical procedure in terms of seriousness, probability of occurrence and feasibility of alternatives, and defendant failed to disclose these risks to plaintiff.

515 P.2d at 1369-70.

Having elected not to reject the doctrine when that would have been the easy means of affirming the trial court, the court then held that plaintiff had failed to make out a case under either of the two applicable informed consent standards. We view that approach, combined with the current trend in other jurisdictions to employ the doctrine,3 as indicating that the Oklahoma Court would adopt the informed consent doctrine in an appropriate case. As will be clarified below, the evidence introduced in this case, unlike the evidence in Martin, would support a conclusion of liability under the doctrine. We therefore decide that if the instant case were before the Oklahoma Supreme Court, the Court would rule the doctrine applicable to it. Failure to instruct on "informed consent" was therefore error.4

Before turning to a discussion of the evidentiary basis warranting instruction on informed consent, we turn to consider which particular standard for instructing the jury the Oklahoma Court would adopt. As noted in Martin, three major tests are extant. The Court classified these standards according to the burden of proof placed on the parties. In "professional standard" or "community standard" jurisdictions the plaintiff usually bears the burden of showing that a doctor in good standing in the community would have disclosed the risk, but that defendant did not. 515 P.2d at 1369. See, e. g., Aiken v. Clary, 396 S.W.2d 668 (Mo.1965); ZeBarth v. Swedish Hospital Medical Center, 81 Wash.2d 12, 499 P.2d 1 (1972); Wilson v. Scott, 412 S.W.2d 299 (Tex.1967).5

In those jurisdictions adopting a "materiality of risk" standard, testimony as to professional standards is not required; rather, plaintiff must bear the burden of proof to establish materiality of the undisclosed risks, which is determined by "the seriousness of the consequence, the probability of occurrence and the feasibility of alternatives." 515 P.2d at 1369. See, e. g., Getchell v. Mansfield, 260 Or. 174, 489 P.2d 953 (1971).

The third general standard discussed in Martin is one requiring the defendant doctor to establish the justification for not disclosing the risks once plaintiff has established non-disclosure, injury, and causation. 515 P.2d at 1369. See, e. g., Cobbs v. Grant, 8 Cal. 3d 229, 104 Cal. Rptr. 505, 516, 502 P.2d 1, 12 (1972).

The Martin court indicated it would, in an appropriate case, "choose between" tests one and two, apparently foreclosing adoption of the third standard. 515 P.2d at 1369-70. Although the Court gave no hint of which of these two standards it would endorse, we believe it would select the second test.6 This test seems preferable because it more fully insures the important social policy underlying "informed consent." That is, a physician should be required to disclose to his patients all material risks of a proposed procedure even if other doctors in the community or specialty would not have made so full a disclosure.7 This is simply an application of the well-known tort doctrine that proof of compliance with the applicable "industry" standard will not insulate a defendant from liability when the standard itself is inadequate. The "material risk" approach is therefore the test to be employed by the trial court on remand, if we conclude the evidence introduced at trial was sufficient to warrant such an instruction.

We have so concluded. The operation in this case, phacoemulsification,8 involved surgical removal of a cataract. The jury could have found that feasible alternatives were available to plaintiff, including the older intracapsular method of cataract surgery (which method is apparently not without its own risks) as well as a decision against surgery of any sort. Testimony of three medical doctors, including the defendant himself, would support a jury conclusion that serious material risks were inherent in the proposed procedure. These risks include a severe allergic reaction to lens material (phaco anti-reaction to the lens), severe inflammation, high ocular pressure, cloudy cornea, cystoid edema, retinal detachments, ititis and secondary glaucoma. Most of these risks result from the possibility that lens material might be left in the eye, Even if the procedure is performed non-negligently by a competent physician. The record would support the conclusion that the possibility of lens material being left in the eye and the possibility of adverse consequences resulting therefrom would not be so insignificant as to make disclosure of these risks unnecessary. Although contested, plaintiff's testimony was that the defendant did not discuss with her or advise her of potential problems that could occur as a result of phacoemulsification and that he did not apprise her of alternative procedures.

The evidence in the record warranted giving instructions appropriate to an "informed consent" case. The refusal of the trial court to so instruct is error, and the case is therefore remanded for a new trial. Plaintiff had an adequate opportunity to present evidence in support of other theories of liability against defendant. No error is claimed in the instructions pertaining to these other theories; the jury, properly instructed, decided these issues against plaintiff. It follows that the new trial will be limited to the issue of liability under the informed consent doctrine, as consistent with this opinion and appropriate additional authorities.