EDITH H. JONES, Circuit Judge:

No satisfactory excuse can be made for the district court's delay of over twelve years in entering findings of fact and conclusions of law after a bench trial. Having carefully reviewed the record, however, we are convinced that the judge's determinations not only are unassailable on appeal but represent the most plausible analysis of the conflicting expert testimony at the heart of this Federal Tort Claims Act (FTCA) case. The judgment is affirmed.

BACKGROUND

On October 9, 1985, Wendell and Patricia Hollis, next friends of their daughter Mariana Hollis, filed suit for damages pursuant to the FTCA alleging that following Mariana's premature birth at the William Beaumont Army Medical Center (WBAMC), in Texas, she received insufficient and negligent medical care, which rendered her blind. A bench trial began on January 30, 1989, and concluded on February 1, 1989. The parties submitted post-trial briefs. There was no further activity until the court rendered judgment more than 12 years later in favor of the defendant.

The district court made the following relevant findings of fact and conclusions of law: At the time of the events from which the claim arose, Wendell Hollis was an Army officer stationed at Fort Bliss, Texas. On February 2, 1984, his wife Patricia Hollis gave birth to a "very premature" daughter, Mariana, at WBAMC in El Paso, Texas. Mariana weighed roughly one pound five ounces at birth (590 grams), and her chances for survival were "very uncertain." While hospitalized, Mariana was monitored and treated for diseases and conditions to which premature infants are subject, particularly lung problems and Retinopathy of Prematurity (ROP), the latter of which gave rise to the instant lawsuit. Suffering ROP is recognized as a risk of premature infants, particularly those like Mariana, who weigh less than 750 grams at birth. Many premature infants are also born with respiratory problems, which require the use of supplemental oxygen. It was known at the time of Mariana's birth that the use of supplemental oxygen increased the risk that the infant would develop ROP. Once ROP developed, it progressed in stages. If the ROP was detected by examination at an early stage, and if the health of the infant permitted, the use of supplemental oxygen could be discontinued. Regardless whether the condition was detected and/or whether the use of oxygen was discontinued, a certain percentage of ROP cases regressed naturally and spontaneously. At ROP's stage five, the retina would completely detach, and the infant would be rendered blind in that eye. At the time of Mariana's birth, a surgical procedure called a vitrectomy existed to reattach the retina; however, the success rate was not high.

When Mariana was approximately seven weeks old, an ROP examination was conducted, and the results were negative. Thereafter, Mariana continued to breathe with the assistance of supplemental oxygen. Roughly six weeks later, when she was 13 weeks old, a second ROP examination was performed. This time, ROP was discovered, with detachment of the retina in both eyes. Mariana was thereafter seen by specialists and underwent a vitrectomy to attempt to reattach the retina of one eye; however, the surgery was unsuccessful, and Mariana sustained total and permanent loss of vision.

The plaintiffs contended that the physicians at WBAMC failed to obtain their informed consent to the use of oxygen to help Mariana breathe. The district court determined under Texas law that the plaintiffs failed to prove a lack of informed consent, because Wendell Hollis testified that doctors did inform him that the use of oxygen could cause Mariana's eyes to suffer retinal detachment and, although the doctors never used the word "blind," the consent which was obtained after this warning was sufficiently informed. The district court further concluded that any lack of informed consent was not the proximate cause of the damage suffered, because Mariana had such severe respiratory problems that withholding oxygen was not a viable option, as the alternative was probably death.

The court found that a more difficult issue raised by the plaintiffs was whether the timing and frequency of the ROP examinations fell below the requisite standard of care. The court recognized that the plaintiffs' experts contended that waiting six weeks to make a second ROP examination fell below the standard of care applicable in 1984. The court, however, made the following findings and concluded that the timing of the follow-up examination did not fall below the applicable standard of care and, further, that the timing of the follow-up examination could not be established as the proximate cause of Mariana's blindness: (1) in 1984, the medical profession recognized that a certain percentage of premature babies would fall victim to ROP and that, in a certain percentage of those cases, detached retinas would occur; (2) cases of ROP were found even in premature babies who were not on supplemental oxygen; (3) many premature babies using supplemental oxygen did not fall victim to ROP; (4) at the time of Mariana's birth, neither the American Academy of Pediatrics nor the Academy of Ophthalmologists had adopted a standard calling for the frequent examinations advocated by plaintiffs' experts — the Academy of Ophthalmologists advocated the examination of a premature infant before discharge and follow-ups of those showing signs of ROP, and the Pediatric Academy called for examinations before discharge and follow-ups every three to six months thereafter; (5) it was impossible to tell from the medical evidence in Mariana's case when the ROP process began or when it reached the stage of retinal detachment; therefore, whether an exam conducted within three weeks of the first exam, as was advocated by plaintiffs' experts, would have disclosed commencement of the ROP process was speculation; and (6) the alternative treatments (cryotherapy, scleral buckling, and Vitamin E therapy) which the plaintiffs argued could have been provided had the ROP been diagnosed prior to stage five were experimental and controversial, and whether Mariana's doctors would have recommended them and whether her parents would have consented were pure speculation.

The court therefore determined that the plaintiffs had not established that the physicians' treatment fell below the applicable standard of care as it existed in 1984 and, further, that assuming arguendo the physicians had been negligent, the plaintiffs had failed to show that the negligence was a proximate cause of Mariana's blindness. The court thus rendered judgment in favor of the government. The plaintiffs filed a timely notice of appeal.

DISCUSSION

On appeal of this judgment rendered after a bench trial, findings of fact are reviewed for clear error, and legal issues are reviewed de novo. Kona Tech. Corp. v. Southern Pac. Transp. Co., 225 F.3d 595, 601 (5th Cir. 2000).

The United States is liable for its torts if a private person would be liable for the same act or omission under local laws. 28 U.S.C. § 1346(b). Under the FTCA, liability for medical malpractice is controlled by state law, the law of Texas in this case. Ayers v. United States, 750 F.2d 449, 452 n. 1 (5th Cir. 1985); see also Urbach v. United States, 869 F.2d 829, 831 (5th Cir. 1989).

I. Informed Consent

Appellants initially complain that the district court "disregard[ed]" the consent issue because its reasons for judgment dealt only with the issue of oxygen, and nowhere did the court discuss the withholding of treatment for the emerging ROP and the failure to inform them of the alternative treatments available.

With regard to the issues of withholding treatment for the emerging ROP or informing the parents of what treatments were available, Appellants have failed to demonstrate that these allegations sustain a cause of action for failure to obtain informed consent. At trial, the plaintiffs argued that if a therapy other than a vitrectomy had been available at stage three of the ROP, the doctors were required to inform them of that alternative treatment, and, in turn, the plaintiffs would then have had the opportunity to give the necessary informed consent; thus, the doctors' omissions caused a lack of informed consent. The definitive response to these contentions was, however, pointed out by the district court: Mariana's disease was not discovered at stage three. Plaintiffs' purely hypothetical argument is not viable in light of the fact that Mariana's ROP was not discovered until stage five.

Moreover, the district court did address the availability of alternative treatments in the context of the allegations of malpractice. The court noted that the plaintiffs' experts contended that had the ROP been discovered in its early stages, other therapies existed which promised a chance for success. The court, however, found that these treatments were "experimental" and "controversial." Consequently, it was speculative whether they would have in fact been recommended or consented to by the parents. Appellants do not challenge this finding. The district court did not ignore the informed consent issue.

On the question of informed consent for Mariana's oxygen treatment, the duty of a physician to fully inform a patient of the risks of medical care is governed by Texas's Medical Liability and Insurance Improvement Act, TEX.REV.CIV. STAT. ANN. art. 4590i (Vernon Supp. 2002). The Act states in pertinent part:

In a suit against a physician or health care provider involving a health care liability claim that is based on the failure of the physician or health care provider to disclose or adequately to disclose the risks and hazards involved in the medical care or surgical procedure rendered by the physician or health care provider, the only theory on which recovery may be obtained is that of negligence in failing to disclose the risks or hazards that could have influenced a reasonable person in making a decision to give or withhold consent.

TEX.REV.CIV. STAT. ANN. art. 4590i, § 6.02 (Vernon Supp. 2002). In a negligence cause of action for the failure fully to inform a patient of risks attendant to a medical procedure, the plaintiff must establish (a) the existence of a duty, (b) the breach of that duty, (c) that the failure to obtain informed consent was a proximate cause of the injury, and (d) damages. McKinley v. Stripling, 763 S.W.2d 407, 409-10 (Tex. 1989). The causation inquiry is an objective one: whether a reasonable person would have refused the procedure had he been fully informed of all inherent risks which would influence his decision. Id. at 410. The plaintiff must additionally establish that he was injured by the occurrence of the risk of which he was not informed. Greene v. Thiet, 846 S.W.2d 26, 30 (Tex.App.-San Antonio 1992, writ denied).

Appellants argue that the evidence showed that the defendants did not inform them of the nature of ROP; that Mariana was at high risk for ROP; that there was a known relationship between the detachment of retinas and the level of supplemental oxygen used and the length of time the child received supplemental oxygen; and that no informed consent was obtained. Appellants further assert that the defendants were under a duty to disclose all risks of oxygen therapy that could have influenced a reasonable person in making a decision to consent to the procedure. They do not argue, however, that a reasonable person knowing all the risks would have refused oxygen treatment. They also fail to show error in the district court's express finding that both parents were aware that the administration of oxygen to Mariana carried a heightened risk of ROP. Moreover, Appellants do not address other than conclusionally the district court's causation determination — that any alleged failure to obtain informed consent was not the proximate cause of Mariana's blindness, because the withholding of oxygen was simply not a viable option, as death was the likely alternative.

Appellants also cite Hall v. Birchfield, 718 S.W.2d 313 (Tex.App.-Texarkana 1986), rev'd, 747 S.W.2d 361 (Tex. 1987), for the proposition that a failure to obtain informed consent for supplemental oxygen treatment proximately caused a premature infant's ROP, but Hall is distinguishable for several reasons. The jury there rejected the doctors' argument that the child would have died had she not received oxygen in favor of the parents' position that, although the infant had had some problems initially, she was not a sick infant who required the continued use of oxygen. Significantly, the infant was kept on oxygen for an extended period, even though she had no signs indicating the need for oxygen. 718 S.W.2d at 333 n. 7 334. Appellants do not contest the district court's observation that Mariana had such severe respiratory problems that the withholding of supplemental oxygen was not a viable option.

Further, Appellants' assertion, which was acknowledged by the district court, that there was a known relationship between the detachment of retinas and the use of supplemental oxygen is simply insufficient to establish causation. See McKinley, 763 S.W.2d at 410; Greene, 846 S.W.2d at 30. Appellants are required to establish based on the evidence presented that, understanding the risk of developing ROP, a reasonable person would have rejected supplemental oxygen treatment and that ROP would not have developed but for the administration of supplemental oxygen. The Appellants did not carry their burden of proof.

II. Standard of Medical Care

A plaintiff in a Texas medical malpractice action must prove four elements to establish liability: "(1) a duty owed by the defendant to the plaintiff, (2) a breach of that duty, (3) actual injury to [the] plaintiff, and (4) . . . [proof that] the breach [was] a proximate cause of the injury." Urbach, 869 F.2d at 831. A physician has a duty to render care to a patient with the degree of ordinary prudence and skill exercised by physicians of similar training and experience in the same or similar community under the same or similar circumstances. Speer v. United States, 512 F.Supp. 670, 675 (N.D.Tex. 1981), aff'd on basis of district court's opinion, 675 F.2d 100 (5th Cir. 1982). Texas tort law "places the burden of proof on the plaintiff to establish by expert testimony that the act or omission of the defendant physician fell below the appropriate standard of care and was negligent." Rodriguez v. Pacificare of Tex., Inc., 980 F.2d 1014, 1020 (5th Cir. 1993).

Appellants complain that in determining the applicable standard of care, the district court ignored the testimony of Mariana's treating physicians in the following respects: (1) Dr. Rowe testified that the local standard at WBAMC was to re-examine the infant for ROP four weeks after the initial evaluation; (2) Dr. Halverson, a pediatric neurologist, testified that the standard follow-up exam at her hospital was conducted every three weeks and as often as every day; and (3) Dr. Alverson, a neonatologist, testified that to meet the standard of care, the follow-up exam should have been done at one to four weeks, or more frequently if necessary, and that in his hospital, examinations occur weekly.

Appellants' characterization of the testimony of these witnesses is inaccurate. The actual testimony by these witnesses is as follows:

Dr. Rowe testified that she was doing her internship in pediatrics at WBAMC when Mariana was born. She further testified that the standard at WBAMC at the time of Mariana's birth was to test premature infants for ROP at six to eight weeks of age and that the timing of the follow-up exam was left to the discretion of the ophthalmologist. Contrary to the Appellants' assertions, she did not testify that it was the acceptable medical standard at WBAMC to conduct a follow-up ROP exam at four weeks. She instead testified that she was unsure when the follow-up exam was typically conducted at WBAMC at the time of Mariana's birth, but believed it was conducted "a little before discharge," which was when Mariana's follow-up exam was conducted. She further testified that as for the applicable standard of care, she would defer to the ophthalmologist as to how often an ROP exam should be conducted; nevertheless, she felt "comfortable" conducting the follow-up exam at discharge as opposed to four weeks after the initial exam.

Dr. Halverson, at the time of her deposition, was the Director of Child Rehabilitation at Santa Clara Valley Medical Center and did not treat Mariana Hollis. It is unclear why the Appellants argue that the district court erred in not considering her testimony as a treating physician. Nevertheless, her testimony does not support Appellants' position. Dr. Halverson responded to the question "what is the hospital policy [at Santa Clara Valley Medical Center] about evaluating [premature neonates on oxygen]?" as follows:

It's my understanding, although I am not directly involved in formulating this ophthalmology policy, that it's about every three weeks. If children are showing problems, it's more frequent than that, on the basis of the ophthalmologist's need to examine the patient and follow up.

In regard to the question regarding what the policy was for a high-risk premature neonate for an ophthalmological examination, Dr. Halverson responded:

I don't know that there's a standard policy on that. I do know the ophthalmologists will see my patients almost every day and they are not neonates, if there is a need for the ophthalmologist to do so. I assume its in the ophthalmologist's degree of concern.

Dr. Halverson acknowledged that her area of expertise concerned rehabilitative medicine costs for Mariana and that she would defer opinions on the ophthalmological standards of care to an ophthalmologist who treated children like Mariana.

Dr. Alverson was also not one of Mariana's treating physicians. At the time of his deposition, he was a faculty member of the University of New Mexico Medical School in the areas of pediatrics and neonatology. He testified that with an infant as premature as Mariana, the first ROP examination should be conducted between six and eight weeks postpartum. Dr. Alverson further testified that the timing of repeat exams would be left to the discretion of the ophthalmologist, which might be as often as once a week or every two weeks, but which could be deferred for three to four weeks. Alverson testified that at his hospital, any subsequent ROP exams

would depend on [the ophthalmologist's] findings and his decision on when he thinks the next exam would be optimal. [The ophthalmologist] would not necessarily examine every infant that has been identified at risk every week. It would depend on the findings of his prior exam.

In sum, Drs. Rowe and Halverson testified that they would defer a determination on the applicable standard of care for follow-up ROP exams to an ophthalmologist. The district court did not err in not considering their testimony on this issue.

Insofar as Appellants assert that the district court ignored Dr. Alverson's testimony that a follow-up exam should have been conducted between three to four weeks after the initial test, his opinion was consistent with the expert ophthalmological testimony that the district court recognized as representative of the plaintiffs' position on the applicable standard of care. The court cited Drs. Kenneth Fox and Melvin Burt, who testified that Mariana's second exam should have taken place within three weeks of the first exam. Thus, Appellants have not shown that the district court erred in failing expressly to mention Dr. Alverson's testimony.

Appellants' root complaint about the court's finding on the standard of care is that the court credited, over their witnesses, the government's expert witnesses, the publications of prominent specialty organizations, and even admissions by witnesses for the Appellants that no agreed protocol existed in 1984 for follow-up ophthalmological exams to detect ROP. Appellants' brief does nothing to undermine the solid basis in the record for the district court's finding.

We also point out, though the government's brief did not do so, that Appellants do not challenge the district court's determination that, since it is unknown when the ROP developed and at what point in Mariana's treatment it reached stage five, it was impossible to determine whether testing performed one, two, or three weeks after the first examination would have disclosed the commencement of the ROP process. Appellants could not prove proximate cause even if they had shown a violation of the standard of care.

III. Due Process

Appellants' last contention is that the district court, in waiting nearly 13 years to issue its opinion, misapplied important facts and failed to consider others, rendering review on appeal nearly impossible. Pursuant to Keller v. United States, 38 F.3d 16 (1st Cir. 1994), they urge this court to review the entire record de novo, taking the place of the trial court. We decline the invitation, but we have reviewed the record with extra care.

In the context of a five-month delay between trial and judgment, this court held that if there was affirmative evidence that the trier of fact had prejudiced the complaining party by abusing its discretionary powers in handling the case, we would order a new trial. Ciccarello v. Graham, 296 F.2d 858, 860 (5th Cir. 1961). While a 12-year delay is certainly more egregious than the one at issue in Ciccarello, the Appellants have nonetheless not successfully pointed out any misstatement of fact made by the district court as a result of the delay or any legal issue the court forgot to address, nor have they shown that the admittedly substantial delay prejudiced them. Appellants express concern over the district court's ability to analyze the case when the transcript was completed after it entered findings and conclusions. The problem was not fatal, however, because most of the expert testimony was offered in deposition form readily available for the court's review. Again, there is no showing of any factual or legal errors by the district court. We cannot help but think that while the district court's unexplained delay — even in the face of a growing criminal caseload — is inexcusable, plaintiffs' counsel also bears a heavy responsibility for never having uttered a word to the trial court or this court seeking a ruling. The delay did not legally prejudice Appellants.

CONCLUSION

For the foregoing reasons, the Appellants have failed to demonstrate error in the district court's judgment or reversible error attributable to the court's delay.

AFFIRMED.